An Introduction to Physical Geography and the Environment [4th ed] 9781292083575, 9781292083612, 9781292134574, 1292083573, 1292083611

This highly respected and best-selling textbook provides an accessible, engaging and comprehensive introduction to the m

1,560 171 52MB

English Pages 841 Year 2017;2018

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

An Introduction to Physical Geography and the Environment [4th ed]
 9781292083575, 9781292083612, 9781292134574, 1292083573, 1292083611

Table of contents :
Cover......Page 1
Title Page......Page 4
Copyright Page......Page 5
Brief Contents......Page 6
Contents......Page 8
Preface to the fourth edition......Page 16
Contributors......Page 18
Editor’s acknowledgements......Page 19
Acknowledgements......Page 20
Part I: The role of physical geography......Page 28
1.1 Introduction......Page 30
1.2.1 Physical geography before 1800......Page 32
1.2.2 Physical geography between 1800 and 1950......Page 33
1.2.3 Physical geography since 1950......Page 35
1.3.1 The positivist method......Page 37
1.3.2 Critique of the positivist method......Page 38
1.3.3 Realism as an alternative positivist approach......Page 39
1.3.4 Benefits of multiple scientific methods in physical geography......Page 40
1.4.1 Approaching data collection from the environment......Page 41
1.4.3 Approaching numerical modelling......Page 43
1.5 Using physical geography for managing the environment......Page 47
1.6 Summary......Page 52
Further reading......Page 53
Part II: Continents and oceans......Page 54
2.1 Introduction......Page 56
2.2.2 The outer layers of the Earth......Page 57
2.3.2 Sedimentary rock......Page 58
2.4.1 Early ideas of global tectonics......Page 60
2.4.2 Evidence that led directly to plate tectonic theory......Page 61
2.5.1 Lithospheric plates......Page 64
2.5.2 Rates of plate movement......Page 65
2.6.1 Divergent plate boundaries......Page 66
2.6.2 Transform faults......Page 68
2.6.3 Convergent plate boundaries......Page 69
2.6.4 Hot spots......Page 77
2.7 The history of the continents......Page 78
Further reading......Page 79
3.1 Introduction......Page 80
3.2.2 Geological structure of the ocean basins......Page 81
3.3.1 Salinity......Page 82
3.3.2 Temperature structure of the oceans......Page 86
3.4.1 Surface currents......Page 87
3.4.2 The deep currents of the oceans......Page 89
3.4.3 The weather of the ocean......Page 91
3.5 Sediments in the ocean......Page 92
3.6.1 Photosynthesis in the ocean......Page 94
3.6.2 Importance of nutrient supply to primary productivity......Page 95
3.6.3 Animals of the sea......Page 96
3.6.4 Pollution......Page 100
3.7 Effect of global climate change on the oceans......Page 101
3.8 Summary......Page 102
Further reading......Page 103
Part III: Past, present and future climate and weather......Page 104
4.1 Introduction......Page 106
4.2.1 Orbital forcing theory......Page 108
4.2.2 Evidence that orbital forcing causes climate change......Page 109
4.2.4 Internal feedback mechanisms......Page 112
4.3.1 Glacial instability......Page 116
4.3.2 The Younger Dryas......Page 117
4.4 Further evidence for environmental change......Page 118
4.4.1 Landforms......Page 119
4.4.2 Plants......Page 124
4.4.3 Insects......Page 125
4.4.4 Other animal remains......Page 126
4.5.1 Age estimation techniques......Page 127
4.6 Pleistocene stratigraphy and correlation......Page 128
4.7 Palaeoclimate modelling......Page 132
4.8 Summary......Page 133
Further reading......Page 134
5.1 Introduction......Page 135
5.2.1 How the Holocene began......Page 136
5.2.2 Drivers of climate change during the Holocene......Page 137
5.2.3 The Little Ice Age......Page 143
5.3.1 Retreating ice sheets......Page 144
5.3.2 Rising seas......Page 145
5.4.1 Responses of ecosystems to the end of the last glacial......Page 147
5.4.2 Tropical Africa and the Sahara......Page 149
5.4.3 European ecosystems......Page 150
5.4.4 Island ecosystems......Page 151
5.5.1 Humans at the end of the last glacial......Page 152
5.5.2 The beginnings of agriculture......Page 153
5.5.3 Social and environmental consequences of agriculture......Page 154
5.6.2 Deforestation......Page 155
5.6.3 Soil erosion and impoverishment......Page 159
5.6.4 Irrigation and drainage......Page 160
5.7 Summary......Page 162
Further reading......Page 163
6.1 Introduction......Page 164
6.2 The basics of climate......Page 166
6.3 The global atmospheric circulation......Page 168
6.4.1 The nature of energy......Page 169
6.4.3 Radiation......Page 171
6.4.4 Thermal inertia......Page 176
6.5.1 Moisture in the atmosphere and the hydrological cycle......Page 177
6.5.2 Global distribution of precipitation and evaporation......Page 178
6.5.4 Drought......Page 180
6.6.1 Convective overturning......Page 181
6.6.2 The Earth’s rotation and the winds......Page 182
6.6.3 Long waves, Planetary Waves and Rossby Waves......Page 183
6.6.4 Jet streams......Page 186
6.7 The influence of oceans and ice on atmospheric processes......Page 188
6.8 The Walker circulation......Page 190
6.8.1 El Niño Southern Oscillation......Page 191
6.8.2 North Atlantic Oscillation......Page 193
6.9.2 A simple climate model of the enhanced greenhouse effect......Page 194
6.9.3 Radiative interactions with clouds and sulfate aerosols......Page 197
6.10 Geoengineering......Page 200
Further reading......Page 201
7.1 Introduction......Page 202
7.2.1 Long-term change......Page 203
7.2.2 Recent climate change and its causes......Page 204
7.2.3 Predictions from global climate models (GCMs)......Page 207
7.2.4 Critical evaluation of the state-of-the-art in GCMs......Page 209
7.3 The carbon cycle: interaction with the climate system......Page 211
7.4 Mitigation......Page 213
7.5 Destruction of the ozone layer by chlorofluorocarbons (CFCs)......Page 214
7.6 The future......Page 216
7.7 Summary......Page 220
Further reading......Page 221
8.1 Introduction......Page 222
8.2 General controls of global climates......Page 223
8.3.1 Equatorial regions......Page 228
8.3.2 The Sahel and desert margins......Page 236
8.3.3 Subtropical deserts......Page 237
8.3.4 Humid subtropics......Page 238
8.4.1 Depressions, fronts and anticyclones......Page 242
8.4.2 Mid-latitude western continental margins......Page 246
8.4.3 Mid-latitude east continental margins and continental interiors......Page 247
8.5 Polar climates......Page 248
8.6 A global overview......Page 249
Further reading......Page 255
9.1 Introduction......Page 256
9.2 Altitude and topography......Page 257
9.2.2 Temperature......Page 260
9.2.3 Wind......Page 261
9.2.4 Precipitation......Page 262
9.3 Influence of water bodies......Page 267
9.4.1 Shelter belts......Page 269
9.4.2 Urban climates......Page 270
9.4.3 Atmospheric pollution and haze......Page 273
9.5 Summary......Page 275
Further reading......Page 276
Part IV: Biogeography and ecology......Page 278
10.1 Introduction......Page 280
10.2.2 The naming of species......Page 281
10.2.4 Biodiversity......Page 282
10.3.1 Potential species distributions......Page 285
10.3.2 Actual species distributions......Page 286
10.3.3 Spatial patterns in biodiversity......Page 287
10.4 Terrestrial biomes......Page 288
10.4.1 Equatorial and tropical forests......Page 289
10.4.2 Savanna......Page 292
10.4.4 Mediterranean-type biome......Page 293
10.4.6 Temperate broadleaf forest......Page 295
10.4.7 Taiga......Page 296
10.4.8 Tundra......Page 297
10.5.1 Marine regions......Page 299
10.5.2 Freshwater regions......Page 301
10.6 Summary......Page 302
Further reading......Page 303
11.1 Introduction......Page 304
11.2.2 Ecological thermodynamics......Page 305
11.2.3 Trophic levels and food webs......Page 306
11.2.4 Biogeochemical cycles......Page 307
11.3 Biotic interactions......Page 308
11.3.2 Herbivory, predation and parasitism......Page 309
11.3.3 Commensalism......Page 310
11.3.5 Competition......Page 311
11.4.1 Short-term changes......Page 312
11.4.3 Succession......Page 313
11.5.1 Degrading ecosystems......Page 316
11.5.3 Conservation......Page 321
Further reading......Page 324
12.1 Introduction......Page 325
12.2.1 River ecosystem geomorphological units......Page 327
12.2.2 Spatial variability of river ecosystems......Page 329
12.2.4 Human alterations to river ecosystems......Page 335
12.3.1 Classification of lake ecosystems......Page 339
12.3.2 Spatial variability of lake ecosystems......Page 342
12.3.3 Human influences on lake ecosystems......Page 346
12.4 Summary......Page 348
Further reading......Page 349
13.1 Introduction......Page 350
13.2.1 Light......Page 351
13.2.3 Temperature......Page 353
13.3.1 The forest/savanna boundary in southern Amazonia......Page 354
13.3.3 Upward march of vegetation in mountains......Page 356
13.3.4 Changes in the timing of flowering......Page 357
13.4 Models for prediction......Page 359
13.5.2 How does fire interact with climate change?......Page 363
13.6 Loss of biodiversity......Page 367
13.7 Agriculture and food security......Page 368
Further reading......Page 370
Part V: Geomorphology and hydrology......Page 372
14.1 Introduction......Page 374
14.2 Environmental and material controls on weathering......Page 375
14.3 Weathering and the role of water......Page 377
14.4.2 Carbonation......Page 379
14.4.5 Biologically related chemical weathering......Page 380
14.4.6 Products of chemical weathering......Page 381
14.5.1 Dilatation – pressure release......Page 383
14.5.2 Thermoclasty......Page 385
14.5.4 Salt weathering......Page 386
14.5.5 Biologically related physical weathering......Page 390
14.6 Climatic controls on weathering......Page 391
14.7 Geological controls on weathering......Page 394
14.8 Urban stone decay and lessons for rock weathering......Page 398
14.8.2 Rates of stone decay are unpredictable......Page 400
14.8.3 Decay is spatially variable......Page 401
14.9 Summary......Page 402
Further reading......Page 403
15.1 Introduction......Page 404
15.2.2 Slope steepness......Page 405
15.3 Hillslope transport processes......Page 407
15.3.1 Chemical transport processes (solution)......Page 408
15.3.2 Physical transport processes......Page 409
15.3.4 Particle movements......Page 417
15.3.5 The balance between erosion processes......Page 423
15.4.1 Concepts......Page 425
15.4.2 Models......Page 428
15.4.3 Interpreting landscape form......Page 432
Further reading......Page 433
16.1 Introduction......Page 434
16.2.1 Classification of clastic sediments......Page 435
16.2.3 Sediment transport and sedimentation......Page 436
16.2.4 Products of sedimentation – bedforms......Page 439
16.3 Biological sediments......Page 442
16.4 Chemical sediments......Page 444
16.5.1 Continental environments......Page 445
16.5.2 Coastal and marine environments......Page 447
16.6 Response of sedimentation to environmental change......Page 449
16.6.2 Mining......Page 451
16.6.3 Urbanization......Page 452
16.6.4 Sediment management......Page 453
16.7 Summary......Page 454
Further reading......Page 455
17.1 Introduction......Page 456
17.2.2 Soil organic matter......Page 457
17.2.3 Soil water......Page 458
17.2.4 Soil air......Page 459
17.3 Soil profile......Page 460
17.4.1 Pedogenesis......Page 461
17.4.2 Factors affecting soil formation......Page 464
17.5.2 Soil texture......Page 468
17.5.3 Soil structure......Page 471
17.6.1 Clay minerals and cation exchange......Page 472
17.6.2 Soil acidity......Page 474
17.7.1 The soil biota......Page 476
17.7.2 Factors influencing soil biodiversity......Page 478
17.8.1 Soil erosion......Page 479
17.8.2 Soil acidification......Page 480
17.8.3 Soil pollution......Page 481
17.8.4 Soil organic matter and carbon......Page 483
17.8.5 Other threats......Page 485
17.8.6 Policy and legislation......Page 489
17.9 Summary......Page 490
Further reading......Page 491
18.1 Introduction......Page 492
18.2.1 Precipitation......Page 493
18.2.2 River flow......Page 495
18.2.3 Evapotranspiration......Page 499
18.2.4 Soil water......Page 500
18.2.5 Groundwater......Page 502
18.3.2 Infiltration-excess overland flow......Page 503
18.3.3 Saturation-excess overland flow......Page 504
18.3.4 Throughflow......Page 505
18.4.1 Stormflow......Page 510
18.4.2 Flow frequency......Page 512
18.4.3 River regime......Page 515
18.5 Flooding......Page 516
Further reading......Page 518
19.1 Introduction......Page 520
19.2.1 Runoff, river regimes and floods......Page 521
19.2.2 Sediment sources and delivery......Page 522
19.3.1 Channel networks and slope......Page 523
19.3.3 Channel planform......Page 524
19.3.4 Channel boundary materials......Page 525
19.4.1 Water flow and flow hydraulics......Page 526
19.4.2 Sediment movement......Page 527
19.5 River channels: linking channel processes and morphology......Page 529
19.5.1 Long profile......Page 531
19.5.3 Channel planform......Page 532
19.5.4 Channel bed morphology......Page 534
19.6 River channel changes: rates and types of channel adjustment......Page 536
19.6.2 Planform change......Page 538
19.6.3 Human-induced change......Page 540
19.7.1 River management and the engineering tradition......Page 542
19.7.2 Living with rivers......Page 543
19.7.3 River maintenance......Page 544
19.7.4 Building new river channels......Page 546
19.7.5 River restoration......Page 547
19.8 Summary......Page 550
Further reading......Page 551
20.1 Introduction......Page 552
20.2.2 pH and redox potential......Page 553
20.2.3 Temperature and pressure......Page 555
20.2.5 Solute fluxes......Page 556
20.3.1 Precipitation......Page 557
20.3.2 Evapotranspiration and evaporation......Page 559
20.3.4 Soil......Page 560
20.3.5 Groundwater......Page 562
20.3.7 Lakes and reservoirs......Page 563
20.4 The role of hydrological pathways in solute processes......Page 565
20.5 Temporal patterns of solutes......Page 567
20.5.1 Patterns of solutes in storm events: short-term changes......Page 568
20.5.2 Annual patterns of solute concentrations......Page 569
20.5.3 Long-term patterns of solute concentrations......Page 571
20.6 Spatial patterns of solutes......Page 575
20.6.2 Regional patterns of solutes......Page 576
20.7.2 Modelling solutes in watercourses......Page 580
20.8 Summary......Page 582
Further reading......Page 583
21.1 Introduction......Page 584
21.2.1 Drylands......Page 587
21.2.2 Causes of aridity......Page 588
21.3.2 Dryland vegetation......Page 590
21.4.1 Dryland landscapes......Page 593
21.4.2 Rock weathering in drylands......Page 594
21.4.3 Hillslope and channel processes......Page 595
21.4.4 Aeolian processes and forms......Page 598
21.5 Environmental change in drylands......Page 605
21.6 Summary......Page 609
Further reading......Page 610
22.1 Introduction......Page 611
22.2 Coastal morphodynamics......Page 615
22.3 Coastal processes: waves......Page 617
22.3.1 Linear wave theory......Page 618
22.3.2 Wave processes in intermediate water......Page 619
22.3.3 Wave processes in shallow water......Page 620
22.3.4 Nearshore currents......Page 622
22.4.1 Storm surge......Page 624
22.4.2 Tides......Page 625
22.5 Coastal classification......Page 627
22.6.1 Barriers......Page 628
22.6.2 Beaches......Page 629
22.6.3 Coastal dunes......Page 632
22.7.1 Wave- and tide-dominated estuaries......Page 635
22.7.2 Estuarine mixing......Page 638
22.7.4 Salt marsh and mangroves......Page 640
22.8 Fluvial-dominated coastal environments......Page 641
22.9.1 Rocky coast processes......Page 644
22.9.3 Shore platforms......Page 646
22.10 Coastal zone management......Page 647
22.11 Summary......Page 650
Further reading......Page 651
23.1 Introduction......Page 652
23.2.1 Types of ice mass......Page 653
23.2.2 Where do glaciers occur?......Page 654
23.2.3 Glacier mass balance......Page 655
23.2.5 Glacier thermal regime......Page 658
23.2.6 Glacier water systems......Page 659
23.2.7 Glacier dynamics......Page 663
23.3.1 Processes of glacial erosion......Page 671
23.3.2 Entrainment and transport......Page 672
23.3.3 Deposition......Page 674
23.4.1 Ice sheet reconstruction......Page 677
23.5 Summary......Page 681
Further reading......Page 682
24.1 Introduction......Page 683
24.2.1 The distribution of permafrost......Page 684
24.2.3 Reconstructing climate change from permafrost temperatures......Page 686
24.2.4 Gas hydrates......Page 688
24.2.5 Hydrology in permafrost regions......Page 690
24.3.1 Ground ice features......Page 692
24.3.2 Slope processes......Page 697
24.4 Summary......Page 700
Further reading......Page 701
Part VI: Monitoring and management......Page 702
25.1 Introduction......Page 704
25.2.1 Satellite positioning systems......Page 705
25.2.3 Environmental sensor networks......Page 706
25.3 Remote sensing date......Page 707
25.3.2 Electromagnetic radiation......Page 709
25.3.3 Image data......Page 711
25.4 Camera sensors......Page 716
25.4.1 Photogrammetry......Page 718
25.5 Multispectral, thermal and hyperspectral sensors......Page 720
25.5.1 Landsat......Page 722
25.5.2 Spot......Page 725
25.5.3 NASA’s Earth Observing System Program......Page 726
25.6.1 Microwave sensors......Page 728
25.6.2 Ranging sensors......Page 732
25.7.1 Digital images......Page 737
25.7.2 Image rectification......Page 738
25.7.3 Image enhancement......Page 739
Further reading......Page 745
26.1 Introduction......Page 746
26.2 Types of environmental hazard......Page 747
26.3.1 The nature of change......Page 749
26.3.2 Rate of change......Page 751
26.4.1 Monitoring......Page 752
26.4.2 Modelling......Page 753
26.6 Management tools......Page 754
26.6.3 Impact assessment......Page 756
26.6.4 Life costing......Page 757
26.6.6 Engagement......Page 759
26.7 Summary......Page 761
Further reading......Page 762
Glossary......Page 763
Bibliography......Page 790
Index......Page 822

Citation preview

An Introduction to

Physical Geography and the Environment Fourth edition

Edited by

Joseph Holden

An Introduction to

Physical Geography and the Environment

At Pearson, we have a simple mission: to help people make more of their lives through learning. We combine innovative learning technology with trusted content and educational expertise to provide engaging and effective learning experiences that serve people wherever and whenever they are learning. From classroom to boardroom, our curriculum materials, digital learning tools and testing programmes help to educate millions of people worldwide – more than any other private enterprise. Every day our work helps learning flourish, and wherever learning flourishes, so do people. To learn more, please visit us at www.pearson.com/uk

An Introduction to

Physical Geography and the Environment Fourth edition

Edited by

Joseph Holden School of Geography, University of Leeds

+DUORZ(QJODQGȏ/RQGRQȏ1HZ300 mm

Dry

Continuous drought Deficit >500 mm

205S 405S

Dry

Super -typhoons Heavy rains

Dry

Stronger jet stream

05

305E

605E

905E

1205E

1505E

Very suppressed hurricane season

1805

1505W 1205W

Very wet Surplus >300 mm 905W

605W

305W

Figure 6.20 Schematic diagram showing the major impacts of El Niño during the period June to December 1997. (Source: after Slingo, 1998)

DROUGHTS IN THE MODERN WORLD Increasing concentrations of greenhouse gases in the atmosphere are expected to modify the global water cycles with significant consequences in the form of droughts for terrestrial hydrology. The concept of drought was introduced in Section 6.5.4. Recent research shows that regional temperature and rainfall worldwide will be affected by big changes that are already underway in key patterns in the global climate system, such as El Niño in the tropical Pacific. Major droughts have occurred in the past. Examples are the droughts of the 1930s in the central and western USA creating the so-called Dust Bowl, when large amounts of topsoil blew away and many people were driven to migrate to California. Because of insufficient understanding of the ecology of the plains, farmers had conducted extensive deep ploughing of the virgin topsoil of the Great Plains during the previous decade; this had displaced the native, deep-rooted grasses that normally trapped soil and moisture even during periods of drought and high winds. Author John Steinbeck, in his novel The Grapes of Wrath (1939), wrote about migrant

workers and farm families displaced by the Dust Bowl. At time of writing, California is now entering its fourth year of drought, leading to unprecedented restrictions on water usage for its 38 million residents. Most of California received less precipitation in 2013 than during any previous calendar year in the 119 year observational record. Scientists from NASA’s Goddard Space Flight Center and other institutions recently found that California has accumulated a ‘rain debt’ of about 50 cm between 2012 and 2015. That is the amount that would normally fall in an entire year in the state. The mix of crops that is traditionally grown in California is changing. Farmers are ploughing up fields where they used to grow vegetables like broccoli, carrots and tomatoes to put in nut and fruit trees, which demand less water. The area of cotton growth has changed from 600 000 hectares to almost nothing. One of the most remarkable aspects of the 2013/2014 Californian drought event is the spatial and temporal coherence of strong north Pacific ridging in the mid-troposphere and associated wind anomalies. In particular, although high-amplitude meridional flow and positive pressure anomalies over the far

north-eastern Pacific are often associated with precipitation deficits in California, the temporal resilience and spatial scale of the positive pressure anomalies were greater in 2013/2014 than during previous droughts in California’s recent past. Modern measurements of drought indicators in the western USA go back only about 150 years, but centuries-old trees allow it to be extended back into the distant past. Tree species like oak and bristle cone pines grow more in wet years, leaving wider tree rings, and vice versa for drought years. By comparing the modern drought measurements to tree rings in the twentieth century for a baseline, the tree rings can be used to establish moisture conditions over the past 1000 years. These show that megadroughts took place between 1100 and 1300 in North America. These medieval-period droughts, on a year-to-year basis, were no worse than droughts seen in the recent past, but they lasted, in some cases, 30 to 50 years. When these past megadroughts are compared side-by-side with computer model projections of the twenty-first century, both the moderate and business-as-usual CO2 emissions scenarios are drier, and the risk of droughts lasting 30 years or longer increases significantly. Connecting

BOX 6.6 ➤

165

Chapter 6 Atmospheric processes

➤ the past, present and future in this way shows that twenty-first century droughts in the Californian region are likely to be even worse than those seen in medieval times, and require considerable forward planning. In the tropics, droughts are often associated with surface temperatures in the Pacific and Indian Oceans. El Niño events were introduced in Section 6.8.1. The Indian Ocean Dipole is a year-to-year see-saw pattern in surface temperature and rainfall across the tropical Indian Ocean. During a positive Indian Ocean Dipole phase, sea surface temperatures off Sumatra and Java in Indonesia are colder than normal. In contrast, off east Africa, surface waters are unusually warm. Like an El Niño, a positive Indian Ocean Dipole brings heavy rainfall to eastern parts of Africa and drought to countries around the Indonesian Archipelago, including Australia. A negative

Indian Ocean Dipole phase tends to do the opposite. Unlike an El Niño, which peaks in summer, Indian Ocean Dipole events form in winter and then peak in spring. This creates a narrower predictability window that gives little warning to industries, such as farming, that depend on rain. In 1997–1998, extremely dry El Niño conditions in Indonesia started large-scale uncontrolled burning, destroying about five million hectares of tropical forest. Much of the burning occurred in carbon-rich peatland forests, releasing large amounts of CO2 into the atmosphere, and clouds of smoke and haze across the region. At the time of writing, conditions in the Pacific Ocean are similar to those in mid-1997, El Niño is set to strengthen, and seasonal weather prediction models forecast an exceptionally dry season in Indonesia. Although some rain falls during a normal

dry season, it is sporadic, leaving many windows of opportunity for burning. During strong El Niño episodes, almost no rain falls during the dry season, so in areas where peatlands have been degraded by logging and draining, fires ignite easily and, once started, the peat is so dry that fires escape underground, and cannot be put out until after the drought ends. These major fires have enormous impacts on carbon emissions, regional haze production, biodiversity and the economy, and are recognised as a serious health risk in Indonesia as well as neighbouring Singapore and Malaysia. During past El Niño years, around one gigatonne of carbon was emitted from peatland forest fires, equivalent to about 10% of annual global fossil fuel emissions, and regional haze from such fires has caused major disruptions to air traffic in nearby Singapore and Kuala Lumpur.

BOX 6.6

Australian vegetation that may be, at least in part, attributable to ENSO’s influence on climate are the following:

t

t

t

Absence of succulents: succulents are almost totally absent from Australian arid and semi-arid regions, because although adapted to arid climates and requiring little moisture, they need regular rainfall. Such plants are therefore unsuited to the high variability ENSO produces over much of Australia. Fire resistance/dependence: much of the Australian flora is fire resistant or even dependent on fire for successful reproduction. Fires are common during drought periods, which are often associated with ENSO events. Fluctuating climax: the high inter-annual variability of annual rainfall in arid and semi-arid parts of Australia affected by ENSO means that vegetation appears adapted to the climate in such a way that demographic components are in a state of unstable equilibrium.

6.8.2 North Atlantic Oscillation A major source of inter-annual variability in the atmospheric circulation over the North Atlantic and western Europe is the North Atlantic Oscillation (NAO), which 166

is associated with changes in the strength of the oceanic surface westerlies. It is often measured by the difference of December to February atmospheric pressure between Ponta Delgado in the Azores (37.8°N, 25.5°W) and Stykkisholmur, Iceland (65.18°N, 22.7°W). When the values are greater than the average the index is said to be positive. Statistical analysis reveals that the NAO is the dominant mode of variability of the surface atmospheric circulation in the Atlantic and accounts for more than 36% of the variance of the mean December to March sealevel pressure field over the region from 20° to 80°N and 90°W to 40°E between 1899 and 1994. The oscillation is most marked during the winter. There can be great differences between winters with high and low values of the NAO index. Typically, when the index is high the Icelandic low pressure is strong, which increases the influence of cold Arctic air masses on the north-eastern seaboard of North America and enhances the westerlies carrying warmer, moister air masses into western Europe. An example was given earlier in Section 6.6.3. Low values of the index are associated with large stationary or slow-moving anticyclones over the North Atlantic or north-western Europe. These block the normal progression of low-pressure systems from the North Atlantic over

6.9  Interactions between radiation, atmospheric trace gases and clouds

north-western Europe, and are therefore often termed blocking anticyclones. Blocking anticyclones can occur at any time of the year; in summer they can give hot dry conditions, but in winter with the warm westerlies absent, they often lead to very low temperatures. The record cold December 2010 in the United Kingdom was partly due to a continuous series of blocking anticyclones over the eastern Atlantic. Thus, NAO anomalies are related to downstream wintertime temperature and precipitation across Europe, Russia and Siberia. High-index winters are anomalously mild while low-index winters are anomalously cold.

Reflective question ➤ How can ENSO affect human activity across the world?

6.9 Interactions between radiation, atmospheric trace gases and clouds 6.9.1 The greenhouse effect The gases nitrogen and oxygen which make up the bulk of the atmosphere neither absorb nor emit long-wave radiation. It is the water vapour, CO2 and some other minor gases present in the atmosphere in much smaller quantities which absorb some of the long-wave radiation emitted by the Earth’s surface and act as a partial blanket. To examine the effect of these absorbing trace gases on surface temperature it is useful to consider an atmosphere from which all cloud, water vapour, dust and other minor gases have been removed, leaving only nitrogen and oxygen. As the Sun’s short-wave radiation passes through the atmosphere, about 6% is scattered back to space by atmospheric molecules and about 10% on average is reflected back to space from the land and ocean surfaces. The remaining 84% heats the surface. To balance this incoming radiant energy the Earth itself must radiate on average the same amount of energy back to space in the form of long-wave radiation. It can be calculated from radiation laws (Box 6.2) that, to balance the absorbed solar energy by outgoing long-wave radiation, the average temperature of the Earth should be - 6°C. This is much colder than is actually observed (15°C). This calculation error occurs because the atmosphere readily absorbs infrared radiation emitted by the Earth’s surface with the principal absorbers being water vapour (absorption wavelengths of 5.397.7 mm and beyond 20 mm), ozone (9.499.8 mm), CO2(13.1916.9 mm) and all

clouds (all wavelengths). Only about 9% of the infrared radiation from the ground surface escapes directly to space. The rest is absorbed by the atmosphere, which in turn reradiates the absorbed infrared radiation, partly to space and partly back to the surface. This blanketing effect is known as the natural greenhouse effect and the gases are known as greenhouse gases. It is called ‘natural’ to distinguish it from the enhanced greenhouse effect due to gases added to the atmosphere by human activities such as the burning of fossil fuels and deforestation. It needs to be stressed that the natural greenhouse effect is a normal part of the climate of the Earth and that it has existed for the whole of the atmosphere’s history. Concern about the greenhouse effect arises over two issues: how the natural greenhouse effect may vary with time, and how human activities might modify and enhance the natural effect. Although CO2 is the most important greenhouse gas, other gases also make a contribution to climate change. The combined effect of the increases to 1990 of the minor greenhouse gases, methane, nitrous oxide and tropospheric ozone, is to add a warming forcing equivalent to that of an additional 60 ppm or so of CO2. Even if there were no further increase in these minor gases, the 1990 forcing would still require to be added to future projections of change driven by CO2. As an example, the effect of this, if turned into equivalent amounts of carbon dioxide (CO2e), would be that the 450 ppm CO2 only level would become about 520 CO2e ppm and the 550 level would become about 640 CO2e ppm. Note that, although the amount of forcing from the minor gases is the same, when turned into equivalent CO2, the amounts added increase with the CO2 concentration to which the amount is added. This is because the relationship between radiative forcing and concentration is nonlinear. It should also be noted that it is not always clear in the literature as to whether the authors are using simple CO2 concentration or the equivalent CO2 value.

6.9.2 A simple climate model of the enhanced greenhouse effect It is possible to construct a simple climate model to illustrate the enhanced greenhouse effect due to the addition to the atmosphere of radiatively active trace gases such as CO2. Figure 6.21 shows a mean temperature profile for the troposphere, calculated assuming convective equilibrium, which gives a lapse rate of 6 °C km - 1. The upper atmospheric temperature is estimated by assuming that the stratosphere is in 167

Height (km)

Chapter 6 Atmospheric processes

10 Average levels where outgoing radiation originates 5

0 -100

-50

0 Temperature (5C)

Figure 6.21 The distribution of temperature in a convective atmosphere

(solid line). The dashed line shows how the temperature increases when the amount of CO2 present in the atmosphere is increased (for doubled CO2 in the absence of other effects the increase in temperature is about 1.2°C). Also shown for the two cases are the average levels from which thermal infrared radiation leaving the atmosphere originates (about 6 km for the unperturbed atmosphere). (Source: after Houghton, 1994)

radiative equilibrium (its temperature is controlled by its radiation balance and not by vertical convection). The height at which these two straight-line temperature

GREENHOUSE GAS STABILIZATION AND GLOBAL TEMPERATURE INCREASES Most pollutants released into the ’atmosphere have very short residence times, often just a few days. Clouds or rainfall washes out the common ones, such as sulfur dioxide, or the larger particles settle out rapidly. The distribution of such pollutants is therefore restricted to near their source regions, usually large urban or industrial regions. Their extent is rarely global; even the sulfur compounds emitted by erupting volcanoes (Box 6.8) rarely stay in the atmosphere for more than a few years. Long-term concentrations of gases such as sulfur dioxide are usually maintained by continuous emissions. If the emissions cease, atmospheric

profiles connect is the tropopause. Viewed from space, the Earth is observed to have a temperature of about - 18°C. This is because most of the infrared radiation to space takes place from the middle atmosphere where the temperature is around - 18°C; most of the greenhouse gases are below this level. As the amount of infraredabsorbing greenhouse gases mixed into the atmosphere is increased, it becomes more likely to absorb infrared radiation. Thus the effective level at which outgoing infrared radiation originates must rise to allow the radiation to escape to space at the same rate as before. However, because the tropospheric lapse rate does not change significantly from 6°C km - 1 then the temperatures at a given height in the troposphere must be dragged up by the changes. Furthermore, since the height of the tropopause is increased, the stratosphere must cool slightly. In particular, and of great importance, the surface temperature is also increased. From this simple model, the doubling of the CO2 content of the atmosphere in the absence of other effects would increase the tropospheric temperature by about 1.2°C. In reality various other effects could increase the predicted temperature changes (Box 6.7).

concentrations drop rapidly and the gases often vanish from the atmosphere. This is the common experience with sulfur dioxide, where closing down pollution sources causes a rapid improvement in atmospheric conditions. Carbon dioxide concentrations in the atmosphere depend on the interactions between the various carbon reservoirs forming the so-called carbon cycle. Consider an emission of anthropogenic CO2 into the atmosphere. Different processes, with very distinct timescales, are responsible for determining the lifetime of anthropogenic CO2. For timescales of particular human interest, ranging from decades to centuries, the responses to excess atmospheric CO2 include ocean uptake, changes in land surface carbon uptake, CO2 fertilization and alterations

to vegetation cover. For scales of centuries to about 5000–10 000 years, ocean uptake becomes dominant. At present, the terrestrial biosphere appears to be a net sink of carbon, in spite of anthropogenic deforestation predominately in the tropics. Long-term numerical modelling studies predict a reversal of present-day net CO2 uptake by the biosphere as the Earth warms, resulting in a net release of carbon to the atmosphere by the end of the century. Studies also indicate that roughly 80% of anthropogenic CO2 input into the atmosphere has an average perturbation lifetime in the atmosphere of approximately 300–450 years. The remaining 20% could remain in the atmosphere for more than 5000 years after emissions cease. Archer (2005) commented that care is required in using 300 years as a lifetime

BOX 6.7 ➤

168

6.9  Interactions between radiation, atmospheric trace gases and clouds

➤ of anthropogenic CO2, because it misses the immense longevity of the tail in the CO2 lifetime. He considered that a better approximation of the lifetime of fossil fuel CO2 for public discussion might be ‘300 years, plus 25% that lasts forever’. The long life of the anthropogenic CO2 released into the atmosphere makes stabilization difficult, and at present nearly impossible in the short term, for levels below the present atmospheric concentration (400 ppm). The global carbon cycle has a strong one-way arrow in the short term directed to increases in atmospheric CO2 concentration. This is because a substantial fraction of the CO2 emitted into the atmosphere by human activity remains there, in effect, for centuries to millennia. This is a very important point to understand: without the use of large-scale atmospheric carbon capture technologies, it is not possible to reduce atmospheric CO2 concentrations in the short term. If the above properties of the carbon cycle are real and enduring, then it is likely that bringing future anthropogenic carbon emissions to zero will not reduce global average surface temperatures except in the very long term. Rather, once temperatures have peaked, they will remain almost steady (see Section 6.4.4 on thermal inertia). Several recent studies have sought to exploit this observation in order to provide a simple link between levels of cumulative anthropogenic carbon emissions and future warming. Allen et al. (2009), for example, considered the cumulative carbon emissions summed between preindustrial times and 2050, linking them to peak global warming. It can be argued that warming by a given date is proportional to cumulative CO2 emissions to that date. The recent Copenhagen Accord contains the aim of limiting warming to no more than 2°C, drawing on earlier targets from the EU and G8. Though not specified in the Copenhagen Accord, this 2°C warming limit is usually presumed to be relative to pre-industrial levels. Using the results in

Allen et al. (2009), a 2°C limit on the most likely peak CO2@induced warming could be achieved by limiting cumulative CO2 emissions to one trillion tonnes of carbon (1 TtC). Present-day human activities, such as burning fossil fuels and changing land use, are estimated to put about 26 000 million tonnes of CO2 a year into the atmosphere. Bowerman et al. (2010) commented that cumulative emission targets represent the sum of emissions over time, and therefore these cumulative emissions could be distributed over time in a number of ways. For example, an early peak in emissions could be followed by a relatively slow rate of post-peak decline, or a later peak could be followed by a much more rapid decline. It may not be technically or politically feasible, or economically desirable, to decrease emissions at rates much in excess of 3 or 4% per year, so that peaking later may not be viable, assuming a 2°C warming target. The so-called A1FI emissions scenario is considered by the IPCC to be one of a number of equally plausible projections of future greenhouse gas emissions from a global society that does not implement policies to limit anthropogenic influence on climate. It assumes that the world is market oriented with fast per capita growth and fossil-fuel-intensive energy production. It also assumes that the global population peaks in 2050 and then declines. Previously, this scenario has received less attention than other scenarios with generally lower rates of emissions. However, there is no evidence from actual emissions data to suggest that the A1FI scenario is implausible if action is not taken to reduce greenhouse gas emissions, and hence it deserves closer attention than has previously been given. The evidence available from recent simulations, along with existing results presented by IPCC (2014), suggests that the A1FI emissions scenario would lead to a rise in global mean temperature of between approximately 3 and 7°C by the 2090s

relative to pre-industrial levels, with best estimates being around 5°C. A temperature rise of 4°C is predicted by the 2070s, but if carbon cycle feedbacks are strong, 4°C could be reached in the early 2060s – this latter projection appears to be consistent with the upper end of the IPCC’s likely range of warming for the A1FI scenario. The Committee on Climate Change (CCC, 2008) has analyzed eight greenhouse gas emission trajectories. Three trajectories had global emissions peaking in 2028, with subsequent reductions in total CO2e emissions of 1.5%, 2% and 3% per annum. Five of the trajectories have global emissions peaking in 2016 with subsequent reductions in total emissions of 1.5%, 2%, 3% and 4%. None of the three trajectories with emissions peaking in 2028 would keep atmospheric greenhouse gas concentrations below 550 ppm CO2e by the end of the century. The CCC (2008) does not believe that a global policy which leaves emissions peaking as late as 2028 is adequate for control of global warming. All the trajectories with emissions peaking in 2016, except the one with 1.5% annual reduction after 2016, would keep concentrations below 550 ppm CO2e by the end of the century. Only emission reductions after 2016 at 3 or 4% per year would limit the chance of reaching 4°C to very low levels, with central model estimates indicating a 2.2°C rise this century with the 3% trajectory, and a 2.1°C rise with the 4% one. Even in these cases the CCC (2008) commented that the chances of exceeding 2°C by 2100 would be 63% and 56% respectively. These conclusions are reflected in the Paris 2015 Agreements. The Paris Agreement (2015) aims to limit the increase in the global average temperature by the end of the century to well below 2°C above pre-industrial levels and to pursue efforts to limit the temperature increase to 1.5°C above pre-industrial levels, recognizing that this would significantly reduce the risks and impacts of climate change. In order to achieve this

BOX 6.7 ➤

169

Chapter 6 Atmospheric processes

➤ long-term temperature goal, Parties to the Agreement aim to reach global peaking of greenhouse gas emissions as soon as possible, and to undertake rapid reductions thereafter in accordance with best available science, so as to achieve a balance between anthropogenic emissions by sources and removals by sinks of greenhouse gases in the second half of this century. The CCC (2008) trajectory described above peaking in 2016 and decreasing afterwards at 4% per year suggests that the Paris Agreement (2015) will be unable to limit a temperature increases above preindustrial to 2°C. This is because little or no attempt is being made at present to meet this target, and future plans barely exist.

It is often argued that it may not be technically, economically or politically feasible to eliminate emissions of all greenhouse gases while, for example, preserving global food security. Crop and animal farming can produce large amounts of CO2 and methane. This limit has been referred to as an ‘emissions floor’. It is difficult to estimate a compelling emissions floor, either in terms of its size (in gigatonnes of carbon per year (GtC yr21)), or in terms of the extent to which it can reduce over time as new technologies become available. Nevertheless, Bowerman et al. (2010) commented that it makes sense to consider the possibility that it may prove prohibitively

expensive to reduce emissions beyond some positive level. They use the following conventions: if the emissions floor is constant, it is referred to it as a ‘hard floor’. If, on the other hand, society is able to continue to reduce residual CO2 emissions, eventually to the point where net emissions are zero, then it is called a ‘decaying floor’. The above discussion strongly suggests that it is unlikely that global temperature increases can be kept at relatively safe levels of 2°C or below without major reductions in greenhouse gas emissions in the next few years and preferably by 2020. Further discussion of contemporary climate change is provided by Chapter 7.

BOX 6.7

6.9.3 Radiative interactions with clouds and sulfate aerosols The terms ‘forcing’ and ‘feedback’ are frequently used in climatology. Forcings are processes that act as external agents to the climate system, such as changes in solar input to the Earth, the loading of the atmosphere with volcanic ash and aerosols (see Box 6.8), or rising levels of CO2 gas. As a result of temperature changes caused by increasing concentrations of greenhouse gases, other changes may take place that could in turn influence temperature. For example, an increase in temperature could result in greater evaporation from the ocean, enhancing atmospheric humidity. Since water vapour is an active greenhouse gas, the increase in atmospheric humidity causes a further increase in temperature. This is an example of a positive feedback process. It also demonstrates that the intensity of the hydrological cycle is closely linked to the greenhouse effect.

6.9.3.1 Clouds Conventionally, the radiative effect of clouds is discussed in terms of ‘cloud radiative forcing’, even though cloud effects are actually feedback processes. The radiative forcing of the Earth’s climatic system is in part determined by the distribution of cloudiness, since clouds strongly influence the distribution of both short-wave and long-wave radiative fluxes within the atmosphere. At short wavelengths,

170

clouds generally increase planetary albedo and so cool the planet by reflecting more solar radiation to space. However, for long-wave radiation, clouds generally add to the greenhouse effect as they absorb upward-moving infrared radiation and reradiate it back downwards. Thus cloudy days are normally cooler than clear-sky days, particularly in summer, because the short-wave radiative input to the surface is restricted. In contrast, cloudy nights are warmer than clear nights, particularly in winter, because the longwave radiative loss from the surface to space is restricted (Figure 6.22). However, observations from satellites show that the albedo cooling effect dominates, and the net effect of clouds at present is to cool the Earth.

6.9.3.2 Aerosols Over the last few years, it has become evident that, when averaged over the global atmosphere, part of the enhanced greenhouse effect has been offset by a negative forcing due to the human-induced emission of aerosols (Figure 6.23). Aerosols have two effects, a direct one and also by modifying cloud properties. Aerosol particles reflect short-wave radiation and therefore increase atmospheric albedo. Aerosols also increase cloud albedo and influence cloud lifetime and precipitation. Sulfate aerosols, formed from sulfur dioxide emitted during the combustion of fossil fuels and partly responsible for producing acid rain, are regarded as particularly important. Sulfate aerosols located in the

6.9  Interactions between radiation, atmospheric trace gases and clouds

VOLCANOES AND CLIMATE Volcanic eruptions are of two main types: effusive eruptions and explosive eruptions (Chapter 2). Effusive eruptions are those in which local lava flows predominate, and are of little meteorological interest. Explosive eruptions can throw vast quantities of rocks, dust and gases to great heights. The volume of material blown into the atmosphere during an explosive eruption can amount to 1 km3 or more. During the eruption of Hekla, Iceland, in 1947, dust fell on ships in the Atlantic Ocean up to 800–1000 km away to the south and east. Major volcanic eruptions can be highly disruptive, mainly due to five significant effects: tsunamis may inundate surrounding low-lying land; sulfate aerosols circulating in the stratosphere may cause globally lowered temperatures for several months or years; toxic elements such as fluorine deposited on the ground or in water may cause death if ingested; tropospheric sulfates, which precipitate within a few days as acid rain, may stunt or kill vegetation over huge areas downwind from the eruption; and volcanic dust may disrupt aircraft movements by damaging jet engines. The gases emitted by eruptions are largely water vapour, CO2, sulfur dioxide, and small quantities of other gases. Volcanoes emit two gases that can have an impact on global temperatures: sulfur dioxide and CO2. They each have very different effects and work on different timescales. Sulfur dioxide When this gas is emitted to high altitudes (about 12–14 km or above) it enters the stratosphere. Here it can form acid droplets that partially scatter and reflect sunlight away from the Earth, causing surface cooling. The droplets have a fairly immediate impact, and in sufficient

concentrations may cool the climate for a few months, or even a year or two, but then the droplets fall out of the stratosphere and temperatures return to normal. Large-scale winds in the stratosphere tend to flow towards the east with a small polar component. Thus stratospheric dust clouds tend to move eastwards and polewards; they also tend to be restricted to the particular hemisphere in which they formed. Typically, volcanic dust takes 2 to 6 weeks to circuit the Earth in middle or lower latitudes, and from 1 to 4 months to become a fairly uniform veil over the whole of the latitude zone swept by the wind system into which it is injected. Thus volcanic eruptions near to the equator can spread stratospheric dust over much of one or, if they are almost on the equator, both hemispheres. In contrast polar eruptions often have only a restricted influence. The final stage of a worldwide dust veil from an equatorial eruption is probably a concentration of the last remaining airborne dust over the two polar caps. Because of their relatively large size compared with the wavelength of the incident short-wave radiation, sulfuric acid droplets and volcanic dust particles predominantly scatter solar radiation in the direction of the incident beam. The result is that scattering reduces the direct radiation much more substantially than is the global solar radiation, since the bulk of the radiation scattered from the direct beam reappears at the surface as diffuse radiation. For example, after the eruption of Mount Agung (Bali) in 1963, the stratospheric dust veil was the most effective since 1902–1903 and possibly since 1883–1886. A marked drop in the direct beam radiation at Aspendale, Melbourne, Australia, was observed which was largely compensated by a similar large increase in the diffuse components. The changes in

the direct and diffuse components almost completely cancelled out, and only a very slight fall was detectable in the global radiation curve. The consequence is that the cooling effect of volcanic eruptions is generally small. Large volcanic eruptions of the calibre of Krakatoa (1883) or Agung (1963) are required to cause atmospheric dust loadings that would affect surface solar radiation. Carbon dioxide This is a greenhouse gas, so when it is emitted in large enough quantities it can have a warming impact on the global climate. The last remnants of CO2 emissions have an atmospheric lifetime of several hundred years, so any impact will be felt over a long timescale. Even so, it is still much smaller than that produced from human activity, such as burning fossil fuels and changing land use. Examples of volcanic eruptions 1. Tambora volcano, Sumbawa, Indonesia: The Tambora volcano on the island of Sumbawa in Indonesia erupted on 10 April 1815, sending a massive cloud of aerosols into the stratosphere. This was one of the largest eruptions of the past 500 years. The year 1816 is often known as the ‘year without a summer’. Snow every month of the year started a mass migration from the US East Coast across the Appalachian Mountains to the Midwest. The effects of the 1816 summer on agricultural productivity in New England were mainly due to a series of killing frosts that reduced the growing season. This, along with a severe drought, reduced agricultural output to record low levels. In Europe, by contrast, there were record low temperatures accompanied by above average rainfall and cloudiness, the combination of

BOX 6.8 ➤

171

Chapter 6 Atmospheric processes

➤ which slowed the growth of crops and produced fungus and moulds. Thus, food production was negatively affected in both regions, but by different mechanisms. This shows the complex relationship between climate and its impacts. 2. Mount Pinatubo, Philippines: Recent history has produced a good case study of the impact of a large volcanic eruption. Mount Pinatubo in the Philippines erupted in 1991, making it one of the biggest volcanic events of the twentieth century. It put about 20 million tonnes of sulfur dioxide into the stratosphere. As expected, this impacted global temperatures. The average for the following year was reduced by between 0.1°C and 0.2°C, although temperatures quickly recovered the year after. In addition, an

estimated 250 million tonnes of CO2 was put into the atmosphere. This is a significant amount, but is still much smaller than that produced from human activity, such as burning fossil fuels and changing land use, which is estimated to put out about 26 000 million tonnes of CO2 a year. 3. Eyjafjallajökull eruption, 14 April–23 May 2010: Volcanic eruptions are not uncommon in Iceland (Swindles et al., 2016). This eruption is of interest because the upper winds were northerly and spread clouds of volcanic ash over north-western Europe, severely disrupting air travel. This was because the ash clouds can damage jet engines and cause them to fail (see Box 26.1 in Chapter 26).

4. Santorini, eastern Mediterranean: The eruption of Santorini (Thera) was one of the largest volcanic events of the last four millennia. Some 30 cubic kilometres of rock was erupted, and ash from the eruption has been identified across more than 2 million square kilometres extending from the Black Sea in the north through Turkey and the south-eastern Mediterranean to the Nile Delta in the south. The best date for this eruption, consistent with all the data, is 1628 BC. 5. Toba, Sumatra, Indonesia: The largest known eruption of the past 100 000 years was the great Toba eruption about 71 000 years ago, which occurred close to the beginning of a major glacial advance, although a causal relationship has yet to be established.

BOX 6.8

anthropogenic contributions to aerosols (primarily sulfate, organic carbon, black carbon, nitrate and dust) together produce a cooling effect, with a total direct radiative

troposphere are rapidly washed out of the atmosphere by rainfall; hence they do not travel far from their industrial sources. The IPCC Report (2014) suggested that

Figure 6.22 Values of incoming short- and long-wave radiation for actual 900

and cloudless sky conditions for a site near Pateley Bridge, North Yorkshire, for two days in June, 1984. During the night of 27/28 June, 2 mm of rain fell in two events of 1 mm each. It is seen that, while cloud makes effectively no contribution to the downward-moving long-wave radiation across the midday period of 27 June (a relatively cloudless day with near-maximum values of incoming short-wave radiation), it contributes around 90 W m -2 throughout the rainfall of the night of 27/28 June and only slightly less during the relatively cloudy 28 June (with much reduced incoming short-wave radiation). (Source: after Lockwood, 1993)

Cloudless sky short-wave radiation

800

700

Observed short-wave radiation

Radiation (W m-2)

600

500 Observed long-wave radiation

400

300 Cloudless sky long-wave radiation

200

100 Rainfall

0 0

12 27 June

172

0 Time (GMT)

12

0 28 June

6.10  Geoengineering

RF terms CO2 Long-lived greenhouse gases

N2O

Anthropogenic

LOSU

1.66 (1.49 to 1.83)

Global

High

Global

High

Continental to global

Medium

0.07 (0.02 to 0.12)

Global

Low

-0.2 (-0.4 to 0.0) 0.1 (0.0 to 0.2)

Local to continental

Medium to low

-0.5 (-0.9 to - 0.1)

Continental to global

Medium to low

Cloud albedo effect

-0.7 (-1.8 to -0.3)

Continental to global

Low

Linear contrails

0.01 (0.003 to 0.03)

Continental

Low

0.12 (0.06 to 0.30)

Global

Low

Ozone

Halocarbons

Stratospheric

Tropospheric

Stratospheric water vapour from methane (CH4) Surface albedo

Land-use

Black carbon on snow

Direct effect

Natural

Spatial scale

0.48 (0.43 to 0.53) 0.16 (0.14 to 0.18) 0.34 (0.31 to 0.37) -0.05 (-0.15 to 0.05) 0.35 (0.25 to 0.65)

CH4

Total aerosol

RF values (Wm-2)

Solar irradiance Total net anthropogenic

1.6 (0.6 to 2.4) -2

-1

0

1

2

Radiative forcing (W m-2) Figure 6.23 Global-average radiative forcing (RF) estimates and ranges in 2005 for anthropogenic CO2, methane, nitrous oxide and other important agents and mechanisms, together with the typical geographical extent of the forcing and the assessed level of scientific understanding (LOSU). (Source: IPCC, 2007a)

forcing of - 0.5 W m-2 and an indirect cloud albedo forcing of - 0.7 W m-2. The combined radiative forcing due to increases in the greenhouse gases of CO2, methane and nitrous oxide is estimated to be + 2.3 W m-2. Therefore IPCC (2014) reports with a very high confidence that the globally averaged net effect of human activities since 1750 has been one of warming, with a radiative forcing of between + 0.6 and + 2.4 W m-2 (Figure 6.23).

Reflective question ➤ Sulfate aerosols originating from industrial areas cool the atmosphere but cause acid rain. Attempts to remove such aerosols from the atmosphere will accelerate global warming.

of climate change. Basically they attempt to modify the various atmospheric processes described in this chapter. None are fully operational at present, but there is growing interest in developing some of them. One class removes greenhouse gases from the atmosphere after they have been released; the other involves solar radiation management, such as deflecting sunlight away from the Earth. The first approach makes use of biological agents, such as land plants or aquatic algae, to produce, for example, biofuels. There is growing interest in this approach, and the planting of forests is already considered in international climate negotiations. The second approach could involve increasing the albedo of land surfaces so that they reflect more sunlight and thus cool the Earth. It is not yet clear how this could be done on a scale large enough to be effective.

What therefore should be the policy on industrial pollution?

6.10 Geoengineering

Reflective question ➤ Are there any planned geoengineering projects intended

Scientists have identified a range of engineering techniques, collectively called geoengineering, to address the control of atmospheric greenhouse gases and reduce the risks

for your area to control greenhouse gas concentrations? Do some web searches to find out.

173

Chapter 6 Atmospheric processes

6.11 Summary

upper ridges. Frontal depressions also move with the upper winds. Embedded within the upper Planetary Waves are strong, narrow

The climate system consists of a closely coupled atmosphere– ocean–ice–land–vegetation system. Short-wave radiation from

currents of air known as jet streams. Hydrological processes play an important part in atmospheric

the Sun is the main energy input that drives the climate system,

and climatic processes. For example, condensation is important

while long-wave radiation from the atmosphere to the cold

in controlling the convective overturning of the atmosphere

of space is the main energy lost. The processes of absorption,

and clouds act as both cooling and warming agents across the

reflection and reradiation of energy, however, are of fundamental

planet. Atmospheric circulation patterns are naturally variable on

importance to atmospheric processes. There is a net radiation

an inter-annual scale. Among the more important are the El Niño

surplus at the surface near the equator and a net radiation deficit

Southern Oscillation associated with the Walker circulation, and

in polar regions, leading to energy transfers. These transfers to

the North Atlantic Oscillation with its influence on winter climate

rectify the equator/pole energy imbalance take place through

in western Europe. On longer timescales there is also natural cli-

atmospheric and oceanic motions.

mate variability.

The basic circulation of the atmosphere is dominated by a

A natural greenhouse effect operates because certain trace

three-cell structure in each hemisphere that is controlled by

gases in the atmosphere such as water vapour and CO2 act to

a pressure gradient force. The Earth’s rotation, however, has a

absorb long-wave radiation emitted from the Earth’s surface

profound influence on air movements across its surface, making

and reradiate it back downwards. They act as a blanket. This

circulation systems much more complex than they would other-

effect is being enhanced by increased atmospheric concentra-

wise be. In the tropical world there are mean north–south circula-

tions of such greenhouse gases as CO2 due to the use of coal

tions, known as Hadley cells, with rising air over the equator and

and oil as energy sources. However, there are also negative

sinking air in the subtropics. There are also east–west circulation

feedback effects from atmospheric aerosol pollution produce

cells along the equator known as the Walker circulation. At the

by various industrial processes. Increased clouds and sulfate

surface in middle latitudes, the predominant features are closed,

aerosols can increase atmospheric albedo and reflect more of

eastward-drifting, cyclonic and anticyclonic systems, while higher

the Sun’s energy directly back into space and thus partly offset

up in the troposphere smooth wave-shaped patterns are the gen-

the greenhouse effect. Nevertheless, increasing atmospheric

eral rule. There are normally five to eight upper Planetary Waves

greenhouse gas concentrations are now dominating global

circling the poles. They are important because frontal depressions

climate change, and unless restricted by the end of the present

tend to form and grow rapidly just downwind of upper troughs

decade have the potential to cause massive damage by the end

while surface anticyclones tend to develop just downwind of

of the century.

Further reading Hansen, J., Sato, M., Kharecha, P. et al. (2008) Target atmospheric CO2: where should humanity aim? The Open Atmospheric Science Journal, 2, 217–231. Good paper about uncertainties and targets for atmospheric emissions and carbon uptake measures. Houghton, J. (2015) Global warming: The complete briefing, 5th edition. Cambridge University Press, Cambridge. This up to date book describes the uncertainties surrounding predictions of the effects of global warming including positive and negative feedbacks. 174

Lockwood, J.G. (2009) The climate of the Earth. In: Hewitt, C.N. and Jackson, A.V. (eds) Atmospheric science for environmental scientists. Wiley-Blackwell, Chichester, 1–25. A detailed introduction to the circulation and climate of the Earth’s atmosphere. MacKay, D.J.C. (2009) Sustainable energy – without the hot air. UIT, Cambridge. An excellent introduction to the physics of energy use. Free copies are available on the Internet at www.withouthotair.com. Stern, N. (2006) The economics of climate change: The Stern Review. Cambridge University Press, Cambridge. The Stern Review is an independent, rigorous and comprehensive analysis of the economic aspects of climate change

C HAPTER 7

Contemporary climate change John Grace School of GeoSciences, University of Edinburgh

Learning objectives After reading this chapter you should be able to:

➤ discuss the evidence for anthropogenic global warming ➤ understand how the use of fossil fuels has impacted upon the climate

➤ describe how the carbon cycle has been perturbed ➤ appreciate how humankind has created environmental problems and perceive how they may be solved

7.1 Introduction Environmental change on a global scale first became a matter of public concern in the 1960s. Before then, the perceived environmental problem was urban pollution, which affected human health and the quality of life of so many people. Although urban pollution became acute during the Industrial Revolution, it was not new. The smelting of toxic metals such as copper and lead was a health hazard in ancient Rome, as revealed by analysis of hair samples from the preserved corpses of Roman soldiers found in bogs, and from traces of metal in Greenland ice cores. Coal was used in London in the thirteenth century. Coal

contains not only carbon but also 1–4% sulfur and traces of heavy metals, and therefore its combustion releases a multitude of pollutants as well as carbon dioxide (CO2). With the onset of the Industrial Revolution in western Europe, around 1780, the use of coal increased dramatically and cities like London became heavily polluted with smog, a mixture of fog and smoke. Domestic coal burning was a major contributor to smog, and the industrial regions around Birmingham in England became known as the Black Country and even non-industrial Edinburgh was known as Auld Reekie, referring to the smell of coal burning. Diseases such as bronchitis and tuberculosis were widespread following the Industrial Revolution, and nearly a quarter of deaths in Victorian Britain (1837–1901) were from lung diseases. In one week of December in 1952, 4000 Londoners were killed by a particularly severe episode of smog. The ensuing public outcry resulted in the Clean Air Act of 1957, which restricted coal burning and resulted in the use of cleaner energy sources such as oil, gas and electricity. Other coal-burning cities of the world such as Pittsburgh in the United States have a similar history. Problems were greatly exacerbated by the growth in use of the motor car, especially in regions receiving high solar radiation, such as Los Angeles, Athens and Mexico City, where the ultraviolet radiation reacts with

Chapter 7 Contemporary climate change

(a)

0 -2 -4 -6 Temperature relative to modern average (5C)

uncombusted hydrocarbons and oxides of nitrogen from exhausts of cars to yield photochemical smog, irritating the eyes, nose and throat. An important milestone in the awakening of environmental concern was prompted by the widespread use of the persistent pesticides that were introduced after the Second World War and the publication of Rachel Carson’s book Silent Spring in 1962. Silent Spring warned against the dangers of pesticides, especially to songbird populations, indicating how persistent chemicals might spread in food chains as well as in the atmosphere, and ultimately damage non-target species. At the same time, other scientists were demonstrating that the pesticide DDT could be found in rainwater in the Antarctic, and that pesticides were responsible for eggshell-thinning in wild birds, threatening especially those species at the end of food chains such as raptors. Thus, the idea of environmental change on a global scale soon became a permanent part of western culture, and part of the international research agenda. The global scale of human influence on the planet is today felt even more strongly, but not because of fears of widespread pollution of the land and sea by pesticides. The global environmental challenge that we face now is climate change, and that is the main focus of this chapter.

0.8

(b)

0.4 0.6 Millions of years ago

0.2

0 -2

Most recent glaciation

Glaciers retreat from Britain

-4 20

10 5 Thousands of years ago

(c) 1.5 1.0 0.5 0

7.2 Climate change 7.2.1 Long-term change The climate has always fluctuated, but usually over very long timescales. There are many sources of information that help in the reconstruction of past climates. These include historical records, evidence from the annual growth rings of trees, deposits of pollen in lakes and bogs, isotopes and fossils (see Chapters 4 and 5). The picture that emerges is quite complex, showing cyclic fluctuations on several scales (Figure 7.1). The longterm cyclic trends in the Earth’s temperature, associated with periods of glaciation known as ‘the Ice Ages’, were attributed by Scottish geologist James Croll and Serbian astronomer Milutin Milankovitch to the irregularities in the orbit and tilt of the Earth, which influence the energy received from the Sun (see Chapter 4). These are known as the Milankovitch cycles. In contrast to these gradual changes, there have also been catastrophic events causing mass extinctions on a global scale. For example, in the Late Permian (245 million years ago) about half the families of marine animals were lost. At the boundary of the Cretaceous and Tertiary (known as the KT boundary, some 65 million years ago) 15% of marine families were 176

-0.5 -1.0

1000

1400 Year (AD)

1800

Figure 7.1 Air temperature over three timescales, relative to modern

records: (a) the past million years; (b) the past 20 000 years; (c) the past millennium. (Source: after Brohan et al., 2006; Hegerl et al., 2007)

lost, perhaps 75% of plant species, and (most famously) dinosaurs became extinct. Such events are now usually attributed to the impact of comets, asteroids or large meteorites, which would have thrown up debris into the atmosphere and greatly reduced the penetration of solar radiation, causing widespread cooling, a reduction in photosynthesis and collapse of food chains. The KT boundary is considered to have been caused by an asteroid 10 km in diameter, which impacted at Chicxulub, northeast Yucatan, Mexico. Geologists recognize five such mass extinction events in the fossil record, all of them global in extent, taxonomically broad, and most evident in marine invertebrates (for which the fossil record is relatively complete). It is against this background that we examine the changes in the climate system which are currently occurring, and their link to anthropogenic activity.

7.2  Climate change

7.2.2 Recent climate change and its causes Over the past century the Earth has warmed by about 0.8°C, particularly in the early part of the twenty-first century (Figure 7.2). Apart from this modern instrumental record from meteorological stations, there are a number of independent sources of information to demonstrate the phenomenon of climate warming: glaciers have been receding, snow cover has declined, polar ice has been melting, sea levels have been rising and spring has been arriving earlier. This period of rapid, human influenced change has been considered as a new geological epoch, the Anthropocene, following the Holocene (see Chapter 5). Many authors describe the present-day temperatures as ‘unprecedented’. We know the temperatures and the concentrations of CO2 and CH4 that have occurred over the last 650 000 years from analysis of deep cores taken from polar ice, and we can compare these with those being experienced now. Although the 650 000 year record does contain large fluctuations, associated with the ice ages and the warm interglacial periods, we see that today’s temperatures and concentrations of CO2 and CH4 are much higher. Moreover, the rate of increase in temperature is faster now than previously.

Difference from 1961 to 1990 (mm)

(a)

14.5

0.0

14.0

-0.5

13.5

50

(b)

0

-50 -100 -150

4

40

0

36

-4

32 1850

1900

1950

Area covered (million km2)

(c) Area difference (million km2)

Temperature (5C)

Temperature difference (5C)

0.5

2000

Year

Figure 7.2 Observed trends on (a) global average temperature, (b) global average sea level and (c) northern hemisphere snow cover for the past century. (Source: IPCC, 2007a)

A causal association between greenhouse gases and temperature is inevitable, ever since the demonstration in 1859 by the Irish scientist John Tyndall that CO2 absorbs infrared radiation. In the Earth’s atmosphere, CO2 and a range of other gases, including water vapour, absorb some of the infrared radiation that would otherwise stream directly out to space, thus causing a heating effect known as the greenhouse effect (the name arises because glass also absorbs infrared radiation and so the glass panes in a greenhouse have exactly the same effect on a local scale). The amounts of three of these gases, CO2, CH4 and N2O (nitrous oxide), have risen sharply in recent times, and the extent of warming to be expected from these rises can be calculated (see the right-hand axis of Figure 7.3). The rise in heat supply to the Earth’s surface caused by these gases, known as radiation forcing (see Chapter 6), amounts to 2–3 watts per square metre (W m-2). This is much smaller than the average incoming solar radiation (averaging about 230 W m-2 at the Earth’s surface) but enough to increase the global temperatures. We know for sure that humans have emitted vast quantities of CO2 by burning fossil fuels and biomass, thus interfering with the global carbon cycle. The rise in CH4 concentrations can be attributed to increases in various types of human activity. Only about 45% of all CH4 emissions are produced naturally: from wetlands, termites, the ocean and from the decomposition of gas hydrates. The remainder is anthropogenic: from energy production, rice fields, landfills, ruminant livestock, waste treatment and biomass burning. The rate of increase in CH4 has in fact been levelling off in the last few years. As for N2O, the causes of its increase are somewhat less clear. It is produced naturally by microbial activity in the nitrogen cycle (see Chapter 13), and at a much faster rate when land is ‘improved’ by the use of nitrogen fertilizer. It is estimated that about one-third of the global emissions of N2O are anthropogenic. Other processes influence global temperatures (see Figure 6.23 in Chapter 6). Some are less well understood, and are the subject of current research. One such case is the influence and general behaviour of aerosols. These are particles in the atmosphere, including fumes and smoke from industrial processes and transport, and naturally produced particles such as pollens and spores. To some extent, they shield the planet from solar radiation, absorbing, scattering and reflecting solar radiation. The aerosol ‘haze’ that we see in the clear sky (especially in the northern hemisphere) effectively reflects part of the incident solar energy back into space, contributing to a cooling effect and therefore offsetting the warming effect of 177

Chapter 7 Contemporary climate change

400

350

300 1800

1900 Year

1

2000

300 0

Radiative forcing (Wm-2)

Carbon dioxide, CO2 (ppm)

350

250 2000 2000

1500

0.4

1000 500 1800

1900 Year

2000 0.2

1000

Radiative forcing (Wm-2)

Methane, CH4 (ppb)

1500

0 500

330 330

270

300 1800

1900 Year

2000

0.1

240

270

10000

0

5000 Time (before 2005)

Radiative forcing (Wm-2)

Nitrous oxide, N2O (ppb)

300

0

Figure 7.3 Trends in three global atmospheric greenhouse gases over the past 10 000 years. Different colours denote different studies. The inset box shows the period since 1750 in more detail. The left-hand axis shows the concentration of each gas and the right-hand side shows the radiative forcing that the concentration implies. (Source: IPCC, 2007a)

greenhouse gases. Periodic changes in the aerosol content of the atmosphere, for example by major volcanoes and by periods of heavy industrialization or biomass burning, have the capacity to change the temperature of the planet. Marine phytoplankton and the vegetation itself contribute 178

to the aerosol content of the atmosphere, by emitting certain volatile organic compounds which form aerosols (Meir et al., 2006). Variations in the Sun’s energy output is sometimes proposed as a possible cause of global warming. Although we talk of the energy incident on the Earth as measured outside the atmosphere as the ‘solar constant’ (and assign it the value of 1366 W m-2), it is not quite constant. The most conspicuous variations are associated with sunspots, which appear as dark marks on the solar surface and arise because of variations in the magnetic properties of the Sun. They occur in an 11-year cycle, but there is a possibility of less conspicuous longer-term trends (see Chapter 5). Changes in the solar output caused by sunspots are only about 0.1% of the solar constant. Changes in solar irradiance since 1750 have caused a radiative forcing of only + 0.12 W m-2, and Lockwood and Fröhlich (2007) have shown that the changes in solar radiation since the preceding 20 years to their paper were in the opposite direction to that required to explain the observed rise in global mean temperatures. Volcanic eruptions are sometimes large enough to have a short-term impact on the climate (see Box 6.8 in Chapter 6). Globally, the CO2 emissions from volcanoes are much smaller than fossil fuel emissions, probably less than 0.3 Gt C yr-1. The significance of volcanoes is in their injection of aerosol particles into the atmosphere. For example, following the June 1991 eruption of Mount Pinatubo the aerosols formed a reflective layer of sulfuric acid haze and global temperatures dropped by about 0.5°C. Likewise, the April 1815 eruption of Mount Tambora in (modern-day) Indonesia is believed to have been the cause of the exceptionally cold conditions everywhere in the world in the following year: 1816 is known as ‘the year without a summer’. Significant volcanic eruptions in recent times were Mount Agung in Bali in 1963 and El Chichonal in Mexico in 1981. The scientific consensus is, overwhelmingly, that the production of greenhouse gases by humans is the primary cause of recent global warming, as outlined in various reports from the Intergovernmental Panel on Climate Change, IPCC (www.ipcc.ch). One of the most compelling lines of evidence is that global climate models (GCMs), in which production of these gases is simulated, show the same pattern of global warming as that observed, and in all parts of the world (Figure 7.4). When run without adding anthropogenic production of greenhouse gases, GCMs show no appreciable global warming. Understanding the causes of climatic variation is still an area of intense research, drawing upon expertise from

7.2  Climate change

Figure 7.4 Comparison of observed and simulated climate change based on three large-scale indicators in the atmosphere, cryosphere and

ocean: change in continental land surface air temperature (yellow panels), Arctic and Antarctic September sea ice extent (white panels) and upper ocean heat content in major ocean basins (blue panels). Time series are decadal averages, plotted at the centre of the decade and are relative to 1880–1919 for surface temperatures, 1960–1980 for ocean heat content and 1979–1999 for sea ice. For ocean heat content three different sets of observations are shown. For temperature panels, observations are dashed lines if the coverage of examined areas is below 50%. For other panels, dashed lines indicate poor data coverage. Model results shown are from Coupled Model Intercomparison Project Phase 5 (CMIP5) multi-model ensemble ranges with shaded bands indicating 5 to 95% confidence intervals. (Source: IPCC, 2014).

many scientific disciplines. One important issue is the behaviour of a myriad of negative feedbacks which tend to dampen any instability, and the extent of the influence of positive feedbacks which might cause run-away warming. For example, as the climate warms and the polar ice melts, the overall albedo of the planet will decline (land

and sea absorb more energy than ice and snow). A decline in albedo will make the planet’s surface absorb more solar radiation, and thus further warming will occur. In this way, warming may give rise to further warming, an example of a positive feedback loop. Some examples of positive and negative feedbacks are given in Box 7.1. 179

Chapter 7 Contemporary climate change

CLIMATE FEEDBACK Anthropogenic activity is believed to be enhancing climate change and encouraging the planet to warm. However, there may be a range of feedbacks that result in different responses to human activity. Positive feedbacks on the climate system will accelerate global warming, whilst negative feedbacks will suppress warming. There are a whole range of interlinked processes that suggest we need to look at environmental change taking a whole-system viewpoint. This box lists some of the hypothesized climate feedbacks which global modellers are investigating. Positive feedbacks

t As the sea surface warms, more water vapour enters the atmosphere. H2O is a strong absorber of infrared radiation and so this increased water vapour is expected to increase the rate of warming.

t Deforestation will lead to an increase of soil erosion, atmospheric aerosols will increase and solar radiation at the surface will decline, causing cooling.

t Warming will melt snow and ice, decreas- t Replacement of coniferous forest by ing albedo and thus increasing warming, melting even more snow and ice.

t Tropical deforestation will cause warming and drying, itself causing a decline in the rainforests of the world.

t Increased cover of woody vegetation in the high latitudes, caused by warming, will decrease the reflectance of the land surface, and thus accelerate warming.

t Warming will cause release of CO2 from t Warming will increase the decomincreased biomass decomposition, primarily in the forest regions of the world but also in the tundra, thus accelerating warming.

Negative feedbacks

position rate of gas hydrates (see Chapter 24) leading to a release of the potent greenhouse gas methane; this will increase warming.

warmth-loving broadleaved forests and by agriculture will decrease planetary reflectance, causing cooling.

t Increased transpiration in a warm world will lead to more clouds, cooling the planet.

t Increased precipitation and ice melt will result in increased runoff into sensitive parts of the oceans altering the balance between freshwater and saline water thereby resulting in a slowing of ocean circulation and allowing northern high latitudes to cool (see Chapters 4 and 5).

BOX 7.1

7.2.3 Predictions from global climate models (GCMs) Global climate models have developed from global circulation models (both abbreviated GCM), which in turn sprang from the application of numerical methods to weather forecasting. Predictions of the climate for the next century are made by running GCMs with specified prior assumptions about the pattern of greenhouse gas emissions. In reality, these patterns will depend on social, political and economic development in the world, and they are patterns which we cannot foretell. So researchers define them as ‘scenarios’ or ‘storylines’, each storyline having a particular pattern of greenhouse gas emissions, and use GCMs to investigate the consequence of each scenario. In the year 2000, the scenarios were defined by the IPCC and are summarized below (in the most recent, Fifth IPCC Assessment report (2014) scenarios are replaced by ‘Representative Concentration Pathways’, RCPs): A1: In this scenario, there is rapid economic growth, an increasing human population until mid-century and 180

thereafter a decline, and the rapid introduction of more efficient technologies. Three A1 groups are distinguished: A1FI is fossil fuel intensive, A1T uses non-fossil energy and A1B uses a mixture of the two. A1B corresponds to what most traditionalists imagine will happen. A1FI leads to the CO2 concentration rising from 380 ppm to around 960 ppm while A1B results in 710 ppm by 2100. A2: Here, the world develops in a more heterogeneous way with emphasis on self-reliance. Fertility patterns have regional characteristics and converge slowly; economic growth and technological uptake are more fragmented. In this scenario, the CO2 concentration rises from 380 ppm to 860 ppm by 2100. B1: Like A1, scenario B1 is a convergent world with a population that peaks in the mid-century but with a strong evolution of a service and information technology, with reductions in material intensity and clean technologies. In B1, global solutions are found to economic, social and environmental sustainability. In this scenario, the CO2 concentration rises from 380 ppm to 540 ppm by 2100.

7.2  Climate change

6.0

Global surface warming (5C)

5.0

A2 A1B B1 Year 2000 constant concentrations Twentieth century

4.0

3.0

2.0

1.0

1900

2000 Year

A1FI

A2

B2

A1T

B1

-1.0

A1B

0.0

2100

Figure 7.5 Predictions of global warming for different scenarios (see text for an explanation). The coloured lines show the assumed socio-economic scenarios and the bands around the lines show the range of model behaviours. Note: the orange line shows the effect of keeping the concentration constant from the year 2000. (Source: IPCC, 2007a)

B2: In B2, local solutions to economic, social and environmental sustainability are found; the population growth rate is slower than A2, with intermediate levels of economic development, and there is less rapid technological change than in A1 and B1. In this scenario, the CO2 concentration rises from 380 ppm to 615 by 2100.

When the GCMs are run, we see a warming by 2100 ranging from 1.8°C in the B1 scenario to nearly 4°C in the A1B scenario (Figure 7.5). There are associated changes in rainfall. In the A1B scenario, the rainfall patterns are substantially different from those today with more rain falling in the polar regions while the mid-latitudes will become drier (Figure 7.6). The Mediterranean region of

A1B

A1B

December, January, February

—20% —10% —5%

5%

10%

June, July, August

20%

Figure 7.6 Relative changes in precipitation for the period 2090–2099, relative to 1980–1999, assuming the A1B scenario for December to February (left) and June to August (right). White areas are where less than 66% of the models agree on the sign of change. (IPCC, 2007a)

181

Chapter 7 Contemporary climate change

(b) (a) 20 Standard deviation

16

A2 B1 A1B

12 8 4 0 -4 1880

1920

1960

2000 Year

2040

2080

—3.75 —3 —2.25 —1.5 —0.75

0

0.75 1.5 2.25

3

3.75

Standard deviation

Figure 7.7 Increase in heatwaves over the rest of the century, expressed as the standard deviation of temperature: (a) trends

from runs of the GCM for three scenarios; (b) global pattern for A1B. (Source: IPCC, 2007a)

Europe and Central America will both become especially dry according to this prediction. These changes are profound, especially so as the models suggest a more variable climate with an increasing frequency of extreme events. News reports of storms, droughts and hurricanes are increasingly shown in the media but these alone should not be taken as evidence of a link between global warming and extreme events. Analyses of reliable long-term records and model predictions are the proper evidence that must be considered. Emanuel (2005) investigated data on hurricanes and found the total power dissipated (longer storms and more intense storms) has increased markedly since the mid-1970s. Moreover, model predictions do indeed show an increased variability with an increased frequency of extreme events. The 2014 report of the IPCC states: Changes in many extreme weather and climate events have been observed since about 1950. Some of these changes have been linked to human influences, including a decrease in cold temperature extremes, an increase in warm temperature extremes, an increase in extreme high sea levels and an increase in the number of heavy precipitation events in a number of regions. Heatwaves, for example, are expected to increase (Figure 7.7). Results such as these prompt economic analysis and receive the attention of the public and of politicians. Insurance companies can no longer base their premiums on the analysis of past data when the climate system is so clearly changing its behaviour. Such ‘extremes’ in temperature, rain and wind will all cause appreciable damage, and the cost of repairing the damage will ultimately consume much of the wealth of the world, as emphasized by the report of Sir Nicholas Stern made to 182

the UK government (Stern, 2006). Box 7.2 provides some examples of potential impacts of global warming.

7.2.4 Critical evaluation of the state-of-the-art in GCMs GCMs have significant weaknesses which are frequently highlighted by sceptics. However, according to the IPCC, most climatologists agree that better models would not materially influence the conclusions of the model runs. Some of these perceived weaknesses are mentioned for consideration here, as follows.

7.2.4.1 Resolution The spatial resolution of models may be too coarse. For example, in the HADCM3 model, the GCM used at the Hadley Centre in the United Kingdom, the grid cells for the global runs made for the IPCC are 2.5 * 3.75 degrees in latitude * longitude, and the time steps are half-hour. There is a practical limitation on spatial and temporal resolution imposed by the speed of the supercomputer as it takes a long time to run a GCM for a 100 year or more simulation. As computing power increases over the next few years there will be improvements in resolution. Some features of the climate system of course have a characteristic size which is small in relation to the grid cells (hurricanes and even clouds have to somehow be represented).

7.2.4.2 Biological and chemical coupling Attempts to represent the impact of the biology and chemistry in GCMs are in their infancy. To some extent this bottleneck relates to the lack of process understanding. For

7.2  Climate change

IMPACTS OF GLOBAL WARMING Most current models suggest a 2–4 °C increase in global temperature over the next century. This rate is 10 times faster than the warming experienced over the past 10 000 years, and substantial impacts on human societies are anticipated. The following is a summary of some of the main effects that have been widely discussed in various publications.

t

t

Extinction rates, already very high, will be increased even more. Species would have to be capable of very fast migration to keep up with the isotherms which would move northwards at about 10 km per year. A recent study suggests that 13–37% of species will be ‘committed to extinction’ by 2050 (Thomas et al., 2004). Although cold countries such as Russia may enjoy an increase in agricultural productivity as a result of warming, at least in the short term, many of the world’s main food-producing regions

may become too hot and dry for crops to grow. This would include major ‘breadbasket’ regions such as central and southern Europe and North America.

t

t

Low-lying ground, including many major cities of the world and some entire small island states, may be inundated as a result of thermal expansion of the ocean and the melting of ice. According to IPCC (2014) sea levels are likely to rise 0.20–0.86 m from 2000 to 2100 (IPCC, 2014), but some recent authors have found much higher rates (e.g. Levermann et al., 2013): 1.6 m for every 1 °C rise in temperature. Some geographical regions will suffer more than others. Temperature rises are currently especially high in the high latitudes (leading to melting of ice). Models show an increase in the extent of El Niño, with high rates of warming and drying in some of the tropical regions, causing replacement of the rainforest of Brazil by savanna.

t

Diseases are likely to spread from the tropics to the temperate and northern regions as the climate warms. Outbreaks of pests may become more extreme, as the natural biological control processes may not always be present. Of particular concern is the northward spread of insect pests which damage crops or transmit disease. One such example is Lyme disease, a lifethreatening disease carried by ticks and found to be more prevalent in the United Kingdom during warm years.

t

The cost of repairing damage caused by extreme events will escalate and occupy a major proportion of the world’s economic production.

Although there is considerable uncertainty in model predictions, partly because the models do not incorporate many of the likely feedbacks that derive from the vegetation itself, there is now agreement that the countries of the world must co-operate to reduce the emissions of greenhouse gases.

BOX 7.2

example, how should we model the effect of warming on the respiratory production of CO2 by the soil microbes? In general, our understanding of the carbon cycle is incomplete, and arguably we are not yet ready to represent it in global climate models, yet its behaviour clearly has the potential to generate ‘surprises’ in the form of new and substantial sinks and sources of carbon. This is touched upon in Section 7.3 below. Similar remarks could be made in the realm of atmospheric chemistry and aerosol science.

7.2.4.3 The behaviour of ice It is very difficult to represent properly the melting of ice, and the existing GCMs fail to deal specifically with the consequences of the possible melting of the Greenland ice sheet. This would cause a massive influx of meltwater into the North Atlantic, changing the ocean circulation patterns and

therefore profoundly altering the distribution of heat over the Earth’s surface. Such abrupt events may have occurred before, as the Heinrich events, which are evident during the last glacial period (Rahmstorf, 2000; see Chapter 4).

7.2.4.4 The human dimension No one has attempted to incorporate models of the social and economic life into GCMs. The most significant change in land use is currently tropical deforestation. When forest is replaced by pasture, as for example, in Amazonia, the land surface becomes more reflective, the pattern of evapotranspiration becomes more seasonal, the cloud cover is reduced, and the surface becomes aerodynamically smoother (Figure 7.8). These changes, on the scale of the Amazon Basin (over 4 million km2), have the potential to change the climate not only in Brazil but elsewhere. 183

Chapter 7 Contemporary climate change

High rainfall

High evapotranspiration

Low sensible heat loss

Low albedo

Deep roots

Low rainfall

Low evapotranspiration

High sensible heat loss

Solar radiation

Pasture Solar radiation

Forest

High albedo

Shallow roots

Figure 7.8 The effect of land-use change (tropical forest to pasture) on the energy and water balance of the landscape. Forests absorb more solar radiation and have a higher evapotranspiration rate than pastures.

Reflective questions ➤ What evidence is there that global warming is caused by human action?

➤ What might be the impacts of climate change associated with enhanced greenhouse gas concentrations?

➤ What are the benefits and limitations of GCMs?

7.3 The carbon cycle: interaction with the climate system The carbon cycle is the circulation of carbon atoms between the ocean, land and atmosphere by physical, chemical and biological processes. Living organisms are of great importance in the carbon cycle, because life on planet Earth is made of carbon compounds such as proteins, carbohydrates and lipids. Thus, the cycle is one of many biogeochemical cycles, whereby organisms reuse vital elements such as carbon, nitrogen, sulfur and phosphorus. The principal processes involved in transfer from the atmosphere are the dissolution of CO2 in the oceans and the uptake of CO2 by the balance between photosynthesis and respiration of living organisms. The processes involved in the return to the atmosphere are the release of CO2 from the ocean in regions where the ocean upwells (see Chapter 3), and the breakdown of organic matter by respiration or fire. We can thus envisage the carbon cycle as a set of fluxes between major pools as shown in 184

Figure 7.9. Greenhouse gases have ‘sources’ and ‘sinks’. In the case of CH4, the main sink is the photochemical process whereby it is oxidized in the troposphere by the radical OH, and its lifetime in the atmosphere is well known (9.6 years). In the case of CO2, the sinks are many and some of them are long-lived compounds such as lignin and the carbonates, which makes it much more difficult to estimate its atmospheric lifetime. An understanding of the carbon cycle is fundamental to our understanding of life itself, as all biomass is carbonbased. The principal biochemical constituents of cells (carbohydrates, proteins, lipids) have a high carbon content, and the overall carbon content of dry biomass is in the range 45–55%. The carbon cycle has become especially topical in recent years, since the realization that warming is large enough to force the cycle out of equilibrium, whereby the concentration of the gas in the atmosphere is rising. As we have seen above, CO2 is by far the most important of the several gases that absorb infrared radiation emitted from the planetary surface, and its continuing rise is capable of causing additional global warming. The annual quantity of carbon emitted by burning fossil fuels and making cement in the period 2005–2014 was 9.0 Gt C yr-1 and clearing forests in the tropics released about 0.9 Gt C yr-1, making a total anthropogenic burden on the atmosphere of 9.9 Gt C yr -1. However, the concentration of CO2 in the atmosphere is increasing by only about 4 Gt C yr-1. The CO2 not appearing in the atmosphere (the ‘missing’ carbon) is being taken up by terrestrial photosynthesis or dissolving in the ocean. It seems that the terrestrial vegetation may be absorbing around half of the ‘missing’ carbon, and the rest is dissolving in the ocean

7.3  The carbon cycle: interaction with the climate system

¢ about +4.4 per year

Total

0.9

9.0

Oceanic sink

Land sink

3.0

Fossil fuels

Terrestrial ecosystems Vegetation 610 1400 Soils 60 Litter

Land-use change

Atmosphere, 750

2.6

2070

Surface ocean 1020

0.9 Land to ocean 40

50

Marine biota 3 6 Fossil fuel reserves 900

DOC 700

100 4

6

92 Ocean 38,100 0.2 Sediment

Carbonate rocks 65 * 106

Figure 7.9 The global carbon cycle, averaged over the period 2005–2014 (units are gigatonnes, i.e. billions of tonnes, 109 tonnes or 1 petagram). The term for fossil fuels includes emissions from cement manufacture. Land and ocean sinks may vary from year to year, especially the land sink. The land sink is the net of very large gains by photosynthesis (around 120 gigatonnes of carbon per year) and losses of a similar magnitude by respiration of plants and other organisms. Data for stocks on the land and in the ocean are very uncertain. ‘DOC’ is an abbreviation for dissolved organic carbon. For regular updates to the global carbon budget please visit: www.globalcarbonproject.org/carbonbudget/

(Friedlingstein et al., 2010). This is reflected in the fluxes shown in Figure 7.9. However, the ocean and terrestrial might not continue to provide this service. According to some models, the terrestrial sink will diminish and then become a source as a result of the impact of high temperatures and droughts in the tropics (Cox et al., 2000). On the other hand, as warming occurs there is a reasonable expectation that the sinks in the northern regions will strengthen, perhaps enough to compensate for the loss of sink strength in the southern regions. This theme is returned to in Chapter 13. There are prospects of enhancing the strength of the sinks in order to slow down the rate of climate warming. On the terrestrial side, this might be done by planting more forests, or by protecting existing forests. It could also be done by modifying agricultural practices (less ploughing, for example) in order to conserve the carbon stocks in the soil or to protect peatlands and other wetlands. The potential of enhancing the terrestrial sink by

land-use changes of this kind is considerable. The ocean sink might also be managed by fertilizing the ocean. The scientific basis for this proposition is the observation that phytoplankton in the deep ocean are short of iron, one of micronutrients that all life requires. When iron is added as ferric ions the productivity of phytoplankton is increased. This provides more food for zooplankton, and more food for the fish that eat the zooplankton, so there should be an enhanced stream of dead biota and carbonate shells that sink to deeper layers of the ocean. This, in turn, should enable more CO2 to dissolve in the surface waters. No one really knows whether this will work on a large scale. Environmentalists have generally opposed all suggestions of increasing the strength of ocean sinks, and sometimes sinks in general, arguing that the mechanisms and processes are imperfectly understood, and the sinks cannot be depended upon in the long term. They argue that there is no alternative but to reduce emissions by reducing consumption and finding alternative ‘clean’ sources of energy. 185

Chapter 7 Contemporary climate change

Reflective questions ➤ Can you draw a diagram of the carbon cycle and explain it?

➤ Why is the total increase of CO2 in the atmosphere currently less than the CO2 emitted by human actions?

7.4 Mitigation The question of ‘what can be done?’ to avoid dangerous climate warming has to be addressed at a global scale. If one country alone were to apply stringent measures at great expense to reduce fossil fuel emissions while other countries go ahead and increase theirs, then no one would gain. Countries therefore must engage in debate and decide on the actions required before it is too late. Following the 1992 ‘Rio Summit’ to discuss environmental change, many countries joined an international treaty – the United Nations Framework Convention on Climate Change (UNFCCC) – to consider what might be done to reduce global warming. Later, in 1997, a number of nations approved an addition to the treaty: the Kyoto Protocol, which imposes powerful (and legally binding) measures to reduce emissions. The Kyoto Protocol is an international agreement to limit the emission of greenhouse gases. Six gases are

mentioned in the protocol, of which CO2 is the most important contributor to warming (Table 7.1). They differ greatly in their residence time in the atmosphere, and in the extent to which they are effective in absorbing infrared radiation. These two factors together are incorporated into an index called the global warming potential (GWP), which measures the relative effectiveness of the gas, on a per molecule basis, in causing global warming over a century. To date, the Kyoto Protocol has not been successful. Most signed-up countries failed to meet their emission targets. The protocol has been largely replaced by the Paris Agreement made in December 2015 at the United Nations Climate Change Conference, COP 21. Members agreed to reduce their carbon output ‘as soon as possible’ and to do their best to keep global warming ‘to well below 2°C’. Meanwhile, the first indication that global carbon emissions may have peaked was published (Jackson et al., 2016). The slow-down of China’s economic growth and fossil fuel usage, rather than any international agreement, seems to be the main cause of the apparent reduction in global carbon emissions. At an individual level, some people opt to take personal responsibility for their carbon emissions. In a developed western society it is possible to make substantial savings in this way, as Reay (2006) has pointed out. But only a few people have so far taken direct control over their ‘carbon footprints’ by changing their lifestyles. Most people in the

Table 7.1 Greenhouse gases in the Kyoto Protocol, lifetime in the atmosphere and global warming potential (GWP) on a scale where CO2 = 1 (see text for definition)

Gas

Sources

Lifetime (years)

GWP

Comment

Carbon dioxide (CO2)

Fossil fuel burning, cement manufacture

Sometimes assumed to be 100 years but probably longer

1

On the increase still, but Annex 1 countries* likely to reduce emissions

Methane (CH4)

Wetlands, rice fields, burning, oil wells, ruminants, termites

12

21

Nitrous oxide (N2O)

Land disturbance, use of nitrogen fertilizers

120

310

Hydrofluorocarbons

Industry, refrigerants

1–300

140–11 700

Substituting for CFCs. They contain only hydrogen, fluorine and carbon, and do not damage the ozone layer (CF3CFH2, CF3CF2H, CHF3, CF3CH3 and CF2HCH3)

Perfluorocarbons (CF4, C2F6)

Industry, electronics, firefighting, solvents

2600–50 000

6500

Also substituting for CFCs

* Annex 1 countries are 37 developed countries and economies in transition who had committed themselves to reducing emissions to 1990 levels by 2000. The term ‘Annex 1 countries’ is still used despite the original date having lapsed. Annex 1 countries pledged in the Kyoto Protocol to reduce emissions to 5.2% below 1990 levels in the 2008–2012 period. (Source: Woodward et al., 2004)

186

7.5  Destruction of the ozone layer by chlorofluorocarbons (CFCs)

developed world, and an increasing number in the developing countries, have a substantial component of emissions from travel; they often ignore options to reduce these by selecting low-carbon-emitting modes of transport (train not plane, bike not car). Currently, the cheapest and most convenient mode of transport is often plane or motor vehicle, both of which have high emissions per journey. In a survey of the travelling habits of the people of Edinburgh a linear relationship was found between personal income and transport emissions (Figure 7.10), suggesting an

(a) 12

CO2 emissions from travel (t CO2 yr-1 capita-1)

10 8 6 4 2 0

0

20 40 60 80 Income (thousands £ per household per year)

(b)

100

Reflective questions

3.0

➤ Can you list some positive and negative climate feed-

2.5 Mean annual CO2 emissions (t CO2 yr-1 capita-1)

inevitable association between wealth and travel emissions. The challenge for governments is how to enable modes of travel which are efficient and affordable yet which do not incur such high emissions of greenhouse gases. At a national level, some countries have set themselves carbon emission reduction targets which go well beyond those prescribed by the Kyoto Protocol. Various countries have passed legislation to limit their own emissions. For example, the UK government’s Climate Change Act states: ‘It is the duty of the Secretary of State to ensure that the net UK carbon account for the year 2050 is at least 80% lower than the 1990 baseline.’ Other European countries have similar legislation, passed or in preparation. These goals will only be achieved by quite radical and potentially unpopular changes which will have to include: carbon taxes, new technologies (especially, the burial of CO2 in geological strata) and a move back to nuclear power. The use of renewable energy sources such as wind power and biomass energy can make a significant contribution. Countries such as China that are developing economically at a very fast rate, and have a large reserve of easily accessible coal, may decide that their economic growth will not be fuelled by coal.

backs and some of the impacts of climate change that could be associated with increased greenhouse gas

2.0

concentrations? 1.5

➤ Using Internet tools work out your own personal carbon footprint and reflect how you might make it smaller.

1.0

Would this impact on your quality of life? 0.5

n

in

Ot

th Va e ca UK tio p n ro ro ab fe ad ss ro fo io ad rp na ro lr e fe as ss on io s na lr ea so ns

he r

ng Le isu re

Sh o

pp i

io at

ab ys

ur ne Jo

Jo

ur ne

ys

in

th

e

UK

fo r

Va c

Ch ild

re n

to

sc h

oo

l

0.0

Figure 7.10 Per capita energy emissions associated with travel for

people of Edinburgh, Scotland, in 2005: (a) relationship with income; (b) type of travel (blue bars are deprived social group, grey bars are affluent social group). (Source: from Korbetis et al., 2006)

7.5 Destruction of the ozone layer by chlorofluorocarbons (CFCs) Electromagnetic radiation from the Sun reaches the Earth’s atmosphere at a rate of about 1366 W m-2. Much of it is scattered back into space, and only a fraction reaches the surface; it drives photosynthesis and evaporation, and warms the planetary surface. The radiation contains ultraviolet (waveband 100–400 nm) radiation, visible radiation, which happens also to be the photosynthetic waveband (waveband 400–700 nm), and near-infrared radiation (waveband 700 nm to a few nanometres). Ultraviolet 187

Chapter 7 Contemporary climate change

radiation is absorbed by the DNA of all organisms, causing damage to the genetic code and consequently interfering with protein synthesis and the control of cell division. In humans the most common effects include reduction in the immune system (all races), skin cancer in Caucasiantype humans and damage to the eyes. For the last billion years, the Earth has been shielded from damaging ultraviolet radiation by ozone (O3) in the stratosphere. This protection has enabled life to develop on the land. Now, the ozone layer is diminishing as a result of a chain of chemical reactions that begins with totally human-made chemicals, the chlorofluorocarbons (CFCs). These CFCs are synthetic non-reactive gases and liquids first made in 1930 and used as refrigerants (later as propellants in spray cans). Being inert under normal conditions, they persist in the atmosphere, and slowly make their way to the stratosphere. Laboratory studies in 1974 established that CFCs could catalytically break down ozone in the presence of ultraviolet radiation to form highly reactive radicals such as ClO and OClO. It is these radicals that catalyze the breakdown of O3 to O2. A ground-based survey of stratospheric ozone was started in Antarctica in 1956, followed by satellite surveys in the early 1970s. In 1985, a British team based in Antarctica reported a 10% drop in the ozone level during the spring, which they attributed to CFCs and oxides of nitrogen. A similar decline was also seen in data from NASA’s Nimbus 7 satellite carrying TOMS (Total Ozone Mapping Spectrometer),

Oct 1979

and it is now evident that a steady decline is occurring over Antarctica (Figure 7.11) whilst a decline has also been detected over the Arctic. In the 1980s, Australians sunbathed much less than before, and sales of sunhats and skin creams to protect against ultraviolet radiation increased. Plants and animals are less able to take such protective measures. The Montreal Protocol of 1987, in which nations agreed to phase out the use of CFCs, has undergone several modifications. Trade sanctions on CFCs have been imposed and a total phasing out is due in 2030. In March 1989, environmental ministers of the European Community announced a total phase-out of CFCs in Europe by the year 2000. More recently, related chemicals which do not significantly destroy ozone have been introduced: these are the hydrofluorocarbons (HFCs) and the perfluorocarbons (PFCs). Unfortunately, these gases are powerful greenhouse gases (Table 7.1), although their concentration in the atmosphere is very low and so they do not presently contribute much to global warming.

Reflective questions ➤ Why is the ozone layer crucial to life on Earth? ➤ Has the Montreal Protocol of 1987 been more successful than the Kyoto Protocol of 1997?

Oct 2015

0

100

200

300

400

500

600

700

Total Ozone (Dobson units) Figure 7.11 Images of the average monthly thickness of the stratospheric ozone layer over the southern hemisphere comparing October 1979 with October 2015. The units are Dobson units named after G.M.B. Dobson, an early investigator. Dark blue and purple indicate thinner ozone. The scale is linear. (Source: NASA Ozone Watch)

188

7.6  The future

7.6 The future Fossil fuel continues to be the main source of energy. Moreover, the developing world, which consists of about five-sixths of humankind, will continue to increase its population and its fossil fuel burning for many years after the rich countries have stabilized and decreased their dependency on fossil fuels. Some poor countries have neither fossil fuels nor any other supply of energy, and so cannot develop. Even fuel-wood is in short supply. Nuclear power was developed enthusiastically by many countries in the 1950s, and 29 countries were running 437 nuclear power plants by 1998. Early optimism about development of an energy economy from nuclear fission faded, following nuclear accidents and leakages such as Chernobyl in the USSR in 1986 and Fukishima in Japan in 2011. Many environmentalists believe that the risks that are inherent in nuclear fission are quite unacceptable. Power from nuclear fission is very expensive, once the costs of handling radioactive waste and decommissioning old power stations are taken into account. Despite all this, many governments are in favour of continuing and even expanding their nuclear power programmes, and for many it is the only practical way to reduce carbon emissions. There are, however, some reasons for optimism. In the period since 1960 considerable progress has been made towards developing alternative sources of energy to replace carbon-based fuels (coal, oil and gas). Governments in the richer countries are setting ambitious targets to decarbonize their energy economies (which means they will use less fossil fuels to produce energy) and are pushing for investment in renewable forms of energy such as wind and biofuels. Solar cells are likely to become increasingly

important. This technology was first developed during the exploration of space in the 1960s. Silicon solar cells convert solar energy to electrical energy with an efficiency of 20%, and the energy may be stored and transported in fuel cells. The construction cost of a solar cell is rather low, as the main elemental ingredient, silicon, is abundant. Unlike wind turbines and wave power generators there are no moving parts and consequently maintenance costs are extremely low. Even in winter in northern countries, solar cells can provide useful quantities of solar energy. In the future, roof tiles incorporating solar cells may be used in all new housing construction, and solar energy ‘farms’ may cover large areas of deserts. The major problems facing the world are related to each other: climate change, energy supply, poverty and the tensions arising from a disparity of living standards between different people. It is difficult to foresee what kinds of environmental change are just around the corner, and therefore hard to plan for the future. A hundred years ago, the problems of today were invisible and quite unpredicted. But there is some evidence from recent history (Box 7.3) that we are at last beginning to grasp the nature of the human– environment interaction. For 200 years humans have been inadvertently damaging the life-support system of the planet. In the last 30–40 years we have realized what is happening, and governments are beginning to take remedial action.

Reflective question ➤ Despite the challenges ahead, why should we be optimistic about dealing with climate change?

IMPORTANT RECORDED EVENTS The most important dates in the relationship between humans and the global environment are listed below. These events and discoveries changed our perceptions of the world we inhabit, and contributed to global environmental change. From 1500 to 2015 the human population increased from 0.5 billion to 7.3 billion. 1400–1450

Chinese fleets arrive in Middle East and East Africa; Arab fleets cross Indian Ocean. Intercontinental trade begins.

1450

Johannes Gutenberg, Germany, establishes printing press, enabling humans to communicate efficiently and to learn from each other more effectively than ever before.

BOX 7.3 ➤

189

Chapter 7 Contemporary climate change

➤ 1490–1500

Decade of European maritime exploration of Asia, Africa and America, paving way for settlement, slavery, trading, further spread of economically important plant and animal species.

1610

Galileo Galilei, Italy, uses the telescope to observe behaviour of the Moon and planets. Other scientific instruments for Earth observation were developed: microscope (1618), thermometer (1641) and barometer (1644).

1780–1820

Industrial Revolution. Dramatic increase in the use of coal. Western Europe sees rapid technological, social and economic transformation, driven largely by the steam engine fuelled by coal. Widespread urban pollution, exploitation of workforce, occupational diseases. Humans begin to alter the composition of the global atmosphere.

1796

First blast furnace opens in Gleiwitz, Poland. Manufacture of iron and steel followed in urban centres: Belgium, Germany, Great Britain. Respiratory diseases increase.

1798

Thomas Malthus, English clergyman, writes Essay on the principle of population, pointing out the natural tendency of human populations to grow exponentially, outrunning the food supply.

1821

William Hart obtains natural gas (methane) from a 9 m well in New York, and provides street lighting.

1827

Jean Baptiste Joseph Fourier (France) first proposed that ‘light finds less resistance in penetrating the air, than in repassing into the air when converted into non-luminous heat’. Possibly the first articulation of the greenhouse effect.

1838

Birth of John Muir at Dunbar, Scotland. First person to call for conservation of wilderness areas, arguably the father of the modern conservation movement.

1839

Antoine-Cesar Becquerel, France, discovers the photoelectric effect, demonstrating that sunlight can be converted into electricity. But practical solar cells were not developed until 1954.

1842

John Lawes, England, invented superphosphate fertilizer.

1851

James Young, Scotland, discovers how to extract hydrocarbons from oil shale, and develops process of refining oil. He establishes a paraffin industry in Scotland (paraffin is called kerosene in the USA) and is nicknamed ‘Paraffin Young’.

1859

Edwin Drake strikes oil at 20 m in Pennsylvania, USA. Oil was soon discovered in North and South America, Mexico, Russia, Iran, Iraq, Romania, Japan, Burma and elsewhere. Oil soon plays its part in the industrialization of the world.

1859

Charles Darwin publishes The origin of species, proposing the theory of evolution by natural selection.

1859

Irish scientist John Tyndall discovers that H2O and CO2 absorb selective wavebands of infrared radiation, and suggested a role for these gases in the regulation of the Earth’s temperature.

1864

James Croll, Scotsman, proposed theory of long-term climate change to account for the Ice Ages (see also the reference to Milutin Milankovitch, 1895)

1866

German engineers Langen and Otto patent the internal combustion engine. The manager in Otto’s factory, Daimler, makes the first petrol engine in 1884.

1868

In Japan, the Meiji Restoration. Japan opens to the west and large-scale industrialization spreads to Asia.

1885

German chemist Robert Bunsen discovers how to make a very hot flame from gas, by mixing it with air before combustion. Gas burners were thereafter much more efficient.

1895

Serbian astrophysicist Milutin Milankovitch describes theory of long-term climate change. Essentially it is the same theory that Croll had proposed in 1864, but Croll’s work was ignored by his peers.

1896

Arrhenius, Swedish chemist, advances theory that CO2 emissions will lead to global warming, and postulates the ocean as a CO2 sink.

1901

Italian Guglielmo Marconi invents radio, achieves first transatlantic radio message.

1903

Henry Ford, USA (1863–1947), establishes the Ford Motor Company, makes Model-T Ford cars in 1908. Others would follow Ford’s idea of mass production, and car ownership would increase rapidly.

1903

Orville and Wilbur Wright, USA, demonstrate a flying machine based on the internal combustion engine. Rapid intercontinental travel by air would follow in 50 years.

1907

Henry Ford completes first tractor, a machine that was to revolutionize agriculture.

1909

Fritz Haber, Germany, shows how to synthesize ammonia from N2 in the atmosphere, and Karl Bosch uses this process for mass production of nitrogen fertilizer. This was to greatly increase the capacity of humans to grow crops.

1915

German geophysicist Alfred Wegener publishes his controversial hypothesis of continental drift, in a book entitled The origin of continents and oceans.

BOX 7.3 ➤

190

7.6  The future

➤ 1917

Chainsaw manufactured for the first time. Enables rapid deforestation.

1926

John Logie Baird, Scotland, invents television, but TV broadcasts did not start until 1936 (Britain) and 1941 (USA), and TV sets were not widespread until the 1950s. Television has allowed ordinary people to see how others live, and thus to understand better their own place in the world.

1927

Alexander Fleming, Britain, discovers the antibiotic effect of the fungus Penicillium. Much later (1940) Florey and Chain, working in the USA, discover how to make the antibiotic penicillin in bulk. This launches golden age of medicine. Infant mortality declines and people live longer.

1928

Mohandas Karamchand Gandhi, India leader, questions the sustainability of the industrial age: ‘God forbid that India should ever take to industrialism after the manner of the West. If it took to similar economic exploitation, it would strip the world bare like locusts.’

1930

Thomas Midgely, USA, invents the gas freon. It was the first of the chlorofluorocarbons (CFCs) which much later (1970s) became widely used and caused thinning of the stratospheric ozone layer.

1938

Guy Stewart Callendar (UK) predicted global warming at a rate of 0.03°C per decade.

1943

Primitive electronic computers constructed: Harvard Mk I (USA) and Colossus (Britain).

1944

Pilotless planes and rockets developed in Germany for use in warfare. Later, the technology formed the basis for space exploration and Earth observation.

1945

Atomic bomb dropped on Hiroshima. Guardian newspaper (UK) comments ‘man is at last well on the way to mastery of the means of destroying himself utterly’.

1948

Agricultural efficiency increases dramatically in the developed world, as a result of mechanization, fertilizers, pesticides, plant breeding and managerial skill. Crop yields increase. Later (1960s and 1970s) the new agricultural technology is taken to the developing countries, and becomes known as the Green Revolution.

1948

First operational stored-program computer, known as Manchester Mk I (Williams, Kilburn and Wilkes, Britain).

1951

Age of nuclear power starts with first commercial nuclear reactor at Idaho, USA. Later there are significant accidents: fire at Windscale, UK, in 1957; meltdown at Three Mile Island, Michigan, USA, in 1979; meltdown and large release of fission products at Chernobyl, USSR, in 1986.

1952

British jet airliner, De Havilland Comet, begins regular intercontinental travel.

1954

Chapin, Fuller and Pearson develop silicon solar cell capable of converting solar energy into electrical energy with a conversion efficiency of 15%.

1957

First spacecraft, Sputnik, USSR; to be followed by first man in space, Yuri Gagarin, USSR, in 1961 and first man on the Moon, Neil Armstrong, USA, 1969.

1958

Charles Keeling, of the Scripps Institute in the USA, begins the first reliable measurements of atmospheric CO2 at Mauna Loa in Hawaii.

1960

Soviet engineers begin large-scale irrigation using rivers flowing to the Aral Sea, the world’s fourth largest lake. Within 40 years the lake would almost disappear, possibly the greatest hydrological change yet engineered by humankind.

1962

Silent spring by Rachel Carson, USA, warns of dangers of pesticides to wildlife. This best-seller inspired a whole generation of environmentalists.

1968

Satellite remote sensing starts. Pictures of Earth from deep space, Apollo 8 mission, USA; followed in 1972 by Earth Resources Satellite ERTS-1 carrying multispectral sensors later called Landsat.

1969

In the USA, the Advanced Research Projects Agency (ARPA) begins the ARPANET. Soon, global communication by e-mail and Internet would become possible.

1970

Establishment of Environmental Protection Agency (EPA), USA.

1971

Formation of Greenpeace. A group of activists sail their small boat into a US bomb-test zone near Alaska to draw attention to the environmental dangers of nuclear war.

1971

Swedish scientists demonstrate long-range transport of sulfur as the cause of acidification of Swedish lakes, and predict that acid rain will damage fresh water ecosystems and forests.

1972

In the UK, publication in The Ecologist of ‘A blueprint for survival’, warning of the extreme gravity of the global situation and criticizing governments for failing to take corrective action.

BOX 7.3 ➤

191

Chapter 7 Contemporary climate change

➤ 1972

Publication of The limits to growth by the Club of Rome, dealing with computer simulation of global environmental change. Fails to identify the threat of global warming; points to resource depletion and pollution as the major threats.

1972

First international conference on the environment, Stockholm, leading to the establishment of the United Nations Environment Programme (UNEP). Acid rain is widely publicized, especially in relation to forest decline, but since then the developed world has been moving to low-sulfur fuels.

1972

The anchovy fishery of Peru collapses because of overfishing and bad weather. Other fish stocks decline sharply, and management of marine resources becomes an important issue.

1973

Organization of Petroleum Exporting Countries (OPEC) restricts the supply of oil, forcing its price to rise five-fold and threatening the global economy.

1979

James Lovelock proposes the Gaia hypothesis (see Chapter 23).

1985

Farman, Gardiner and Shanklin, a British team working in the Antarctic, report thinning of stratospheric ozone, attributable to CFCs.

1986

Nuclear accident at Chernobyl, USSR, creates radioactive fallout everywhere in the northern hemisphere, reminding people that environmental problems cross political boundaries. The expansion of nuclear power in the west falters.

1987

First appearance of the word biodiversity in the scientific literature (by E.O. Wilson, USA).

1987

Ice core from Antarctica, taken by French and Russian scientists, reveals close correlation between CO2 and temperature over the last 100 000 years.

1987

Montreal Protocol signed, an agreement to phase out CFCs.

1987

United Nations World Commission on Environment and Development produce the Brundtland Report, dealing with definitions of sustainability.

1988

Intergovernmental Panel on Climate Change (IPPC) is established.

1990

IPPC’s first Scientific Assessment Report, linking greenhouse gas emissions to warming.

1992

Implementation of the International Geosphere Biosphere Programme (IGBP) to predict the effects of changes in climate, atmospheric composition and land use on terrestrial ecosystems; and to determine how these effects lead to feedbacks to the atmosphere.

1992

Earth Summit, Rio de Janeiro. Leaders of the world’s nations meet in Rio and set out an ambitious agenda to address the environmental, economic and social challenges facing the international community. Heads of state sign the UN Framework Convention on Climate Change. The UNFCCC was one of three conventions adopted. The others – the Convention on Biological Diversity and the Convention to Combat Desertification – involve matters strongly affected by climate change.

1997

Kyoto Protocol is drafted, the first international agreement to limit greenhouse gas emissions.

1997–1998

Particularly severe El Niño causes drought and widespread forest fires in Indonesia, Malaysia, Brazil and Mexico. In Indonesia alone, 97 000 km2 of forest is destroyed, releasing 2.6 Gt CO2.

1998

The warmest year of the century, and probably of the millennium.

2000

International Coral Reef Initiative reports that 27% of the world’s corals reefs are lost, mainly a consequence of climate warming.

2001

US President George Bush announces that the USA will not ratify the Kyoto Protocol.

2002

As warming of Antarctica proceeds, some 3200 km2 of the Larsen B ice shelf collapses.

2002

Schools in Seoul, S Korea, are closed when a dust storm originating from China sweeps over the country.

2003

Gates of China’s Three Gorges Dam are shut, and the world’s largest hydropower reservoir is created, destroying archaeological sites and forcing the relocation of nearly 2 million people.

2003

European heatwave causes premature death of 35 000 people.

2004

Indian Ocean earthquake causes large tsunami, killing a quarter of a million people.

2005

Kyoto Protocol comes into force, 16 February.

2005

Hurricanes sweep the US Gulf Coast, causing widespread damage and loss of life. New Orleans evacuated.

2005

Worst Amazon drought in the last 100 years.

BOX 7.3 ➤

192

7.7  Summary

➤ 2006

The film An Inconvenient Truth (Director Davis Guggenheim, Presenter Al Gore) presents the science of global warming in a manner accessible to non-scientists. It is a box-office success and wins awards.

2006

Nicholas Stern, an economist, in a report for the UK Government, suggests that global warming will cause the greatest and widest-ranging market failure ever seen, and proposes environmental taxes as the best remedy.

2006

China becomes the top CO2 emitter, overtaking the USA.

2007

IPCC Fourth Assessment Report predicts that temperatures will rise by 1.8–4.0°C by the end of the century.

2008

Forests of western North America devastated by epidemic of mountain pine beetle.

2009

Worse wildfires in Australian history, 181 killed in the state of Victoria.

2009

Global fossil fuel emissions fell by 1.3% to 8.4 Pg C per year, as a result of the economic recession; emissions in China and India rose, nevertheless, by 8% and 6% respectively.

2009

United Nations Climate Change Conference ‘Copenhagen Climate Summit’ fails to reach legally binding agreement on curbing greenhouse gas emissions or reducing deforestation. In the resulting Copenhagen Accord many governments pledge to reduce emissions by 2020, including the USA.

2010

Deepwater Horizon oil leak discharges the equivalent of 4.9 million barrels of oil into the Gulf of Mexico, causing a 6500 km2 oil slick and harming fisheries. It became known as the BP oil disaster.

2010

Severe Amazon drought, thought to be the second (after 2005) most severe drought on record.

2010

Arctic sea ice now covers its smallest area since records began.

2011

Canada becomes the first signatory to announce its withdrawal from the Kyoto Protocol.

2011

Two papers in the journal Nature present compelling evidence that an increase in the frequency of heavy rain and floods in the northern hemisphere is linked to the increase in greenhouse gases in the atmosphere.

2014

Speaking for the two largest greenhouse gas emitting nations, the President of the USA, Barack Obama, and his Chinese counterpart, Xi Jingping, pledge to reduce their nation’s carbon emissions.

2014

IPCC Fifth Assessment Report states that the effect of increases in greenhouse gas concentrations are ‘extremely likely’ to have been the dominant cause of observed warming since the mid-twentieth century.

2015

Strong El Niño impacts are recorded in Brazil: severe drought and water shortages. Carnival festivities are scaled back.

2015

At the 2015 United Nations Climate Change Conference, COP 21, the members agreed to reduce their carbon output ‘as soon as possible’ and to do their best to keep global warming ‘to well below 2 degrees C’. This is called ‘The Paris Agreement’.

2015

The global annual average surface temperature reaches a record level, 1 °C above the pre-industrial average (as represented by the 1850–1900 reference period).

2016

First indication that global carbon emissions may have peaked, published in the journal Nature Climate Change and discussed further in Earth System Science Data.

BOX 7.3

7.7 Summary

have an impact on the environment. The enhanced greenhouse effect appears to be causing accelerated climate change which

The overwhelming scientific consensus is that the production

will have major impacts on humans and global ecosystems.

of greenhouse gases by humans is the primary cause of recent

However, there are a number of positive and negative feedback

global warming, as outlined in various reports from the IPCC. The

effects that global circulation models are only just being able to

burning of fossil fuels and the creation of human-made chemicals

predict and model.

such as CFCs and techniques of creating nitrogen fertilizer all



193

Chapter 7 Contemporary climate change

➤ World population growth has been associated with increased

health and biodiversity. Cities often experience harmful smogs,

utilization of the land for agriculture. Increased domestication of

and the ozone layer which protects the Earth from vast amounts

plants and animals and use of wood for fuel have resulted in vast

of harmful ultraviolet radiation is suffering severe damage due

amounts of deforestation, particularly over the past 200 years

to CFC use.

when the human population has increased from 1 billion to

However, humans are an inventive species and technologi-

7.3 billion. Humans now appropriate 40–50% of the land’s bio-

cal improvements are continuously being made that may help us

logical production. Large-scale change in the land surface from

cope with and mitigate environmental change. The development

forest to farm influences regional and global climates in many

of safe nuclear fusion and increased use of solar cells may allow

ways that are still not properly understood. Conversion of forest

us to harness the world’s resources in a more sustainable manner.

to pasture involves release of CO2 to the atmosphere, as well as

In addition, the international recognition that global environ-

changes to albedo, air movements, and the water, carbon and

mental change is taking place has been achieved and there is

nitrogen cycles. Atmospheric pollution from industry, combus-

a willingness around the world to try to combat environmental

tion engines and agricultural practice is impacting on human

problems.

Further reading Hegerl, G.C., Crowley, M. Allen, W. et al. (2007) Detection of human influence on a new 1500-year temperature reconstruction. Journal of Climate, 20, 650–666. Pulls out human impacts on the recent temperature record. Houghton, J.T. (2015) Global warming: The complete briefing, 5th edition. Cambridge University Press, Cambridge. An authoritative and lucid account. If you want to read only one book about global warming then it should be this. Jackson, R.B., Canadell, J.P., Le Quéré, C. et al. (2016) Reaching peak emissions. Nature Climate Change, 6, 7–10. The first indications that global CO2 emissions have started to decline. See also the related in-depth paper by Le Quéré et al. in Earth System Science Data 7, 349–396.

194

Levermann, A., Clark, P.U., Marzeion, B. et al. (2013) The multimillennial sea-level commitment of global warming. Proceedings of the National Academy of Sciences 110, 13745–13750. This paper estimates much faster rises in sea levels than the IPCC (2014) report, especially as a result of ice melting in Antarctica. Lovelock, J.L. (1979) Gaia: A new look at life on Earth. Oxford University Press, Oxford. This is a very readable book, and a seminal work, proposing that the biosphere behaves as a homeostatic system. In a more recent book, The Revenge of Gaia, published in 2006, the author argues that it is now too late to avoid substantial global heating which will make large regions of the Earth inhospitable. In 2014, his new book, A Rough Ride to the Future, takes a rather different viewpoint. But read the 1979 book first.

C HAPTER 8

Global climate and weather John McClatchey Late of Environmental Research Institute, University of the Highlands and Islands

Learning objectives After reading this chapter you should be able to:

➤ understand how major features of the general atmospheric circulation play a role in climate

➤ describe the types of climates that exist across the Earth ➤ identify the type of weather associated with different types of climate

➤ recognize the importance of seasonality in different climates ➤ understand why there is a distinct geographical distribution of climates

➤ explain why there are differences between the climates of the northern and southern hemispheres

8.1 Introduction The climate at any place has a critical influence on the natural vegetation, the landscape and human activity. For the natural world, three components of climate are crucial, these being the precipitation, the thermal conditions and the wind conditions. For example, in desert climates the landscape is usually dominated by aeolian (wind-related) features, while in much of the middle

latitudes, fluvial (water-related) features dominate the landscape. The vegetation cover is also strongly controlled by the thermal conditions and by the availability of moisture. The climate at any one location arises from the types of weather that are experienced there over a period of years. However, the climate is not just about average values of weather elements such as temperature or precipitation but includes the range and extremes of those elements and the frequencies of types of weather. For example, in desert regions the average monthly precipitation total may be almost meaningless, as precipitation might occur only on one or two days a month. Of much more importance is the number of days on which precipitation falls and the amounts that fall in each event. Of course in all climates there can be substantial local and regional variations from the average conditions (see Chapter 9) but despite such local variations, the types of weather and the weather systems in each major climate regime will be similar. Therefore, in order to understand why a location has a particular climate, the starting point should always be to understand what types of weather are a feature of that climate. This chapter therefore starts by outlining the general controls of weather conditions and climate across the Earth, before discussing the features of the major global climate zones.

Chapter 8 Global climate and weather

8.2 General controls of global climates The Sun drives weather systems and the global climate. As a result, the latitude of any location plays an important role in determining the climate. The decline of the solar radiation input that takes place in moving from the equator to higher latitudes eventually leads to a negative net radiation balance at the top of the atmosphere (the balance between the incoming solar and outgoing terrestrial radiation). This occurs at close to 45°N and 45°S and would lead to a cooling at higher latitudes if energy were not transported there from lower latitudes by ocean currents and the general circulation of the atmosphere.

METEOROLOGICAL AND CLIMATOLOGICAL OBSERVING NETWORKS It is important to be aware of the availability and type of climatological records that can be used to assess climate. There are a number of different types of observing station but the three principal types are synoptic stations, climatological stations and precipitation stations. They are not spread uniformly across the planet. There are very few in oceanic locations or in less developed countries, deserts, polar regions and other harsh or inaccessible places. Such stations tend to be concentrated in developed countries.

Therefore, although important, solar radiation alone does not control the climate. The distribution of the oceans and continents, ocean currents and the general circulation of the atmosphere also play an important role (see Chapter 6). Hence despite having a similar solar radiation input, places at the same latitude may have mild winters in one case and very cold winters in another. For example, Scotland has much milder winters than Labrador at the same latitude in north-east Canada. Climate is often assessed by using a network of observation stations that record meteorological and climatological data. There are a number of types of such records that can be used to assess climate and these are discussed in Box 8.1.

and past (over past 3 hours) weather are also recorded but recent developments in equipment now allow visibility and some cloud information to be recorded at some automatic weather stations (Figure 8.1). Synoptic stations record weather elements at least every 6 hours at the same time at every station in the world (at 00:00, 06:00, 12:00 and 18:00Z, the

Z denoting Universal Time, equivalent to Greenwich Mean Time) and many stations make hourly observations. These synoptic stations are located to provide measurements representative of a wide area. Many are at airports or military aviation bases and therefore the local area surrounding the station is generally flat and open.

Synoptic stations Synoptic observing stations provide surface weather observations as an input into numerical weather forecasting models. The principal elements recorded are temperature, humidity, wind speed, wind direction and atmospheric pressure (and change over past 3 hours). These elements are recorded at all synoptic stations, including automatic stations. For stations at which meteorological observers are based, observations of cloud cover, cloud heights, visibility, precipitation, current weather

Figure 8.1 An automatic weather station.

BOX 8.1 ➤

196

8.2  General controls of global climates

➤ Solar radiation is recorded at only a few stations, mostly at what are termed ‘agrometeorological stations’ (used to provide information specifically for agriculture). At some of those stations both the direct and diffuse solar radiation are recorded and at a few stations there will also be an observation of the net radiation, which is the balance between the short- and long-wave radiation (see Chapter 6). Climatological stations Climatological observations are generally taken at least once per day, although at some sites it is twice per day. The time at which observations are taken at an ‘ordinary’ climatological station is decided by each country (e.g. 09:00 GMT in the United Kingdom). All climatological stations make an observation of the maximum and minimum temperatures and precipitation total over the previous 24 hours. The ‘average’ temperature is just the sum of the maximum and minimum temperatures divided by 2. Note that for synoptic stations, however, a more representative mean can be obtained from the hourly values. Some stations record soil temperatures (at various depths) and some record a night ‘grass’ (surface) minimum temperature. Some non-instrumental observations of weather phenomena are also recorded. There are far more climatological stations than synoptic stations (which also record climatological observations), as they are used to provide an indication of

the variation of climate within a region (Chapter 8). The number of stations required to show the local variations of climate within a region depends on the topography and the proximity to the sea or very large lakes (such as the North American Great Lakes). Precipitation stations In addition to climatological stations there are also a large number of precipitation stations at which only precipitation is recorded. In areas of frequent winter snowfall, special snow gauges are used to obtain a rainfall equivalent measure. World Meteorological Organization standards Both synoptic and climatological stations have to conform to certain standards set out by the World Meteorological Organization (WMO). These ensure that observations taken at any place in the world are directly comparable. For example, thermometers are sited in a screen at between 1 and 2 m above a grass surface (this may be a snow surface during winter). This screen has to be above the surface as at night there can be differences of over 10°C between the (colder) grass (surface) minimum and screen minimum temperatures as a result of surface radiational cooling. The height at which wind is recorded is also important as the friction of the surface slows the wind and causes the wind direction to change. The standard height for an anemometer (wind speed recorder)

and wind vane is 10 m (although a range between 8 and 12 m is acceptable). A ‘climate normal’ (average) is established over a 30 year period for a standard station updated every 10 years (e.g. 1971–2000, 1981–2010), which is the 30 year average of the individual monthly values. Often, however, meteorological and climatological observations are made as part of micrometeorological, ecological, hydrological and even geomorphological investigations. Such measurements may not conform to WMO standards but may be appropriate in terms of the particular investigation. However, non-standard observations cannot be directly compared with standard measurements and therefore they need to be used with care if used in any local climate assessment. With modern technology it is now becoming much easier to collect automated data from synoptic stations that are at WMO standards. In some countries there are networks for public enthusiasts to share data and for their data to be used by weather forecasters as part of synoptic station data collection. The UK Met Office for example, runs the Weather Observation Website, http://wow.metoffice.gov .uk/home, which allows users to upload their data from their stations and website visitors can click on each station and see that live data. These datasets from private individuals now supplement the Met Office’s own official synoptic and climatological networks.

BOX 8.1

A key factor determining global climate is the presence of zones of ascent or descent of air. In areas of the world subject to high-pressure anticyclonic conditions, air is subsiding (descending) and is being warmed by compression. That warming leads to the formation of a temperature inversion at a few hundred to perhaps 2000 m above the surface whereby warmer air overlies colder air. This prevents any further rise of the lower colder air because it is more dense and less buoyant than the overlying warm air. This inversion therefore inhibits the growth of clouds, and as clouds need to be deep to give substantial falls of

precipitation, areas in which anticyclones are located tend to be dry. Conversely, in areas of the world with low pressure, air is ascending (as a result of convergence in the lower atmosphere) and clouds can often grow to considerable depths. This convergence can enhance thermal convection or encourage slantwise convection (movement of air upwards and north or south) if there are noticeable thermal contrasts between regions of air (e.g. at fronts). A map of average surface pressure (Figure 8.2) therefore provides an initial indication of likely areas of ascent (relatively wetter areas) and descent (relatively drier areas). 197

Chapter 8 Global climate and weather

1205W

605W

05

605E

1205E

1805E

High

(a) July

1011

1014

1008

Low

605N

H

05

101

1

Low

05

14

1011 1023

101

10

10

4 101 1 1008 1005 1002 999

605S

205N

11

10

205S 405S

10

999 1002 1005 1008

405N

7

205N

High

14 10

1026 1023 1020 1017 1014

101

405N

605N 1011 1014 7 101 0 102 23 10

17

20

1020 1023

H

High

1017

1014

10

17

205S

H 405S 1014 1008 1011 1002 1005

605S 1205W

605W

05

605E

1205E

1805E

1205W

605W

05

605E

1205E

1805E

(b) January

High 1011

605N

405N

0 102 3 102 6 102 29 10 32 10

Low 996 1002 100181 10 4 101 10170 102

Low 2 05 100 10 008 1 11 10 14 10 17 10

1020 1020

High

High

High

1020 1017 1014

205N

605N 10 Low 0 10 5 101 08 1 1014

H

205N

1014 1011

05 205S

05 1008

1011

101

405S

7

Low 1008 1011

H

High

605S

405N

1017 1014

1020

1011 1008

1205W

205S

1 101 1020 H 1017 1014 1011 1008

1014

1005

8

100

405S 1005

605W

05

605E

1205E

605S 1805E

Figure 8.2 A map of average surface pressure over the Earth in (a) July and (b) January in hPa (or millibars). High-pressure areas represent zones of descending

air (air at the surface is diverging away from that point) while low-pressure areas are zones of ascending air (air at the surface is converging towards that point). (Source: after White et al., 1992, Environmental Systems: An Introductory Text, 2nd edition, by D.N. Mottershead, S.J. Harrison and I. White, Fig. 4.5, p. 84. Published by Taylor and Francis Books Ltd (Nelson Thornes) 1992. Reproduced by Permission of Taylor & Francis Books UK)

In addition, because the middle and high latitudes have an overall energy deficit (more radiation is lost to space at the top of the atmosphere than is gained from solar radiation) the weather systems help transport energy from lower to higher latitudes. Therefore, the average position of the weather systems can give an indication of regions for which air is generally being brought into from lower latitudes. These regions will be milder than might otherwise be 198

expected and the circulation of air around pressure systems means that these regions will be on the southern and eastern sides of low-pressure systems in the northern hemisphere (northern and eastern sides in the southern hemisphere). The type of weather experienced at a particular location depends on latitude, the position and movement of pressure systems and the presence or absence of areas of ascent or descent (subsidence) of air. The thermal climate

8.2  General controls of global climates

will depend on latitude (height of the Sun in the sky), the principal prevailing wind directions (from warmer or colder areas) and the cloudiness (thick clouds reduce daytime maximum temperatures but keep night-time minimum temperatures higher). Cloud and precipitation depend on the presence or absence of ascending air (giving cloud) or subsidence (giving clear skies). Therefore, although the following discussion will be separated into latitudinal zones, there will be substantially different climates experienced within each latitudinal zone. A number

of classifications of climate have been suggested, with probably the most commonly used being Köppen’s classification that was developed in the early twentieth century. The classification scheme was modified by Köppen’s students (Geiger and Pohl, 1953) and is shown in Table 8.1. It is, however, important to realize that the boundaries between different types of climate are not distinct and that climates merge into each other. Nevertheless a very general map of climatic zones is provided by Figure 8.3 based on Köppen’s classification.

Table 8.1 Köppen’s climate classification with additional modifications by Geiger and Pohl (1953) Letter code 1st

2nd

3rd

A

Basic description

Classification criteria†

Humid tropical f

tropical wet (rainforest)

Wet all year

w

tropical wet and dry (savanna)

Winter dry season

m

tropical monsoon

Short dry season

B

Dry

Potential evapotranspiration 7 precipitation

S

semi-arid (steppe)

Mean annual precipitation between 250 and 760 mm

W

arid (desert)

Mean annual precipitation 6 250 mm

h

hot

Mean annual temperature Ú 18°C

k

cold

Mean annual temperature 6 18°C

C

Moist with mild winters

Coolest month mean temperature between 18°C and -3°C

w

dry winters

Wettest summer month 10 times rain of driest winter month

s

dry summers

Driest month 6 40 mm and wettest winter month Ú 3 times driest month

f

wet all year

Criteria for w or s not met

a

summers long and hot

Warmest month 7 22°C with more than 4 months 7 10°C

b

summers long and cool

All months below 6 22°C with more than 4 months 7 10°C

c

summers short and cool

All months 7 22°C with 1 to 3 months 7 10°C

D

Moist with cold winters

Coldest month … -3°C warmest month 7 10°C

w

dry winters

Same as Cw

s

dry summers

Same as Cs

f

wet all year

Same as Cf

a

summers long and hot

Same as Cfa

b

summers long and cool

Same as Cfb

c

summers short and cool

Same as Cfc

d

as c with severe winters

Coldest month 6 -38°C

E

Polar climates

Warmest month 6 10°C

T

tundra

Warmest month 7 0°C but 6 10°C

F

ice cap

Warmest month … 0°C



Temperature limits given for warmest and coldest months are for average monthly temperatures.

(Source: Geiger and Pohl, 1953)

199

200

1605

1405

1205

Csa

Tropical wet-and-dry

Semi-arid or steppe

Highland

Polar ice cap

EF

H

Polar tundra

Dry winter

Subpolar

Humid continental

Dry winter

ET

Cw Dfa Dfb Dfc Dfd Dw

Cfb Marine Cfc Mediterranean Cc (dry summer)

Cfa Humid subtropical

BS

BW Arid desert

Aw

E

BWh

105

05

105

BSh

H

Bwh

Denver

ET

205

305

Equator

Aw

Aw

Aw

Af

Aw

Dfb

EF

505

1005

605

Tropic of Capricorn

405

Af

Mobile

Cfa

Des Moines

Aw

Dfa

Dfb

Winnipeg

Dfc

Bsk

Phoenix

BS H Csb H Bsk BWk BWk San Francisco

Port Hardy

Cfc Cfc

Af Tropical rainforest Am Tropical monsoon

Af

E

ET Fairbanks

805

Cfb

ET EF

Bsk Cfc

Cfa

Cw

Aw

Am

405

Tropic of Cancer

Cfc

ET

BSh

BWk Bsk

805

ET

Csb

Bsk

BWk

H

Aw

BW

Iquitos

Arica

At

BSh

ET

Af

205

Cw

Am

Af

205

ET

05

Cfb

Csa

BSh

Aw

Af

Cfa

Cfb

H

Cfc

205

BWh BWk BSk Csb

Aw

BWh

BSk BSh

BSk

Cfb

Cfc

BSh

Tropic of Capricorn

Timbo

H

EF

405

Cfb

Cwb

EF

Cfa

Cwa

605

BWh

BSh

H

H

Aw Af

Aw

BWh

BSh

BWh

Dfa

BSh

BSk

BSk

Dfb

ET

1005

BSh

Csb

BSh

EF

Aw

Cfb

1605

Dfc

1005

1805

605

505

Cfb

Cfa

Cwa

Dwa

Dwb

705

Aw

Cwa

Dwb

Am

Cfb

Cfa

Dfd

BSk

Dwb

Aw

Af

Dwc

ET ET

ET

Af Aw BSh Cw

H

1405

Csb

Am

Aw

H

Af

Cwa

BSh

Am

Dwd

BWk

BSh Aw

BSh

BWh

1205

Equator

60 5

BSk

BSk BWh BSh

BWk

Dfc

Dfd

805

405

305

Aw

Af

Cfa

Am

205

1205

Aw

105

05

105

205

305

405

505

605

Figure 8.3 A map of the Köppen–Geiger–Pohl climate classification. (Source: from Meteorology today: Introduction to weather, climate and the environment, 6th edition by AHRENS, 2000. Reprinted with permission of Brooks/Cole, a division of Thomson Learning: www.thomsonrights.com, Fax 800-730-2215)

H Highland climates

E Polar climates

D Moist climates with severe winters

C Moist climates with mild winters

B Dry climate

A Moist tropical climate

205 1805

305

405

505

605

705

Barrow

Chapter 8 Global climate and weather

8.3  The tropics and subtropics

905N

Reflective questions

605N

➤ What are the main factors controlling the location of climate zones?

➤ What would happen to our climate zones if motions in

305N July 05

January

the atmosphere and oceans did not redistribute energy around the planet?

➤ Which Köppen climate zone do you live in?

305S 605S 905S 1805W

8.3 The tropics and subtropics 8.3.1 Equatorial regions Horizontal air movement occurs if there is a pressure gradient force caused by a pressure difference between different places. Over much of the Earth, once air moves as a result of a pressure gradient force, the Coriolis effect becomes apparent (see Chapter 6) and the winds deviate to the right in the northern hemisphere and to the left in the southern hemisphere. Close to the equator, however, the Coriolis effect is negligible and therefore air follows the pressure gradient from high to low pressure. This is important as it means that near the equator, the weather is not dominated by the movement of large circulatory weather systems. However, outflow from the large subtropical anticyclones (the north-east and south-east trade winds) does play an important role leading to convergence of air into a region of somewhat lower pressure sometimes termed the equatorial trough. This trough can be seen as a distinct intertropical convergence zone (ITCZ), particularly over the oceans, and is part of the Hadley cell circulation (see Chapter 6 for further details on the formation of the ITCZ). The ITCZ is not located along the equator but moves quite well north of the equator in some places in the northern hemisphere summer and a little south of the equator in the southern hemisphere summer, with an average position somewhat north of the equator (Figure 8.4). This movement is in response to seasonal differences in pressure gradients caused by changes in solar energy received during the year. The movement of the ITCZ north and south is greater over land because the seasonal cooling and warming are more pronounced over land (the oceans warm and cool at a much slower rate than land). The ITCZ is typically from 4° to 8°N (in March and September respectively) in the Pacific. There are, however, extensions of the ITCZ further south of the equator (Linacre and Geerts, 1997),

1205W

605W

05

605E

1205E

1805E

Figure 8.4 Average position of the intertropical convergence zone in

January and July. North-east trade winds and south-east trade winds converge at the ITCZ. Here solar heating and convergence of air results in ascension and instability. Thus precipitation rates are high. (Source: Critchfield, H.J., General climatology, 3rd edition, 3rd © 1974. Adapted (or electronically reproduced in case of e-use) by permission of Pearson Education, Inc., Upper Saddle River, NJ)

particularly in the southern hemisphere summer. These are into southern Africa, east of the Andes in South America and a larger and more consistent extension called the South Pacific convergence zone, roughly south-south-west from the north-east of Australia (Figure 8.4). Low-level convergence of air such as at the ITCZ leads to rising air and the formation of clouds (Figure 8.5).

Figure 8.5 A satellite image showing a distinct narrow region of cloud aligned across the Pacific close to the equator, representing the location of the ITCZ. (Source: NASA-Goddard Space Flight Center, data from NOAA GOES)

201

Chapter 8 Global climate and weather

Therefore, the weather close to the equator is dominated by frequent convectional clouds and rain. This convectional activity may be relatively random with many individual cumulus clouds being formed but there are also more organized cloud clusters. For example, in some tropical coastal areas, more organized features sometimes occur diurnally as part of a land and sea breeze circulation (see Chapter 9). Equatorial climates are therefore dominated by the movement of the ITCZ with many places close to the equator having no distinct dry season. However, some equatorial areas show a clear peak in the monthly precipitation totals at each of the equinoxes with lower totals when the ITCZ is at its northern and southern limits (summer in the northern and southern hemisphere respectively). Total annual precipitation is high and in some parts of the Amazon and West Africa totals can be from 2500 to over 4000 mm. As the Sun is always high in the sky, average temperatures in the equatorial regions are fairly constant throughout the year, although maximum temperatures may be a little lower during the wetter periods. Humid

35

conditions keep night temperatures high, giving a relatively small diurnal range with average annual temperatures around 27 °C. Typical monthly temperature and precipitation values for equatorial stations are given in Figure 8.6. Equatorial climates are not exactly alike in every place, particularly over the continents of Africa and South America, as topography and proximity to the sea do play a role in altering thermal and precipitation regimes (see Chapter 9). Moving north and south away from the equator into the trade wind belt, between about 5° and 20° latitude, the ITCZ still plays a key role in the climate but as the ITCZ is only close to these areas in the summer (in that hemisphere), there is a distinct rainy (summer) and dry (winter) season. This greater seasonality gives a greater range of temperatures through the year, but still much less than in many middle and high latitudes, with the highest temperatures tending to be observed just before the rainy season. There can be quite large average diurnal ranges of temperature in places (15°C or more), particularly in the dry season. The trade winds also mean

(a)

35

25 Temperature (5C)

30

Temperature (5C)

30 25 20 15 10 5

20 15 10 5

0

0

-5

-5

-10

J

M

A

M

J J Month

A

S

O

N

-10

D

(b)

450

350

350

300

300

Precipitation (mm)

Precipitation (mm)

400

F

(c)

250 200 150

50 M

A

M

J J Month

A

S

O

N

D

A

M

J

F

M

A

M

J J Month

A

S

O

N

D

(d)

150

50 F

M

200

100

J

F

250

100

0

J

0

J J Month

A

S

O

N

D

Figure 8.6 Example temperature and precipitation graphs for equatorial stations. Mean monthly temperatures (°C) and mean monthly rainfall (mm) for (a) and (b) Belem, Brazil (1.5°S), and (c) and (d) Tamavate, Madagascar (18°S).

202

8.3  The tropics and subtropics

High 1022 mb 1020 mb 1018 mb 1016 mb 1014 mb 1012 mb 0

500 km

1010 mb

Low

Figure 8.7 An easterly wave in the subtropical easterly flow. Easterly waves occur within the trade wind belt and represent waves of air (moving north and south) for air travelling in an easterly direction. (Source: after Critchfield, H.J. General climatology, 3rd edition, 3rd © 1974. Adapted (or electronically reproduced in case of e-use) by permission of Pearson Education, Inc., Upper Saddle River, NJ)

that there is a fairly steady but not particularly severe wind regime.

8.3.1.1 The easterly wave and tropical cyclones One important type of more organized cloud is related to easterly waves in the trade winds. Such waves represent moving troughs of slightly lower pressure denoting zones of low-level convergence. This convergence is concentrated along the trough line resulting in a zone of heavier rain. These wave features (Figure 8.7) are important as they can develop into tropical depressions and tropical storms (Malkus, 1958) and some of those become tropical cyclones. These cyclones are called hurricanes in the Atlantic and typhoons in the west Pacific. The WMO has produced a classification system that allows us to define when tropical depressions become severe enough to be called tropical storms or tropical cyclones (Table 8.2).

Tropical cyclones require high sea surface temperatures (at least 27°C) and do not form over land as the energy of the system comes from the release of latent heat when the clouds are formed. The moisture evaporating from the sea is therefore a key factor in sustaining tropical cyclones and when that source of moisture is cut off (when hurricanes move inland) the intensity declines. Although tropical cyclones occasionally move inland, the majority are confined to the oceans. Even across the oceans the tracks of tropical cyclone systems show a distinct geographical distribution (Figure 8.8). Tropical cyclones do not occur at the equator as there needs to be a sufficient Coriolis effect to help develop the circulation of the initial depression. As well as requiring high sea surface temperatures, the mid-troposphere has to be moist in order to allow tropical cyclone development. There must also be no strong temperature inversion inhibiting mixing of the warm moist surface air with the mid-troposphere air. This is why tropical cyclones are extremely rare in the South Atlantic. In the South Atlantic there is a noticeable extension of the trade wind inversion from south-west Africa towards Brazil and hence there is insufficient opportunity for the build-up of a deeper moist layer. A strong temperature inversion also extends out into the Pacific from northern Chile and Peru and the sea surface temperatures are also relatively cool. This again inhibits the development of tropical cyclones. However, the Pacific is much wider than the Atlantic and therefore a deeper moist layer has time to build up and as a result tropical cyclones are found in the central and west Pacific south of the equator (Brasher and Zheng, 1995). Figure 8.8 shows the tracks of tropical cyclones in both the northern and southern hemispheres. It is also of note that there are fewer tropical cyclones in the southern hemisphere as the summers are somewhat cooler than those in the northern hemisphere and therefore the area of the ocean with high sea surface temperatures is

Table 8.2 World Meteorological Organization classification of tropical storms Wind speed m s -1

knots

Beaufort scale

6 17.2

6 34

7

Moderate tropical storm

17.2–24.4

41–47

8–9

Severe tropical storm

24.5–32.6

48–63

10–11

7 32.6

7 63

12

Tropical depression

Hurricane (tropical cyclone)

203

Chapter 8 Global climate and weather

605N

305N

Typhoon

05

Bangladesh

Hurricane

Cyclone

Hurricane

Cyclone

305S

605S

905E

1805W

905W

05

905E

Figure 8.8 Typical tracks of tropical cyclones, hurricanes and typhoons. Note that hurricanes and typhoons are also tropical cyclones but that the names depend on their location (called hurricanes in the Atlantic, typhoons in the west Pacific). (Source: from Meteorology today: Introduction to weather, climate and the environment, 6th edition by AHRENS, 2000. Reprinted with permission of Brooks/Cole, a division of Thomson Learning: www.thomsonrights.com, Fax 800-730-2215)

Amazon Basin and in North Atlantic sea surface temperatures. The current phase of the cycle has involved lower wind shear (change of wind speed and/or direction with height) and warmer sea surface temperatures across the tropical Atlantic. In addition, there have been weak low-level easterly (trade) winds and a westward expansion of upper-level easterly winds from Africa, the African Easterly Jet. These conditions are favourable to

smaller and confined to north of 20°S. In the northern hemisphere, high sea surface temperatures of 27°C and above extend to 30°N. Tropical cyclone activity in the Atlantic has been above ‘normal’ since 1995 (Figure 8.9). This has been largely in response to the active phase of the ‘multi-decadal signal’. The multi-decadal signal relates to cycles of about 20–40 years in monsoon rains over West Africa and the 30 Other named storms Other hurricanes (category 1, 2) Major hurricanes (category 3, 4 5)

Number of storms

25 20 15 10 5 0 1855

1875

1895

1915

1935 Year

1955

1975

1995

2015

Figure 8.9 Total recorded Atlantic tropical storms in each year since 1851. Note that prior to 1966 some storms may have been missed as many tropical storms may not have hit land and gone unobserved. However, from 1966 onwards satellite data has provided a more reliable record.

204

8.3  The tropics and subtropics

the development of hurricanes. This phase is expected to continue for the next few years. Figure 8.9 shows a peak in Atlantic tropical cyclones in 2005. Of the 28 named storms in 2005, 15 were hurricanes and 7 of these were major hurricanes. Hurricanes are classified on the Saffir–Simpson scale between 1 and 5 with hurricanes that are category 3 and higher being classed as major. A category 5 hurricane called Katrina made landfall during August 2005 with devastating consequences (Box 8.2). The strongest winds recorded in a tropical storm (215 mph, 96 m s-1) occurred in October 2015 during Hurricane Patricia which struck Mexico (Figure 8.10). Fortunately it weakened just before landfall and hit a sparsely populated region, but it still resulted in a track of devastating damage and two deaths.

8.3.1.2 Monsoons One other important feature of the climate in the northeast and south-east trade wind areas is that there are regions that experience an exceptionally wet rainy season with a very distinct dry season (Figure 8.13). These are the regions that experience monsoon conditions (monsoon is from the Arabic, meaning season). The largest and most intense monsoon is in Asia but there are monsoon-type climates in West and East Africa, and in Australia (Webster, 1981). There is a much less well-defined monsoonal

HURRICANE KATRINA Hurricane Katrina was the most costly natural disaster in US history and also caused the highest number of deaths from a single hurricane since 1928. Tropical storm Katrina was designated on 24 August 2005 at which time it was located in the central Bahamas. The track of the storm is shown in Figure 8.11. Katrina began strengthening rapidly and became a category 1 hurricane 24 km east-northeast of Fort Lauderdale at 17:00 EDT (21:00 UTC) on 25 August. At 18:30 EDT (22:30 UTC), the hurricane made landfall

Figure 8.10 Satellite image of hurricane Patricia approaching the Pacific

coast of Mexico on 23 October 2015. The classic ‘eye’ of the hurricane can easily be seen. (Source: NASA, Jeff Schmaltz, LANCE/EOSDIS Rapid Response).

between Hallandale and North Miami Beaches with sustained winds estimated at 36 m s - 1 gusting to over 40 m s - 1. Katrina moved south-west across the tip of the Florida Peninsula during the night but the landfall did little to reduce the intensity as the storm re-intensified as it moved back to sea over the warm waters of the Gulf. The sustained winds over Florida were never higher than 36 m s - 1 but the heavy rain and gusty winds caused substantial damage and flooding, and 14 people lost their lives. Katrina moved west after entering the Gulf of Mexico and then over the next few days

gradually turned to the north-west and then north (Figure 8.11). The high sea surface temperatures and an upper-level anticyclone over the Gulf encouraged the rapid intensification, which led to Katrina attaining ‘major hurricane’ (category 3) status on the afternoon of 26 August. By 07:00 CDT (12:00 UTC) on 28 August, hurricane Katrina reached category 5 status with wind speeds of 72 m s - 1 or more. At 16:00 CDT (21:00 UTC), Katrina’s minimum central pressure dropped to 902 hPa, one of the lowest pressures ever recorded. At this time Katrina was at its peak strength with hurricane force winds

BOX 8.2 ➤

205

Chapter 8 Global climate and weather

➤ Tropical depression Tropical storm Category 1 Category 2 Category 3 Category 4 Category 5

30/08/05 : 0900

29/08/05 : 1500 29/08/05 : 0900 28/08/05 : 2100 25/08/05 : 1500 24/08/05 : 1500 27/08/05 : 2100

26/08/05 : 1530

23/08/05 : 2100

Figure 8.11 The storm track of hurricane Katrina.

extending outwards up to 168 km from its centre and tropical storm force winds (up to 33 m s - 1) extending outwards up to nearly 370 km. Sustained tropical storm force winds were already battering the south-east Louisiana coast and the 16:00 CDT (21:00 UTC) Bulletin from the National Hurricane Center warned of coastal storm surge flooding of 5.5 to 6.7 m above normal tide levels, locally as high as 8.5 m, and stated ‘some levees in the Greater New Orleans area could be overtopped’. At 06:00 CDT (11:00 UTC) on 29 August Katrina made landfall in Plaquemines Parish just south of Buras (between Grand Isle and the mouth of the Mississippi River) and by 10:00 CDT (15:00 UTC), the eye of Katrina was making its second northern Gulf coast landfall near the Louisiana– Mississippi border. Katrina affected over 15 million people and caused enormous damage to homes and businesses in both Louisiana and

Mississippi estimated at around US$125 billion. The loss of human life was even more catastrophic with an official death toll of 1836 with the majority of the deaths in Louisiana (1577) and Mississippi (238). This made Katrina the third deadliest hurricane since 1900, after the Galveston hurricane of 1900 (at least 8000 deaths) and the Lake Okeechobee Hurricane of 1928 (over 2500 deaths). While the wind damage caused by Katrina was significant, the bulk of the devastation was caused by flooding, largely due to the very substantial storm surge which peaked at 8.5 m at Pass Chritian, Mississippi. A surge of 7.3–8.5 m was estimated along the western Mississippi coast across a path of about 32 km. The surge was 5.2–6.7 m along the eastern Mississippi coast, 3.0–5.8 m along the Louisiana coast and 3.0–4.6 m along the Alabama coast. A number of factors contributed to the extreme storm surge:

t the massive size of the storm; t the strength of the system (category 5) just prior to landfall;

t the 920 hPa central pressure at landfall; and

t the shallow offshore waters. In the delta country south-east of New Orleans, a number of towns were completely flooded, with Plaquemines and St. Bernard Parishes particularly badly affected. The levee system protecting New Orleans was put under severe pressure due to the rise in the level of Lake Pontchartrain caused by the surge. Damage and high-water marks indicated that the surge reached up to 19 km inland in some areas, especially along bays and rivers, and in New Orleans there were significant failures in the levee system on 30 August on the 17th Street Canal, Industrial Canal and London Avenue Canal levees (Graumann et al., 2005). As much of

BOX 8.2 ➤

206

8.3  The tropics and subtropics

➤ New Orleans lies below sea level, the failure of the levees led to drainage of water into the city, leading to 80% of the city being underwater to depths of over 6 m. While much of the flood waters had been cleared by 20 September, the storm surge from hurricane Rita on 23 September caused a new breach in the repaired Industrial Canal levee and many of the areas of the city were flooded again. Much of New Orleans is below sea level and the Mississippi Delta will always be liable to flooding whether due to river floods, extreme rainfall events or storm surges. While levees can be built up and strengthened, deltas are notoriously unstable and there is still the danger that levees may be overtopped or undermined. The loss of life and the cost of damage caused by Katrina provide a warning about possible future events and the need for disaster systems to be designed to help vulnerable people – more than half of the victims of Katrina were senior citizens. Tropical cyclones give very high rainfall totals and intensities. The rainfall

Figure 8.12 Heavily damaged homes from the Katrina storm surge flood-

ing in the ninth ward of New Orleans. (Source: Brian Nolan)

falling over two to three days from a single hurricane can even be close to the average annual total in some places. The relatively low frequency of tropical storms, and the even lower frequency of tropical cyclones at any individual location, mean that the standard climate statistics do not provide any real indication

of the importance of these events in the climate of these areas. They are highmagnitude, low-frequency events. In any one year, two locations that have similar mean rainfall totals may well have widely differing totals if one of the locations had experienced a severe tropical storm or tropical cyclone.

BOX 8.2

climate in South America, partly as a result of the relatively cool sea surface temperatures to the west of South America that increase stability in the lower layers of the atmosphere. The Andes also play a role as they obstruct the trade winds and air descent in the lee of the mountains increases atmospheric stability. Monsoon regions are all subject to the switching of the wind direction as the ITCZ moves north and south. During the Asian winter, the winds are generally north-easterly (although they may be north-westerly over western India). At high levels in the troposphere (above 11 000 m) there is a distinct westerly jet stream (see Chapter 6) located to the south of Tibet. This jet stream is the southern and stronger branch of the subtropical westerly jet stream, the northern branch being located to the north of Tibet. In spring, this southerly branch of the jet stream weakens but remains south of Tibet while the northern branch strengthens and becomes extended. At the same time, the north of India is warming, with temperatures reaching a maximum in May. This warming creates a ‘heat low’ beginning a process that

encourages the inflow of warm moist air from the south (Robinson and Henderson-Sellers, 1999). In summer, the upper-level westerly jet stream to the south of Tibet breaks down and then moves north across Tibet. As it moves across the mountains and the Tibetan Plateau, the high ground blocks the flow and the lower portion of the jet is deflected and re-established to the north. The strong convection over India (enhanced by the heat low) creates an outflow of air aloft and the southerly outflow develops into an easterly jet (under the influence of the Coriolis force). This upper air flow reversal from a westerly to an easterly jet is associated with the onset of the monsoon season in India and South-East Asia (Figure 8.14). As noted earlier, the south-east trades are located to the south of the ITCZ. In some parts of the world, these winds remain basically south-easterly as the ITCZ moves north during the northern hemisphere summer. Over India and Asia where the ITCZ moves much further from the equator than in most regions, the south-east trades move far enough north to become 207

Chapter 8 Global climate and weather

(a)

35

30

30

25

25 Temperature (5C)

Temperature (5C)

35

20 15 10 5 0

20 15 10 5 0 -5 -10

J

F

M

A

M

J J Month

A

S

O

N

D

J

F

M

A

M

J J Month

A

S

O

N

D

J

F

M

A

M

J J Month

A

S

O

N

D

(d)

(b)

Precipitation (mm)

Precipitation (mm)

-5 -10

650 600 550 500 450 400 350 300 250 200 150 100 50 0

(c)

J

F

M

A

M

J J Month

A

S

O

N

D

650 600 550 500 450 400 350 300 250 200 150 100 50 0

Figure 8.13 Example temperature and precipitation graphs for monsoon climate stations. Mean monthly temperatures (°C) and mean monthly rainfall

(mm) for (a) and (b) Bombay, India (19°N), and (c) and (d) Manaus, Brazil (3°S).

affected by the Coriolis force. As a result, they are deflected to the right and become south-westerly. The warm moist winds of the monsoon are therefore southwesterly. The return of the upper westerly jet takes place in October but the cessation of the monsoon rains is less distinct than their start. October and November also have the most frequent occurrences of tropical cyclones in the Bay of Bengal, and the rains from those cyclones give rise to a rainfall maximum at this time of year in south-east India. Monsoon rains are not continuous throughout the monsoon summer period, as there are breaks between more active phases. In some places there is orographic enhancement of rainfall (forcing air to rise leading to further condensation, see Chapter 9) and the alignment of hills can also lead to increased low-level convergence (creating zones of ascending air). In such places the rainfall can be exceptionally high, as for example in Assam (northeast India) where a number of places have exceptionally high rainfall totals (annual totals in excess of 10 000 mm). Cherrapunji, at 1340 m above sea level, is the most famous

208

of the wet places. These exceptionally high totals compare with more typical values of between 1500 and 2000 mm close to the Bay of Bengal coast (∼300 km to the south of Cherrapunji). Despite such examples of orographic and convergenceinduced enhancement of rainfall, the dominating influence on rainfall in monsoon regions is the large-scale circulation system creating the monsoon itself. The Asian monsoon therefore comes about as a complex interaction of the formation of the heat low, the changes in the upper air flow patterns, the movement of the ITCZ and the topographical barrier of the Himalayas and the Tibetan Plateau. It is also important to note that there is great variability from year to year. For example, El Niño years can be associated with the failure of the monsoon rains, resulting in food shortages (Kumar et al., 2006). Northern Australia has some monsoon rains but there is no topographical barrier to the south, and the land mass is not as large as in Asia and so the ‘heat low’ is not as intense. There is a less intense monsoon circulation over West Africa but the lack of a major mountain

8.3  The tropics and subtropics

Subtropical jet

(a) Winter

Air descends beneath jet

Cold surface H

INDIAN SEA

TIBETAN PLATEAU

Winter rains HIMALAYAS

Subtropical jet

(b) Summer

Equatorial jet Jet draws up air from surface

Major surface flow from south-west

ARABIAN SEA

Heated surface

L

Summer rains

TIBETAN PLATEAU

BAY OF BENGAL HIMALAYAS

Figure 8.14 Upper air flow reversal and the onset of the Asian monsoon. (Source: after Robinson and Henderson-

Sellers, 1999)

barrier allows a more steady movement of the ITCZ. However, as with Asia, the northward movement is sufficient to allow the Coriolis force to deflect the winds round to the south-west, bringing warm moist air in from the Atlantic Ocean. Researchers are aiming to develop better predictions of the monsoon so that farmers can grow the right crops for the year ahead (e.g. see Fitzpatrick et al., 2016). A major long-term collaborative research project called AMMA 2050 (African Monsoon Multidisciplinary Analysis 2050) is currently operating and aims improve understanding of how the West African monsoon will be affected by climate change in the coming decades – and help West African societies prepare and adapt. Further detail can be found at www .amma2050.org. In recent decades, West Africa has experienced some of the most extreme rainfall variability anywhere in the world. This is extremely problematic for its growing population and we are uncertain as to how climate change may influence the seasonal monsoon,

which makes adaptive planning very difficult for the region. Hence significant research is required.

8.3.2 The Sahel and desert margins The influence of the poleward movement of the ITCZ declines further away from the equator. The effects of this are seen most clearly in the Sahel region of Africa that lies on the southern side of the Sahara Desert (Figure 8.15). Most of the year the region is under the influence of the north-east trade winds blowing out of the subtropical anticyclone to the north. Temperatures are high during the day (possibly higher than 40°C in early summer) and there is a substantial diurnal range of temperature (10–15°C). The Sahel region still has a rainy season (when the ITCZ is at its furthest north), but the amounts are generally modest (300–600 mm) and in some years the rains fail to come (Nicholson et al., 1996). During normal years, the rainfall is usually sufficient to sustain the vegetation and

209

Chapter 8 Global climate and weather

0

2000 km

Tunisia Morocco

S A H A R A

D E S E R T

Algeria

N

Libya

Western Sahara

Egypt

S

Senegal Gambia Guinea Guinea Bissau Sierra Leone Liberia

Niger

Mali

Mauritania

A

H

E

L

Ivory Coast

Eritrea Chad

Burkina Faso

Sudan

Djibouti Somalia

Nigeria

Togo Benin

Ethiopia

Central African Republic

Ghana Cameroon

Democratic Republic Uganda of Congo

Kenya

Figure 8.15 The Sahel region of Africa.

any grazing by animals (Figure 8.16). In dry years, however, there is an encroachment of the desert from the north (although that encroachment may be partly due to overexploitation by people – see Chapter 21). There are several regions of the world with a similar climate to that of the Sahel, but they are not as extensive and the problems of drought are less severe. For example, in northern Argentina and south-east New South Wales in Australia the water supply is better because rivers flow into them from more humid areas.

Figure 8.16 Sahal landscape in Mali with sparse vegetation around a tra-

ditional tribal village. (Source: Scott S. Brown) 210

8.3.3 Subtropical deserts To understand the geographical pattern of these deserts, it is necessary to examine the positioning and intensity of the subtropical anticyclones that form the descending part of the Hadley cells north and south of the equator. These anticyclones are large features extending across the North and South Atlantic and Pacific Oceans centred at about 30°N and S. The intensity of the temperature inversions is greater on the eastern side of these anticyclones and the inversions are at lower altitudes: 300–500 m, as opposed to 1500–2000 m on the western extension of the anticyclones. In part, this is due to stronger subsidence in the east but it is also due to the circulation of the ocean currents, with cool currents being present under the eastern end of these anticyclones. The lower sea surface temperatures in these currents help to increase atmospheric stability and reduce convection. The driest hot deserts are therefore found in the western coastal regions of the continents where the subtropical anticyclones are most intense (see Figures 8.3 and 21.7). In the southern hemisphere they are found in Namibia in south-west Africa (Namib Desert), in western Australia (Great Sandy and Gibson Deserts) and in northern Chile (the Atacama Desert). In the northern hemisphere they are found in southern California (Sonoran and Mojave Deserts), in Africa (Sahara Desert), in Arabia (Arabian Desert) and in north-west India and southern Pakistan

8.3  The tropics and subtropics

(Great Indian Desert). The extent of the hot desert region from the western Sahara, through to the Arabian Desert east of the Red Sea and then again in southern Iran, Pakistan and north-west India, is not mirrored elsewhere in the world, except in Australia. This very large extent of the hot desert region in Africa, the Middle East and southern Pakistan arises as the large land mass allows a much greater eastward elongation of the subtropical anticyclone from the east Atlantic. This helps damp down convection over these regions. In addition, unlike the western side of the Pacific and Atlantic where warm ocean currents bring warm moist air to the east of the continents, warm currents do not move polewards in the Indian Ocean but move along the equator and then cross the equator off the east coast of Africa. The Australian deserts have greater annual precipitation totals than some of the other deserts at similar latitudes; the driest regions have annual totals close to 90 mm, while stations in the Sahara may have less than 15 mm. Part of the reason for greater precipitation is that the Australian anticyclone is not a constant feature but an average of individual anticyclones moving eastwards across the continent, allowing occasional inflow of moister air from the oceans to the north and south of the continent. In North and South America, there is only a relatively small area of desert due to major mountain barriers to the east: the southern Rockies and Mexican mountain ranges in the north and the Andes in the south. However, as the deserts in North and South America are situated to the lee of the mountains, the descent of the trade winds as they cross the mountains further dries the air, thereby intensifying the aridity. It is also important to note that variations in weather patterns over time can lead to extension or contraction of desert regions (Tucker et al., 1991). The main features of the weather in most desert regions are the wind and the high daytime temperatures (typically over 45°C in the summer in Libya). The wind increases aridity and causes considerable aeolian erosion (by sand and other particles carried in the wind) in some places (see Chapters 16 and 21). The dry air and clear skies of the anticyclones give large diurnal ranges of temperature (as much as 20°C in some places). During winter, night temperatures can even drop below freezing in parts of these deserts. In South America, the northern parts of the desert are narrow and rainfall increases rapidly inland. This is due to the sea breeze circulation that can trigger thunderstorms if it reaches the edge of the Andes where forced ascent of the moist air can penetrate the temperature inversion and trigger the potential instability aloft. South of 10°S the desert widens and even the western Andes are dry. The aridity is

most noticeable in the Atacama Desert in northern Chile. While it does rain on a few days per year in most deserts (even in the Sahara), rainfall is very rare in the Atacama Desert, which has the lowest annual precipitation totals of any place in the world (with the possible exception of the central Antarctic). Close to the coast, typical maximum temperatures in summer may be 25°C with diurnal ranges of 5–10°C and although temperatures do increase somewhat inland they are never extreme owing to the moderating influence of the ocean. The desert areas in southern California are similar, but inland from the Californian coast, in the lee of both the coastal ranges and the Sierra Nevada, temperatures in Death Valley do reach the very high values found in Libya. Examples of temperature and precipitation data from desert climates are given in Figure 8.17, which provides monthly mean temperature and precipitation totals for Khartoum (Sudan) and Baghdad (Iraq).

8.3.4 Humid subtropics On the western side of the subtropical anticyclones there is a deep moist layer and convective activity is stronger than on the eastern side giving a higher likelihood of the development of rain clouds. A particular feature of this type of climate is the hot (often over 32°C), very humid summers associated with the tropical maritime air. These uncomfortable conditions can occasionally be interrupted if cooler air moves in from higher latitudes as part of the return flow of the Hadley or Ferrel circulation cells (see Chapter 6). However, if a ridge in the upper westerlies becomes established, very hot and humid conditions can last for weeks. The winters are generally mild, although there can be outbreaks of cold polar air into these regions. These unusually cold periods can cause major damage to sensitive crops such as citrus fruits and coffee beans (e.g. in Florida or southeast Brazil). Further into the continental interiors winters become more severe and at the same latitude winters in China are colder than winters in the United States. Annual precipitation totals are typically between 1100 and 1700 mm in this climate regime. Examples of areas with this type of climate include south-east Australia (Figure 8.18a and b), Taiwan and south-east Brazil, Paraguay, Uruguay and northeast Argentina in South America and the south-east states in the United States (e.g. Florida, Georgia, Alabama and Louisiana; Figure 8.18c and d), North and South Carolina. There is also a small zone of this type of climate in the east of southern Africa. The climate of eastern China is similar except that in central and southern China there is a winter minimum of precipitation (increasingly evident towards the north) rather than the more evenly spread rainfall in, 211

35

35

(a)

30

30 Temperature (5C)

25

25 Temperature (5C)

(a)

20 15 10 5

20 15 10 5 0

0

-5

-5

-10

-10

J

F

M

A

M

J J Month

A

S

O

N

J

F

M

A

M

J

F

M

A

M

J

F

M

A

M

J

F

M

A

M

D

J J Month

A

S

O

N

A

S

O

N

D

(b) 225

(b)

200

200

175 Precipitation (mm)

225

Precipitation (mm)

175 150 125 100

125 100 75

75

50

50

25

25

0

0

35

J

F

M

A

M

J J Month

A

S

O

N

D 35

(c)

J J Month

D

(c)

30

30 Temperature (5C)

25

25 Temperature (5C)

150

20 15 10

20 15 10 5

5

0

0

–5

–5

–10

–10

J

F

M

A

M

J J Month

A

S

O

N

D

J J Month

A

S

O

N

D

(d) 225

(d)

200

200

175 Precipitation (mm)

225

Precipitation (mm)

175 150 125 100

125 100 75

75

50

50

25

25

0

0

J

F

M

A

M

J J Month

A

S

O

N

D

Figure 8.17 Example temperature and precipitation graphs for desert sta-

tions. Mean monthly temperatures (°C) and mean monthly rainfall (mm) for (a) and (b) Khartoum, Sudan (15.5°N), and (c) and (d) Baghdad, Iraq (33.5°N). 212

150

J J Month

A

S

O

N

D

Figure 8.18 Example temperature and precipitation graphs for humid

subtropical climate stations. Mean monthly temperatures (°C) and mean monthly rainfall (mm) for (a) and (b) Sydney, Australia (34°S), and (c) and (d) New Orleans, USA (30°N).

8.3  The tropics and subtropics

for example, Georgia and Alabama. This winter minimum is a result of cool dry winds circulating around the winter Siberia high-pressure region. Areas with a winter precipitation minimum can also be found in Africa (e.g. east Zimbabwe) and South America (south-east Brazil and northern Paraguay). Humid subtropical climates do not generally suffer from severe winds but the more coastal areas (in both the United States and China) can be hit by tropical cyclones (hurricanes) as they turn northwards and westwards. In

Texas, Oklahoma and Kansas, tornadoes are another localized feature of the climate which is important. Here warm moist air from the Gulf of Mexico moves northwards inland and initially becomes trapped under a temperature inversion in the westerly winds aloft. If the temperature inversion is penetrated, substantial instability is released, leading to the growth of very large storm clouds, some of which will have associated tornadoes (Box 8.3). Although rare at any one place, these tornadoes are an important feature of the climate of this part of the United States.

TORNADOES Tornadoes are violently rotating columns of air extending to the ground (Figure 8.19). They are capable of causing great damage with wind speeds of over 300 km h - 1 (Figure 8.20). A tornado can be either very narrow and only a few metres across or very large with some over 500 m wide. They can often travel long distances, with some causing damage over 75 km. Tornadoes are often thought of as a phenomenon of the mid-west United States. However, they occur all over the world (e.g. Holden and Wright, 2004) and even in the United Kingdom the Tornado and Storm Research Organisation (TORRO) reports an average of 33 tornadoes per year (mainly in southern Britain, rarely in Northern Ireland or Scotland). Nevertheless, the largest ones tend to be concentrated in the Plains of the United States where atmospheric conditions often occur that suit their formation. On 25–28 April 2011, a ‘super-outbreak’ of tornadoes, described by the US National Weather Service (NOAA) as the most severe in recorded history, tore across the southeastern United States leading to a state of emergency in seven states. In total, 362 tornadoes were reported, killing more than 320 people and leaving a million people without power. Figure 8.20 shows

Figure 8.19 A tornado in Texas rampaging across fields. Tornadoes form when two air masses of different temperatures and humidity meet. If the lower layers of the atmosphere are unstable, a strong upward movement of warmer air is formed. This starts to spiral as it rises, and intensifies. Only a small percentage of these systems develop into the narrow, violent funnels of tornadoes. Wind speeds can reach up to 400 km h - 1 and they can damage an area 1 mile (1.6 km) wide and 50 miles (80 km) long. Tornadoes come in many shapes and sizes. (Source: lafoto, Shutterstock)

the damage left in just one small area from a EF4 category tornado (Table 8.3) during this outbreak. Tornadoes form when temperature and wind flow patterns in the atmosphere can cause enough moisture, instability, lift and wind shear for tornadoes to form in association with

thunderstorms (Figure 8.21). The most destructive and deadly tornadoes occur from supercells which are rotating thunderstorms with a well-defined radar circulation called a mesocyclone. As well as tornadoes, supercells can also produce damaging hail and strong winds. All thunderstorms tend to produce lightning

BOX 8.3 ➤

213

Chapter 8 Global climate and weather



Figure 8.20 Severe damage being cleared up two weeks after an EF-5

tornado that killed 189 people in Joplin, Missouri, 2011. (Source: FEMA / Alamy Stock Photo)

Mesocyclone Cloud base

Warm updraft

Figure 8.21 Formation of a tornado. Warm rising air meets cooler air aloft

creating turbulence. These winds start to rotate because of wind shear which is when the wind direction changes and the wind speed increases with height. The larger mesocyclone which then develops aloft can generate sufficient strength to extend a funnel cloud down to the ground.

and heavy precipitation and in supercells the lightning is often more frequent and the precipitation can lead to flash floods. The rotating in the storms is due to wind shear which is when the wind direction changes and the wind speed increases with height. This kind of wind shear and instability usually exists only ahead of a cold front and depression system. The rotation of a tornado partly stems from updrafts and downdrafts caused by the unstable air interacting with the wind shear. Cyclonically flowing air which is already slowly spinning to the left (in the northern hemisphere) converges towards the centre of the thunderstorm, causing it to spin faster due to the conservation of angular momentum. This is a similar process to one you can try for yourself. If you sit on an office chair and spin round with your arms outstretched you spin slowly. If you then pull your arms towards your chest you will suddenly start spinning faster. It is this conservation of angular momentum that creates the very high wind speeds within tornadoes. Most tornadoes rotate cyclonically (counterclockwise in the northern hemisphere and clockwise in the southern hemisphere) but the Coriolis force does not play any real part in the rotation as the size of tornadoes is too small for the Coriolis force to have any real effect. In fact there have been observations of anticyclonic rotating tornadoes, usually in the form of waterspouts (which are essentially tornadoes over water and also mainly have a cyclonic rotation), nonsupercell land tornadoes, or anticyclonic whirls around the rim of a supercell’s mesocyclone. Tornado intensity is measured by the Fujita scale (F-scale named after Theodore Fujita), which was replaced in February 2007 by the enhanced Fujita scale (EF-scale) (Table 8.3).

BOX 8.3 ➤

214

8.4  Mid and high-latitude climates

➤ Table 8.3 Enhanced F-scale for tornado damage (1 mph = 1.6 km h - 1) Fujita scale

Derived EF-scale

Operational EF-scale*

F number

Fastest 1/4-mile (mph)

3 second gust (mph)

EF number

3 second gust (mph)

EF number

3 second gust (mph)

0

40–72

45–78

0

65–85

0

65–85

1

73–112

79–117

1

86–109

1

86–110

2

113–157

118–161

2

110–137

2

111–135

3

158–207

162–209

3

138–167

3

136–165

4

208–260

210–261

4

168–199

4

166–200

5

261–318

262–317

5

200–234

5

Over 200

* Note that the EF-scale still is a set of wind estimates (not measurements) based on damage. It uses 3 s gusts estimated at the point of damage based on a judgement of eight levels of damage related to 28 indicators.

BOX 8.3

Reflective questions ➤ What is the main feature influencing the climate in equatorial regions and how does it affect temperature and rainfall?

➤ What are the necessary conditions for the formation of tropical cyclones?

➤ Why are there not as many tropical cyclones in the southern hemisphere as the northern hemisphere?

➤ What factors make the Asian monsoon a much more

makes the weather both much more variable and also much less predictable, particularly in areas close to the oceans. The middle latitudes of the southern hemisphere are dominated by oceans. The only land in the southern hemisphere middle latitudes is the southern tip of South Africa, the most southern parts of Australia and New Zealand, and central and southern Argentina and Chile. In the northern hemisphere, however, as well as the North Atlantic and Pacific Oceans, there are the major land masses of the North American, European and central Asian continents. These span all of the middle latitudes and extend into the polar regions, and (in the case of North America and Asia) into the subtropics.

marked feature than in other parts of the world at similar latitudes?

➤ Why are the deserts located where they are and what factors help increase aridity in many of these deserts?

➤ What is the dominant feature of summer weather in humid subtropical regions?

8.4 Mid and high-latitude climates In the tropics and subtropics, the weather is often relatively predictable but the middle latitudes are dominated by weather systems that move across the planet. This

8.4.1 Depressions, fronts and anticyclones The equatorial and subtropical climates are dominated by the weather systems associated with the thermally direct Hadley cell circulation (convection along the ITCZ and subsidence in the subtropical anticyclones) that transfers energy from lower latitudes. In the middle latitudes, however, there is also an energy transfer from lower latitudes to higher latitudes but this is not achieved by direct convection via heating at the surface. Instead the transfer is accomplished through the movements of large weather systems. However, just as in the tropics and subtropics, the key to understanding the development, movement and

215

Chapter 8 Global climate and weather

dissipation of mid-latitude weather systems is not what happens at the surface but what happens in the middle and upper troposphere. The key factor is the positioning and movement of the westerly polar front jet stream. The jet stream is a ‘thermal wind’ related to sharp thermal gradients in the atmosphere. In the case of the polar front jet stream, the thermal gradient is created by the temperature difference between polar and tropical air where the two air masses meet (the polar front). A range of types of air masses exist and are described in Box 8.3, and it is the interaction and modification of these air masses that may determine the weather conditions experienced in the middle and high latitudes. The polar front is just a steeper part of the normal low- to high-latitude temperature gradient (it is where there is a sharp transition between warm and cold air). The polar front jet stream is not fixed both in terms of its location as it moves north and south and there are marked waves along its length which vary in amplitude with shallower waves giving what is called a zonal flow (west to east) and larger amplitude waves giving a meridional flow (a more marked north to south component as shown in Figure 8.22). Air accelerates into anticyclonically curved waves creating a zone of upper-level divergence while the flow slows down as it enters a cyclonically curved wave creating upper-level convergence. If the upper-level divergence is greater than any low-level convergence this leads to a fall in surface pressure and creates a zone of ascending air. Such a situation is one of cyclogenesis (development of a depression). Conversely, if there is strong upperlevel convergence and weaker lower-level divergence, there is a zone of subsiding air and this situation is one of anticyclogenesis. The thermal gradient in the upper westerlies is therefore of more significance in the development of weather systems than temperatures close to the surface. Waves along the upper westerly polar front jet stream (Figure 8.23) can develop into frontal depressions. Air generally rises at fronts (by slantwise convection, as warm air is forced to rise above the cooler air it meets), leading to the formation of clouds (and precipitation) as the air cools on ascent. Fronts therefore mark areas of general precipitation (which may fall as snow in winter), although there are bands of more intense precipitation embedded within those areas. Some less active fronts may produce little or no precipitation. Jet streams are not simple continuous features. They have marked entrances where the flow becomes more concentrated into a stronger jet. They also have exits where 216

Zonal flow (generally west to east)

Meridional flow (flow with a large north-south component)

L

H

L

Cut-off high and cut-off lows (giving ‘blocked’ conditions) Figure 8.22 Zonal and meridional flow.

L

L

Upper westerly jet stream typically found at about 300 hPa (300 mb)

L–Depression (low-pressure system with fronts) Figure 8.23 Jet stream development. Waves along the jet stream can

develop into frontal depressions.

the jet spreads out and the flow rate reduces. Associated with jet entrances and exits are marked zones where the formation of depressions and anticyclones (or anticyclonic

8.4  Mid and high-latitude climates

AIR MASSES The term ‘air mass’ is given to a body of air that has a very large horizontal extent and in which its potential temperature and moisture content are similar through most of the troposphere (close to the surface there may be differences). An air mass develops over a source region where it has remained for a period of days (Barry and Chorley, 2010). There are four basic types of air mass according to their source regions. These are tropical maritime, tropical continental, polar maritime and polar continental. Essentially the tropical air masses are from low latitudes or the subtropics and polar air masses are from high latitudes. There are also extreme versions of polar air masses called Arctic maritime and Antarctic continental. Continental air masses are relatively dry and maritime air masses are relatively humid. The term ‘relatively’ has to be used as warm air holds much more moisture than cold air. A polar maritime air mass with 90% relative

humidity holds 3.9 g of water vapour per kg of air at 0°C while a tropical continental air mass with a 30% relative humidity has 8 g of water vapour per kg of air at 30°C. Tropical maritime air is common in both hemispheres, but tropical continental air is less common as the only really large land mass in the subtropics is northern Africa and to a lesser extent in Australia. India is a source region for tropical continental air in winter but the intense winter high pressure over Siberia and the mountains to the north act as a barrier. In summer, central Asia can be a source for tropical continental air although strictly speaking it is in the middle latitudes. Polar maritime air is common in both hemispheres with source regions in the high-latitude oceans. As with all air masses, there is not a single set of characteristics defining this air mass as the values change according to the actual source region and the time of year. Polar continental air is found only in the northern hemisphere as there is no

continent at higher latitudes in the southern hemisphere other than the Antarctic where the air is classed as being Antarctic continental. It is important to note that all air masses are modified by the underlying surface. If the surface is colder than the air mass then low-level stability will be increased. If it is warmer than the air mass, low-level stability will be decreased. The best example of this is Arctic maritime air moving south across the North Atlantic. The sea surface is much warmer than the air mass and this warms the lowest layers, decreasing stability and encouraging convection. This leads to the formation of frequent instability showers, a characteristic of this type of air mass. By studying source areas for air and tracking their movement over a particular region and predicting how the air masses might be modified by local conditions and how they might interact with other air masses, it is possible to provide short-range weather forecasts.

BOX 8.4

ridges) is favoured through divergence or convergence of the upper winds (leading to cyclogenesis and anticyclogenesis respectively) (Figure 8.24). It is important to remember that if there is an upper-level outflow (divergence) the pressure will fall at the surface and air will therefore converge Diffluent thermal trough

Confluent thermal trough CON Entrance

DIV Jet core

Exit

DIV Confluent thermal ridge

CON Diffluent thermal ridge

CON – Convergence (leads to anticyclogenesis and descent of air) DIV – Divergence (leads to cyclogenesis and ascent of air from below) Figure 8.24 Jet stream entrance and exit, cyclogenesis and

anticyclogenesis.

at lower levels and rise up to replace the ouflow. As a result there is a general (slow) upward motion of air in depressions in the conveyor belts moving through the system. In anticyclones or ridges the air flows inwards in the upper atmosphere, leading to an increase in pressure with descent and outflow at lower levels. The descent of air in anticyclones increases the temperature (adiabatically) and dries out the air, which means anticyclones bring clear conditions. However, the descending air does not fall all the way to the land surface and due to the adiabatic warming a temperature inversion forms. While the air above may be clear, on some occasions the inversion can trap moister air below and a layer of cloud can form below the inversion, giving a condition termed ‘anticyclonic gloom’. This is more common in winter than summer. This is because during summer solar radiation warms the air below the inversion sufficiently to cause the cloud to dissipate. On occasions the upper westerly flow 217

Chapter 8 Global climate and weather

can be interrupted by what is termed a blocking anticyclone generally formed by an intensification of a ridge in the upper westerlies into a closed circulation. This can lead to quite long periods of anticyclonic weather in areas where normally the weather would be characterized by the passage of mid-latitude depressions. The ascent of air in depressions is concentrated to some extent along the warm and cold fronts. Once aloft, the air rotates to become more parallel with the upperlevel flow (Figure 8.25). The ascent of the air is at perhaps 20 cm s-1 compared with the 5920 m s-1 that is typically found in large convective clouds. An ascent of air will eventually produce cloud as water vapour condenses out of the atmosphere in cooler conditions. The ascent of warm moist tropical air will quickly lead to the formation of layers of cloud (stratiform cloud) and continued ascent of air is likely to lead to precipitation. The ascent of air in depressions takes place over a much wider area

and over a much greater time than is the case for a convective cloud, and therefore although precipitation rates (e.g. how heavy it is raining) may be less than those found with convective clouds, there can be substantial amounts of precipitation arising from a frontal depression. It is important to note that even within a frontal precipitation zone there will be areas of more intense precipitation so that fronts do not produce simple areas of steady precipitation. Fronts tend to slope gently at a rate of 1 m vertical rise for every 80–150 m of lateral distance (slope of 1 : 80 to 1 : 150) with cold fronts being steeper than warm fronts. Over time, the cold front tends to overtake the warm front, leading to what is termed an occluded front, which is classified as warm or cold depending on whether the air ahead of the warm front is colder or warmer than the air following the cold front (Figure 8.26). The complex threedimensional nature of depressions and fronts is important

Jet core

400 hPa

500 hPa

600 hPa Warm front

700 hPa WCB Isobar in WCB Cold front 800 hPa

WCB – Warm conveyor belt hPa – hectopascals (1 hPa = 1 mb)

Figure 8.25 Conveyor belts in depressions. Slantwise convection (i.e. with a strong horizontal motion as well as conductive ascent) in the warm conveyor

belt carries sensible and latent heat polewards. 218

8.4  Mid and high-latitude climates

Tropopause

Tropopause –605C

–605C

Warm air mass Co ld

ld Co

oc n sio clu

Cold air mass

t fron rm a W

t alof Cool air mass

–405C

Warm air mass

al

t of

fro nt alo ft

Co ld fro nt

–205C

Cold air mass

05C

rm Wa

t

lof

ta

n fro

rm Wa Cold air mass

on usi

cl oc

–405C –205C 05C

(b) Warm occlusion

(a) Cold occlusion

Figure 8.26 Occluded fronts: (a) cold occlusions are where air behind the cold front is colder than air ahead of the warm front; (b) warm occlusions

are where air behind the cold front is less cold than air ahead of the warm front.

and is one reason why weather forecasting in mid and high latitudes is a very complicated science.

8.4.2 Mid-latitude western continental margins The Mediterranean-type climate is the first subtype of climate found polewards of the subtropical desert regions and is characterized by a mild wet winter half-year and a hot dry summer half-year. It is found in the far south-west of South Africa, in central Chile, on south-west-facing coastlines in the south of Australia, in California, as well as in the Mediterranean itself. In winter, mid-latitude frontal depressions bring rain to these areas, although in the Mediterranean itself, most of the depressions are not the frontal depressions of the Atlantic, the latter accounting for only 9% of Mediterranean depressions. A significant proportion of Mediterranean depressions develop as a result of dynamic effects on air flow over the Alps and Pyrenees that can lead to the formation of ‘orographic’ low-pressure areas (Barry and Chorley, 2010). These lows can develop frontal characteristics, particularly if the air flow across the mountains has a cold front embedded within it. This does not happen in the other Mediterranean climate zones of the world as they comprise only relatively narrow coastal areas. The Mediterranean Sea provides the mechanism for extending the climate type much further into the continent. Typically average winter temperatures in Mediterranean climates will be between 5 and 12°C with summer daytime maximum temperatures between 25 and 30°C. Rainfall totals will typically be between 400 and 750 mm with a distinct summer minimum. It is also

of note that the summers in the Mediterranean climate zones of California and central Chile are drier than in the Mediterranean owing to the upwelling of cold water off those coasts. This stabilizes the air and inhibits convection. Further polewards, mid-latitude western continental margins are most extensive in the northern hemisphere, although a similar climate regime is experienced by southern Chile, Tasmania and New Zealand. These climates have unusually mild winters for their latitude (e.g. the British Isles, western Europe and the west of Norway). These mild winters are particularly marked in the northeast Atlantic as the North Atlantic Drift pushes relatively warm water a long way north. There is a similar climate in much of New Zealand where again there is a warm current off the western coast. Generally temperatures in the southern hemisphere are lower than those at similar latitudes in the northern hemisphere owing to the large extent of the southern oceans and the paths of their ocean currents. As well as unusually mild winters for their latitude, these climates have a remarkably small range of annual temperature, have precipitation distributed throughout the year and there is considerable orographic enhancement of precipitation in the coastal mountain ranges (see Chapter 9). The mountain ranges of North and South America keep this climate confined to a relatively narrow coastal strip while in Europe the relatively low-lying ground from the Netherlands to Russia allows this climate type to extend to Poland in the east (Robinson and Henderson-Sellers, 1999). Average winter temperatures are typically between 2 and 8°C with average summer maximum temperatures between 15 and 25°C. Precipitation 219

Chapter 8 Global climate and weather

totals are generally in the range 500–1200 mm. The midlatitude depressions that are a feature of this climate can bring strong winds which can cause considerable damage. Windiness is a feature of this type of climate, particularly in coastal areas.

8.4.3 Mid-latitude east continental margins and continental interiors On the eastern side of North America and Asia, the eastern continental margin climates merge into the continental interior climates. Being within the mid-latitude westerly belt, the winds experienced on these continental margins have generally had a considerable passage across land. For that reason, the climate is more closely related to the continental interior than to the oceans, although some weather systems do come from the oceans to the east. Winters are much colder, with frequent snowfall, than those experienced in the subtropical humid climates to the south. These mid-latitude east continental margin climates do not exist in the southern hemisphere as they require large land masses. Furthermore, in South America east of the Andes, where such a climate might exist, the Andes act as a block to the westerlies and descent in the lee of the mountains dries the air and creates a climate that is more like a mid-latitude continental interior (semiarid) climate. The main extent of the humid continental type of climate is in North America, China and eastern Russia. In China and North America, this climate merges from the humid subtropical into a humid continental maritime margin, with increasingly severe winters, although at lower latitudes the summers are hot and long. As noted earlier, in China the cool winds circulating around the winter Siberian high pressure mean that there is less precipitation in winter than at other times of the year. The three winter months (December, January and February) have a total of close to 13 mm precipitation in Beijing while in New York the total for the same period is over 230 mm. Summers are similar in both Asia (eastern China, Korea and central Japan) and the eastern United States (south of New England), being hot and humid. Average temperatures in July for Beijing, China (40°N) are 26°C and those in New York (41°N) 24.5°C. Mean winter temperatures are -40°C and 0°C respectively. In summer, the June, July and August precipitation total is over 460 mm in Beijing while it is less than 320 mm in New York. Overall Beijing is drier (annual total close to 620 mm) than New York (over 1110 mm) owing to the aridity of Asia north and 220

east of the Tibetan Plateau. Some examples of conditions experienced at mid-latitude climate stations are given in Figure 8.27. Winters become increasingly colder as you move north or west into the mid-latitude continental interiors and summers also become milder and less humid (north of 45°N). Further north still (north of 50°N) the climate becomes subpolar with severe winters and relatively short summers (only three, or fewer, months with average temperatures above 10°C). For much of the mid-latitude continental interiors, precipitation is distributed throughout the year but generally with a distinct summer maximum. This summer precipitation is mainly in the form of convective showers although there are some weak frontal systems. Winter precipitation tends to fall as snow and as temperatures are low it often does not melt until the spring thaw. Total precipitation amounts are relatively low (below 500 mm), but the cold winter period and summer maximum of precipitation ensure that in most years there is sufficient moisture for plant growth. The main wheat-growing areas of North America have this type of climate. In Asia, however, the southern part of these mid-latitude continental interiors is semi-arid (as are those states in the United States just east of the Rockies). East of the Caspian Sea the climate becomes truly arid. These are cold desert regions in which winters are cold, although summers may still be warm. The Gobi Desert is an example of this type of desert. The further north you go the shorter the summer and the growing season. Winters are cold with average temperatures below - 12°C in the coldest month and below - 25°C in northern regions. In the coldest regions of Siberia the average temperatures in the coldest month can even be as low as - 50°C There is a very large range in temperatures with the warmest months having average temperatures of over 21°C, and even in the coldest parts of Siberia, July average temperatures reach over 13°C (an annual range of over 60°C). In North America there can also be some extreme diurnal ranges in temperature, particularly in areas prone to Chinook winds (see Chapter 6) or if warm moist air from the south pushes much further north than usual. Diurnal changes in temperature have even exceeded 50°C. These regions are influenced by midlatitude weather systems. However, in winter, high pressure dominates, especially in Siberia where pressures can reach over 1080 mb. Precipitation totals tend to fall as you move north within the mid and high-latitude continents of the northern hemisphere, with totals below 400 mm in the northern United States and southern Canada and falling below 300 mm further north. In eastern Siberia, annual totals can even be less than 150 mm.

8.5  Polar climates

(a)

35

30

30

25

25 Temperature (5C)

Temperature (5C)

35

20 15 10 5

(c)

20 15 10 5

0

0

-5

-5

-10

-10

J

F

M

A

M

J J Month

A

S

O

N

D

225

200

200

175

175

150 125 100 75

M

A

M

J

F

M

A

M

J J Month

A

S

O

N

D

150 125 100 75

50

50

25

25

0

F

(d)

225

Precipitation (mm)

Precipitation (mm)

(b)

J

J

F

M

A

M

J J Month

A

S

O

N

0

D

J J Month

A

S

O

N

D

Figure 8.27 Example temperature and precipitation graphs for mid-latitude climate stations. Mean monthly temperatures (°C) and mean monthly rainfall

(mm) for (a) and (b) Paris, France (49°N), and (c) and (d) Berlin, Germany (52.5°N).

Reflective questions ➤ Why are the waves in the upper westerlies important in encouraging the formation of depressions and anticyclones?

➤ Why are depressions associated with cloud formation and precipitation?

➤ What is an occluded front? ➤ Why are there no polar continental air masses in the southern hemisphere?

➤ Why are the winter precipitation Mediterranean systems different from those in the other areas with a Mediterranean-type climate?

8.5 Polar climates Polar climates are split into tundra and polar ice cap types (see Figure 8.3 for a map of their extent). Polar tundra is found in North America, in northern Labrador, the far

north of Quebec, North West Territories (north and east of the Great Slave and Great Bear Lakes), part of central and northern Yukon and north Alaska. Polar tundra can also be found in Europe and Asia in northern Sweden, Finland and Russia and also in northern Iceland. Coastal Greenland is also classed as having a polar tundra climate as is the northern part of Graham Land in the Antarctic. The polar ice caps are found in central Greenland and all of the Antarctic (except the northern part of Graham Land). In the polar tundra climate, the temperature of the warmest month will be above 0°C but below 10°C. Winter temperatures are generally extremely low (average temperature in January below - 25°C), although in coastal Greenland winter temperatures are higher ( - 7°C in January). In North America and Asia, the annual precipitation will typically be less than 300 mm and even below 120 mm in parts of north Alaska and northern Siberia, but in coastal Greenland annual totals are higher from 750 to over 1100 mm. In the tundra regions, weather is dominated by the prolonged winter season characterized by anticyclonic conditions (particularly in Siberia). However, this provides 221

Chapter 8 Global climate and weather

(a)

(b) 225

30

200

25

175 Precipitation (mm)

Temperature (5C)

35

20 15 10 5

150 125 100 75

0

50

-5

25

-10

J

F

M

A

M

J J Month

A

S

O

N

0

D

J

F

M

A

M

J J Month

A

S

O

N

D

Figure 8.28 Example temperature and precipitation graphs for a polar climate station at Ivigut, Greenland (61°N).

a sharp contrast to the short growing season, which although not very warm does provide an opportunity for the local flora and fauna to survive, if not actually flourish. The polar ice cap climates are extremely cold and weather is dominated by high pressure. Data from a polar station at Ivigut, Greenland, are given in Figure 8.28. Summer temperatures are generally below 0°C and winter temperatures below - 40°C. In parts of the Antarctic, the mean annual temperature can be close to - 50°C and an extreme minimum of - 89.6°C has been recorded at the Vostok research station (21 July 1983). There is little precipitation with annual totals typically less than 100 mm. These areas can actually be classified as cold deserts. In the central Antarctic, there is almost no precipitation. Air with a temperature below - 40°C contains almost no water vapour (even when saturated the amount of water vapour held at such low temperatures is very small), and therefore even if clouds form there is unlikely to be any precipitation. Polar ice caps cool the air in contact with them and as a result there can be strong winds blowing off the centre of the ice caps towards the coasts. This climate type is therefore dominated by the cold and by the frequent strong winds that produce extreme wind chill.

El Niño Southern Oscillation (ENSO) and the North Atlantic Oscillation (NAO) but there are other teleconnections and indices which try to characterize variability or differences across large regions through time. Key indices are:

t

8.6 A global overview The previous sections have outlined the major climate regions of the world but it is worth taking a global overview to consider how the various parts of the global climate interact and how there can be changes in opposite directions at the same time in different regions of the world. Chapter 6 provided an introduction to the

222

t

Southern Oscillation Index (SOI): Mean sea-level air pressure anomaly difference for the pressure at Tahiti minus the pressure at Darwin, normalized by the longterm mean and standard deviation of the mean sealevel pressure difference. Data are available from the 1860s. Data from Darwin can be used alone, as these data are more consistent than those for Tahiti prior to 1935. An El Niño event involves warming of tropical Pacific surface waters from near the International Date Line to the west coast of South America. This limits the upwelling of cold water near South America and typically occurs every 3 to 7 years. Periods of below-average temperatures in the eastern tropical Pacific are called La Niña events. El Niño events reduce the sea surface temperature gradient across the equatorial Pacific and are linked with the atmospheric Southern Oscillation, which brings changes in trade winds, tropical circulation and precipitation. El Niño Southern Oscillation events are coupled ocean–atmosphere phenomena with global implications that have extratropical teleconnections characterized by changes in the jet streams and storm tracks in mid-latitudes (particularly in winter months) as well as mean sea-level pressure anomalies (Box 8.5). North Atlantic Oscillation (NAO) Index: The difference in normalized mean sea-level air pressure anomalies between Lisbon in Portugal and Stykkisholmur in Iceland has become the most widely used NAO Index

8.6  A global overview

t

t

t

t

and extends back in time to 1864. The data can go further back to 1821 if Reykjavik is used instead of Stykkisholmur and Gibraltar instead of Lisbon. The NAO has a strong link to the alternation of westerly and blocked flow across the Atlantic and is present from the surface up into the stratosphere. Northern Annular Mode (NAM) Index: The amplitude of the pattern defined by a mathematical term known as the ‘leading empirical orthogonal function’ of winter monthly mean northern hemisphere mean sea-level air pressure anomalies polewards of 20°N. The NAM has also been known as the Arctic Oscillation (AO), and is closely related to the NAO. Southern Annular Mode (SAM) Index: The difference in average mean sea-level air pressure between the southern hemisphere middle and high latitudes (usually 45°S and 65°S), from gridded or station data (Gong and Wang, 1999; Marshall, 2003), or the amplitude of the leading empirical orthogonal function of monthly mean southern hemisphere 850 hPa height polewards of 20°S. This was formerly known as the Antarctic Oscillation (AAO) or High Latitude Mode (HLM). The principal mode of variability of the atmospheric circulation in the southern hemisphere extratropics is the SAM Index. It is essentially a zonally symmetric structure, but with a zonal wave pattern, and reflects changes in the main belt of subpolar westerly winds. Enhanced Southern Ocean westerlies occur in the positive phase of the SAM. Pacific-North American pattern (PNA) Index: The mean of normalized height at which the air pressure is equal to 500 hPa at 20°N, 160°W and 55°N, 115°W minus those at 45°N, 165°W and 30°N, 85°W. As with the NAO, the PNA appears related to periods of blocked flow, particularly in the Gulf of Alaska, and periods of stronger westerlies. It is associated with changes in the Aleutian Low, the Asian jet, and the Pacific storm track, and affects precipitation in western North America and the frequency of Alaskan blocking events and associated cold-air outbreaks over the western United States in winter. Pacific Decadal Oscillation (PDO) Index and North Pacific Index (NPI): The NPI is the average mean sealevel pressure anomaly in the Aleutian Low over the Gulf of Alaska (30°N–65°N, 160°E–140°W) and is an index of the PDO, which is also defined as the pattern and time series of the first empirical orthogonal function of sea surface temperature over the North Pacific north of 20°N. The PDO broadened to cover

t

the whole Pacific Basin is known as the Inter-decadal Pacific Oscillation (IPO). Decadal to inter-decadal variability of the atmospheric circulation is most prominent in the North Pacific, where fluctuations in the strength of the winter Aleutian low-pressure system co-vary with North Pacific sea surface temperature in the PDO. These are linked to decadal variations in atmospheric circulation, sea surface temperature and ocean circulation throughout the whole Pacific Basin in the IPO. Atlantic Multidecadal Oscillation (AMO): The AMO is related to the North Atlantic sea surface temperatures that show a 65–75 year variation (0.4°C range), with a warm phase during 1930–1960 and cool phases during 1905–1925 and 1970–1990. A warm phase could peak around 2020.

Some of the above indices and their teleconnections provide a means of making long-range weather predictions. Note that these are not forecasts which use numerical weather prediction models, but they give an indication of very generalized patterns of weather that might be expected, for example, for a winter seasonal prediction. However, there also seems to be some relationship between variations in the indices and global climate (see Section 6.8.1 in Chapter 6). In addition, there is concern about how climate change may influence many of the teleconnections. For example, Wang et al. (2015) showed that the severe Texas and Oklahoma flooding of May 2015 was linked to El Niño, and may have been exacerbated by global warming. Modelling suggested that precipitation anomalies might increase in association with El Niño as the teleconnections are strengthened in the future under climate change. It is also important to recognize that there is considerable inter-annual variability in global climate and that inter-annual variability will be even greater at a regional level and variations in precipitation are much greater than those for temperature. The extent of inter-annual variability can clearly be seen in the Central England Temperature (CET) series (Figure 8.29), which runs since 1659 for the period when instruments have collected data. The CET data can be found at: http://hadobs.metoffice .com/hadcet. It is also important to note that this interannual variation in global as well as the CET series is less than the variation at a seasonal or monthly scale. There are two particularly cold years in the CET series, 1740 and 1879, with average temperatures of 6.84 and 7.42°C respectively. There is no obvious explanation as to

223

Chapter 8 Global climate and weather

11

Temperature (5C)

10

9

8

7 1665

1715

1765

1815

1865

1915

1965

2015

Year Figure 8.29 Central England mean annual temperatures 1659–2015. (Source: http://

www.metoffice.gov.uk/hadobs/hadcet/index.html)

why those two years were so much colder than the years immediately before and after. During the last century, the cooler period between the 1940s and 1970s was characterized by more frequent negative ENSO indices and a more frequent occurrence of a negative winter NAO Index. That period was marked by colder winters in northern Europe. This inter-annual variability is why it is important not to assume that a few years when the average temperature increases or decreases is evidence of a warming or cooling trend. There is the danger of ‘cherry-picking’ data and the standard practice is to produce a ‘climate normal’,

THE WINTER OF 2009–2010 IN NORTH-WESTERN EUROPE The winter of 2009–2010 was unusual in the northern hemisphere with cold and snow affecting the eastern United States and northern Europe but with much milder conditions than normal in Canada, which caused problems with a lack of snow for the 2010 Winter Olympics in Vancouver.

which is an average calculated over a 30 year period that is updated every 10 years. Hence, data from 2011 to 2020 will typically be compared to the 1981–2010 period. Global temperature series comparisons have been criticized as it may be that some stations have been affected by urbanization which will increase local temperatures (see Chapter 9), although those developing the series have tried to eliminate any such effect by changing the stations used to compile the series. In the case of the CET series there has been very careful quality control to ensure the values are not affected by urbanization.

This pattern arose due to a strongly negative phase of the NAO (Figures 8.30 and 8.31). At the same time there was a positive phase of the ENSO and this led to a different pattern than would have been the case with a negative NAO alone, as shown in Figure 8.30. In the United Kingdom, the winter was particularly cold in central and northern

Scotland (Figure 8.32) and normal precipitation patterns (more in west, especially in the mountains of north-west Wales, England and Scotland) were reversed so that much more precipitation fell in the east of the United Kingdom with much as snow from a generally easterly or northeasterly direction rather than the usual westerly direction (Figure 8.33).

BOX 8.5 ➤

224

8.6  A global overview



El Niño

Negative NAO Temperature anomalies (5C) = Land mass -3

0

El Niño + negative NAO

3

Winter 2009–2010 observed

Temperature anomalies (5C) -6

0

6

Figure 8.30 The combination of ENSO and negative NAO.

BOX 8.5 ➤

225

Chapter 8 Global climate and weather

➤ NAO negative mode

L

ea m st r

Cold & snowy

Jet

Warm & less sea ice

NAO positive mode

Cold & more sea ice

Cold & dry

‘Blocking’ H

Warm & wet

Warm

L

Warm & wet Cool & dry

Jet stream

H

Figure 8.31 Phases of the NAO.

Anomaly value (5C) 7-1.5 -2.0 to -1.5 -2.5 to -2.0 -3.0 to -2.5 6-3.0

Figure 8.32 Winter 2010 mean temperature anomaly from the 1971–2000 average.

(Source: Met Office, 2011)

BOX 8.5 ➤

226

8.6  A global overview



% of average 7170 150–170 130–150 110–130 90–110 70–90 50–70 30–50 630

Figure 8.33 Winter 2010 rainfall amount (% of 1971–2000 average). (Source: Met Office, 2011)

BOX 8.5

Reflective question ➤ The raw data for the global temperature series can be found on the Climatic Research Unit website (www.cru .uea.ac.uk/cru/data/temperature/#datdow) with the temperatures being given as anomalies from the 1961– 1990 average of 14.0°C. The ENSO data can be found at www.esrl.noaa.gov/psd/people/klaus.wolter/MEI/table

(Hint: do not look at absolute values but at the direction of change: are the anomalies getting larger or smaller as the ENSO Index changes in particular years? Look especially at the El Niño years of 1982–1983, 1986–1988, 1997–1998, 2009–2010 and 2015–16 and in the La Niña years of 1988–1989, 1998–2001, 2006–2008 and 2010– 2012. Do not assume that any signal will be clear.)

.html. Does there appear to be a relationship between the ENSO values and the global temperature anomalies?

227

Chapter 8 Global climate and weather

8.7 Summary

boundaries between air masses are known as fronts and represent sharp contrasts in temperature and moisture contents of the

This chapter has shown that the climate of any region is a result

air. At fronts the warm air rises above the cooler air, often result-

of the type and frequency of the weather systems found in that

ing in condensation and precipitation. Continental interiors tend

region. Major features of atmospheric circulation including zones

to have different climates from those close to oceans even at the

of ascending and descending air such as the ITCZ (ascending),

same latitude.

the subtropical anticyclones (descending) and the slantwise con-

Individual climate elements and their changes through the

vection at the polar front (ascending) play an important role in

year vary in their importance between different climates. For

climate.

example, the distribution of precipitation during the year as well

Equatorial climates are dominated by movements of the ITCZ

as annual precipitation totals is important because a region may

whereas at higher latitudes in the tropics the easterly wave, mon-

not be arid if there is a winter rainfall maximum, even if it has

soonal conditions and tropical cyclones may be more important,

a fairly low annual total and a hot summer. In some regions the

although it is still the movement of the ITCZ that partly controls

heat and humidity of the summer are the dominating features

these. Thus, some regions may have a fairly constant climate but

of the climate while in others it may be precipitation totals or

are subject to occasional extreme events that have a substantial

winter cold. There are also some climate types where there may

impact (e.g. regions prone to tropical cyclones). At still higher

be large differences between individual locations in one or more

latitudes in the tropics, desert conditions are prevalent associated

climate element (this is discussed further in Chapter 9). It is also

with the anticyclonic conditions related to the descending limb of

important to recognize that climate types do not have distinct

the Hadley cell.

boundaries, as unless there is a major mountain barrier one type

The distribution of land masses and oceans play an important role in global climates. Thus the southern hemisphere experi-

of climate usually merges gradually into another. Finally, it is clear that far from being constant, the climate,

ences different mid and high-latitude climate conditions to the

even as measured over the normal 30 year climate period, is

northern hemisphere owing to the lack of land masses within

not constant. During the instrumental record before the twen-

these regions. Air masses are important in the middle and high

tieth century these climate changes were relatively small, but

latitudes. They are distinguished by the source area from which

the last 100 years have been marked by a period of increasing

they originate (continental or maritime, polar or tropical). The

temperatures.

Further reading Ahrens, C.D. (2015) Meteorology today – An introduction to weather, climate, and the environment, 11th edition. Cengage Learning, Boston. This American textbook provides a good clear overview and is very nicely illustrated with colourful figures. There are lots of reflective and essay-style questions. Barry, R.G. and Chorley, R.J. (2009) Atmosphere, weather and climate, 9th edition. Routledge, London. This book contains useful chapters on air masses, fronts and depressions and on climates of temperate and tropical zones. It has been a very popular book over the years. 228

O’Hare, G., Sweeney, J. and Wilby, R. (2005) Weather, climate and climate change. Prentice Hall, Harlow. Excellent and accessible introduction to the area. Robinson, P.J. and Henderson-Sellers, A. (1999) Contemporary climatology. Pearson Education, Harlow. This book contains good sections on tropical and mid-latitude climates. Wang, S.-Y., Huang, W.-R Hsu, H.-H. and Gillies R. R. (2015) Role of the strengthened El Niño teleconnection in the May 2015 floods over the southern Great Plains. Geophysical Research Letters, 42, 8140–8146, doi:10.1002/2015GL065211 An important research paper highlighting teleconnections that have been strengthened due to global warming focusing on the case of the severe flooding in Texas and Oklahoma in May 2015.

C HAPTER 9

Regional and local climates John McClatchey Late of Environmental Research Institute, University of the Highlands and Islands

Learning objectives After reading this chapter you should be able to:

➤ understand how local factors can modify regional climate ➤ understand how altitude and topography control local and regional climates

➤ describe how large water bodies influence local and regional climates

➤ recognize how human activity can have a deliberate or inadvertent impact on local climate

9.1 Introduction There is substantial variation at the local and regional scales within the global climate zones discussed in Chapter 8. In some places the variation is part of a relatively gradual change from one climate type to another. For example, in the mid-west states of the United States, moving north from Tennessee through Kentucky to Illinois (Figure 9.1) there is a steady change in climate. The whole of this large area north of the southern half of Missouri, Illinois and Indiana and north of the whole of

Ohio (i.e. the coverage of Figure 9.1) is within the humid continental climate type. However, there are more gradual regional variations. Tennessee has a humid subtropical climate but from Illinois to the north into Canada the climate is described as humid continental (see Chapter 8). In moving northwards the winters become increasingly severe. Northern Iowa has summers that are noticeably cooler than the hot humid summers experienced in the south of Missouri. This cooling of the summers continues into Canada. North of Winnipeg the summers become much shorter with less than four months having temperatures above 10°C. Iowa, Wisconsin and Michigan are within the same climate subtype with at least four summer months with temperatures above 10°C. However, even within this area, there are marked regional variations in climate on top of the gradual change northwards. The most obvious of these differences are associated with the areas bordering the Great Lakes of Michigan and Superior. In any climate region, there can be very marked local variations in certain climate elements. For example, over a distance of little more than 100 km across Scotland there are places in the west where the total annual precipitation is nearly 10 times greater than that in parts of the east. However, on a global or even European scale the whole of Scotland falls well within the limits of a single climate

Chapter 9 Regional and local climates

9.2 Altitude and topography

500 km

CANADA

Winnipeg

N. Dakota

N Su p e ri o r L ak e

Minnesota Michigan L ak e Mi c h i gan

IT

S. Dakota

UN

Wisconsin

ED

Nebraska

S

T

Iowa

A

TE

S O F

e

ke

La

AM E R I C A

Illinois Kansas

Lake Huron

E ri

Ohio

Indiana West Virginia

Missouri

Figure 9.1 The mid-west United States and the Great Lakes.

type. The classification or description of a climate at a particular place therefore depends on the scale at which that climate is being considered. The smaller the geographical extent of the area of interest, the more important it becomes to have detailed climatological observations and statistics in order to specify the climate of an individual location. It is also important to note that regional and local climates are a feature of the land and not the oceans because altitude, topography and proximity to the sea (or large bodies of water) are the key features that lead to the development of local and regional climates. These factors and associated processes will be discussed in this chapter as it is important to understand how both local and global processes interact at different scales. This interaction helps to explain the nature of local and regional climate across the Earth’s surface.

The climate of mountains has always been of interest to scientists. Early studies helped establish that pressure and temperature fell with height and in the latter part of the nineteenth century a number of mountain observatories were established in Europe and North America to support astronomical studies and weather forecasting. Examples include Mount Washington (New Hampshire, established 1870), Sonnblick (Austria, 1886) and Ben Nevis (Scotland, 1883). Some of these mountain observatories closed after 10–20 years of observations (e.g. Ben Nevis, 1883–1904) but many are still in existence. The rate of fall of temperature with altitude (the lapse rate, see Box 9.1) varies in different parts of the world. The amount of solar radiation that can potentially be received by the ground surface actually increases with height. This is because less radiation has been absorbed or reflected by components of the atmosphere back into space by the time it reaches the ground at higher altitudes. The lower the altitude, the thicker the layer of atmosphere that can reflect solar radiation back into space. However, whether solar radiation received at the ground surface actually increases with height in any particular mountain range depends upon local cloudiness. Harding (1979) reported that in the mountains of the United Kingdom, which tend to be cloudy, solar radiation decreased by 2.5 to 3 million J m-2 day-1 km-1, which was regarded as typical by Grace and Unsworth (1988). Nevertheless, even in those areas where received solar radiation increases with altitude, temperature is still likely to decline upwards because of adiabatic processes (see Table 9.1). In many places, wind speed and precipitation increase with altitude but this is not true everywhere. Substantial mountain ranges can also act as a barrier to the movement of

Table 9.1 Predicted change of pressure and temperature with height based on a lapse rate of 6.5°C per 1000 m (pressure P in hPa or millibars and temperature T in °C)

Height (m) (above mean sea level)

Mid-latitude

High-latitude

P

T

P

1013

25.0

1013

15.0

1013

5.0

1000

902

18.5

899

8.5

894

- 1.5

2000

801

12.0

795

2.0

788

- 8.0

3000

710

5.5

701

- 4.5

691

- 14.5

4000

627

- 1.0

616

- 11.0

605

- 21.0

5000

552

- 8.5

541

- 17.5

527

- 27.5

0

230

Tropical

T

P

T

9.2  Altitude and topography

LAPSE RATES The rate at which temperature falls with increasing altitude is known as the environmental lapse rate. An air parcel will rise if it is warmer than the surrounding environment. Once the air parcel reaches the same temperature as the surrounding environment it will stop rising (Figure 9.2). When air rises (ascends) it expands. This is because the air pressure decreases. Conversely if air sinks (descends) it is compressed as the pressure increases. If no energy is added to or lost from that air as it rises (or falls), the changes in pressure and temperature are the result of what is termed an adiabatic process. When air rises and the pressure falls, the energy for the expansion of the air comes from the air itself. As temperature is a measure of the energy of air, if energy is removed by expansion of the air then the temperature

of the air decreases. The reverse is true if air sinks and is compressed, in which case the temperature of the air increases. If air expands adiabatically as it ascends, the temperature of the air falls at a constant rate of 9.8°C km - 1 (Figure 9.2). If air is compressed adiabatically as it sinks the temperature of the air increases at the same rate. This rate of temperature change is called the dry adiabatic lapse rate and applies only if the atmosphere remains unsaturated. Air can hold only a certain amount of water vapour at any temperature and if the temperature falls or if more water is evaporated into the atmosphere then once saturation is reached water vapour will condense out of the atmosphere. When water condenses out of the air, energy is released. This energy is the latent heat of vaporization and is the energy released when gaseous water vapour condenses into liquid water. When

Height (m)

3000

SALR 2000 Condensation level 1000 DALR ELR Air parcel 0 -5

0

5

10

15

20

Temperature (5C) Figure 9.2 Air parcel buoyancy. A parcel of air warmed at ground level will rise if it is warmer than its surroundings and will cool at the environmental lapse rate (ELR). Note that the top of the cloud occurs where the air parcel temperature is the same as its surroundings. DALR – dry adiabatic lapse rate (9.8°C/1000 m); SALR – saturated environmental lapse rate.

water evaporates or ice melts, this uses up energy in order to change the state of the water but without changing the temperature of the water (see Chapter 6). When the reverse occurs energy is released. This energy release reduces the cooling rate of the air as it rises and expands. In the absence of any loss of total water content (gaseous, liquid or solid water) from the air, the rate of change of temperature with height is given by what is termed the saturated adiabatic lapse rate. Unlike the dry adiabatic lapse rate, the saturated adiabatic lapse rate depends on the amount of water vapour the air can hold at any temperature. The amount of water vapour air can hold more than doubles for each 10°C increase in temperature (Table 9.2). This means that the amount of latent heat released is much less at low temperatures than at high temperatures as less water vapour will condense out of the air at lower temperatures. The saturated adiabatic lapse rate therefore varies from around 0.3°C per 100 m close to the surface in the tropics, where air temperatures are over 30°C, to close to the dry adiabatic rate at temperatures below -40°C. Temperatures of below -40°C are normally found at heights of between 5 and 10 km in mid-latitude regions. When saturated air ascends, the temperature decreases at the saturated adiabatic lapse rate and the water that condenses out of the air forms a cloud. However, when saturated air descends, it will warm at the saturated adiabatic rate only if all of the liquid or solid water (cloud droplets or ice crystals) is evaporated back into the air. If any of the water has fallen out of the cloud as precipitation there will be less water to evaporate and the air will therefore warm more on descent than it cooled on ascent. The difference between the dry and saturated adiabatic lapse rates is crucial

BOX 9.1 ➤

231

Chapter 9 Regional and local climates

➤ Table 9.2 Water vapour saturated vapour pressures Temperature (°C)

- 40

- 30

- 20

- 10

0

10

20

30

Water vapour pressure (hPa)

0.19

0.51

1.25

2.86

6.11

12.27

23.37

42.43

in understanding why some atmospheric conditions are unstable (or conditionally unstable) (Figure 9.3). This instability occurs because if rising air becomes saturated, any further ascent will cause the air to cool at the saturated adiabatic rather than dry adiabatic lapse rate. Such saturated air will therefore be warmer than the surrounding air and as warmer air is less dense than colder air, the saturated air will ascend further as a result of this density difference. This leads to strong upward convection and the growth of shower clouds. A situation is described as conditionally unstable if the environmental lapse rate is steeper than the saturated adiabatic lapse rate through the lower atmosphere (around 10 km). If clouds form and grow into this layer with a steep lapse rate they are then going to continue to grow to great depths. The instability is ‘conditional’ as clouds have to form and reach the height at which the cloud temperatures become warmer than the surrounding air. Although adiabatic processes are common

(a)

in the atmosphere, particularly when there is widespread ascent or descent of air such as in frontal ascent or anticyclonic subsidence, energy can be gained or lost from air by a number of processes including the loss of energy associated with precipitation that falls out of the atmosphere. Radiative exchanges can also be important in the atmosphere. While air is largely transparent to solar radiation, certain atmospheric gases absorb terrestrial (longwave) infrared radiation (see Chapter 6). Liquid or solid water completely absorbs and reradiates long-wave radiation. This means that clouds play an important role in radiative exchanges. Radiational losses at the top of clouds can cause localized cooling. Radiational heating from the ground surface is also important. Strong heating can create steep lapse rates in the lowest layers of the atmosphere. These steep lapse rates can be greater than even the dry adiabatic lapse rate and are termed super-adiabatic. Such steep lapse rates cause rapid local convection which tends

(b)

to mix the atmosphere. This effect is greatest in summer and at lower latitudes. At night when long-wave infrared radiational emission from the surface cools the ground, the lowest few hundred metres of the atmosphere can be cooled as a result. This creates a temperature inversion perhaps a few hundred metres deep. This is where warmer air overlies cooler air so that it is no longer the case that temperature declines with altitude. Once warmer air overlies cooler air, the lower layer is trapped because it is denser. Thus, pollutants from fossil fuel combustion (e.g. fires, car engines) may not be able to escape from the lower air layer and thus a longlasting ‘smog’ can develop. During these times public health can be at severe risk. However, while adiabatic lapse rates are important in the atmosphere, other processes, such as mixing of air, affect temperatures. If rising air mixes with its surroundings its energy will be shared with the surroundings and the temperature changes will no longer be adiabatic.

(c)

ELR ELR

SALR Height

SALR Height

Height

SALR

DALR

DALR

DALR

ELR

Temperature

Temperature

Temperature

Figure 9.3 Atmospheric stability relationships between the environmental lapse rate (ELR), dry adiabatic lapse rate (DALR) and the saturated

environmental lapse rate (SALR): (a) stable; (b) conditional instability; (c) absolute instability.

BOX 9.1

232

9.2  Altitude and topography

weather systems and even smaller ranges of mountains and hills can give rise to noticeable differences in the weather (and hence in climate) on the lee side of those mountains. Hills and mountains are therefore important as they can create substantial regional and local modifications to the general climate of that part of the world. The following discussion will explain how individual climate elements are modified by hills and mountains, and how a specific regional climate may be developed on the leeward side of upland areas.

Figure 9.4 Temperature inversion in the early morning. Cooler air lies below warmer air and a mist has formed close to the ground surface in the valley. (Source: JPL Designs / Shutterstock.com)

9.2.1 Pressure The fall of air pressure with height is the most consistent feature of mountain climate. Up to around 3000 m the fall in pressure is close to 10 millibars per 100 m. The rate of fall is more rapid in colder (denser) air and therefore pressures are higher at the same altitude in the tropics compared with the middle and high latitudes (Table 9.1).

9.2.2 Temperature The change of temperature with altitude is known as the lapse rate. Box 9.1 describes the important characteristics of lapse rates and should be read in order to understand fully the following section. The values in Table 9.1 give an indication of the differences in the fall of pressure with altitude in tropical, mid-latitude and high-latitude regions, but they are based on an assumed lapse rate of 6.5°C per 1000 m. However, there can be substantial diurnal variations in lapse rates and these will influence altitudinal pressure gradients. During the day strong solar heating warms the air close to the ground, steepening the lapse rate in the first few hundred metres, of the atmosphere. In the first few tens of metres, the lapse rate can exceed the dry adiabatic lapse rate of 9.8°C per 1000 m (see Box 9.1), although the lapse rate will depend on how well mixed the atmosphere is. The stronger the wind, the greater the turbulent mixing and the closer the lapse rate will be to the dry adiabatic rate (as long as there is no condensation of water vapour). In cloudy conditions with strong winds, the atmosphere is so well mixed in the lowest few hundred metres that below the cloud base the lapse rate will approximate the dry adiabatic lapse rate. Within the cloud the lapse rate will be at the saturated adiabatic rate. At night, light winds and clear skies allow the ground to cool through long-wave radiation loss and a strong temperature inversion can

form close to the ground (Figure 9.4). Katabatic drainage can further strengthen the night-time inversions (see Box 9.4 below). These diurnal variations in the fall, of temperature with height are influenced by cloud cover and the presence or absence of vigorous mixing caused by strong winds. Together these play an important role in controlling temperatures of mountain regions. As well as diurnal variations, the lapse rates also vary according to air mass. As air masses move they can be warmed or cooled by the underlying surface, leading to a steeper or a shallower lapse rate respectively. Hence Arctic or Antarctic maritime air masses, which are always warmed from below as they move into lower latitudes, have the steepest lapse rates, which are often close to the dry adiabatic lapse rate. Tropical air masses, however, are cooled from below as they move away from lower latitudes. This reduces the lapse rate. The affect of warming and cooling from below is illustrated in Figure 9.5,

Inversion:

Wind

(a)

top base

Lake or sea

Land

(b) 3 Wind 1 Lake

2

4 Land

Figure 9.5 Temperature inversion: (a) the modification of an unstable temperature profile to give a surface-based inversion over a cold surface (e.g. lake/ice); (b) elevated inversion due to the advection of stable lake air across the shoreline to a warmer land area on a spring afternoon. (Source: after Oke, 1987)

233

Chapter 9 Regional and local climates

234

Approximately 10 km

Dry adiabatic lapse rate Altitude

which shows an air mass being cooled as it moves across a cool lake and one being warmed as it moves from a cold lake to a warm land area. Temperature inversions are temporarily formed as the air is warmed or cooled from below. Polar maritime air may also have steep lapse rates if there is a large contrast between the air and sea surface temperatures. As the contrast between the air and sea surface temperatures varies with the seasons, there are some seasonal differences in lapse rates. The average lapse rates in mountains of the British Isles are steeper than in most mountainous regions as a result of the frequency of polar and Arctic air masses and the relatively warm surrounding sea surface temperatures (Harding, 1978). Air with steep lapse rates is conditionally unstable (see Box 9.1) and instability showers are therefore common. Air with a shallow lapse rate is stable, damping down convective activity. However, such air can be subject to forced ascent over mountains or dynamically induced ascent such as in a mid-latitude depression. If the ascent is sufficient to cool the air to its dew point temperature, clouds will form and this can lead to enhancement of precipitation. Tropical continental air moving towards higher latitudes is cooled from below but in summer that cooling may only extend through a relatively shallow layer. Tropical continental air is dry and is formed over an area in which there was strong heating from below. Therefore above the layer of air that is cooled from below, the lapse rate will be much steeper. This means that there can be strong conditional instability. Convection that rises above the cooled layer can trigger the rapid formation of deep convectional clouds. Anticyclonic conditions are also important in determining temperature variations with altitude. The large-scale descent of air in anticyclones creates a dry adiabatic lapse rate aloft, above a strong temperature inversion. For example, in the trade wind belt flowing out of the subtropical anticyclones, there is a marked temperature inversion, generally at a few hundred metres above the surface on the eastern side of the anticyclone rising to 2000 m in the west. Mountain ranges such as the Atlas in Africa, the Andes in South America and the Sierra Nevada in California often penetrate well above these subtropical inversions. This leads to distinct changes in temperatures and therefore local climate on ascending such mountain ranges. Subsidence inversions as shown in Figure 9.6 are a feature of all anticyclones. Air is forced to descend under the anticyclonic conditions but part of this descending air may be warmer than the air below it. Thus the temperature might vary with altitude as shown in Figure 9.6. Depending on the height of the inversion, mountain ranges in any part of the world may be above the height of the inversion

Subsidence inversion

Temperature Figure 9.6 Lapse rate under anticyclonic conditions with a subsidence

inversion.

base. This will lead to unusual conditions on those mountains with different lapse rates above and below the inversion. There may also be a very low humidity. The descending air in anticyclones leads to exceptionally low values of humidity close to the base of the inversion. However, as anticyclones found in the middle and high latitudes are short-lived compared to subtropical anticyclones, on some days there will be a sharp change in the weather conditions experienced on ascending the mountains. Such changes will be incorporated into the averages of the climate elements.

9.2.3 Wind As well as changes in temperature with height, the wind regime in mountains can be quite different from that at lower levels. It is not, however, altitude that is necessarily the key factor. It is the topography itself that is important. For example, the wind can be funnelled through valleys or even gaps between individual peaks, particularly if the orientation is in the direction of the prevailing wind. This will result in much greater local wind speeds. In addition, individual peaks and exposed ridges will experience higher winds as there will be less surface friction acting to reduce wind speeds as the wind approaches. Friction reduces surface winds by about 30% compared with the ‘free atmosphere’. In some circumstances, the compression of air between mountain summits and a temperature inversion aloft can lead to wind speeds above free atmosphere values (Figure 9.7). This frequently occurs around the Cairngorm Mountains in Scotland and at Mount Washington in New Hampshire, where maximum gusts have been recorded at 76 and 103 m s-1 respectively. The wind regimes in midlatitude mountains are also influenced by the westerly

9.2  Altitude and topography

Acceleration of flow over mountain top Height

Inversion

Temperature Figure 9.7 Acceleration of winds over ridges and mountain tops where the air is compressed by a temperature inversion aloft. Because the space for the air to pass through is narrower then air is forced through at greater speeds.

winds. The westerly winds are generally faster aloft. However, in the tropical and subtropical trade wind belts, the north-east and south-east trade winds generally weaken with height. Therefore, mean wind speeds can be low on tropical and subtropical mountains. For example, typical wind speeds are 2 m s-1 during the period December to February at 4250 m in New Guinea and an annual mean of 5 m s-1 at 4760 m in Peru. In the Himalayas, the monsoon circulation gives strong westerly winds through the winter half-year (October–May) with more moderate easterly winds in the summer (June–September). The westerly winds decrease from over 25 m s-1 at 9 km in the winter half-year to only 10 m s-1 by the end of May (being replaced by easterlies in the second half of June). These wind speed changes in the Himalayas emphasize the importance of the weather systems experienced in mountain regions in determining the wind regime. One other local and regional climate feature of winds is a warm and dry wind that blows down lee slopes of hill and mountain ranges. These winds tend to warm and dry as a result of the compression and adiabatic warming of the air in the lee of the hills and mountains as it descends from higher levels. It is called a Föhn wind, although it has other names in different parts of the world such as in Canada where it is called the Chinook. The onset of the wind is typically accompanied by a sharp rise in temperature often with a substantial decrease in relative humidity. In Canada, in the lee of the Rockies, temperature rises of over 20°C have been recorded in just a few minutes. Evidence of smaller rapid rises in temperature has been found in Scotland and even with winds across the Pennines in England, where the hills are typically only 500–700 m high (Lockwood, 1962). Föhn winds in the Alps and other mountainous regions can cause rapid snow melt, greatly increasing avalanche risk and flooding (Barry, 1992). It has already been noted that surface cooling can lead to katabatic drainage into valley bottoms. Over glaciers

and particularly over the Antarctic ice sheets more substantial katabatic winds can form due to local cooling (e.g. Renfrew and Anderson, 2002). These winds, which can be extreme, will flow into hollows and valley bottoms and are a special case of mountain winds (see Box 9.4 below).

9.2.4 Precipitation The amount of moisture the air can hold is strongly dependent on temperature and as temperatures fall with height, generally the moisture content of air does so too. It might therefore be expected that precipitation would also decrease with height as the moisture content declines. However, in the lowest 3000 m of the atmosphere this is certainly not the case. Air forced to rise over mountains cools at the dry adiabatic lapse rate until the dew point temperature is reached. At this point clouds form and temperature then decreases at the saturated adiabatic lapse rate if there is any further ascent of the air. Therefore, while temperature is reduced by adiabatic expansion as the pressure falls, the moisture content of the air does not change until saturation is reached. Hence, even in a very dry region, if air is forced to rise, it will eventually reach saturation. For example, if dry air (say 30% relative humidity) at 30°C is forced to rise, it will become saturated at about 2000 m and cloud will form. Unless there is no wind, the forced ascent of air over hills will therefore provide a supply of moisture from lower levels, even though there may be a decrease of vapour pressure with height in the free atmosphere. This forced ascent of air means that in moist airstreams, clouds will form over relatively low hills (Figure 9.8). Even in dry airstreams

Figure 9.8 Cloud formation over Great Dunn Fell, northern England. As the air is forced to rise over the hill it expands adiabatically and saturation of the air occurs aloft. The water vapour can then condense to form clouds.

235

Chapter 9 Regional and local climates

clouds will form if mountains are sufficiently high. The mere formation of clouds will not, however, lead to precipitation. Precipitation has to be initiated. This occurs either through the formation of ice crystals in the upper parts of a cloud which then fall through the cloud leading to aggregation of crystals and production of supercooled water (Bergeron process), or by coalescence of smaller droplets onto larger droplets falling more quickly through the cloud. In a convective cloud, the maximum rate of precipitation will be close to the cloud base, as once rain falls out of the cloud the raindrops begin to evaporate. If there are strong updrafts even the raindrops may be transported upwards and, if that is happening, the zone of maximum precipitation may be above the cloud base (Figure 9.9). Thunderstorms can often occur when the atmosphere is very unstable, producing very rapid falls of precipitation. Box 9.2 provides more details of these hazardous features of the atmosphere. In the tropics and subtropics, precipitation is often as a result of convective activity and therefore the highest rainfall totals are found at typically between 1000 and 1500 m, at or just above the average cloud base. This is very

Strong updrafts

common in the trade wind belts where the air above the tradewind inversion is very dry. For example, rainfall on Mauna Loa in Hawaii is over 5500 mm at 700 m but only 440 mm on the summit at 3298 m, well above the inversion (Barry, 1992). In the moister equatorial regions, rainfall generally tends to decrease with height. For example, in equatorial Africa rainfall on mountains above 3000 m is only 10–30% of the highest totals which are observed lower down the mountains. In the middle latitudes, however, precipitation totals increase with altitude above 3000 m. Thus, there are distinct latitudinal differences in the change of precipitation with height in mountains (Figure 9.10). The presence of mountains in the middle latitudes enhances precipitation in a number of ways. The most important effect is that low-level cloud is formed as air is forced to rise over the mountains (Box 9.3). Although convective precipitation can form a significant proportion of the rainfall totals in some mid-latitude locations, much of the precipitation arises from frontal activity associated with depressions. Orographic enhancement through the feeder–seeder mechanism (Box 9.3) can be substantial at warm fronts and in warm sectors, and to a lesser extent with cold fronts. The forced ascent of air over hills and mountains may also intensify vertical motions in depressions and troughs or even trigger conditional instability in polar or Arctic airstreams. The general increase of wind speeds with height also ensures there is a supply of moist air brought in to replace any loss of water content through

3 Gr

Drops break up Drops coalesce

Drops break up

Sp H (km) above mean sea level

Drops coalesce

2

P

M

T

E

1000

1500

1

0

0

500

2000

Annual precipitation (mm)

Figure 9.9 Raindrop formation where there are strong updrafts.

236

Figure 9.10 Precipitation changes with height in different parts of the world: E, equatorial; T, tropical; M, middle latitude; P, polar (Sp, Spitzbergen; Gr, Greenland). Only in the middle latitudes does precipitation increase with altitude over the 3000 m range shown. For equatorial and polar areas precipitation decreases with altitude and for tropical areas precipitation increases to about 1500 m and then declines above this height. (Source: after Lauscher, 1976)

9.2  Altitude and topography

THUNDERSTORMS Thunderstorms form when significant condensation of water vapour occurs, resulting in the production of many water droplets and ice crystals. This happens when the atmosphere is in an unstable condition that supports fast upward motion. Although thunderstorms often happen during warm weather when heating of the ground surface causes sufficient moisture to accumulate in the lower atmosphere, and the warm surface causes there to be a steep adiabatic lapse rate, what is required for thunderstorm formation is an unstable atmosphere (see Box 9.1) through a considerable depth of the atmosphere (possibly right up to the tropopause). The unstable atmosphere means that considerable energy is released as water condenses out of the atmosphere to form the deep cumulonimbus clouds that characterize thunderstorms. That energy can range from the equivalent of a small nuclear bomb (say 10 kt of TNT equivalent) in a small thunderstorm to over 100 times more in a severe thunderstorm. Thunderstorms occur anywhere in the world but are most frequent in the tropics where they can be an almost daily occurrence. Thunderstorms are more common in summer in the mid-latitudes, although winter thunderstorms can occur, due to low-level convergence along a cold front, while at high latitudes thunderstorms are fairly rare, and form only in the summer. At high latitudes in winter, the air is so cold at the surface and throughout the atmosphere that there is insufficient moisture in the air to provide enough energy for a thunderstorm. Thunderstorms are usually accompanied by heavy rainfall, often with strong winds and possibly hail. It is worth noting that large hail (Figure 9.11) (which in supercell storms can be up to 10 cm in diameter, although large hail is more

Figure 9.11 Large hailstones next to an orange golf ball. Such large

hailstones can be very damaging and have been known to smash car windscreens and cause damage to buildings. (Source: Jack Dagley Photography Shutterstock.com)

typically 2–4 cm across) can cause major damage to crops, property and vehicles. NOAA notes that hail causes damage to crops and property amounting to $1 billion per year in the United States. Thunder is caused by the explosive expansion of a narrow column of air which is heated by a lightning discharge. Therefore lightning precedes all thunder. Large cumulonimbus clouds form, often extending to great heights during thunderstorms, although the storm cloud is not normally larger than a few kilometres in diameter. Thunderstorms can be single or multicellular and a series may form a squall line with an associated gust front of strong winds. Supercell storms are severe storms characterized by wind shear with height creating a rotating updraft or mesocyclone. Severe tornadoes are associated with supercell storms (see Box 8.3 in Chapter 8). Lightning occurs when a large charge is built up within a cloud and is then discharged. In tall cumulonimbus clouds electrical charge is built up as water droplets, hail and ice crystals collide with

one another in the strong air movement (a bit like when you rub a balloon on your sweater you can create a charge that, when you put the balloon near your hair, makes your hair stand on end). The positive and negative electrical charges in the cloud separate from one another, the negative charges dropping to the lower part of the cloud and the positive charges staying in the middle and upper parts (Figure 9.12). Positive electrical charges also build upon the ground below. When the difference in the charges becomes large, a flow of electricity occurs in a discharge event. Within cloud lightning is most common but lightning does strike the ground or strikes from the ground to the cloud. A lightning strike occurs in less than a millionth of a second. The temperature of a lightning bolt can be hotter than the surface of the Sun. Although lightning is extremely hot, its short duration means it is not always fatal – around 10–15% of those struck will die. It is uncertain how many people are killed by lightning each year but it could be more than 24 000. In June 2016, more than

BOX 9.2 ➤

237

Chapter 9 Regional and local climates



+

+ + + + + + + + + + + + + + + + + + + + - + - - - - - +

+

-

Rain and hail

+

+

+

+

+

+

+

+

+

+

+

+

+

+

topped metal vehicle with all windows closed may be the next best thing if there are no buildings to shelter inside. If you find yourself in open terrain during a thunderstorm then it is best to avoid areas of open water and avoid isolated tall trees, towers or metal poles. These all act as focal points for lightning strikes (Figure 9.13). If you have nowhere to shelter, make yourself as small a target as possible by crouching down with your feet together, hands on knees and head tucked in. Lightning at the ground is the most common natural cause of forest fires, which can cause considerable damage and present a major hazard to both people and the environment, particularly after long periods of drought.

Figure 9.12 Electrical charge in a thunderstorm.

80 people were injured when lightning struck a music festival in Germany and in India 93 farm workers were killed in one day across Bihar, Uttar Pradesh, Jharkand and Madhya Pradesh from separate lightning strikes. If you experience an approaching thunderstorm and are outdoors you should try to get indoors to a fully enclosed building with wiring and plumbing as these provide the best protection – if lightning strikes then most of the charge will be directed along the wiring or plumbing systems into the Earth. Tents and sheds will not provide suitable protection. Getting inside a hard

precipitation. As a result of orographic enhancement of precipitation, mid-latitude hills and mountains have unusually high annual precipitation totals. In New Zealand, average precipitation totals on the windward side of the Southern Alps can reach 10 000 mm yr-1. As well as enhancing precipitation on the windward side of mountains, there can be substantial reductions of precipitation in lower-lying areas to the lee of the mountains. These ‘rain shadow’ areas are found in many places such as northern Chile (south-east trade wind belt), Patagonia in Argentina and east of the Rocky Mountains in the United States (mid-latitude westerlies) and on a 238

Figure 9.13 Lightning striking a tall building during a thunderstorm.

(Source: Cpaulfell / Shutterstock.com)

BOX 9.2

smaller scale in many hilly locations. This occurs because the air which can now descend down the lee side of the mountains warms adiabatically. This in combination with the earlier loss of moisture through precipitation on the windward and summit parts of the mountains makes the air less saturated and thus less likely to produce precipitation. The other feature of mountains is that precipitation may fall as snow which can accumulate over time. Over the winter substantial amounts of snow can accumulate in mountains. Often snow melt in the spring can produce large river flow peaks downstream even when there is no

9.2  Altitude and topography

precipitation at the time. Precipitation in mountains is therefore not just enhanced but its hydrological impact may occur a number of months (or years) later. Given sufficient depth of snow (many metres) surviving over many summers it is possible for a glacier to form. At present nearly all mid-latitude mountain glaciers are retreating as annual melt is greater than snow accumulation. Despite that retreat, as the glacier surface ice is at 0°C even in midsummer, glaciers have an impact on their local summer climate as they act as heat sinks. Glaciers cool air in contact with the surface and depending on the moisture content of the air they can act as either a local moisture source or sink. If air is dry then water vapour can sublimate (change directly from the solid to gaseous state) into the air above the glacier. If the air is moist then water can condense out

of the air onto the surface of the glacier which is then acting as a moisture sink. Even in mountains where no glaciers exist, snow patches can survive over the summer. The Observatory Gully snow patch on Ben Nevis in the Scottish Highlands is an example of this, having only melted completely a few times over the past 120 years. Although snow may last for many years before melting, the Observatory Gully patch is too shallow to form a glacier. Like glaciers, such snow patches can also modify the microclimate close by. However, this is to a much smaller extent than glaciers. The snow can also have local ecological impacts creating a niche for certain alpine species. Again the local topography is important with snow patches forming and surviving in local shaded depressions or gullies in the mountains.

OROGRAPHIC ENHANCEMENT OF PRECIPITATION Precipitation totals increase with altitude in the tropics (up to around 1500 m) but the increase with altitude is greatest in the middle latitudes. This mid-latitude orographic enhancement of precipitation is related to the feeder–seeder mechanism. As air is forced to rise over the mountains the adiabatic cooling reduces air temperature to the point at which the air becomes saturated with water vapour. Further, ascent leads to the formation of cloud, a common feature of mid-latitude mountains. This cloud acts to increase precipitation falling from higher ‘seeder’ clouds because the lower ‘feeder’ cloud droplets are swept into the precipitation falling through the cloud (Browning and Hill, 1981). This feeder–seeder enhancement of precipitation as illustrated in Figure 9.14 can be substantial at warm fronts and in warm sectors and to a lesser extent with cold fronts. This is most apparent in coastal mountain ranges such as the New Zealand Alps, the Coast Mountains of British Columbia and the western Highlands of Scotland. Orographic enhancement of precipitation is difficult to measure as in the middle

Pre-existing seeder cloud Raindrops Feeder cloud Orographic enhancement

Figure 9.14 Schematic diagram of the feeder–seeder mechanism. The frontal cloud aloft produces precipitation which falls and hits the water droplets in the orographic cloud. These then combine to increase the overall precipitation totals on the mountain/hilltop.

latitudes winter precipitation often falls as snow and the turbulent flow around mountains means that the deposition of snow can be into gullies and hollows. Standard measurements of precipitation using rain gauges are therefore impossible. In addition, orographic enhancement is not constant in all precipitation events. Enhancement tends to be greatest at warm fronts and in warm sectors and therefore as weather events are not distributed evenly through time, there are seasonal differences in enhancement. In addition, even across a relatively narrow mountain

range such as the Scottish Highlands, typically 80 km wide and only up to around 170 km at their widest (from the west coast to the east of the Cairngorms), there is a rapid decline in enhancement from west to east. While enhancement rates can be over 4.5 mm m - 1 in the western Highlands, in the east of the Cairngorms the rate is only of the order of 1.33 mm m - 1 (McClatchey, 1996). This enhancement gives annual totals of just over 2000 mm on the highest tops of the eastern Cairngorms, as compared with 6000 mm in parts of the western Highlands.

BOX 9.3

239

Chapter 9 Regional and local climates

9.2.5 Frost hollows The density of air is inversely proportional to temperature (r a 1/T; density increases as temperature falls). In light winds, mechanical mixing of the air is very limited, and at night there is a lack of convective turbulence that is normally created by solar radiation warming the ground during the day. As a result, the air close to the surface is cooled during the night as the surface temperature drops. This cooling is greatest when the sky is cloud free and the air is dry. As this cooled air is now denser than the air aloft, if it is on a slope, the air can start to move downslope in what is called katabatic drainage. This flow of air is not fast and is not like that of water but more like the flow of something like porridge. Very rarely genuine katabatic winds can occur but this really only happens in the Antarctic when cold air flows off the main ice sheets (which can be at over 3000 m above sea level) down to the coastal ice shelves. In most parts of the world, katabatic drainage is slow and the cold air can pond up behind restrictions in the flow such as walls (on a small scale) and on a larger scale where a wide valley becomes constricted at a lower point in the valley. Anticyclonic conditions are most likely to give rise to stronger katabatic drainage, as winds tend to be light and the sky cloud free (calm and clear conditions). Such nighttime conditions are sometimes called radiation nights as the surface has its maximum radiational loss under such conditions. Anticyclones have a marked temperature inversion aloft formed by the subsidence of air from higher levels and on such radiation nights the surface temperature inversion, formed by the cooling at the surface, can extend up to the anticyclonic inversion aloft. The drainage of cold air into lower-lying areas can give unusually high occurrences of frost and low temperatures in certain locations. Such frost hollows are much more a feature of middle and high latitudes than other locations. This is because low temperatures occur when there is strong radiational cooling at the surface resulting in a surface temperature inversion. As water vapour is a strong absorber of long-wave radiation, a significant amount of which it re-emits down to the surface, the relatively moist air in most subtropical and tropical regions (deserts are the exception) reduces the amount of this surface radiational cooling. Geiger (1965) described an extreme example of local surface cooling at the Gstettneralm sinkhole in Austria where temperature inversions from the bottom to the top (about 150 m) can be over 27°C and extreme minima of below - 50 °C have been recorded in the valley bottom. Low-lying areas with well-drained soils such as 240

sands or gravels, or thin soils on chalk, are more likely to experience increased frequencies of frosts or unusually low temperatures. Outside the tropics and subtropics, the lowest surface air temperatures are, however, almost always recorded when the ground is snow covered as the snow insulates the air from the soil heat flux (Robinson and HendersonSellers, 1999). In addition, the cold air above snow surfaces contains less water vapour and therefore the surface radiational cooling will be greater.

Reflective questions ➤ While an average temperature lapse rate for the troposphere may be around 6.5°C km -1, under what circumstances would the lapse rate in mountains (i) show an increase in temperature with height from the valleys to well up the hillsides; (ii) be close to the dry adiabatic lapse rate; (iii) show a fall in temperature followed by a rise and then a further steep fall in temperature on ascent?

➤ Why is it generally windier on the top of mid-latitude mountains than at lower elevations?

➤ Why is the change of wind speed with height in tropical and subtropical mountains different from that in midlatitude mountains?

➤ Why do rainfall totals in the middle latitudes increase much more greatly with height than those in the tropics?

➤ What conditions are needed for thunderstorms to form and what hazards do they pose?

9.3 Influence of water bodies Unlike land surfaces, water bodies have little diurnal change in surface temperature except in very shallow water close to the water’s edge. Surface temperatures are fairly constant for a number of reasons. Solar radiation is transmitted through water to a considerable depth and is not absorbed at the surface as is the case for land surfaces. The high specific heat of water (the energy required to increase water temperature) means that it requires more energy to be absorbed for any given temperature change than other substances. Furthermore, the surface layers of water bodies tend to be well mixed, which helps spread any temperature change through a substantial depth of water. In addition, energy at the surface is used largely for the latent

9.3  Influence of water bodies

heat needed for evaporation rather than sensible heat that would cause a change in water temperature. Over the land there are much more substantial diurnal changes in air temperature particularly in the summer half-year when solar radiation is stronger. In the middle latitudes, sea surface temperatures (and the air in the layers close to the surface) are therefore cooler than land surfaces during the day in the summer half-year. They are warmer than land surfaces at night. The same is true in the high latitudes but in the winter half-year the temperature of the snow-covered land may remain colder than sea surface temperatures during both day and night. Such differences in local temperature result in sea and land breezes as shown in Figure 9.15. Features like this can also develop where there are large inland bodies of water, such as the Great Lakes in North America, when they are called lake breezes. Sea breezes form only when there are light wind conditions (typically anticyclonic conditions) as the stronger winds of more active systems help reduce land–sea temperature differences through vigorous mixing of air. Summer sea breezes are a feature of many coastal areas and

(a) mb Daytime 950 970 990 1010

Sea

Land

(b) Cumulus develop and migrate seawards

Counter flow

i n fl ow B ou n d a ry of

Lake or sea air

typically have speeds between 2 and 5 m s -1. Sea breezes exist from the surface to 2 km above ground and may penetrate 30 km or more inland (occasionally as far as 100 km). The sea (or lake) breeze brings cooler (occasionally up to 10°C cooler) more humid air inland and a shallow sea breeze front may be evident. Uplift takes place along this front and can trigger the development of cumulus cloud (Figure 9.16). These clouds can be carried seawards by the counterflow present aloft. Occasionally sea breezes from different directions can converge enhancing uplift, which can lead to convective showers. The sea breeze dies off as night falls and is often replaced by a weak land breeze (Figure 9.15). In the introduction to this chapter, the regional change in climate across the mid-west United States was highlighted as an example of how, although within a single climate type, there would be differences in local climate. A further illustration of this can be seen in the temperatures and rainfall of Milwaukee and Madison in Wisconsin, USA (Madison is about 120 km west of Milwaukee). Both are in the same climate type but Milwaukee comes under the influence of Lake Michigan, giving it slightly milder winters and slightly drier summers than Madison. This is illustrated by climate data shown in Figure 9.17. Another feature of middle- and higher-latitude large inland water bodies such as the Great Lakes in North America is lake-effect snow. When cold polar or Arctic air moves across a warmer underlying water body the temperature of the lowest layers of the air is increased, which causes instability and the formation of convective activity. As well as warming the lowest layer of the air, some additional moisture is also added, but the main effect is

yer ary la ound b l a n Inter Sea breeze

Land air

Sea breeze front (c) mb Night-time 950 970 990 1010

Sea

Land

Figure 9.15 Circulation of sea and land breezes for (a) and (b) day time

Figure 9.16 Cumulus clouds develop at the coast where air rises over the

and (c) night-time. (Source: after Robinson and Henderson-Sellers, 1999)

land and migrates seawards. 241

Chapter 9 Regional and local climates

(a) 35 Milwaukee

30

Madison

Temperature (5C)

25 20 15 10 5 0 -5 -10

J

F

M

A

M

J

J

A

S

O

N

D

Month (b) 150

Milwaukee

125 Rainfall (mm)

January or early February when the lake freezes over. The Lake Ontario lake-effect snow season, however, continues into March as it does not freeze completely. As well as occurring over lakes a similar snow effect can be found with polar or Arctic air flow over bays, some more enclosed seas and oceans. Examples include the eastern Black Sea region, the Aegean and Athens, eastern Italy and the Adriatic, and in the United Kingdom cold, fairly dry, easterly air flow from Europe can be warmed and moistened over the North Sea giving snow over eastern England and eastern Scotland. An ocean effect is also seen in the Sea of Japan and can occur in Nova Scotia and Cape Cod and in northern Scotland on the other side of the Atlantic.

Madison

100 75

Reflective questions

50

➤ Why do sea and land breezes form?

25 0

J

F

M

A

M

J

J

A

S

O

N

D

➤ Why is there more snow in the vicinity of some large lakes?

Month Figure 9.17 Mean monthly temperature (a) and precipitation (b) at

Milwaukee (43°N) and Madison (43°N). Milwaukee is close to Lake Michigan and so has milder winters owing to the influence of the warmer water. The lake water in winter is warmer than the surrounding land because it cools more slowly than land. Thus air over the lake becomes warmed by the lake water. The lake also acts as a moisture source and so winters in Milwaukee are wetter than those in Madison, which is too far away from Lake Michigan to be affected.

the relative warming (the air may still be very cold, just not as cold as upwind of the water). The convective activity causes precipitation which, as the air is cold, falls as snow. The convective activity can be enhanced at the edge of lakes as the greater surface friction over the land can cause slowing down of air and hence some convergence which then forces air to rise. Convection can be also enhanced with across-lake flow if there is an upslope (or orographic) effect when the air reaches land. Lake-effect snow results in annual snowfall totals of over 250 cm to the lee of the lakes, and exceeds 500 cm in the Tug Hill Plateau in New York (to the lee of Lake Ontario) and on the Keweenaw Peninsula of northern Michigan (to the lee of Lake Superior). In areas where lakes become frozen (often with snow cover) the lake-effect season ends at this point. For example, at Lake Winnipeg there can be lake-effect snow in November but then the lake freezes and the effect ends. For Lake Erie the lake-effect snow season often ends in late 242

9.4 Human influences 9.4.1 Shelter belts It was noted earlier that coastal areas are subject to stronger winds than inland areas when winds are blowing off the sea. This is a result of reduced friction over the relatively smooth sea surface. This suggests that any alterations in surface roughness have an impact on the local wind. However, unless there is a permanent change to a new surface (as from sea to land) individual roughness elements such as trees or buildings will have an effect for only a relatively short distance downwind. If a line of trees or hedges is planted upwind of a field sown with sensitive crops it is possible to reduce the local wind speeds in the field to provide some protection. The same thing can be done around a garden to provide shelter. The ideal shelter belt is slightly permeable as the lower layer of air acts somewhat like a cushion and extends the reduction in wind speed over a longer distance as shown in Figure 9.18. An impermeable shelter belt will produce a greater reduction in wind speed close to the belt. However, recovery of the wind speed to upwind values is more rapid for impermeable belts, taking place over a distance of about 10–15 times the height of the barrier. Low-density shelter belts have an impact 15–20 times the height of the barrier downwind with

9.4  Human influences

Lee eddy High density

‘Cushion’ Low density

120 Figure 9.19 Shelter belts in The Netherlands protecting fruit trees and

% open wind speed

100

open fields. (Source: Shutterstock.com) 80 High

60 40

Medium

density

9.4.2 Urban climates

Low 20 0 10

5 0 5 10 15 20 25 Distance equivalent to the number of times greater than height of obstacle

30

Figure 9.18 The role of shelter belts in reducing wind speed downwind. Low- and medium-density belts offer better shelter than high-density belts (e.g. walls) for a greater distance away from the belt. This is because the small amount of air flowing through the low-density belts cushions the air flowing over the top of the belt. Without this the air flowing over the top immediately subsides, causing fast-moving eddies to form and thus rendering high-density belts less useful. (Source: after Nägeli, 1946)

medium-density shelter belts having the greatest impact of up to 20–25 times the height of the barrier. The use of shelter belts is best when there is a particular wind direction from which damaging winds come. Figure 9.19 shows an example where a tall tree shelterbelt in the background protects the crops and smaller fruit trees in the foreground from winds coming towards the photographer. For example, in the Rhône Valley of southern France, shelter belts are planted to protect crops from winds that can come down the valley from the north. Snow fences set back from roads and railway lines are also a type of shelter belt as they are used to reduce air flow to allow snow to fall to the ground before the air reaches the road. This keeps roads and rails more clear of snow than would otherwise be the case. In many ski resorts, snow fences are also used to reduce local air flow and help allow snow to accumulate on the pistes.

The fabric of towns and cities substantially alters surface characteristics compared with surrounding rural areas. The urban surface is much rougher than most vegetation. An indication of roughness can be given by what is called the roughness length. This is of the order of 5–20 cm for agricultural crops but up to 10 m for tall buildings. There are also important changes to the radiation and energy fluxes in urban areas. The urban fabric (stonework, road materials, roofs and so on) strongly absorbs solar radiation. There is also a substantial release of energy into the urban atmosphere as a result of humans heating their environment (domestic and industrial). This is especially the case for mid- and high-latitude cities in winter. The urban atmosphere is also affected by air pollution with increased levels of carbon monoxide, oxides of nitrogen and various hydrocarbons. Although the local climatic impact of urban areas is always present, it is reduced in strong wind conditions as the vigorous mixing spreads any impact through a greater depth of the atmosphere and rapidly transports effects away from the urban area. The greatest impact of urban areas on the local climate is therefore found during light wind conditions. While the most important impact of urban areas on climate is linked to air pollution, there are local climate changes in both the temperature and wind regimes experienced by towns and cities. The most commonly discussed climate modification in urban areas is the urban heat island effect. This is so called as the urban area is an ‘island’ of warmer air within the surrounding cooler rural air (Figure 9.20). The urban heat island occurs mainly at night when the urban atmosphere 243

Chapter 9 Regional and local climates

Temperature (°C)

30 29 28 27 26 25

Farmland

Factories

Town center

Town park

Suburbs

Rural/Farmland

Figure 9.20 The theoretical urban heat island effect.

Air temperature

cools more slowly than that in the surrounding rural areas (Figure 9.21). It is more strongly developed in generally light wind conditions with clear skies when long-wave radiational loss is greatest (at night). Oke and East (1971) recorded a maximum heat island of up to 12°C in Montreal in winter, which is unusual for mid-latitude cities where the maximum heat island is normally observed in summer. However, in high-latitude cities the anthropogenic heat losses in winter can be very large (owing to the extra internal heating switched on in offices and homes),

Urban

25C

Heat island intensity

Rural

Urban minus rural temperature

12

18

24

06

12

Time (h) Figure 9.21 Typical temporal variation of urban and rural air temperatures

under clear skies and weak air flow. (Source: after Oke, 1987) 244

leading heat island maxima to occur at that time of year. Even in London, a mid-latitude city with relatively mild winters, anthropogenic heat losses reach over 200 W m-2 in winter, which is greater than typical solar radiation amounts at that time of year. Infrared technology can be attached to aircraft to examine which buildings emit most heat and to determine those areas most in need of additional insulation (Figure 9.22). There can be quite rapid changes in temperatures in moving from urban parks into built-up areas (Jauregi, 1991). Oke (1976) identified two parts to the modification of the urban atmosphere: the urban canopy layer and the urban boundary layer (Figure 9.23). Oke (1987) also suggested how an urban heat island would develop these two layers. In the case of the urban canopy layer he suggested that the following would all play a role: (i) greater absorption or direct solar radiation due to ‘canyon geometry’ (width of the road plus the height of the buildings on either side); (ii) greater daytime heat storage due to properties of urban materials; (iii) anthropogenic heat release from buildings (largely due to heating losses); and (iv) decreased evaporation. In the urban boundary layer, however, he suggested entrainment of air from the canopy layer, anthropogenic heat from roofs and chimneys, and downward flux of sensible heat (see Chapter 6) from the overlying stable layer would be the principal causes of the urban boundary layer heat island. There appear to be critical wind speeds above which the urban heat island disappears. Examples are 12 m s-1 for London, 11 m s-1 for Montreal and 4–7 m s-1 for Reading (England) and these critical speeds are related to size

9.4  Human influences

High

Low Heat loss

Figure 9.22 Heat loss from an urban area detected using infrared sensors on an aircraft. It is possible to see which individual buildings are in need of increased insulation.

Regional wind

Rural

Suburban

Urban boundary layer

Urban canopy

Urban

Suburban

Urban ‘plume’ Rural boundary layer

Rural

Figure 9.23 The urban canopy and boundary layer. The canopy layer con-

sists of the spaces around the local buildings beneath the mean height of the buildings, whereas the boundary layer is the layer affected by the urban environment. (Source: after Oke, 1976)

of the urban area. Even in relatively light wind conditions, the turbulence caused by air flow over buildings is enough to maintain a well-mixed atmosphere (up to an altitude of a few hundred metres in the largest cities). This mixing

establishes an adiabatic lapse rate and as a result a stable layer is formed aloft. The heat island formed in this wellmixed urban boundary layer is strongly related to the city size and types of building in the city and therefore the largest urban boundary layer heat islands are found in cities such as New York (e.g. Gedzelman and Austin, 2003). The failure to differentiate between the local urban canopy layer and the more general urban boundary layer heat islands can lead to erroneous conclusions being drawn in observational studies. In strong winds, urban buildings can create powerful gusts as the wind is forced to flow round tall buildings. In a zone close to tall buildings gusts may reach 2.5 or even 3 times the mean wind speed upwind of the building. The Venturi effect occurs when winds are forced to funnel between two buildings increasing localized wind speeds (Figure 9.24a). These winds can cause difficulty in walking and opening doors and may put severe stress on the buildings. Transverse currents can be generated 245

Chapter 9 Regional and local climates

(a) Venturi effect

(b) Transverse currents

(-)

(+) (+)

Zones of increased wind speed (up to 2.5 to 3 times regional wind)

(+)

+ Increased pressure - Decreased pressure

Figure 9.24 Wind regime around buildings: (a) flow can be funnelled into narrow passages between buildings resulting in the Venturi effect; (b) transverse currents can develop when buildings are at right angles to the wind. (Source: after Thurow, 1983)

when buildings are at right angles to the wind. Here pressure differences between the upwind and downwind sides of the buildings can lead to unexpected strong gusts (Figure 9.24b). Overall, however, large urban areas tend to reduce overall mean wind speeds as their greater roughness slows wind speeds to below their rural upwind values. This modification of wind flow around buildings also plays a role in the dispersion of air pollution and can lead to areas where the local air pollution is unusually high as a result of the trapping of pollution between buildings. Although there is some evidence that convectional rainfall can be enhanced by urban areas (but perhaps downwind of the city), there is no real indication that urban areas have any influence on precipitation events. Therefore, the climate modifications created by urban areas are largely the canopy and urban boundary layer heat islands and the modification of the wind regime. In addition, there is increased air pollution in cities but levels at any one place are related to emissions, dispersion and local geography and not just to any change in the local climate. Urban areas do also have some influence on local atmospheric moisture conditions related to lower evapotranspiration (Deosthali, 2000).

246

9.4.3 Atmospheric pollution and haze Most air pollution events are localized in urban or industrial areas where traffic emissions or pollutants from factories occur. There can be periods when the atmospheric conditions of urban areas are a danger to human health (Figure 9.25). In regions with high solar radiation such

Figure 9.25 Smog over Taipei City, Taiwan. Air quality index levels were

classified as ‘Beyond Index’ (PM 2.5 of over 500 mm per cubic meter). Smog blocks out the sunlight and promotes respiratory and other health problems. (Source: Chen Min Chun / Shutterstock.com)

9.4  Human influences

as Los Angeles, Athens and Mexico City, the ultraviolet radiation reacts with the uncombusted hydrocarbons from vehicle emissions and produces a photochemical smog which irritates the eyes, nose and throat. However, often these conditions are localized and do not persist for long periods. Recently, however, in some places there have been haze pollution events that have lasted months and spread over many hundreds of kilometres (Box 9.4). Many of the large haze pollution events have resulted from forest fires. Forest fires can occur both through natural action and by human intervention. Forest fires in Indonesia are very largely the result of human activity as fire is used to clear land for agricultural purposes. Fire is cheap and, as well as reducing vegetation cover, enriches what are often very poor soils. Indigenous tribes, such as the Dayak people in Kalimantan, have traditionally used shifting cultivation (slash and burn) techniques and their use has been in tune with the natural environment with

strict traditional rules of using fire. Unfortunately, the large number of settlers who came from other islands and new plantation companies do not follow rules that help the long-term maintenance of the environment. Plantation companies (or people hoping to profit from providing services to them) have a particular responsibility as they are largely aware of potential environmental damage and yet place a higher value on their own profits. The Indonesian Government has banned the use of fire for clearing land for a number of years but fires continue to be lit as Indonesia expands its wood pulp, palm oil and rubber industries. In addition to fires, the extensive logging of the rainforests, particularly selective logging, plus other land-use changes have played an important role in making the Indonesian rainforests more susceptible to fire. Rainforests are humid and fires do not naturally take hold. Selective logging and other agricultural land uses open up the forest and allow it to dry out

2015 SOUTH-EAST ASIAN HAZE A major air pollution crisis occurred during June to October 2015 affecting Indonesia, Brunei, Malaysia, Singapore, Thailand, Vietnam, Cambodia and the Phillipines. Fires were started illegally in parts of Indonesia such as Sumatra and Kalimantan by those trying to clear peatland forests for production of palm oil or wood pulp. The smoke caused significant haze across the region (Figure 9.26). Many businesses and schools had to close in Malaysia and Singapore. Transport was affected with air and sea travel affected. Major tourist areas in Singapore and Thailand were affected and life became intolerable in many locations with serious health consequences. For example, there were states of emergency in six Indonesian states with 500 000 people reporting respiratory illnesses, and evacuations of school children from the worst polluted areas. It is estimated that the 2015 fires in Indonesia,

Figure 9.26 Smoke from fires on Sumatra and Borneo, spreading across South-East Asia on

24 September 2015. Image is approximately 1700 km across. (Source: NASA image by Adam Voiland, NASA Earth Observatory, and Jeff Schmaltz, LANCE MODIS Rapid Response)

caused by greedy businesses clearing forest for financial gain, may have cost more

than US$50 billion in economic losses for the South-East Asian region.

BOX 9.4

247

Chapter 9 Regional and local climates

more easily. The forest fires have caused massive damage within Indonesia but, due to the smoke from the fires, damage has also been caused to neighbouring countries, such as Malaysia and Singapore. As well as human activity, natural climate cycles can increase the susceptibility of forests to fires. For example, during El Niño events there tends to be reduced rainfall over Indonesia which leads to drier than usual vegetation. However, fires now occur every year in Indonesian forests and are not simply coincident with El Niño events.

9.5 Summary

Reflective questions ➤ Why is a permeable barrier more effective as a shelter belt than an impermeable barrier?

➤ What is the urban heat island effect? ➤ If the mean wind speeds in cities are slower than in rural areas why are urban wind gusts sometimes faster?

reduction of wind speed close to the surface is much less than over land. Therefore, if the wind direction is from the sea to land,

Local climate variations vary from extremely localized microcli-

coastal areas are subject to stronger winds. In addition, there

mates to more generalized regional climates within a general

can often be substantial differences between air temperatures

climate type. Regional and local geographies are the key factors

over land and sea partly due to the high specific heat of water

in determining the magnitude and importance of these climates.

but also due to the ability of solar radiation to penetrate to

A particular climate classification type can therefore be affected

considerable depths before being fully absorbed. This can lead

more locally by a number of different influences, some of which

to cooling sea breezes in coastal regions particularly in summer

will depend on particular weather conditions or on the time of

(see below). Lakes can also have an influence on the climate

year. The most obvious and rapid climate gradients result from

such as lake-effect snow, but except in the case of very large

changes in altitude. With increasing altitude atmospheric pressure

lakes (such as the North American Great Lakes that behave like

decreases, and associated with this are decreases in temperature.

inland seas) any effect is limited to a very narrow strip around

Precipitation totals often increase substantially with altitude and

the edge of the lake.

the wind regime in hills and mountains can be much more severe

Human activity can also alter climate on a regional and even

than that for nearby low-lying ground. Small topographic features

on a global scale. Examples include the haze pollution caused by

can also be important. At night, radiational cooling takes place,

forest fires, and emissions of CO2 into the atmosphere through

reducing the temperature of the air close to the ground. This

the burning of fossil fuels (see Chapters 6 and 7). However, there

cooler air is slightly denser and tends to move down slopes to

are more localized changes to climate that are a result of human

lower ground and into depressions (this is termed katabatic drain-

activity. The most obvious example is the creation of a distinct

age). There are also microclimate variations introduced by veg-

urban climate in built-up areas. Urban areas affect the urban

etation and these include both forests and deliberately planted

atmosphere and in lighter winds strong heat islands can develop.

shelter belts.

The urban atmosphere can be split into a canopy (between build-

The climate near sea coasts is also modified as a result of the

ing) layer and a boundary (above roof) layer and each will have its

influence of the sea, particularly in middle and high latitudes.

own heat island. Buildings also modify wind flow. Towns and cities

Coastal areas are often subject to stronger winds than more

tend to have lower wind speeds but greater gustiness than sur-

inland areas. Winds over the sea are stronger as the frictional

rounding rural areas.

248

Further reading

Further reading Barry, R.G. (1992) Mountain weather and climate. Routledge, London. Excellent book on topographic controls of local, regional and global climate with a whole section dedicated to case studies. Beniston, M. (2006) Mountain weather and climate: a general overview and a focus on climatic change in the Alps. Hydrobiologia, 562, 3–16. An interesting paper looking at change over time in gradients of rainfall and surface temperature with altitude. Geiger, R., Aron, R.H. and Todhunter, P. (2003) The climate near the ground. Rowman and Littlefield, Oxford.

This is republication of a book from 1965 which is still of great relevance today. It details a whole range of ground– air interactions ranging from soil and vegetation energy balances to the deposition of dew on ponds. It is a very detailed text. Robinson, P.J. and Henderson-Sellers, A. (1999) Contemporary climatology. Pearson Education, Harlow. Chapters 10 and 11 are particularly relevant to the material discussed above. Rosenberg, N.J., Blad, B.L. and Verma, S.B. (1983) Microclimate: The biological environment. John Wiley & Sons, New York. Only Chapter 9 covering the subject of shelter belts is relevant here.

249

PA RT I V

Biogeography and ecology Figure PIV At different levels from the atmosphere, the top of the tree

Part contents

canopy and down into the soil and water bodies there are interactions of species, matter and energy. However, humans can have a major impact on these interactions.

➤ Chapter 10: The biosphere ➤ Chapter 11: Ecosystem processes ➤ Chapter 12: Freshwater ecosystems ➤ Chapter 13: Vegetation and climate change

Part IV Biogeography and ecology

Scope The various features of the biosphere can be associated with all aspects of geography, from the climate system, oceanography, geology, hydrology, social issues and even global tectonics. These interactions result in distinctive regions of plants and animals (biomes) that differ depending upon the controlling variables, and often phase into one another along a gradient of change. Chapter 10 discusses the characteristics of these various biomes. It is not enough, however, simply to describe their features, but it is necessary to explain their nature and form. Chapter 11 is therefore concerned with studying the processes behind the spatial distribution of plants and animals and their change over time. By studying ecosystem processes, we become aware of the very dynamic nature of ecosystems and the checks and balances operating as drivers of ecological change. The planet’s biomes are not static. Closer observation shows that important links between plants, animals and soils are related to processes involving the movement of energy and organic and inorganic materials through the system so that ecosystems are in a constant state of change.

252

Chapter 12 deals with living things in freshwater. Freshwater bodies hold around 6% of the Earth’s species of plants and animals despite covering less than 1% of the Earth’s surface. Factors that control the spatial variability of aquatic life are outlined within Chapter 12 as are the feedback effects between aquatic life, water quality and the geomorphology of freshwater aquatic systems. The biogeographical system also plays an important role in the climate system and there are twoway interactions between climate and vegetation. Chapter 13 considers these interactions and demonstrates how vegetation on Earth is changing in response to climate change. The alteration and fragmentation of major biomes have been features of the human impact on the environment for thousands of years. However, the increasing density of human populations and improvements in technological capabilities put added pressure on the biosphere including freshwater bodies. The requirement of environmental managers to balance the preservation of ecosystems with human needs for the ecosystem necessitates a sound understanding of the major ecological, biogeographical and environmental processes that characterize the living portion of our planet.

C HAPTER 10

The biosphere Julie Peacock School of Geography, University of Leeds

Learning objectives After reading this chapter you should be able to:

➤ describe the organizational hierarchy in ecology ➤ explain what biodiversity is and how it is calculated ➤ summarize the patterns of biodiversity across the biosphere ➤ describe the vegetation of the major terrestrial biomes based on climatic variables

➤ outline the main aquatic biomes ➤ discuss the issues caused by fragmentation of ecosystems

10.1 Introduction The biosphere is the living part of the planet. It stretches from the upper troposphere at an altitude of 15 km, where living microorganisms such as bacteria have been found (DeLeon-Rodrigueza, 2013), down to over 1.5 km below the seafloor, where cells of microbes have been extracted (Roussel, 2008). The majority of the biomass in the biosphere is around the Earth’s surface in the oceans and on land. In the biosphere, life interacts with physical and chemical processes in the atmosphere, water and land with complex feedbacks between the systems (Figure 10.1),

some of which are well known, while other feedbacks are not yet fully understood. Altering one part of the biosphere can have profound consequences for other parts. Therefore, humans have caused substantial impacts by altering the nature of the atmosphere, water and land via pollution, modifications for food supply, and urban or industrial development. It has been proposed that the Earth is now approaching a ‘state shift’ in the biosphere, where it will alter rapidly and irreversibly to a new state (Barnosky et al., 2012). While ecologists try to understand what factors will cause the tipping point for such a transition, scientists who study genes and gene function are trying to develop ‘the biocode,’ the genetic variation in the whole of the biosphere (Davies et al., 2012). Perhaps by 2050 a DNA-based code for most of the biosphere will exist. This will vastly improve our understanding of the interactions within and between ecosystems. With knowledge of the DNA codes, exciting possibilities to reintroduce extinct species arise. However, knowledge of all genetic information seems unlikely when it is thought, for example, that there are millions of microbes yet to be discovered. It is also worth considering reducing our impact on the biosphere now rather than risking trying to recreate extinct species using as yet unproven techniques or face the ethical debate surrounding the revival of ecosystems from DNA coding.

Chapter 10 The biosphere

Atmosphere Precipitation

Insolation Habitat (e.g. microbes)

Alteration of microclimate: transpiration, albedo changes

Geochemical movement and storage

Biosphere

Destructive effects of burrowers and plant roots

Addition of nutrients to soil, alteration of pH balance

Soil nutrients Lithosphere

Habitat (e.g. burrowing animals)

Figure 10.1 Interactions between the atmosphere, biosphere and

lithosphere.

This chapter starts with an overview of key functions and processes that operate in the biosphere with the aim of ensuring that the rest of the chapter then follows from this overview. Many of these processes are dealt with in more detail in Chapter 11, which deals with ecosystem processes. The biomes, which are large ecological areas, with characteristic vegetation usually determined by the climate, will be outlined. This chapter will consider issues caused by fragmentation of ecosystems.

10.2 Biological concepts 10.2.1 What is a species? The biological species concept states that a species is a group of individuals, which has the potential to interbreed in nature and produce fertile offspring, and cannot reproduce with other groups. Therefore, species are not defined by their appearance, although this may help in identifying species. Appearance can cause confusion. For example, in 2001, zoologists discovered a new species of salamander in the hills of Mexico (Parra-Olea and Wake, 2001). The soil dwelling salamander appeared completely identical to a salamander found hundreds of miles away. However, their DNA showed they were not descended from the same 254

species, and in addition they did not reproduce with the other salamander species. The definition of a species may at first seem quite straightforward. However, there are a number of complications. For example, in nature some plants and animals will form hybrids. When this happens separate groups interbreed and the offspring will be fertile, leaving an unresolved question of how to classify these groups. Perhaps more difficult to determine are species definitions for microbes that do not undertake sexual reproduction, but reproduce asexually so that offspring arise from a single organism. Defining a species as a group of interbreeding individuals cannot really be applied to organisms that reproduce mainly or exclusively asexually. Thus there are still major questions about defining species but the biological species concept is at the very least a reasonable working definition, which will be used in this chapter.

10.2.2 The naming of species In Chapters 10 and 11, species will be identified by their Latin binomial, which is conventionally written in italics, although their common names will be given as well. This system of naming was originally developed by Carolus Linnaeus in the 1700s, and is still the accepted method of naming species used by scientists today. It uses a combination of two names. The first is the genus name and the second the specific epithet. For example, for Homo sapiens, Homo is the genus name and sapiens is the specific epithet. Why not just use a common name that is easy to remember and say? The reason is that the Latin binomial can be understood universally. A species with a large geographic range may have different common names in different areas. Taraxacum officinale, for example, is known as dandelion, blowball, puffball, cankerwort, milk-witch and yellow-gowan in different places. Further, some species may have different common names depending on their life stage and some common names can lead to confusion. For example, the flying fox is actually a species of bat and a mantis shrimp is neither a mantis nor a shrimp. The science of the classification of species is known as taxonomy. Species are classified hierarchically, with two examples given in Table 10.1. The left-hand column in Table 10.1 lists the hierarchical classification levels from the highest level at the top of the table (domain) to lowest level at the bottom (species). When an organism is said to belong to a taxonomic group this may refer to any level from kingdom to genus, because it means a group of related species. A debate remains about how to divide life, particularly at the higher levels of classification. However, many will divide

10.2  Biological concepts

Table 10.1 The classification of sunflower and humans based on the hierarchical system developed by Linnaeus

Table 10.2 Levels of organization in ecology Cells

Domain

Eukaryota

Eukaryota

Kingdom

Plantae

Animalia

Phylum

Tracheophyta

Chordata

Class

Magnoliopsida

Mammalia

Order

Asterales

Primates

Family

Asteraceae

Hominidae

Genus

Helianthus

Homo

Species (common name)

Helianthus annuus (sunflower)

Homo sapiens (human)

it into three domains – eukaryota, bacteria and archaea – with the first being multi-celled organisms and second two being single celled organisms. Eukaryota are usually divided into four kingdoms – Plantae, Animalia, Fungi and Protista. The Plantae, Animalia and Fungi kingdoms are all very well defined, with all the organisms placed within them descended from a single ancestor. For example, all the organisms in Plantae are plants that have evolved from a single ancestral plant; all the organisms in the Kingdom Animalia are animals descended from a single animal ancestor; and all the organisms in the Fungi Kingdom are descended from one fungus. The Kingdom Protista, however, contains organisms that are now known to be related more closely to organisms in the other Kingdoms than they are to each other. Grouping into the Protista Kingdom was based partly on the absence of certain characteristics such as cell differentiation not being extensive and a lack of complex development from embryos. However, it is likely that in the future this Kingdom will be separated into several smaller kingdoms. With only an estimated 14% of the Earth’s 8.7 million eukaryotic species identified (Mora et al., 2011) there may be a need to change classification schemes in the future and there is still a huge task of species identification to be undertaken.

10.2.3 Levels of organization It is worth considering the different levels of organization that there are within and beyond the biosphere (Table 10.2). An organism is the smallest unit within the ecological hierarchy. Researchers will often focus on one level of organization. For example, a study of ‘What triggers flowering time in Silene dioica (red campion)?’ is focused at the organism level; while a study to answer ‘How does pond size effect the diversity of freshwater invertebrates within it?’ focuses on the community level;

Tissues Organs Organization levels of interest in ecology

Organisms

An individual

Populations

A group of individuals from the same species, occupying the same area

Communities

Communities are the different groups of populations that interact together

Ecosystems

All the communities in an area and all the abiotic factors that affect it

Biome

A group of connecting ecosystems

Biosphere

All the biomes together

Earth Solar System

and a study examining ‘What is the ability of tropical forests to store carbon?’ is focused at the biome level. These levels of organization are not scale-dependant and although they are extremely useful in understanding ecology and framing research questions the levels of organization do not hold when you consider that there is an entire ecosystem within a human gut.

10.2.4 Biodiversity The key feature of the biosphere is life. Perhaps life’s most remarkable characteristic is its diversity. Biological diversity, shortened to biodiversity, is a term widely reported by the media, perhaps without many readers fully understanding its meaning. At the 1992 convention on Biological Diversity, Rio de Janeiro, biodiversity was defined as, ‘the biological diversity within species, between species, and of ecosystems’. So there are in fact three levels of biodiversity: genetic diversity – variation in individual genetics within a population and between populations; ecosystem diversity, the variety of ecosystems within the biosphere; and species diversity, which is usually what the media refers to, and is the variety of species in a given area. Biodiversity is important as it provides a wide range of products and resources from different food crops, to medicine, to rubber and building materials. Biodiversity has intangible value to people for recreation, aesthetic and cultural reasons. In addition, biodiversity has been shown to increase 255

Chapter 10 The biosphere

ecosystem resilience, making them more stable when faced with environmental stressors (Tilman et al., 1996; Naeem and Li, 1997). Biodiversity also ensures that a wide range of ecosystem services, such as carbon storage, water resources and pollination for food production, can be provided. Biodiversity is made up of two components, namely richness and evenness. When considering species, biodiversity richness is simply the number of species present, known as species richness. Many important theories in ecology including island biogeography theory (see Box 11.3 in Chapter 11) and the keystone concept (see Box 11.1 in Chapter 11) are based on species richness. Evenness takes into account the abundance of the different species present. Where the total number of individuals is spread equally between species in an area, the area has a greater evenness than one in which a particular species dominates and the other species have just a couple of individuals. An area with the greatest biodiversity has high species richness and high evenness scores. An example of this is shown in Table 10.3. There are a number of techniques that quantify diversity which take account of both richness and evenness, and some of these are described in Box 10.1.

MEASURING DIVERSITY Species richness, the number of species in an area, is a basic measure of diversity, but it does not account for the evenness of species in that area. In addition, species richness can also be difficult to estimate accurately if only small areas of the study region are sampled and also when there is extremely high diversity (see Gotelli and Chao, 2013 for a full discussion). Diversity indices are commonly used as a measure of the biodiversity of an area. They can be used to compare the species diversity of similar sites, or of the same site over time. Diversity can be calculated using the number of individuals of each species, but where this is difficult to accurately measure in the field

Table 10.3 Examples of biodiversity at two sites. Site 1 and 2 have the same species richness (5) and the same number of individuals (100). However, because the individuals in site 1 are more evenly distributed than site 2, where Lasius flavus dominates, site 1 is considered to be more biodiverse

Numbers of individuals Invertebrate Species

Site 1

Site 2

Glomeris marginata (millipede)

21

2

Philoscia muscorum (common woodlouse)

23

9

Trichoniscus pusillus (small woodlouse)

18

7

Stenobothrus lineatus (stripe-winged grasshopper)

12

1

Lasius flavus (yellow meadow ant)

26

81

100

100

Total

Whittaker (1970) introduced the concept of explaining biodiversity at different spatial scales and termed the diversity at different scales, alpha, beta and gamma diversity

then the percentage cover of plants, or biomass can also be used. Diversity can be calculated for any taxonomic group, so the focus could be on genera rather than species, if that was appropriate for the study. The most popular measure of species diversity used in ecology, which takes into account richness and evenness, is the Shannon–Weiner Index. This index is also known as Shannon’s Index, the Shannon–Weaver Index or the Shannon Entropy (it is a measure of entropy). The Shannon–Weiner Index is widely used in other disciplines including Information Theory, Computer Science, Chemistry and Physics. The Shannon–Weiner Index makes no assumptions about data being used and all species are weighted

proportionally to their frequencies in the sample (unlike other indices used which give extra weighting to rare or common species). Some of the advantages of the Shannon–Weiner Index are that it is relatively easy to calculate and fairly sensitive to site differences. However, there can be instances where it produces very similar results for sites which are in fact quite different. The formula for the Shannon–Weiner Index (H′) is: s

H′ = - a [pi ln(pi)]

(10.1)

1

where a is the symbol for ‘sum of’, S is the number of species, pi is the proportion of each species (the ith species), and ln is the natural logarithm. Equation (10.1) can

BOX 10.1 ➤

256

10.2  Biological concepts

➤ Table 10.4 Field data with species information and abundances from two woodlands

Woodland 1

Woodland 2

Quercus robur (Oak)

4

0

Fraxinus excelsior (Ash)

7

2

Betula pendula (Silver Birch)

6

23

Acer pseudoplatanus (Sycamore)

7

2

Fagus sylvatica (Beech)

1

3

25

30

Total individuals

be expanded so that it is easier to understand how it is calculated: ni ni H′ = - a J ¢ ≤ * ln¢ ≤ R N N

(10.2)

The symbols in (10.2) are the same as in (10.1) but in addition N is the total number of individuals, and ni is the number of individuals of each species (the ith species). As an example, Table 10.4 provides

data collected from the same sized area in two woodlands. Note the number of individuals and the presence or absence of species in each wood are different. Table 10.5 shows how to calculate H′ for these woodlands. H′ is 1.48 for woodland 1 and 0.81 for woodland 2. Therefore biodiversity is higher in woodland 1 than in woodland 2. It is important to remember two things about results from the Shannon– Weiner Index. First, the results are relative values, so you can say that one is higher than another, but not by how much. Second, they are not diversity values, but indices of diversity and the results cannot be compared with results of other indices of diversity such as the Simpson’s Index.

Table 10.5 Calculating the Shannon–Weiner Index (H′) for woodlands 1 and 2. Note that as no individuals of Quercus robur (Oak) were found for woodland 2 the data fields are left blank

i

Number of individual trees (ni)

ni N

ln¢

ni N



ni ni * ln¢ ≤ N N

-a

Woodland 1 Oak

1

n1

4

0.16

- 1.83

- 0.29

Ash

2

n2

7

0.28

- 1.27

- 0.36

Silver Birch

3

n3

6

0.24

- 1.43

- 0.34

Sycamore

4

n4

7

0.28

- 1.27

- 0.36

Beech

5

n5

1

0.04

- 3.22

- 0.13

Sum

ni ni a J ¢ N ≤ * ln¢ N ≤ R = - 1.48

N = a ni = 25

1.48

Woodland 2 Oak

1

n1

0

Ash

2

n2

2

0.07

- 2.66

- 0.19

Silver Birch

3

n3

23

0.77

- 0.26

- 0.20

Sycamore

4

n4

2

0.07

- 2.66

- 0.19

Beech

5

n5

3

0.10

- 2.30

- 0.23

Sum

N = a ni = 30

ni ni a J ¢ N ≤ * ln¢ N ≤ R = - 0.81

0.81

BOX 10.1

257

Chapter 10 The biosphere

Woodland B

Woodland C

Woodland A

Woodland Woodland A

Alpha diversty 4

Species present

Woodland B

8

Woodland C

5

Gamma diversity 11

Diversity of woodlands A, B and C combined

A compared to B A compared to C B compared to C

Unique Species 8 9 6

Species Overlap 2 0 4

Figure 10.2 Species richness in three woodlands demonstrating alpha, beta and gamma diversity. Each symbol represents a different species, recording

only presence, rather than the number of species. Woodlands A and C have the greatest beta diversity. For gamma diversity there are three woodland habitats with a total species diversity of 11.

(Figure 10.2). Alpha diversity is the species richness at a particular site. Beta diversity is a measure of the variation of species between two sites and considers how many species are present at both sites and how many species are unique to just one of the sites. Studies of beta diversity have increased dramatically recently, and there has been much discussion as to the detailed definition of beta diversity (Anderson et al., 2011). Finally, gamma diversity is concerned with diversity at a landscape level and is basically the species richness of the entire landscape.

Reflective questions ➤ Why are there issues with the commonly recognized definition of a species?

➤ Why is it important to use the Latin binomial when using species names?

➤ What are the components of biodiversity?

258

10.3 Patterns of distribution There are three factors which determine where species are found: biotic (living), abiotic (non-living) and historical. Below we cover how these factors interact to affect spatial patterns of species distributions.

10.3.1 Potential species distributions All species have optimum levels at which growth and reproductive capacity is greatest for any physical variable. At the edge of a species distribution there are threshold levels beyond which an organism’s metabolism cannot be sustained. The most important physical limiting factors are temperature and moisture availability. In general terms, moisture and temperature control what plants and animals can survive at a particular location and therefore control the global distribution of biomes. Temperature is so important because enzymes, which are catalysts of biological metabolic reactions, have adapted to particular

10.3  Patterns of distribution

PHILIPPINES

South China Sea

Andaman Sea

Celebes Sea

THAILAND

Pacific Ocean

PAPUA NEW GUINEA

Halmahera I.

BRUNEI Irian Jaya

MALAYSIA

Sumatra N

MALAYSIA SULAWESI

SINGAPORE

BORNEO

Medan

Banda Sea

Lesser Sunda Is.

Timor

Bali Java

300 miles 300 km

Indian Ocean

Timor Sea

Lombok Wallace’s line

AUSTRALIA

Figure 10.3 Wallace’s Line. This line marks a distinct change in biogeography and is coincident with an ocean trench.

environmental conditions. If the temperature becomes too hot or too cold, the enzyme function will reduce. When the threshold level is reached the enzyme will denature (its structure alters or breaks up) and so can no longer work. Water is fundamentally important for all life. Water transfers substances between cells of living things and their external environment. Most plants would die if their cellular water content drops below 50%. These physical threshold levels of temperature and moisture availability define the potential range or distribution of a species, determining where a species can exist. Potential distributions depend on the ability of species to tolerate physical stress, with stress defined as ‘any deficiency in light, moisture, nutrients, and optimum temperature’ (Grimes, 1979). It is important to remember that organisms do not experience annual means of temperature and precipitation nor monthly means, but seasonality and extremes including frost, heat or drought.

10.3.2 Actual species distributions Abiotic factors, particularly temperature and moisture, set the limits of the potential range of a species. However, biotic and historical factors limit where that species does exist, its actual species distribution. Biotic factors, including competition for resources are discussed in Chapter 11. Species distributions are also limited by barriers, such as oceans, mountains and rivers, preventing dispersal to suitable habitat on the other side. A barrier may even no longer exist, but still have an effect as species disperse slowly. This is known as the lag effect. Such distribution limits are called historical limits. Rather than being determined by the contemporary environment, these limits are a result of past environments. A classic example of such

a limit is the so-called Wallace Line that starts between the islands of Bali and Lombok and extends between Borneo and Sulawesi (Figure 10.3). Alfred Russel Wallace noticed on an expedition in the 1850s that the fauna on the island of Bali represented the Asian continent whereas the fauna on Lombok, just 30 km away, much more closely resembled that belonging to Australia. What was unknown to Wallace is that this imaginary line corresponds approximately with a deep ocean trench separating the continental plates of Asia and Australia, and it was not until Alfred Wegener developed the theory of continental drift in 1912 and then in the 1960s when the theory of plate tectonics was developed (Chapter 2) that this could be explained. The species were from land on two separate plates which had been separated for more than 200 million years but which had now moved closer together around the Wallace Line. The strong distinction between both sides of the Wallace Line remains for both plants and animals, though it is much more distinctive for animals for which migration over even short stretches of water may be impossible. Using transplant experiments, either deliberately or semi-naturally, is one way to try to disentangle the different factors to understand what is limiting the distribution of a species. When Sciurus carolinensis (grey squirrel) was ‘successfully’ introduced into Europe from North America it was clear that it had been historical factors (the Atlantic Ocean) that had previously limited its distribution. If transplantation experiments are not successful, it is possible that either abiotic or biotic factors are limiting the distribution of a species. Removal investigations can be used in conjunction with transplantation. For example, when transplanted beyond its northern range limit, Chthamalus fragilis (small barnacle) was shown to survive, but only 259

Chapter 10 The biosphere

if Semibalanus balanoides (acorn barnacle) had been removed (Wethey, 2002). This shows that competition, a biotic factor, was limiting the range of Chthamalus fragilis.

10.3.3 Spatial patterns in biodiversity Understanding the patterns of species diversity is important for issues of conservation management, managing invasive species and understanding the impact of climate change on biodiversity (see also Chapter 13). In ecology it is usually true that the larger an area is, the greater the species richness (Preston, 1962a, 1962b; Rosenzweig, 1995). Studies of species–area relationships have usually focused on a particular type of organism, such as birds or butterflies, or species that perform a particular role in the community. The species–area relationship may seem obvious, as a larger area is likely to have more resources available to species but it is important and has implications for conservation, and it can be used to help plan appropriate sizes of protected areas. However, possibly the clearest spatial pattern regarding species richness is that between richness and latitude, with richness being greater at the tropics and declining towards the poles. Figure 10.4 shows species richness for five degree

latitudinal bands plotted for mammals, amphibians and threatened bird species. The figure also shows the land area for each band, and although this is clearly linked to the pattern in richness, changes in latitude account for a lot more of the difference. High temperatures and moist conditions in the tropics provide good conditions for photosynthesis, so there will be more net primary productivity (amount of energy produced by photosynthetic organisms) (Figure 10.5), which in turn is thought to increase species richness. Gillman et al. (2015) found a negative relationship between latitude and annual net primary productivity of forests using several datasets. They also found that vascular plant richness was positively correlated with net primary productivity (Figure 10.6), although for some areas in the tropics there may be extremely small potential primary productivity due to low rainfall. In addition to the global patterns of species richness, it is important to understand how richness is modified at regional scales. Soils are particularly important, especially in relation to rooting depth. For example, in Peru where tropical rainforest with a high species diversity is typically found, small patches of savanna with lower species diversity can also be found where the soil is much shallower and cannot support rainforest.

Mammals (n = 4,734)

Land area

905N

Amphibians (n = 5,743)

Threatened birds (n = 1,208)

905N

805N

805N

705N

705N

605N

605N

505N

505N

405N

405N

305N

305N

205N

205N

105N

105N

05N

05N

105S

105S

205S

205S

305S

305S

405S

405S

505S

505S

605S

605S

705S

705S

805S

805S 905S

905S 0

5000

10 000

Area (10 000 km2)

0

1000

2000 0

500 1000 1500

0

100 200 300

Number of species

Figure 10.4 Variation in species richness with latitude for terrestrial mammals, amphibians and threatened bird species. (Source: from Millenium

Ecosystem Assessment, 2005) 260

10.4  Terrestrial biomes

0—2.5 2.5—5 5—10 10—15 15—20 20—25 (t ha–1 yr–1)

Figure 10.5 The potential primary productivity of the Earth. (Source: from Schultz, 1995, after Leith, 1964)

1500 Annual net primary productivity (g C m-2 yr-1)

Vascular plant species richness

4000

3000

2000

1000

1000

500

0

0 0 (a)

500 1000 Annual net primary productivity (g C m-2 yr-1)

0

1500

20 40 60 Absolute latitude (Degrees)

(b)

Figure 10.6 Spatial patterns in vascular plant species richness using data combined from different forest stud-

ies: (a) the relationship between vascular plant species richness and net primary productivity; (b) the relationship between latitude and total net primary productivity. (Source: Gillman et al., 2015)

10.4 Terrestrial biomes Reflective questions ➤ Define what a potential species distribution is. ➤ What are the main patterns of species richness found globally?

Biomes are the Earth’s major ecosystems. Biomes are usually defined by the vegetation that characterizes them. Although the hydrosphere makes up around 70% of the biosphere, over 90% of the biomass is found in the terrestrial area, and around 97% of the biomass on 261

Chapter 10 The biosphere

-15 Arctic–alpine -10 Tundra etc. -5 Cold temperate Taiga 5

10

Woodland

0

Temperate rainforest

50

Warm temperate

(Savanna)

b

25

30

and

20

Temperate forest

(Grassland)

Thorn scru

Semi-desert

15

odl Wo

Shrubland

Mean annual temperature (5C)

0

100

Tropical

Tropical seasonal forest

150

200

Tropical rainforest

250

300

350

400

450

Mean annual precipitation (cm) Figure 10.7 The distribution of the major terrestrial biomes with respect to mean annual precipitation and temperature. Within regions delimited by the

dashed lines a number of factors, including proximity to oceans, seasonality of drought and human land use, may affect the biome type that develops. Boundaries are, of course, approximate. (Source: after Whittaker, 1975)

land is made up of plants. The major climatic variables determine where the different species assemblages can survive, and adaptations to environmental variables are part of any biome’s character. Whittaker (1975) classified the major terrestrial biomes with respect to mean annual temperature and precipitation as shown in Figure 10.7. However, where these biomes occur is not just related to latitude but also varies with altitude as this also controls temperature and precipitation. This altitudinal pattern in biodiversity was first noted by Alexander von Humboldt and Aimé Bonpland in 1799 after an expedition to the Andes. The global distribution of the biomes can be seen in Figure 10.8. In reality, the distinction between biomes is not usually abrupt but occurs across an environmental gradient known as an ecotone. Ecotones are transitional areas, which have species representatives from both

262

neighbouring biomes and sometimes have distinct species of their own. Species often reach the edge of their physiological range in an ecotone and so these areas may well be greatly affected by environmental change. Ecotones can sometimes be more diverse than the surrounding biomes.

10.4.1 Equatorial and tropical forests Temperature: The temperature of tropical forests is uniformly high and ranges from 20°C to 35°C with little seasonal difference in temperature. Precipitation: Rainfall is high and annually tends to exceed 2000 mm. Distribution: These forests are found near the equator, ranging from 23.5°N to 23.5°S.

10.4  Terrestrial biomes

TROPICAL BIOMES Tropical forests Savanna region Hot deserts TEMPERATE BIOMES Mediterranean chaparral Temperate evergreen Temperate deciduous woodland Temperate grassland COLD BIOMES Taiga Tundra Ice/Mountains

Figure 10.8 The distribution of the world’s major biomes.

Important species: Tropical forests are composed of species rich, dense forests with tall, broadleaved trees competing for light (Figure 10.9). Moisture and temperature are ideal for growth and light is often the limiting factor. The forest canopy typically reaches a height of 25–35 m with a few emergent trees reaching heights of 50 m. The advantage of the additional height is the access to light for photosynthesis. Below the main canopy there may be a second (about 15 m) and even a third canopy level (about 5 m), which consists of immature trees, ready for a branch or tree fall in the upper canopy into which they can quickly grow and win the competition for light. If there are larger disturbances in the forest creating bigger gaps, pioneer species that germinate quickly from seed may fill the gaps and will gradually be replaced by slower growing, old growth trees as small gaps appear again. Lianas, which are climbing plants, are important within the tropical forest. They need structural support and so grow up trees. Once they reach the top of the canopy they grow across it, ensuring their leaves capture as much light as possible. If a tree falls, lianas may pull down several other trees as they are tangled together. Epiphytes (plants that grow on another plant) are a common feature of rainforests: orchids, bromeliads, ferns and mosses are all adapted to grow high in the branches of trees. The rainforest floor is very dark. By growing higher up in the trees epiphytes are more likely to get the light they need. The forest soils are deep, yet infertile and there is a rapid cycling of nutrients for plant growth, from leaf or branch fall. This means there is not a build-up of leaf litter in the soil.

(a)

(b) Figure 10.9 Rainforest, Queensland, Australia, showing the upper (a) and

lower (b) canopies. (Source: Treasure Dragon / Shutterstock.com) 263

Chapter 10 The biosphere

The variety of plant species and structure within the tropical forests enables a huge diversity in animal and birds. Birds tend to be fruit eaters, rather than seed eaters. In tropical forests it is easier to hear the birds than to see them. Animals may stick to one level within the forests, some never coming onto the forest floor. For this reason they are adapted to tree climbing, rather than walking and have adaptations such as suction padded toes (Figure 10.10), grasping feet and prehensile tails. Threats: This ecosystem is under threat as it is logged, cleared for mining, agriculture and urban development. Recently there have been many roads built across tropical forests which greatly fragment the remaining habitat. Between 1990 and 2015, the global tropical forest area decreased by 9.9% (RBG Kew, 2016). There is some positive news though, as for the last few years zero-deforestation agreements have been signed by major corporations who were previously clear-felling areas for commodities such as soya, palm oil and beef. However, not all these agreements have been adhered to, but it is a positive step for this greatly reduced biome. Figure 10.10 A rainforest lizard with suction padded toes.

(Source: Nicholas Berry and Despina Psarra)

STUDYING A RAINFOREST Understanding the vast, complex, biodiverse tropical rainforest is clearly of huge importance, but how do scientists study such a large and complex biome? The Amazon Forest Inventory Network (RAINFOR) was set up in 2000 to gain a greater understanding of the dynamics of Amazon ecosystems and is coordinated by Professor Oliver Phillips at the University of Leeds. It is an international collaboration working with over 150 researchers. The network now has over 750 permanent forest sample plots across Amazonia. The typical size of these is one hectare. Each plot is usually sampled every three to four years, with all trees of a diameter greater than 100 mm having their diameter and height recorded (Figure 10.11), along with species names, details of the condition of the tree, and whether it has any associated lianas and

264

Figure 10.11 Researchers measuring the diameter of a rainforest tree in Bolivia.

(Source: Roel Brienen, University of Leeds)

BOX 10.2 ➤

10.4  Terrestrial biomes

➤ crown illumination. Crown illumination is a measure of how much of the tree crown is exposed to direct light. Although the measurements appear straightforward, it is critical that consistent methods are used across the network so these plots can be compared accurately. Field campaigns may last for several months as researchers measure plots in a particular area. Field teams combine local experts and international scientists. Developing and sharing the knowledge of all in the network is key. Excellent communication is needed because researchers are based around the world and use different languages and have different cultures. The data built up over years, with some plots having been

studied for 40 years, enables researchers to understand the ecology of the forests, and determine how the forests are changing in response to global climate change. Following the 2005 Amazon drought, the network was able to gain real insight into responses to global change through the data analyzed. Usually, tropical forests store carbon, slowing climate change by absorbing some of the carbon from the atmosphere. When conditions for growth are not favourable the forests can emit carbon, which is what happened during the 2005 drought (Phillips et al., 2009). As the forests are so vast, these drought emissions can have a huge impact on world carbon emissions.

AfriTRON (the African Tropical Rainforest Observation Network), coordinated by Professor Simon Lewis of the University of Leeds, was set up more recently following the same protocols as RAINFOR. It enables comparisons between the forests on the two continents and provides much greater information of the biome as a whole. With both networks, there is huge international scientific effort and collaboration. Scientists are sharing unpublished data which has taken time and energy to collect, often in very difficult field conditions. Therefore, careful data sharing agreements need to be made. Publishing science is very important to researchers, so data sharing on this scale requires trust and cooperation.

BOX 10.2

10.4.2 Savanna

Distribution: Equatorial and subtropical regions.

Temperature: The temperature is similar to that of a tropical forest with a mean monthly temperature above 18°C in all months. The temperature range in winter is usually 18–25°C and in summer 25–30°C.

Important species: The savanna biome is characterized by clumps of grasses surrounded by bare ground and occasional trees and shrubs (Figure 10.12). Typical species include Chloris gayana (Rhodes grass), Themeda triandra (red oats grass) and Pennisetum purpureum (elephant grass), which can grow up to 3 m tall. Grasses can remain dormant throughout the dry season, to aid water conservation, which therefore enables them to

Precipitation: Savannas have a mean annual rainfall of 500–1200 mm, which falls seasonally. The long dry season is characterized by rainfall which may be less than 100 mm over a 5–9 month period.

Figure 10.12 An acacia tree among short grasses and bare ground in the savanna of Botswana. (Source: Artush, Shutterstock.com)

265

Chapter 10 The biosphere

survive. A few tree species have evolved to survive the drought of the savanna, such as the evergreen Acacia tortilis (umbrella thorn acacia tree), which has a long tap root to access groundwater, and trees in the Adansonia genus (baobab trees), which survive despite having a shallow rooting system, partly by shedding leaves in the dry season to reduce water loss and by storing water in their stems (Chapotin et al., 2006). Stem succulent trees are also prominent in arid tropical regions. The storing of water enables the plant to maintain turgor pressure in its cells (which maintains the rigidity of the plant and stops it wilting) during the dry period, rather than to support further growth and photosynthesis (Chapotin et al., 2006). In addition to drought, plants have evolved to withstand grazing and browsing. With more than 40 different large herbivores in savannas (Owen-Smith, 1982), the dynamic characteristics of savannas are in part due to the impact of these animals (Sinclair and Norton-Griffiths 1979). Most herbivores both browse and graze, but there are exclusive grazers, such as zebra, and exclusive browsers, including giraffes (Bergström, 1992). Grasses are protected from grazing as their apical and axillary buds are near the base of the plant, enabling them to regrow. Plants protect themselves from browsing with tough leaves and spines on stems. Threats: As with other biomes, the dominant threats are conversion to agriculture, and expanding urban areas which destroy or fragment the habitat. Overstocking of domestic grazing animals can damage the ecosystem as well; animals will compact the soils with their hooves and high levels of grazing means plants cannot recover and regenerate and species will be lost from the area. A model version of this biome has been created as part of research and visitor attractions under glass (see Box 10.3).

reaching the ground. Fog and dew, however, can provide important additional water. Distribution: Deserts cover about one fifth of the Earth’s surface, with hot deserts found from 20° to 35° north and south of the equator. Important species: Soils are freely draining and coarsely textured as smaller particles are blown away. Plants are sparse with a low profile and have extremely reduced leaves to protect against water loss (see Figure 21.9 in Chapter 21). Cacti species have evolved spines and photosynthesis takes place on the stems. Many plant species only open their stomata (openings in leaves for gas exchange) at night to reduce water loss. Physiological adaptations include using the alternative photosynthetic pathways, C4 and CAM (see Section 13.2.1 in Chapter 13). Oases, where water from natural springs reach the surface, can result in denser plant cover and some palm species can survive here. Animals of the hot desert tend to be nocturnal to avoid the very high temperatures during the day. They are predominately burrowing animals, with burrows providing further protection from the extremes in temperature. Insects, arachnids, reptiles and birds can also survive in the desert and the sparse plants provide shelter for them. This results in greater organic matter in the soil around the base of plants. Threats: Global warming poses a threat to hot deserts as extended and prolonged drought may cause seasonal water pools to dry up. Hot deserts face destruction of habitat for mining and for oil and gas extraction which may cause irreversible damage. Irrigation for agriculture may lead to salt levels becoming too high for plant growth (see Chapter 21). The increasing use of off-road vehicles can also damage the fragile desert environment.

10.4.4 Mediterranean-type biome 10.4.3 Hot Desert Temperature: Mean annual temperatures range from 20 to 25°C. Extreme high temperatures can reach 49°C. At night it can become very cold, with minimum temperatures reaching as low as - 18°C. Precipitation: Deserts occur where rainfall is less than 500 mm annually. In hot deserts, such as the Atacama Desert in Chile, mean annual precipitation is less than 15 mm. Evaporation rates regularly exceed rainfall rates. Sometimes rain starts falling and evaporates before

266

Temperature: This temperate biome has hot, dry summers, with mean temperatures of around 30°C and highs over 40°C, and mild, wet winters with temperatures of 10–12°C. Precipitation: The majority of the rainfall occurs in the winter and ranges from 300 mm to 500 mm annually. The dry season is in the summer. Distribution: This biome occupies five separate areas on the west coasts of continents in the mid-latitude region. The area around the Mediterranean Sea is by far the

10.4  Terrestrial biomes

largest of this type of biome. The four smaller areas are in California, Chile, the South Western Cape Province of South Africa and south-western and southern Australia. The Mediterranean biome has different local names including the garigue and maquis in France, matorral in Spain and Chile, Chaparral in North America and fynbos in South Africa. Important species: The Mediterranean biome has high levels of plant species richness and endemism, greater than the combined floras of tropical Africa and Asia (Cowling et al., 1996). Plant species of Mediterranean biomes are ecologically very similar despite their independent history and recent evolution. Plants found in all areas are adapted to drought. Mixed woodland would have been the original vegetation, but human impact from grazing and felling trees has resulted in much of the area being scrubland. Woody species are often sclerophyllous, which means they have hard, thick, small leaves and short internodes between the leaves, which are all adaptations for conserving water. Typical species include Quercus suber (cork oak), Pinus pinaster (maritime pine) in woodland and Ulex europaeus (common gorse) and Myrtus communis (myrtle) in shrubland. Other plants are deciduous and have adapted to the drought by losing their leaves during the dry season. Other adaptations to conserve water in this biome include plants changing the position of their leaves to reduce exposure to the sun or having hairy leaves. The hairs on the surface of the leaf help reflect sunlight and also reduce air movement near the surface, which in turn reduces transpiration and therefore water loss. While leaves in these regions have evolved strategies to reduce water loss, roots like those found in savannas have evolved strategies to aid water uptake. Typically, plants in the Mediterranean biome have a dual root system with a long deep taproot, and a dense network of lateral roots close to the surface. Hellmers et al. (1955) excavated the roots of all Californian shrub species from the Mediterranean biome and found the roots of some species growing at a depth of 9 m, which were presumed to be growing deeper still. The deep root system enables plants to access water for most of the year. Drought-evading annuals are very diverse in this biome. They germinate in late autumn growing slowly through the winter, ready to use the last of the water when the warm weather returns, before setting seed in late spring and surviving the summer droughts in seed form. In addition to adaptations to withstand drought, plants also need to be adapted to withstand fire to survive

Figure 10.13 New shoots appearing after a wildfire on Kythira, Greece.

(Source: siete vidas, Shutterstock.com)

in the Mediterranean biome. Many woody species will burn quickly and then resprout (Figure 10.13). They often have woody underground tubers, burls, which are thick enough to protect a store of sugar and hidden buds, which resprout quickly after a fire. Quercus suber is one species whose stem does not burn completely in a fire. The thick bark of the cork oak protects the stem and crown buds so these can resprout (Pausas, 1997). Annuals and perennials will survive fires through their seed bank in the soil. The fauna of the Mediterranean biome is not particularly characteristic only to this biome. The animals must be adapted for drought and are often nocturnal to avoid the extreme heat during the day. Although quite diverse, there are few species which are endemic to these regions. For example, in areas which have some forest, the same species can be found in woodlands outside the Mediterranean biome as well, such as deer and hedgehogs. Desert species including lizards and scorpions are also found. In grassier areas, ungulate herds, such as mountain sheep and goats, comprised of the same species that are present in the steppes, can be found. Threats: Damage caused through overgrazing and changes to the fire regime continue to threaten this biome. Sala et al. (2000) predicted that the Mediterranean biome, due to its sensitivity to land use and climate change, will experience the greatest proportional change in biodiversity by 2100 of all biomes. Between 2001 and 2012, this biome had an 18% change in land cover type (RBG Kew, 2016). Increases in urbanization and population density as well as a conversion of land to agriculture are the main threats (Underwood et al., 2009).

267

Chapter 10 The biosphere

10.4.5 Temperate grassland

10.4.6 Temperate broadleaf forest

Temperature: Summer temperatures can be hot, reaching over 40°C, and winter temperatures are cold, down to - 40°C.

Temperature: Temperate forests are seasonal with a distinct winter. Temperatures range from - 30°C to + 30°C.

Precipitation: Annual mean precipitation is between 510 mm and 880 mm, with the majority falling between late spring and early summer. Distribution: The major areas of temperate grasslands are the veldts of South Africa, the puszta of Hungary, the pampas of Argentina and Uruguay, the steppes of Russia, and the plains and prairies of central North America. Important species: As its name suggests, grass is the dominant vegetation in this biome. Unlike savannas, trees and scrub are generally absent. Fires, drought and grazing all act to prevent trees becoming established. Grasses are usually perennial and adapted to drought, fires and also to cold. Fires are common and the frequency of fires means that most plants have underground stems, bulbs or can regenerate quickly from roots to survive all but the most severe fires. Moisture variation determines the height of the vegetation. In wetter regions grasses can reach heights of three metres, whereas in drier areas heights may less than half a metre. Perennial forbs also grow here, particularly those in the Asteraceae (those with a daisy-like flower) and Leguminoseae (nitrogen fixing) families. There are few places for fauna to hide from predators owing to the limited structure of the vegetation. Smaller grazing animals have adapted by burrowing, larger ones by forming herds and developing speed (deer) or large size (elk and bison). Predators include wolves, coyotes, foxes and polecats. Threats: The soil of temperate grasslands is nutrient rich from decomposing grass roots, and the rotting roots provide a good structure to the soil. For this reason much of the biome has been developed into grazing land by farmers or into arable agriculture. Removal of the root network holding the soil together for arable agriculture (e.g. after ploughing) leads to soil loss through water and wind erosion. Dust storms in the USA in the 1930s and 1940s when drought occurred after ploughing were environmentally disastrous. Domestic grazing has led to severe damage of temperate grasslands. Unlike farmed animals, large native species migrate and rarely overgraze areas. Encroachment from urban areas and invasive species also threaten what remains of the temperate grasslands. 268

Precipitation: Mean annual precipitation in temperate forests ranges from 750 mm to 1150 mm. Distribution: Temperate forests occur in eastern North America, north-eastern Asia, and western and central Europe, across latitudes ranging from 25o to 60o both north and south. Important species: In the northern hemisphere this temperate broadleaf forest biome is deciduous with trees shedding leaves in the winter. In the southern hemisphere the forests are usually evergreen, probably as a result of their separate evolution. In these evergreen forests epiphytes and climbing plants are commonly found, but there is limited ground flora due to the lack of available light. In favourable environments, with rich brown earth soils, in the northern hemisphere tall deciduous trees grow with a canopy height of 10–30 m. Below this tree canopy, a shrub layer develops at a height of about 5 m, with a ground flora of herbaceous plants, mosses and liverworts. The nature of the ground flora is dependent on the canopyforming species and the light that can penetrate the canopy. Much of the ground flora has adapted by flowering early in the season before the canopy species have regrown their leaves (Figure 10.14). Ground species include Hyacinthoides non-scripta (bluebell), Galanthus nivalis (snowdrop), Primula vulgaris (primrose) and orchids. In some areas the soils may be acidic and nutrients may have been washed out. On these soils heathland may develop. Here Calluna vulgaris (common heather) and Erica tetralix (cross-leaved heath) dominate. They typically reach heights of half a metre, but can get to one metre. With the heather, species of grass, rushes and mosses grow. Large mammals which were once found in this biome have become extinct or are much reduced in range and numbers due to human interference. Deer are found throughout the biome and browse on the trees and shrubs. Smaller mammals have adapted to the cold winter by either hibernating (hedgehogs) or burrowing (foxes, rabbits, rodents and badgers). Threats: There is very little temperate forest that has not been altered by humans and a classic image of a natural temperate woodland is actually one heavily managed by people. Forest size has been eroded by the expansion of urban areas and agriculture and it has been extensively fragmented. Invasive species (see Chapter 11), have also changed the ecology of a lot of the biome. For example, Rhododendron ponticum was introduced to western Europe and New Zealand as a garden ornamental, but it has spread widely

10.4  Terrestrial biomes

Figure 10.14 Hyacinthoides non-scripta flowering in spring (April) before the woodland canopy is closed, Belgium.

(Source: Anneka, Shutterstock.com)

in the biome. Its evergreen leaves shade out the ground flora and prevent seeds of native trees germinating.

10.4.7 Taiga Temperature: The winter temperature range is - 54°C to - 1°C and the summer temperature range is - 1°C to + 21°C. There are only 50–100 frost-free days in the year. Precipitation: Taiga has a mean annual precipitation of 300–850 mm, with most of this falling as rain in the summer. Distribution: This is the largest land biome and spans Eurasia and North America in the subarctic region.

Important species: This biome is also referred to as boreal forest and the predominant vegetation is coniferous forest (Figure 10.15). The soil is nutrient poor and below the surface layer is permafrost (permanently frozen subsoil; see Chapter 24) or bedrock. The predominant trees are Picea species (spruce), Abies species (fir) and Pinus species (pine). These are all evergreen species. In some areas Larix species (larch), which is deciduous, survives. The reason the majority of species are evergreen is that growing leaves requires high energy and resource input. There are limited resources with few nutrients in the soil and low light levels. By not shedding leaves annually the plants do not use up the limited resources, but can use these resources for structural

Figure 10.15 Boreal forest, northern Finland. The trees have wider branches near the foot and

narrower branches at the top – this minimizes potentially damaging snow accumulation. (Source: Catalin Petolea / Shutterstock.com) 269

Chapter 10 The biosphere

support instead. As the needles are present on the tree then photosynthesis can start as soon as conditions are suitable. The dark green colour of coniferous trees is an adaptation to absorb as much light as possible for photosynthesis. Needle leaves have a low sap content and this adaptation helps prevent freezing in the winter, which would damage cells. The needle shape also reduces water loss, because although there is relatively high precipitation, when the ground is frozen plant roots cannot access water. Even the shape of the trees has evolved to ensure survival in the frozen conditions; they are conical so snow sheds easily, thereby avoiding the weight of snow causing damage to branches (Figure 10.15). The canopy is dense so little light reaches the forest floor and this, together with the needle leaves which maintain an acidic soil, means ground flora is limited. Mosses and lichens do survive, as do fungi, and all these provide food for grazing animals. In wet depressions bogs may form and species of sphagnum moss form mats on these. Occasionally after a disturbance (see Chapter 11), and if conditions are suitable, deciduous trees including Alnus species (alder), Betua species (birch) and Populus species (aspen) can survive. These trees can support larger browsing animals such as Alces (elk) and Cerus elaphus (red deer). In some areas of this biome Castor fiber (Eurasian beaver) and Castor candensis (North American beaver) survive. These large rodents dramatically change the landscape with the dams they build (see Figure 5.16 in Chapter 5). Predators include Felis lynx (lynx) and members of the Mustelidae (weasel) family. Threats: Over-harvesting of the taiga biome is a threat. If managed carefully, logging could be sustainable, but the

forest takes a long time to regenerate, as growth rates are slow. Unfortunately, logging is often intensive and clear cutting is usual. Vast areas of the biome are also being destroyed by exploration for oil and gas. Global warming may cause the permafrost to thaw, changing the ecosystem, and rising temperatures could cause deciduous trees from the temperate zone to migrate northwards, outcompeting current species.

10.4.8 Tundra Temperature: Tundra is the coldest biome with a mean winter temperature of - 34°C and a mean summer temperature of 3–12°C. Precipitation: The yearly precipitation in the tundra biome is 150–250 mm. Distribution: Arctic tundra encircles the North Pole extending down to the coniferous forest of the taiga. Alpine tundra is located on mountains throughout the world at high altitude where trees cannot grow, usually above 3000 m. Important species: The growing season for Arctic tundra is just 50–60 days; it is longer for alpine tundra, with approximately 180 days a year. Tundra comes from the Finnish word, tunturia, meaning treeless plain (Figure 10.16). The soil of alpine tundra is free draining, but soil in the Arctic tundra has limited drainage and a permafrost layer that prevents deep plant roots. Despite the harsh growing conditions there are approximately 1700 species of plants growing in the tundra, which include around 400 species

Figure 10.16 Alpine tundra, Mount Cook National Park, New Zealand.

(Source: Ariadne Van Zandbergan / Alamy Stock Photo) 270

10.4  Terrestrial biomes

of flowering plants and mosses, lichens, grasses, liverworts and low shrubs. The plants grow close together for protection from the cold temperatures and strong winds, and snow offers protection from the extreme temperatures in the winter. Due to the short growing period most plants do not produce seeds or spores, but reproduce vegetatively, where new individuals are asexually produced by a single parent plant. For example, new shoots will grow from a tuber, or horizontal stems will produce new vertical stems and roots. Considering the severe conditions there is a reasonable diversity of animal life in the tundra. Many species migrate south in the winter to avoid the worst of the weather. Grazing mammals include Lepus arcticus (arctic hare) and Rangifer tarandus (reindeer) and predatory

mammals such as Vulpes lagopus (arctic fox) and Canis lupus arctos (artic wolf) breed and raise their young in the short growing period. Some mammal species hibernate in the winter months. Others such as Rangifer tarandus survive on fungi and lichens through the winter, clearing the snow with their hooves and antlers to access them. Threats: The main threat to the tundra biome is global warming, which as in the taiga could cause the permafrost to melt, changing the biome irreversibly. Other threats come from oil and gas exploration fragmenting or destroying areas of the biome. In addition, air pollution is reducing lichen growth, which, particularly in the winter, is the main food source of grazing animals. Ozone depletion over the Arctic results in stronger ultraviolet rays reaching the tundra which can damage the organisms living there.

GLASS BIOMES Enabling a completely different way to research the changing biomes of the world, Biosphere 2, named to denote that it is a model of the Earth’s biosphere, is essentially a huge glass enclosure (Figure 10.17), recreating major biomes and facilitating unique large-scale experimentation in each of the biomes represented. Space Biospheres Ventures bought the property in 1984, and in 1986 began developing the current facility to gain an understanding of how adept to survival such an enclosure was. Between 1991 and 1994, two missions were run with people sealed in Biosphere 2. Columbia University managed it from 1996 to 2003, changing the emphasis of the scientific research to include a study on the effects of different carbon dioxide concentrations on plants. From 2007, the University of Arizona ran Biosphere 2. In 2011, they became the owners, continuing the experimental research into ecosystem science and global climate change. The model ecosystems which the University of Arizona run experiments on are

Figure 10.17 Some of the buildings forming Biosphere 2 in Tuscon, Arizona.

(Source: Purestock / Alamy Stock Photo)

a model city and urban ecosystem, made up of the Biosphere 2 campus and associated buildings, three desert hillslopes, a coastal desert, a savanna, swamp forest, 2600 m3 of ocean, and a rainforest with over 90 tree species. The experiments

within these ecosystems aid the development of computer models to simulate how ecosystems will respond to environmental change. The models then help determine the direction of the experiments needed to refine the models further.

BOX 10.3 ➤

271

Chapter 10 The biosphere

➤ In addition to being a research facility, Biosphere 2 is also a highly rated visitor attraction for people wishing to experience the ecology of the Earth and understand the different ecosystems created there. The Eden project in Cornwall, UK, is a visitor attraction which houses the largest rainforest in captivity. It aims to inform people about the living world and the threats faced by

different ecosystems. It does this in an exciting way and is clearly successful. In 2014–2015 it was visited by over 47 000 school children, and in 2015 it was awarded ‘Best UK Leisure Attraction’ in the British Travel Awards as voted by the general public, an award it has held for several consecutive years. In addition to the rainforest biome, the Eden Project also has a Mediterranean biome as well as a ‘wild Cornwall’ and a Prairie

grassland. These biomes are managed and because it is not possible to recreate the biomes completely, not all species can be represented, and also some plants must be pollinated by staff. Yet, every effort is made to be authentic. Agricultural systems are also represented. Both Biosphere 2 and the Eden Project show the appetite for biogeography as leisure activities and also play an important role in educating people about the biosphere.

BOX 10.3

Reflective questions ➤ What are the key plant adaptations to limited water availability?

➤ What are the similarities and differences between temperate grassland and the Mediterranean-type biome?

➤ What are the major differences between the tundra and taiga biomes?

10.5 Aquatic biomes When considering the Earth’s biomes, it is important to consider the aquatic biomes since water covers nearly 75% of the Earth’s surface and it is, after all, where life began. Aquatic biomes can be separated into two main regions: marine and freshwater. The important distinction between these is the difference in dissolved salt concentration, which makes up less than 1% of the water in freshwater ecosystems but around 3% in marine ecosystems.

10.5.1 Marine regions 10.5.1.1 Oceans The oceans represent the largest biome within the biosphere. The ocean biome can be spilt into zones based on depth: intertidal, pelagic, benthic and abyssal zones, which all have different abiotic conditions to which life has adapted. 272

The intertidal zone fluctuates between being fully submerged and completely exposed. Organisms that survive here have to be adapted to these regular changes in environment, with the upper shore experiencing more exposure to air, with a greater variation in temperature as shown in Figure 10.18. This gradient in environmental conditions is reflected in a change in organisms, from those adapted to almost dry conditions to organisms that are only exposed during unusually low tides. Species will vary depending whether the substrate is rocky or sandy and whether it is exposed to strong wave action or it is a quiet bay. In sandy areas animals which bury into the sand as the tide comes in are typical. These include worms and clams. In a protected area with little wave action, marine algae (seaweed) and seagrass, flowering plants, of which there are 58 known species, will flourish. There is a lack of attached plants in areas with strong wave action. In rocky intertidal zones, animals tend to have adaptations to enable them to attach to the rock to withstand the strong tidal currents. The pelagic zone is the open ocean, not reaching the bottom or a coast, but with a depth of about 4000 m. There is a steep light gradient in this layer with plenty of light for photosynthesis until around 200 m depth. Algae and photosynthetic bacteria are abundant in this upper zone and provide food for the fish and zooplankton. Below 200 m depth animals rely on plankton floating down to provide nutrients and some animals swim higher at night to where food is more plentiful. Between 200 m and 1000 m depth, bioluminescence, where light generated from biochemical reactions is emitted by organisms, is common, with around 90% of animals displaying it. The bioluminescence, where there is little or no sunlight,

10.5  Aquatic biomes

High Splash Zone

Amount of sunlight

Low Competition

High Time exposed to air

High Salinity variation

Temperature variation

High

Upper Shore

Middle Shore Low

Low

Low

High

Low Lower Shore

Subtidal Zone

Figure 10.18 The intertidal zone on a rocky shore. (Source: Marine Biological Association of the UK)

is used to attract or find prey, and as a defence against predation and also as sexual communication (Haddock et al., 2010). The benthic zone consists of the areas of the seafloor, below the pelagic zone and also in coastal areas. Some definitions include the abyssal zone (see below), as it is also on the seafloor, but ecologically these are quite different. The bottom is usually sand or silt, and dead organisms may build up forming a detrital litter. There is a wide variety of seaweeds and filamentous algae and there is a huge biodiversity of bacteria, fungi, sponges, sea anemones, worms, starfish and fish. The abyssal zone is the coldest and darkest area with water temperature around 3°C. It has a low nutrient content and is under high pressure. Despite this, many species of fish and invertebrates have adapted to life here. Plant life cannot be sustained in the abyssal zone due to the lack of light. Instead chemosynthetic bacteria, which are found near hydrothermal vents on the sea floor, are able to convert hydrogen sulfide into energy and it is these bacteria that the fish and invertebrates eat (see Box 3.4 in Chapter 3).

10.5.1.2 Coral reefs Found in tropical and sub-tropical oceans, coral reefs form in shallow water covering approximately 1% of the ocean floor (Figure 10.19). Despite their small area, they have high biodiversity, accounting for over a quarter of the oceans’ total species biodiversity (Plaisance et al., 2011). Individual corals from the order Scleractinia

are known as polyps. Polyps are stationary and feed on plankton and other tiny organisms, using stinging tentacles which extend at night. Scleractinia species live in colonies. Each polyp excretes calcium carbonate forming an exoskeleton beneath it. This is how the structure of a coral reef develops over time. Reefs, however, are one of the most endangered habitats on the planet. It is thought that one third of reef-building corals face extinction through climate change and other human impacts (Carpenter et al., 2008). The nature of how the reefs develop means that once they become degraded or removed, it will take a very long time to replace them. In addition to climate change, reefs face threats from overfishing, pollution and invasive species.

10.5.1.3 Estuaries Estuaries are transition zones where the freshwater of the rivers meets the salt water of the oceans (see Chapter 22). This creates a range of changing habitats that support a wide variety of organisms. These habitats include oyster reefs, kelp forests, coastal marshes, mud flats, mangrove forests and deep water swamps. Organisms that are adapted to survive rapidly changing salt concentrations in water are known as euryhaline species. Plants, which cannot generally move to avoid the fluctuations, have often evolved special structures to cope with the salinity changes. Spartina alterniflora (cord grass) has filters on its roots to remove salts from the water and can also expel salt from its leaves. Animals have also adapted to the changing conditions. For example, bivalves (mussels 273

Chapter 10 The biosphere

Figure 10.19 The location of major coral reefs. Most occur between 30oN and 30oS. (Source: National Oceanic and Atmospheric Administration)

and clams) close their shells during low tide, when there are low saline conditions, and switch to anaerobic (without oxygen) respiration. With high tide, high salinity conditions return, so they open their shells, start to feed and switch back to aerobic respiration. Estuaries are extremely important areas for supporting nesting birds and juvenile fish, and so have been termed ‘nurseries of the ocean’. Gadus morhua (cod) is one fish species whose juveniles find shelter in these habitats (Seitz et al., 2013). Like many other species, it requires estuarine habitats to be protected to boost species numbers as they battle decline caused by factors such as overfishing elsewhere. Estuaries are also incredibly important for birds. The Severn Estuary, one of the largest in the UK, had a mean peak bird population of 93 986 (excluding gulls) between 1991/92 and 1995/96 (Stroud et al., 2001), which had reduced to 66 022 between 2002/03 and 2006/07 (Austin et al., 2008). This decline has been linked to climate change (Burton et al., 2010). Many of the world’s largest cities are built on estuaries. Such development has been detrimental to estuaries environmentally, with impacts of overfishing, dams, pollutants and the introduction of invasive species. The degradation of these habitats has meant that they can no longer fulfil the nursery, feeding and reproductive functions with which they have been associated (Beck et al., 2011). In other areas, conflicts with cockle and mussel fisheries have had impacts on bird species such as oystercatchers (Burton et al., 2010). 274

10.5.2 Freshwater regions Freshwater ecosystems are dealt with in detail in Chapter 12. Therefore only a brief overview is provided below of freshwater biomes. Freshwater can be separated into three distinct regions: streams and rivers, ponds and lakes, and wetlands. As with other water bodies, streams and rivers are zoned vertically and organisms are found throughout the zones, with special adaptations to that layer. Rivers are often rich in phytoplankton or rooted aquatic plants, though many fish and invertebrates will also feed on terrestrial vegetation growing into the river. Runoff from agriculture and industry pollutes these environments and can kill many of the organisms found there. Rivers and streams are also under threat from dams and flooding control, which can affect their functioning and the environment for native species. Ponds and lakes can range in size from a square metre to thousands of square kilometres. Some may only exist seasonally, and others have been present for thousands of years. Often lakes can be isolated from other water bodies, which has led to many species being endemic to a particular lake. Nutrition in lakes can vary greatly and the biodiversity they support also varies. The top zone of the lake, or that close to shore, the littoral zone, can support a range of rooted and floating aquatic plants, as light levels are high enough for photosynthesis to take place. Where light can penetrate there will be a variety of phytoplankton and

10.6  Summary

cyanobacteria. Depending on the size of lake or pond and the nutrient content, a range of zooplankton will graze on the phytoplankton, and fish and invertebrates are found in all zones including the deeper profundal zone which lacks enough light for photosynthesis. As with rivers and streams, pollution from agriculture and industry can be a huge threat to lake ecosystems, along with climate change. Unlike for terrestrial habitats, cold-adapted species cannot easily migrate from lakes. Wetlands are areas which have standing water for at least some of the year. Marshes, swamps, bogs and fens are all considered wetlands and these may form in hollows or large basins (or even on very wet gently sloping ground where there is poor drainage), along the edge of a river which is periodically flooded or the edges of lakes or oceans which may be flooded by tides, or rising water levels. Those flooded by seas such as salt marshes are classed as marine environments. Wetlands are one of the most productive and biodiverse of all ecosystems. Plants are adapted to the wet conditions and support a variety of

10.6 Summary

invertebrates, amphibians, reptiles and birds. Many wetlands have been destroyed throughout the world over the last 200 years, through drainage for agriculture and development and the diversion of water. In some areas, efforts are being made to re-establish wetlands, particularly as they often support water purification and flood reduction (Acreman and Holden, 2013). However, wetlands are still under threat from invasive species and climate change.

Reflective questions ➤ What is the key difference between marine and freshwater biomes?

➤ Which is the most productive aquatic biome and what are the reasons for this?

➤ Why is it difficult to recreate damaged or destroyed coral reefs?

in temperate areas, with species numbers declining still further towards the poles. The higher rates of primary productivity which

The biosphere is the part of the Earth where all life exists and this

occur in the tropics are likely to be at least partly responsible for

chapter has attempted to explain its main features. However, with

this latitudinal gradient in species diversity. Terrestrial biomes can

only 14% of the Earth’s 8.7 million eukaryotic species identified

be mapped based on climatic factors, predominately mean annual

there is clearly still much to learn. Ecologists, in trying to under-

temperature and precipitation. Biomes have characteristic veg-

stand the complexity of the biosphere, study it across different

etation and the plants and animals that are found in them have

hierarchical tiers ranging from individual organisms, through

adapted to the environmental conditions. Just as environmental

populations and communities to ecosystems, biomes and finally

conditions do not generally change abruptly, but change along a

at the level of the whole biosphere. Species, groups of individu-

gradient, there are transitional zones between different biomes

als with the potential to interbreed in nature and produce fertile

known as ecotones. Aquatic biomes account for approximately

offspring, are the unit commonly used when studying biological

75% of the Earth’s surface and are divided into freshwater and

diversity, although biodiversity also covers genetic diversity within

marine biomes.

species and the diversity of ecosystems. When measuring the bio-

All biomes, whether aquatic or terrestrial, are under threat

diversity of an area it is important to consider the evenness of the

from human actions including habitat destruction and fragmen-

species (or other taxonomic group) as well as the richness.

tation due to urban or agricultural expansion. Invasive species,

Species are found within a range of environmental variables

climate change, over-exploitation and pollution are other impacts

they are able to tolerate and where interactions with other spe-

which the biosphere as a whole has to adapt to. Understanding

cies and local history enable them to exist. There are general

the important patterns and functions within the biosphere and

patterns to biological diversity with more species being found

how they are changing is key if the impacts of global change and

in large areas and more species found near the equator than

predictions for the future are to be accurately made.

275

Chapter 10 The biosphere

Further reading Allison M., Doak, D.F. and Angert, A.L. (2015) Where and when do species interactions set range limits? Trends in Ecology and Evolution, 30, 780–792. A good review of the theory that abiotic forces set species range limits at high altitude and latitude and species interactions set range limits in less hostile regions. A good paper to read to enhance your understanding of species distributions. Cox, B., Ladle, R.J. and Moore, P.D. (2016) Biogeography: An ecological and evolutionary approach, 9th edition. Wiley-Blackwell, Chichester. A good undergraduate text covering all aspects of biogeography. The book has a greater emphasis on marine life than most books covering the subject. Gaston, K.J. and Spicer, J.I. (2004) Biodiversity: An introduction, 2nd edition. Blackwell, Malden, MA.

276

This is a good overview of biodiversity, with some interesting case studies, which would serve well as a short introduction to the subject at an undergraduate level. Lomolino, M.V., Riddle, B.R., Whittaker, R. and Brown, J.H. (2010) Biogeography, 4th edition. Sinauer Associates, Sunderland, MA. This book provides an excellent summary of all disciplines that are categorized as biogeography including the distribution of species and communities, biodiversity and island biogeography. Reece, J.B., Urry, L.A., Cain, M.L., Wasserman, S.A., Minorsky, P.V. and Jackson, R.B. (2011) Campbell biology, Global Edition, 9th edition. Pearson Education, Harlow. This book gives an excellent overview on all aspects of bioscience at undergraduate level. Chapters 52 to 56 are particularly relevant to the study of the biosphere. I recommend the Global Edition unless you want your focus to be predominately the USA and Canada.

C HAPTER 11

Ecosystem processes Julie Peacock School of Geography, University of Leeds

Learning objectives After reading this chapter you should be able to:

➤ summarise what is meant by the terms ‘ecosystem’ and ‘ecosystem processes’

➤ explain how energy and nutrients transfer through the ecosystem ➤ classify the effects of different biotic interactions on organisms ➤ describe temporal processes in ecosystems ➤ explain the main human impacts on ecosystem processes

11.1 Introduction An ecosystem consists of all of the organisms within an area and their interactions with the physical and chemical environment in which they are located. Ecosystem processes involve the transfer of substances between the different components, both living and non-living, within the ecosystem. These substances include resources needed for life such as carbon, water, nitrogen, phosphorus and also pollutants from agriculture, industry or roads. The processes include, among others, primary production, decomposition, soil formation and nutrient mineralization (Currie, 2011).

Ecosystem processes are the major foci for scientists working on ecosystems. This research requires an understanding of factors external to the ecosystem such as climate, topography and the underlying geological material as these will all affect the processes within the ecosystem. In addition, the factors and interactions within the ecosystem, which will both affect the processes and be affected by them, must be understood. These internal factors include the biotic community and the community dynamics, the human influence, disturbance regime, and resource availability (Figure 11.1). Understanding ecosystem processes will permit accurate predictions of the responses of ecosystems to global climate change and to other human impacts. In addition, conservation effort will be more successful if the processes within ecosystems are understood. Finally, humans rely on ecosystem processes for survival; from the provision of food and water, to the regulation of disease pathogens, flood mitigation and for well-being, these are all linked to ecosystem processes. Understanding how best to preserve and utilise these services that ecosystems provide even in exceptionally modified environments can only be of benefit to people throughout the world. This chapter will start by considering the flow of energy and resources through an ecosystem. It will cover ecological thermodynamics and the impacts this has on food webs, as well as the cycling of major nutrient

Chapter 11 Ecosystem processes

Climate

Time Disturbance regime Micro-environment

Human activities

Ecosystem processes

Topography Resources

Biotic community

Parent material

Potential biota

Figure 11.1 The relationship between the external conditions, which

are independent of the ecosystem (outside the circle), and the interactive controls (words inside the circle), which affect and are affected by ecosystem processes. These effects are mediated by feedback systems (internal arrows). (Source: Chapin III et al., 2002)

resources. These cycles are impacted by different organisms and the different types of biological interactions will be discussed. Changes to ecosystems over time, with a focus on succession will be covered, as well as the impacts humans have had on ecosystems. Here we will also consider the most rapidly expanding ecosystem at present – the urban ecosystem, an ecosystem where alteration of all parts of the biosphere, the atmosphere, oceans and land have been most obviously manipulated by humans, yet a place where ecological processes continue regardless. Finally, conservation efforts and how ecological science can aid these efforts will be addressed.

11.2 The flow of energy and resources 11.2.1 Energy entering an ecosystem The majority of energy received by the biosphere is from solar energy which plants and photosynthetic bacteria and archaea use to convert carbon dioxide (CO2) and water into carbon sugars, through photosynthesis (see Section 13.2.1 in Chapter 13). In the deep oceans and other places where light cannot penetrate, some bacteria can produce sugars utilizing the energy from chemical reactions using hydrogen sulfide, sulfide or methane. This is called chemosynthesis and is particularly important around hydrothermal vents on the ocean floor (see Box 3.4 in Chapter 3). Organisms which can produce food from inorganic compounds and energy are known as autotrophs and they are ‘self-feeders’ and primary 278

producers. The total amount of energy produced (as biomass) in a given time and area is called the gross primary productivity (GPP). Photosynthesis produces a store of energy in the form of sugar, while respiration transfers the energy into forms suitable for growth and other metabolic processes, with some energy given off as heat. All living things undergo respiration. The chemical process for respiration results in the combination of sugar and oxygen producing CO2, water and energy. Much of the energy produced by photosynthesis is used to maintain plant functioning. The remaining fraction of GPP is stored in plant tissue and this is called the net primary productivity (NPP) of an area. The NPP of an ecosystem is important as it represents the amount of energy available as food to other organisms, the consumers, within the ecosystem. A low NPP value would mean there was only a small amount of food available and few organisms could be supported. Conversely, a high NPP value represents a lot of food available for other organisms, so greater numbers can be supported. Conditions for photosynthesis are improved when there are warm temperatures, good light availability and plenty of water. In ecosystems with these conditions (e.g. a tropical rainforest) NPP will be higher. In ecosystems where one or more of these factors are limited (e.g. tundra) the rate of photosynthesis will be reduced and therefore net primary productivity will be reduced as well.

11.2.2 Ecological thermodynamics When considering the flows of energy and nutrients through an ecosystem it is important to remember that the laws of physics and chemistry apply: knowledge of these helps to explain some basic concepts in ecosystem dynamics. The first law of thermodynamics, also known as the law of conservation of energy, states that energy is not created or destroyed but may be transformed from one type to another. Therefore, energy inflow to an ecosystem must be balanced by energy storage plus energy outflow. The second law states that as energy is transformed from one form to another there is a dispersion of energy into unavailable forms. It will, for example, dissipate as heat. The law of conservation of mass states that matter cannot be created or destroyed. Therefore, in chemical reactions the mass of an element at the beginning of a reaction will be the same as the mass of that element at the end of the reaction. All living things are composed of elements: organisms obtain them and then use them before disposing of them. The main elements important for life are calcium, carbon, hydrogen, nitrogen, oxygen and phosphorus and

11.2  The flow of energy and resources

these cycle through the biosphere. Ecosystems are not closed systems and some elements will be lost from them through moving water or wind-blown materials and others will escape as gas to the atmosphere. The principle of conservation of mass can be used to calculate all inputs into and outputs from an ecosystem to determine whether it is a source or a sink for a given element. For example, much recent research has focused on understanding which ecosystems are storing or emitting net amounts of carbon, which is important for climate change.

11.2.3 Trophic levels and food webs In an ecosystem, an aphid may feed on a plant. The aphid in turn may be eaten by a ladybird, the ladybird eaten by a small rodent which may be eaten by a hawk. In this simple food chain, the plant is the primary producer and represents the first trophic level. The aphid is at the second trophic level in the chain. It is a heterotroph, or consumer, as it does not produce its own food, but feeds on organic material. It is at the second level of the food chain and is

Producers

Primary consumers

also referred to as the primary (or first-order) consumer. The ladybird, also a heterotroph is at the third trophic level and is the secondary (second-order) consumer. The rodent that eats the ladybird is at the fourth trophic level and is the tertiary (third-order) consumer and finally the hawk is at the fifth trophic level and is the quaternary (fourth-order) consumer. This is an example of a grazing food chain as the primary consumer feeds on a living plant. In a detritus food chain the primary consumer, a decomposer, feeds on dead plant matter. Decomposers may be detritivores which ingest material before breaking it down or saprovores which secrete chemicals to breakdown the material extracellularly, before absorbing them: these are usually fungi or bacteria. Decomposers enable the cycling of material through an ecosystem, making the nutrients available again to other organisms. How a detritus food chain and a grazing food chain interacts can be seen in Figure 11.2. Food chains are limited in length usually to four or five trophic levels. This is due to the laws of thermodynamics as energy is lost from the food chain at each trophic

Secondary consumers

Tertiary consumers

Grasshopper Grazing food chain

Lizard

Hawk

Rabbit

Mouse

Vegetation

Snake

Shrew

Detritus food chain

Dead organic matter Fly maggot

Bacteria

Dermestid beetle Ant Dung beetle

Toadstool

Millipede

Bacteria

Bacteria

Figure 11.2 A simplified example of nutrient flows within a temperate ecosystem, indicating trophic levels and interactions between a grazing food

chain and a detritus food chain. 279

Chapter 11 Ecosystem processes

by zooplankton there is never very much available in the ecosystem. In terrestrial ecosystems, however, much of the vegetation is never consumed by grazers. The pyramid of net production in marine environments, however, does conform to a true pyramid as energy is still lost from the system as described above. Clearly, in nature, energy and nutrients are not passed through ecosystems in simple chains, but rather in complex food webs, such as the one shown in Figure 11.4. In these webs it is clear that some animals operate at more than one trophic level. For example, a mouse would be classed as operating on the second trophic level when eating seeds and the third trophic level when eating insects.

Terrestrial ecosystem

rodent ladybirds aphids tree

dolphin Marine ecosystem sardines zooplankton phytoplankton Pyramids of biomass

Pyramids of energy

Figure 11.3 A representation of a typical pyramid of biomass and a pyra-

mid of energy transfer through a terrestrial and a marine ecosystem.

level. Typically only 10% of energy from one trophic level is passed to the next, although trophic efficiencies have been calculated to range between 5 and 20% depending on the ecosystem. This transfer of energy is often represented by a pyramid of net production (Figure 11.3), with the net production at each trophic level represented by a band proportional to the size of net production. The band for the first tropic level is always large but by the fourth trophic level the band will be very small. Net production pyramids in terrestrial systems are mirrored by pyramids of biomass at each tropical level; the biomass of the primary producers is much greater than the biomass of each consumer level. In some marine systems, however, the biomass pyramid may be inverted as the life cycle of phytoplankton is very rapid and consumed so quickly

11.2.4 Biogeochemical cycles Although ecosystems receive solar energy as an ongoing supply, elements needed for growth are limited in their availability and therefore need to be recycled in ecosystems. The term biogeochemical cycles refers to the fact that biological, geological and chemical factors are involved in the recycling of matter. Approximately 10 major nutrients are vital to the survival of plants and animals, but the four most important cycles for life are the water, carbon, phosphorus and nitrogen cycles. Some of these will cycle locally within the ecosystem, others will cycle more globally. For example, phosphorus is consumed at the different trophic levels and then decomposers will return it to the soil. In the soil it will be available for plants to take up again, thereby resulting in local cycling. However, some phosphorus may be washed out of the ecosystem in water, and move on its journey in a more globalized cycle. There

Tertiary carnivores Secondary carnivores Primary carnivores

Herbivores

Foxes

King snakes

Road-runners

Raccoons

Quail

Gopher snakes Insects Deer mice

Pocket mice Pack rats

Plants

Insects

Woody plants

Figure 11.4 The complexity of a typical food web. This example is from the chaparral scrub of California. The various trophic levels indicate the animals

of different groups and it is noticeable that not all animals are confined to one trophic level. Raccoons, for example, are both herbivores and carnivores. However, they are ultimately dependent on the plants. 280

11.3  Biotic interactions

Table 11.1 The different interactions between species and their effect on the two species involved

Animal skeletons Phosphate within rocks

Nutrition for animals

Animal detritus

Geological weathering

Nutrition for vegetation Water-soluble phosphate in soil water

Species 1

Species 2

Interaction

+

+

Mutualism

+

-

Herbivory/parasitism/predation

+

0

Commensalism

-

0

Amensalism

0

0

Neutralism*

-

-

Competition

*Neutralism could theoretically exist, but it is not thought that an association between two species would have zero impact on either.

Intraspecific interactions take place between individuFigure 11.5 The phosphate cycle, as an example of a biogeochemical cycle.

may be points in the cycle where phosphorous is stored and accumulates for long periods, such as in ocean sediments, before the cycle continues. The main components of the phosphate form of the phosphorus cycle are shown in Figure 11.5. The phosphorus cycle does not have a gas phase to its cycle and it cycles slowly, but water, carbon and nitrogen all have gas phases.

Reflective questions ➤ Why are food chains usually limited to five trophic levels? ➤ What is the net primary productivity of an ecosystem and why is this value important?

➤ What are the definitions of autotrophs, heterotrophs, detritivores and saprovores?

11.3 Biotic interactions In ecology it is necessary to try to understand biotic interactions, as these will have a major impact on the processes within an ecosystem and in turn the processes which will affect them. The interactions take place at many levels, within a population and between individuals from different populations. Some interactions are transitory (brief). Others last for the lifetime of the organisms involved. Some, but certainly not all, are obligate, which means the individuals need the interaction to survive.

als of the same species. Interspecific interactions take place between individuals from different species. Interactions between species (Table 11.1) are known as a symbiosis, which is the long-term interaction of individuals of two different species. Symbiosis has previously been limited in its definition to an interaction between two species in which both species gain from the interaction (mutualism). However, the term is now usually used in the wider sense, regardless of whether the outcome of the interaction is positive or negative for the species involved. Symbiosis can shape whole ecosystems as well as the evolution of the species within them. Symbioses are classified in terms of where the interaction takes place. Ectosymbiosis describes one species living on another, whereas endosymbiosis is the term used when one species lives within the tissues of another species. As well as direct species interactions being important in ecological communities, indirect interactions also play an important role. Sanders and van Veen (2012) showed through experiments that extinctions of Aphidius ervi, one species of parasitoid wasp, led to the extinction of Lysiphlebus fabarumand, another species of parasitoid wasp. This was despite there being no direct connections between them – in fact they were four trophic levels apart. The wasps lay eggs inside aphids and when the wasp larvae hatch they feed on the aphid. The wasps only use certain aphid species as hosts. A. ervi only lays eggs in Acyrthosiphon pisum (the pea aphid) while L. fabarum only uses Aphis fabae (the black bean aphid) as a host. Sanders and van Veen (2012) set up cages with broad bean plants, the two species of aphid and the two species of parasitoid wasp. With this food chain set up all the species survived. In cages without L. fabarum, A. ervi populations continued to survive. Yet in cages without A. ervi, L. fabarum 281

Chapter 11 Ecosystem processes

populations crashed and went extinct in about six weeks. This is caused by competition between the aphids. A. pisum outcompetes A. fabae, therefore when A. ervi is not present the A. pisum population rapidly increases and the A. fabae population crashes. Without A. fabae to host its eggs L. fabarum populations become extinct too. This shows that it is key when studying ecosystems and modelling extinctions to consider the ecosystem as a whole because individual extinctions can have huge implications for other species.

11.3.1 Mutualism Birds eating berries are good examples of mutualism. The birds benefit by gaining food from the plants and, in turn, the plants benefit because the seeds inside the berries are dispersed away from the plant. Lichens are a mutualistic symbiosis between a photoautotropic partner, known as a photobiont and lichen forming fungi, called a mycobiont. The photobiont is usually a green algae or a cyanobacteria. The fungi provide the photobiont with nutrients and will often anchor the lichen to the substrate, such as a tree or rock. In turn, the mycobiont benefits from receiving carbon sugars. Most lichen partners are facultative so when a symbiosis forms, the morphology, along with the biochemistry and physiology, is very different from either of the photobionts or mycobiont when they are not forming a symbiosis. This symbiosis is considered to be ancient, although there is little fossil evidence. There have been 415 million-year-old fossil cyanobacterial and algal lichens in rocks in Wales that have been identified (Honegger et al., 2013), although it is likely that these mutualistic symbioses formed much earlier. Lichens are important colonisers of new surfaces. Lichens, which can withstand extremes of cold and aridity, are found in their crustose form in deserts and the Polar Regions. In more moist areas, many lichens have a foliose or sheet-like structure, whilst others have a fruticose structure, meaning it is branched and either erect or hanging. In urban areas, lichens are regularly found growing on the sides of buildings and trees, and on gravestones and pavements. Attempts to remove lichens and to prevent regrowth often fail. Another fungi–photoautotropic mutualistic relationship is formed between mycorrhizal fungi and land plants, with arbuscular mycorrhiza being the most common type of mycorrhizal fungi (Figure 11.6). It forms associations with around two thirds of all land plant species. The fungi exchange nutrients from the soil for carbon sugars, which are photosynthetically fixed by the plant. This provides energy for the fungi. It is thought that this symbiosis was 282

C

Hyphae

Vesicles

5 mm Figure 11.6 A cross section of a plant root (Ceratocarpus arenarius)

under a microscope showing arbuscular mycorrhizal fungi stained blue inside the root. The fungi create vesicles and thin hyphae as part of their structure. (Source: Zhang et al., 2012, under Creative Commons Attribution License)

critical in the evolution of plants on to land, with associations of moss-like plants occurring before the evolution of higher plants and plant roots (Brundrett, 2002). There is evidence for these associations from 400 million year old fossils (Remy et al., 1994), showing their importance in plant evolution. The predominant role of the arbuscular mycorrhiza was until recently thought to be phosphate uptake. However, more recently it has been shown that the fungi are also important in plant nitrogen uptake. Given that nitrogen is a key limiting nutrient in terrestrial ecosystems, a symbiosis with such fungi would give the plant a competitive advantage (Hodge and Storer, 2015). Arbuscular mycorrhiza are an example of obligate symbionts because they could not survive without the host plant providing carbon sugars.

11.3.2 Herbivory, predation and parasitism Herbivory, predation and parasitism are all interactions, which are positive for one individual and negative for the other. Predation differs from parasitism and herbivory because with predation the negative impact is death of the individual. However, with parasitism and herbivory the individual often survives, albeit with reduced fitness. Plants have adapted to herbivory in different ways. Species in the Poaceae (grass) family have their apical meristems from which primary growth of the plant originates, near the base of the plant. As such it is protected from grazing. Other plants use defence mechanisms such as spines and thorns, which make it difficult or uncomfortable for grazing or browsing animals to eat them. Some species are toxic or taste bitter, so that herbivores avoid them. A parasite reduces the fitness of a host organism by consuming nutrients from it, so the parasite benefits and its host is negatively impacted. Parasites may live inside their hosts and are known as endoparasites. These include, for example, tapeworms, protozoa, and flukes. Other parasites stay on the outside of their hosts including ticks

11.3  Biotic interactions

and lice, plants, protozoa, bacteria and fungi. Animals and plants are usually the hosts. Species in the genus Rafflesia are very unusual plants. They produce no leaves, stems or roots, but obtain all of their nutrition from the Tetrastigma vine on which it is parasitic (Beaman et al., 1988). Rafflesia produces threadlike growths which penetrate the tissues of the vine and become embedded into its xylem and phloem, where water and nutrients are transported (Rubiales and Heide-Jørgensen, 2011). The first time the Rafflesia species can be seen is when it is ready to reproduce and a bud bursts through the host’s bark. After around nine months the flower opens, and it can be over a metre in diameter and weigh more than 10 kg (Figure 11.7). Most species of Rafflesia smell of rotting flesh. This odour attracts flies to pollinate it. The fitness of the vine on which it is parasitic is dramatically reduced. Predation, which occurs when an individual, a predator, kills and eats another individual, its prey, usually takes place interspecifically, although intraspecific predation (cannibalism) does take place as well, usually when resources are very low. Cannibalism can occur across a broad range of taxa and will ultimately benefit the survival of the population, as not only does this provide a food source but there will also be reduced intraspecific competition for scarce resources. There is some evidence that certain species, such as Ursus maritimus (polar bear) (Amstrup et al., 2006) and Larus glaucescens (glaucouswinged gulls) (Haywood et al., 2014) are becoming cannibalistic in response to climate change as resources become scarce. The ecology and evolution of the predator species and the prey species are closely linked. To survive and reproduce, individuals must both feed and avoid being eaten.

Prey species have adapted to avoid being eaten, perhaps through camouflage or defensive spines. Some species are brightly coloured to show they are toxic, while others use colour to imitate toxic species to avoid being eaten. Others, such as Aglais io (peacock butterfly), have false eyes on their wings, which must appear threatening to its avian predators. Likewise, predators have adapted to catch their prey, evolving acute senses, claws, sharp teeth or venom. They usually catch their prey by pursuing them, so bursts of speed have evolved, or by ambushing them, and species have adapted by becoming camouflaged. Although the majority of predators are animals, carnivorous plants, as the name suggests, are predatory as well. These plants get nutrients, mostly nitrogen, by trapping insects and in some cases frogs and small mammals. Most plants absorb nitrogen from the soil through their roots, but carnivorous plants are most commonly found in bogs and fens where soil nitrogen concentrations are very low and they absorb the nitrogen from the prey through their leaves. These plants evolved numerous times and are an example of convergent evolution. Less well known perhaps are the 200 species of predatory fungi, which use special adaptations to trap nematodes from which they then absorb nutrients (Pramer, 1964). Predators are often central to maintaining the community structure of which they are part and those that do hold the community together are termed keystone species, an idea first conceived by Paine (1966). Paine (1969) showed experimentally that removing Pisaster ochraceus (ochre starfish) from a rocky intertidal ecosystem had a dramatic effect on the habitat. The starfish were the main predator, selectively feeding on mussels. Without the starfish the mussels were able to outcompete other species and took over the area. In under a year of removing the starfish, species diversity in the area had significantly reduced from 15 to 8 species. Another example is provided by the grey wolf of Yellowstone National Park as outlined in Box 11.1. The term keystone species is also sometimes applied to species which are not predators, such as species of the Trochilidae (hummingbird) family. They are important in the survival of several plant species through pollination. As both the plant species and the hummingbird species benefit then they are termed keystone mutualists.

11.3.3 Commensalism Figure 11.7 Rafflesia flower, Sumatra, Indonesia. These flowers are the

largest in the world and the individual flower can weigh 1 kg. (Source: Alexander Mazurkevich / Shutterstock.com)

Commensalism is when one species benefits from an interac-

tion and the other is unaffected. It is likely that commensalism began very early in evolution. The earliest example 283

Chapter 11 Ecosystem processes

WOLVES AS KEYSTONE SPECIES IN YELLOWSTONE At the end of the nineteenth century there were between three and four hundred Canis lupus (grey wolf) individuals in the Yellowstone National Park. The prey of C. lupus is predominately Bison species and Cervis elaphus (elk), which were also hunted by human settlers (Beck and Meier, 2004). As C. lupus and humans were in competition and with the additional concern that the wolves may start

hunting livestock, programmes to eliminate the wolves began. By 1924 there were no wolves remaining in Yellowstone National Park. The population of C. elaphus exploded once C. lupus was eradicated, which led to overgrazing. This caused a decline in many plant species including Salix (willow), which in turn led to loss of habitat for other species and a degradation of the environment, including increased stream erosion (Ripple and Beschta, 2004). By removing the keystone species

the whole ecosystem was threatened. Attitudes towards C. lupus changed and it was added to the Endangered Species list in the United States in 1966. In 1995 C. lupus was reintroduced to Yellowstone National Park (Bangs and Fritts, 1996) and a population became established over the next 10 years (Smith, 2005). The population of C. elaphus declined and vegetation communities have started to re-establish. Song bird populations have also started to increase, suggesting that the ecosystem as a whole may be gradually recovering.

BOX 11.1

of commensalism found in the fossil record to date is from the Cambrian, series 3, Burgess Shale (about 500 million years ago). This was Nisusia burgessensis, a Brachiopod, which is a marine animal with shell on its upper and lower surfaces, attached to a species of Wiwaxia, a soft-bodied marine animal covered in scales and spines (Topper et al., 2014). This is thought to be an example of commensalism as the sessile Nisusia burgessensis benefits from being attached to the mobile Wiwaxia species, with the result that it might have been carried to a more favourable habitat. Contemporary examples of commensalism include species of the genus Arctium (burdock) whose seed heads, known as burs, are prickly and attach easily to fur of passing animals. In this example, the seeds of the Arctium species are dispersed and the animal is not affected.

11.3.4 Amensalism Amensalism is where one species is unaffected and the other species is negatively affected. An obvious example of this is when animals trample vegetation. The vegetation being crushed is negatively affected, but the animals are not impacted. Allelopathy, the phenomenon whereby plants release chemicals that have a negative impact on other plants, but do not have a positive or negative effect on themselves is another example of amensalism,

284

although some argue that there is a cost to the plant in producing these chemicals and consider it to be a form of competition.

11.3.5 Competition Competition can be both interspecific and intraspecific and occurs when a particular resource is limited, such as food, water, shelter, or in the case of intraspecific competition, mates. When competition takes place both individuals will suffer. If two species have a very similar set of requirements, a similar niche, they will not be able to exist in the same area as one will outcompete the other. This is known as the Competitive Exclusion Principle and was developed by Gause (1932, 1934). Gause demonstrated the principle in experimental studies using Paramecium species, a single-celled protozoa. As shown in Figure 11.8, when the two species are grown separately both populations establish and stabilise. When grown together, Paramecium Aurelia population grows more slowly, due to competition with Paramecium caudatum, but it is P. caudatum which is outcompeted and does not survive. To co-exist, one species would need to evolve so that its requirements were not exactly the same as the other’s and it would therefore have a different ecological niche. This adaption for many species could involve altering the feeding time, changing the main food source, or moving reproduction earlier or later.

11.4  Temporal change in ecosystems

60

P. caudatum alone

P. caudatum in mixed culture

Volume of population

30

0 P. aurelia alone

80

P. aurelia in mixed culture

40

0

2

4

6

8

10

12

14

16

18

Days Figure 11.8 The impact of competition on population densities of P. caudatum and P. Aurelia. Solid lines show population growth rate when species are

grown individually; dotted lines when they are grown in a mixed culture. It is clear that the growth rate of both species is reduced, and P. caudatum goes extinct within 18 days in a laboratory culture experiment. (Source: Gause, 1934)

Reflective questions ➤ What are the terms used for biotic interactions that have a positive effect on organisms?

➤ Why are keystone species so important for the functioning of an ecosystem?

➤ What is the difference between intraspecific competition and interspecific competition?

➤ Why would cannibalism be beneficial to a population?

11.4 Temporal change in ecosystems Ecosystem processes in an area cannot be studied at one point in time and then thought to be understood. The processes are affected by past events as well as responding to current changes in the environment. The change may be short term, such as annual variation in weather or variation in population size at different trophic levels. Changes may also be longer term. For example, some species are still advancing towards higher latitudes in response to continental ice sheet melt more than 10 000 years ago. Hence these slow species movements will slowly alter the

processes of the ecosystems they occupy. In addition, some individuals have lived for thousands of years. A Pinus longaeva (Great Basin bristlecone pine) in Wheeler Peak, Nevada has been recorded at over 4800 years old and several Sequoiadendron giganteum (giant sequoia) trees in Sierra Nevada, California have been recorded at over 3000 years old (Brown, 1996). The condition of the ecosystem they germinated in may be very different to those occurring now; and the trees themselves may have played a role in changing local ecosystem processes. Understanding how ecosystem processes respond over time to cyclical and one off changes is important if predictions regarding impacts of global change are to be made accurately.

11.4.1 Short-term changes Annual cycles within an ecosystem cause great fluxes in ecosystem processes over the year. For example, if primary productivity in temperate areas was measured in winter it would give a very different value than if it was measured in summer. For other ecosystems the annual cycle may vary between dry and wet seasons. Migration is another factor which causes ecosystem processes to fluctuate through the year and as ecosystems have evolved with these fluctuations they play an important role in the normal functioning of those areas. Climate change is already affecting annual

285

cycles (see Chapter 13) with many species of plants flowering earlier in the UK than fifty years ago (Fitter and Fitter, 2002) and migratory birds are altering the timing of their autumn migration (Jenni and Kéry, 2003). These changes will occur over longer periods and will change ecosystem processes profoundly as pollinators and flowering times are no longer in sync or predator–prey cycles alter. Ecosystem processes within a habitat should not just be considered on a yearly scale though, as some cycles which affect them will occur on an inter-annual basis. Some of these factors will be predictable such as fluctuations in herbivore populations which in turn determine plant biomass and nutrient cycling. For example, Olofsson et al. (2012) showed that fluctuations in lemming and vole densities (grazers in the ecosystem), drove fluctuations in plant biomass and these fluctuations occurred on three yearly cycles. Other predictable inter-annual events which will cause changes in ecosystem processes include El Niño events, which cause drought and flooding recurring every 2 to 10 years (see Chapters 6 and 8), and these inter-annual events are part of the ecosystem processes in these areas.

11.4.2 Disturbance and resilience Disturbance in ecosystems is caused by a distinct event which changes the usual functioning and processes of the ecosystem. Disturbances may be small scale, such as a single tree fall within a forest or much larger events such as a landslide, volcano or fire. Natural disturbance is integral to the processes of most ecosystems, but disturbance may also be caused by humans, for example, by felling trees or overgrazing with domestic animals. In either situation, disturbance often leads to a sudden decline in biomass. Natural disturbance is key to maintaining high biodiversity in an ecosystem where competition would otherwise exclude some species. Ecosystems with an intermediate disturbance regime are likely to have highest biodiversity. Ecological resilience was first defined by Holling (1973) and is the ability of an ecosystem to withstand disturbance by resisting change. High biodiversity is thought to increase ecosystem resilience, as with greater diversity there are more options for that ecosystem to recover. Therefore, disturbance and resilience are intertwined as disturbance maintains higher biodiversity and high biodiversity is needed to recover from disturbance. Human activities can reduce resilience as humans often try to control variability in ecosystems. For example, reducing the frequency of natural fires in an ecosystem often leads to more occasional but extremely severe fires as combustible material builds up and once the fire occurs it can become 286

Species numbers

Chapter 11 Ecosystem processes

1 Disturbance event

2 Disturbance event

Time Figure 11.9 Effects of repeated disturbance on species numbers.

so intense that ecosystems cannot always recover (Folke et al., 2004). In addition, if disturbance events occur in quick succession it may be harder for an ecosystem to recover, as is shown in Figure 11.9.

11.4.3 Succession After a disturbance event, which leads to the sudden loss of biomass and/or structure of an area, ecological succession will occur. Succession is the sequential change in species over time following a disturbance event. Succession is an important concept in ecology and has been debated and discussed for over one hundred years (Walker and del Moral, 2003). Succession does not consider seasonal changes in vegetation and similar regular disruptions, nor does it address change on a long-term, palaeoecological scale. However, an understanding of succession can aid in the prediction and interpretation of disturbance responses of ecosystems as a result of increased human interference and global change and aid conservation efforts to recreate habitats. Two types of succession are recognized: primary and secondary succession. Primary succession starts in areas that are completely devoid of life such as after a volcano eruption or glacier retreat. There is no soil – species must colonise the area from the outside and soil must slowly develop. Secondary succession takes place after less dramatic disturbances, such as after a forest fire or the abandonment of agricultural fields. After these events soil is present, though probably degraded, but there will be seeds present in the soil, ready to germinate and there may be plant shoots or roots which can resprout. Secondary succession can therefore progress much quicker than primary succession. In the field, succession has been studied using longterm study sites and by studying chronosequences, which are similar sites where disturbance took place at different points in the past (Figure 11.10). In chronosequences the

11.4  Temporal change in ecosystems

(a)

(c)

(b)

(d)

Figure 11.10 Different stages of succession of a Panamanian forest after clearance by humans. (a) Cattle pasture, the forest has been recently

cleared and grasses predominate. The seedling trees in the foreground are around two months old and farmers cut these to maintain the pasture, some remnant mature trees survive. (b) A forest that has been recovering from clearance for three years. (c) A forest 26 years since it was cleared. (d) A mature forest. The differences between these forests can be studied and researchers can gain an understanding of the successional processes. (Source: Sarah Batterman, University of Leeds)

287

Chapter 11 Ecosystem processes

sites are at different successional stages and the study uses space as a substitute for time. Succession has also been studied by manipulating sites experimentally (Wardle et al., 1999) and using theoretical models of succession. Succession has been classified into stages: pioneer, early, mid and late. In the example of a forest developing after glacial retreat, the pioneer community is made up of lichens and mosses and other autotrophic species which do not require soil, but can cling to the rock. These species provide food for grazers and decomposers which will then colonise the site. Gradually soil builds up as decomposers break down dead plant and animal matter. The soil and organisms retain water in the ecosystem. Eventually vascular plants will invade the area once the soil is suitable. These will create more litter for the developing soil and more resources for grazing species and decomposers. Pioneer species may be gradually replaced. With a lot more time, conditions will become suitable for woody plants to become established. Woody plants generally require more soil and are slow growing so their establishment is delayed. Their establishment also relies on seeds dispersing to the area. The further away species are from the disturbed area the longer this succession can take. Eventually, enough trees will grow to produce a forest canopy and

LIFE HISTORY TRADE-OFFS In evolutionary terms an individual is successful if it reproduces and has offspring that successfully reproduce themselves. The strategies to increase an individual’s chance of survival, reproduction and the successful reproduction of its offspring, are its life-history strategies. Putting a lot of energy into a large number of offspring would clearly increase the success of an individual, yet this is not possible as the cost to the individual in terms of resources and energy needed to do this would be too high. Instead there is a trade-off. For example, an organism will produce either a few, ‘high cost’ offspring, or a lot of ‘low cost’ ones. The strategy that has evolved depends on the environment in which

early pioneer tree species (see Box 11.2) will be replaced by slow growing species. This final community is termed the climax community and is considered to be quite stable. Some ecosystems will never reach a climax community if the time between disturbances is less than the time needed to reach a climax state. Natural grazing and fire regimes may prevent a climax from being reached, as may human disturbances. Mechanisms that enable change through the successional stages are still being understood and proposals include facilitation, random colonization, competition and inhibition. Facilitation, where one successional stage ‘paves the way’ for species typical of the next successional stage by altering the environment, such as light or water availability, or changing the soil, is the most common mechanism proposed to explain succession. For example, Alnus species (alder) are abundant in an intermediate stage of glacial succession. As they drop their leaves they lower the soil pH. The change in pH facilitates the entry of Picea species (spruce), which require acidic soil. Using the same example can show how species might inhibit the next successional stage. Alnus species increase light competition and reduce the germination success of Picea species and in doing so inhibit the next stage of succession.

the species lives. In environments where disturbance is likely and many individuals will not survive to reproductive maturity, producing a lot of offspring will increase the chances of one surviving and producing offspring of their own. The trade-off is that limited resources can be put into that offspring: a plant would produce a lot of small seeds, animals may be of small size and gain little or no care from parents. Species which display these characteristics are known as r-selected or pioneer species and will be the first to colonise an area after a disturbance event. Typical r-selected species include rodents and ‘weedy’ plant species. If, however, the environment is stable and the majority of individuals reach reproductive age, greater investment in each

offspring is most likely to result in reproductive success. Individuals with these characteristics are known as K-selected or climax species and a full list of typical traits is shown in Table 11.2. K-selected species will only be found in communities which are reaching, or at, their climax. In these communities there will be strong competition for resources and the extra energy put into each offspring will give it an advantage. Examples of K-selected species include large mammals and slow growing trees such as Quercus species (oak). The terms r and K are taken from mathematical models used to predict population growth and size. In these models, r represents the population growth and K the carrying capacity (the number of individuals an area can support based

BOX 11.2 ➤

288

11.5  Human impact

➤ Table 11.2 The characteristics of extreme r-selected species and extreme K-selected species

Characteristic

r-selected (pioneer) species

K-selected (climax) species

Environment

Unstable

Stable

Energy required to make an individual

Low (in flowering plants this would be reflected in low seed size, which need few resources to produce)

High (in flowering plants this would be reflected in a large seed, which use a lot of resources to produce, but will provide a lot of energy for the new plant)

Size

Small size

Large size

Survivorship

Most of the individuals die within a short time but a few live much longer

Most individuals live to near the maximum life span

Maturation

Early

Late

Reproductive episodes

One

Multiple

Number of offspring

Many

Few

Life expectancy

Short

Long

on resources available) of a habitat. An r-selected species means that it is one where the population is growing and there are plenty of resources available. A K-selected species is one which is found when the habitat is at carrying capacity

and there is not the potential for populations to grow further. In reality, there is a continuum between r-selected and K-selected species with some species displaying some characteristics of each strategy, or moderate versions

of one or other. For example, Sequoiadendron giganteum is a large tree which lives a very long time and takes decades to reach maturity. However, it produces thousands of tiny seeds and suffers high pre-reproductive mortality.

BOX 11.2

Reflective questions ➤ What is the definition of ecosystem resilience? ➤ Why can disturbance be beneficial to ecosystems? ➤ Why might a climax community never be reached by an ecosystem?

change, alien invasions, fragmentation, pollution and over-harvesting of plant and animal species are all major threats to ecosystems. Global climate change is altering ecosystems world-wide and Chapter 13 focuses on that topic, so it will not be considered further here.

11.5.1 Degrading ecosystems 11.5.1.1 Overharvesting

11.5 Human impact Understanding human impacts and being able to predict the consequences to ecosystem processes is one of the most fundamental issues faced by ecologists today (Murphy and Romanuk, 2014), primarily because human life is dependent on the functioning of ecosystems (Millennium Ecosystem Assessment, 2005). Habitat destruction clearly has the most detrimental impact on ecosystems, but remaining ecosystems are degraded through our actions, often resulting in a loss of biodiversity. Global climate

Overharvesting, or over exploitation, occurs in natural populations when the rate of harvest of individuals from a population is greater than the rate of replacement. This leads to a decline in the population size of that species. If overharvesting continues, that species will eventually become locally extinct. Species with a low reproductive rate, such as old growth trees, elephants and whales, are particularly prone to over harvesting, due to the long time period it takes to replace the individuals taken and the age at which replacement individuals reach reproductive maturity. 289

Chapter 11 Ecosystem processes

Logging and fishing are typical industries which can cause problems with overharvesting and studies have shown that it is not just individual species that are suffering from their impacts. Muler et al. (2014) studied Euterpe edulis (juçara palm), which is selectively logged for palm heart harvesting. They found that after harvesting of E. edulis individuals, seed density of light dependent climbers, pioneer tree species and bamboo increased. This may be an expected result of a disturbance event. However, the species richness of the seed rain (seeds falling from canopy) had declined by half in a five-year period, particularly of animal dispersed species, suggesting a significant alteration of the ecosystem dynamics, rather than it just being the population of E. edulis that had been impacted. Selective logging has additional environmental impacts as tracks are made into the forest to reach the trees, which if not done carefully can cause soil compaction and fragmentation of the forest.

11.5.1.2 Fragmentation Habitat fragmentation occurs when a continuous ecosystem is separated into patches, due to the destruction of some of the habitat, perhaps by road building, conversion to agriculture or other development. Although only a small area of the original ecosystem may have been destroyed, the remaining patches are usually badly degraded and respond with a rapid reduction in species richness. Quammen (1996) used the analogy of a Persian carpet, if you cut the original into squares, you are not left with something of equal value, but unravelling rags. Degradation through fragmentation is caused firstly because there is some reduction in size of the overall habitat. Mobile species may have been able to move to another surviving patch but this can cause an increase in competition for the remaining resources. Immobile (sessile) organisms will have been destroyed. Secondly, the remaining patches may have been degraded during the fragmentation. In addition, the patches are effectively islands and following island biogeography theory (Box 11.3) a smaller island can support fewer species. Some species which can move between patches, such as bird species, may not be affected as much by fragmentation. Other species which cannot move between patches and have a large range size, may struggle to find enough food to survive. Divided populations may also be too small to remain viable and may die out. Fragmentation can also lead to increased edge effects. When fragmentation takes place the boundary between the remaining patch and the destroyed habitat can be 290

abrupt. The edge of the remnant patch will be impacted by the environmental conditions of the newly created ecosystems and changes in biotic interactions as a result of this. If a forest is fragmented, often the adjacent area may be agriculture or roads where light levels will be higher, temperature will fluctuate more and moisture levels will be altered. At the boundary of the patch there will be a gradient between the environmental conditions, so that the environment at the edge of the patch will be quite different from that of the interior. In the short term this will lead to leaf flush in the edge vegetation, which may attract additional herbivores and predatory animals and birds. In the longer term, different tree species may be favoured again, altering the species dynamics within the ecosystem. In addition, the external biota will interact with species from the fragment along the fragment edge. These may be domestic predators such as cats entering the edge of the forest, or pests from crop species. There may also be problems with invasive species entering the fragment along its edge. Patch size and shape determines how much of the fragment is impacted by edge effects, with smaller patches and long thin patches more greatly impacted as is shown in Figure 11.11. Habitat corridors can, to some extent, overcome some of the impacts of fragmentation. Habitat corridors, such as a hedge on a field margin running between two woodland patches, connect patches of an ecosystem. The hedge provides a route where species can migrate along, there can be mixing of separate populations and animals can look for food in both patches. Other corridors include the provision of tunnels under major roads joining two habitats and even rope bridges connecting the canopy of forest fragments to enable arboreal species, such as monkeys, to cross roads.

11.5.1.3 Alien invaders Species that have been introduced to ecosystems outside their natural range and are not only naturalized (consistently reproducing) but increasing their range are termed, invasive, alien or exotic species. Some species have been introduced deliberately, perhaps as horticultural specimens, such as Impatiens glandulifera (Himalayan balsam), Heracleum mantegazzianum (giant hogweed) and Fallopia japonica (Japanese knotweed), which were all introduced in the UK in the nineteenth century as garden ornamentals (Coombe, 1956; Beerling et al., 1994; Tiley et al., 1996). Others were introduced as biological control. For example, Rhinella marina (cane toad), which is native to Central and South America, was introduced

11.5  Human impact

Edge Edge Edge Interior habitat

Interior habitat Interior habitat

(a)

Interior habitat (b)

Edge

Interior habitat

Figure 11.11 Edge effects: (a) edge effect on patches of a similar size but different shape; (b) edge effects have a greater impact on smaller areas.

ISLAND BIOGEOGRAPHY THEORY MacArthur and Wilson (1963) first proposed the theory of island biogeography as a predictor of species richness on islands. They developed the theory based on oceanic islands, although islands in ecology are not always surrounded by water, but can be a habitat surrounded by a very different area, such as a forest fragment surrounded by arable fields, or ponds, mountain tops and grasslands

surrounded by urban development. Island biogeography theory can be applied to these areas as well and the theory has therefore become important for conservation and the creation of reserves. MacArthur and Wilson (1963) were able to show that larger islands had a greater species richness than smaller ones, and that there was a consistent relationship between island size and number of species. Figure 11.12 shows how the species richness of bird species on the various

islands of the Sunda group, in the Malay archipelago, increases with an increase in island size. Larger islands would probably have more resources and different habitats for species to occupy. These findings have been repeated many times by different researchers and for different groups of species. MacArthur and Wilson also proposed that distance to the island from the mainland was a factor. They proposed that species richness of the island was

BOX 11.3 ➤

291

Chapter 11 Ecosystem processes

➤ 1000

Number of species

100

10

10

100

1000

10000

100000

Area in square miles Figure 11.12 The numbers of land and freshwater bird species on various islands of the Sunda group, together with

the Phillipines and New Guinea. The islands and grouped close to one another and to the Asian continent and the Greater Sunda group, where most of the species live. (Source: after MacArthur and Wilson, 1963).

Large Island

Number of species on island

Far Island

b)

an ds

greater change of a species arriving there, than to a small island. Therefore, large islands close to the mainland will have much greater species richness than small islands a long way from the mainland. Not all islands, however, are newly formed volcanic islands to be colonized

Far

isla

all

ar

isl

Ne

isl

Sm

Rate of extinction or immigration

n tio tin c Ex

Rate of extinction or immigration

(L Extin ar ge ctio Isl n an d)

E (Sm xtin all ctio Isla n nd )

ion at d) gr an mi Isl Im all m (S Small Island

a)

survive on the island until competition for resources led to extinction of some species. The balance between migration from the mainland and extinction represents the species richness equilibrium for that island (Figure 11.13). The rate of migration to a large island will be higher, as there is a

on ati igr and) Imm ar Isl ion (ne at d) gr mi lan Im ar Is (f

n atio igr and) Imm ge Isl r (La

Rate of extinction or immigration

composed of species able to migrate to the island from the mainland. For islands closer to the mainland, species are more likely to migrate there by chance. The further away from the mainland an island is the less likely migration to it will be. More species could continue to migrate and

an

d

nd

g Lar

ds

lan

e is

s

Near Island

Number of species on island

c)

Number of species on island

Figure 11.13 Species richness at equilibrium based on island biogeography theory. Where immigration and extinction intersect will be the

species equilibrium for the island: (a) the point of equilibrium will be different for large and small islands, with lower species richness on small islands; (b) species equilibrium is different for islands close to the mainland compared to those further away, due to different rates for immigration; (c) considering island size and distance from the mainland together, species equilibrium is lowest for small islands far from the mainland, and largest for large islands, close to the mainland.

BOX 11.3 ➤

292

11.5  Human impact

➤ by species migrating from the mainland. Some islands will previously have been connected to the mainland and a rise in sea-level has cut them off resulting in what is known as a continental island.

These islands will already have a number of species on them. However, as a result of becoming an island they may no longer be able to sustain as many species because the limited size of the island reduces the

resources available. Therefore, continental islands are likely to undergo a period of extinctions before their equilibrium is reached.

BOX 11.3

to Queensland, Australia in 1935 in an attempt to control the beetles Dermolepida albohirtum and Lepidiota frenchi (Phillips et al., 2007). It rapidly expanded its range and is now found in over one million square kilometres of subtropical and tropical Australia (Phillips and Shine,

2004) (Table 11.3). Unfortunately, it is unclear if it actually helped reduce crop damage caused by sugar beetles. Finally, some species will have been transported accidentally in different ways including as seeds and microorganisms in soil, either around plants or on footwear;

Table 11.3 Case studies of the invasive species Heracleum mantegazzianum, Rhinella marina, Dreissena polymorpha Species name

Heracleum mantegazzianum

Rhinella marina

Dreissena polymorpha

Common name

Giant hogweed

Cane toad

Zebra mussel

Description

A large plant with stems that can grow to more than 2 m high. It has an umbrellashaped flowering head which can be up to 600 mm in diameter

These are large amphibians with averagesized adults measuring 100–150 mm long and weighing around 1.5 kg

Zebra mussels are small animals, 10–40 mm long at maturity and have D-shaped shells with yellow and brown coloured stripes

Natural range

Southern Russia and Georgia

Central and South America

Eastern Europe and southern Russia

Invasive location

Europe, North America and Canada

Australia

North America originally in the Great Lakes, they have also spread to the Gulf of Mexico and southern Canada

How introduced

Introduced into Britain via the Royal Botanic Gardens Kew as a horticultural plant in the early 19th century

Introduction as a biological control agent of pest beetles in the sugar cane industry in 1935

Accidently introduced in the ballast water of ships

Reproduction

Each plant can produce 1500 to 100 000 seeds per year

They are reproductively very successful. Females lay 8000 to 35 000 eggs at a time and may produce two clutches a year. Although only around 0.5% of cane toad individuals that hatch from eggs survive to reach sexual maturity and reproduce, this can still lead to a dramatic increase in population size

Female zebra mussels can produce 1 to 5 million eggs in their life time and around 100 000 offspring will reach reproductive maturity

Problems caused

The plants form dense stands, outcompeting native species through vigorous growth and shading and reduce the biodiversity of plant and animal species locally.

Poisonous throughout its life cycle, which can lead to the death of predators.

They filter feed on plankton, reducing the availability of food for fish and native mussels. As fish numbers reduce, this can have implications for the rest of the food web. They can also smother native mussels by attaching to them.

Additionally, furanocoumarins in the sap cause extreme photosensitivity which leads to severe blistering of the skin. The hollow nature of the stem is tempting for unsuspecting children to blow through, causing severe blistering in the mouth and around lips

It is rapidly extending its range in Australia. It has been linked to decline in predators as they extend into new ranges, but there is also evidence that predators are adapting behaviourally and physiologically to avoid poisoning. They are insectivorous and have impacted the populations of native insect species, in addition to outcompeting native insectivores for food. They also outcompete native toads for shelter.

The reduction in phytoplankton results in clearer water, which can increase the growth of aquatic vegetation. They cause problems by blocking water intakes for power plants and residential properties

There are also no natural predators or diseases in Australia to limit numbers

293

Chapter 11 Ecosystem processes

invertebrates in packing crates; and aquatic species in the ballast water of ships. Dreissena polymorpha (zebra mussel) is native to the Black and Caspian Seas and was introduced to the North American Great Lakes, released as larvae in ship ballast water (Hebert et al., 1989). It has now increased its range to several major rivers and inland lakes of eastern North America (Table 11.3). Invasive species can alter ecosystem processes by altering species interactions, such as through competition. Hemidactylus frenatus, an invasive gecko throughout the Pacific, has been shown experimentally to outcompete Lepidodactylus lugubris, the native gecko species (Petren and Case, 1996). Invasive plant species can also alter soil chemistry and nutrient dynamics by having different leaf tissue chemistry and phenology (Ehrenfeld, 2003) and different root morphology can lead to soil erosion. In addition, ecosystem processes can be altered by invasive plants changing fire regimes as they have different fuel properties to native ones (Brooks et al., 2004) and if an invasive species establishes after a disturbance event then the natural progress of succession will be altered. Although many species introduced to an area will not survive, or will become naturalized in low densities, in 2013 there were 751 species in a global database of invasive trees and shrubs (Rejmánek and Richardson, 2013) and it is likely there will be many more added. While trees and shrubs and their ranges are well documented for much of the world, it will be much harder for some groups of organisms, particularly microbes, to put together such a list of invasive species.

11.5.2 Urban ecology Recently, there has been increasing interest in urban ecology, which is not surprising given that since 2010 over half of the world’s population lives in cities rather than rural areas (United Nations, 2011). Although cities only account for 3% of the Earth’s terrestrial surface (Schneider, 2010), this is where the majority of people experience ecology on a daily basis. The ecology of these created environments is fascinating. Urban habitats are, unlike many others, rapidly expanding. Cities tend to coincide with areas that originally had high biodiversity (Cincotta et al., 2000; Luck, 2007). By conserving the biodiversity of urban areas, global biodiversity could be greatly preserved. A global study by Aronson et al. (2014), showed that species richness was drastically reduced in cities, with only 25% of native plant species and 8% of native bird species found compared to non-urban environments. Despite this, urban areas retained endemic native species, thus maintaining regional biodiversity. 294

Although increased native biodiversity in cities would be a positive step, retaining endemic species is hugely important. Urban ecology originally focused on showing how urban environments differ in ecology from non-urban areas (McPhearson et al., 2016). The tools it uses are those of the ecology of rural areas, applied to green spaces in cities. This has been useful in understanding the biodiversity of cities and helping develop the conservation of wildlife within cities. However, an approach was needed that took urban ecology further and there was a shift in ecological studies to recognize that the city is an ecosystem and the processes of this ecosystem are strongly influenced by humans and the social, political and economic pressures that drive human actions (Grimm et al., 2000; McPhearson et al., 2016). Hence, an interdisciplinary approach is needed when studying urban ecology (Figure 11.14). Urban ecology is therefore at the interface between human decision making and ecological processes and the feedbacks and complex interaction between the two. It is a topic which is probably best studied using an interdisciplinary approach, with expertise in social sciences as well as the sciences; a social ecology. However, urban ecosystem processes are still poorly understood (Elmqvist, 2015). The ecology of the urban area can bring many benefits both to the functioning of a city and also to the wellbeing of people who live and work there. These include a reduction in job stress (Stigsdotter et al., 2007), a reduction in the urban temperatures caused by pollution (Gill et al., 2007), and reducing the toxicity of runoff from roads (Spromberg et al., 2015). These benefits are termed ecosystem services (Box 11.4). Science and policy are often considered at a national or perhaps international scale, but planning and green infrastructure decisions are usually taken at a local level. Scientists and local planners need to work carefully together to ensure that biodiversity and ecological benefits are enhanced when developing our urban spaces. Small differences in a local urban ecosystem when upscaled globally may have a huge difference on the impacts of global change and on the species richness of the Earth as a whole.

11.5.3 Conservation Dasmann (1968) first introduced the concept of conservation biology. As a discipline it is concerned with developing practical approaches to protecting species and ecosystems. Often conservation policies put in place are static, focusing on an ecosystem at a particular point in time. However, as has been discussed through this chapter, ecosystems are not static but fluctuate in space and time, responding to changes in, for example, climate,

GN

ECO LO

E SI

GY I N

ND BA UR

CITI ES

11.5  Human impact

GO VER

NA

NC

CES

E IAL SOC

COMPLEXITY

N SCIE

ECOLOGY

OF CITIES ING

ANN N PL

SYSTEMS THINKING

IP

H DS

AR

EW ST

URBA

Figure 11.14 The ecology of cities approach incorporates several disciplines, interdisciplinary collaborations and interactions

with other ecosystems. (Source: from McPhearson et al., 2016)

ECOSYSTEM SERVICES Ecosystem services can be defined as ‘the direct and indirect contributions of ecosystems to human well-being’ (Braat and de Groot, 2012; TEEB, 2012). This definition is widely cited and had evolved from the 1980s when the concept was introduced for conservation (Ehrlich and Mooney, 1983). Since that time the term has been redefined and a framework for ecosystem services was set out by the Millennium Ecosystem Assessment which took place between 2001 and 2005. The Millennium Ecosystem Assessment involved 1,360 experts from 95 countries and it split ecosystem services into four

categories – provisioning, cultural, regulating and support as shown in Figure 11.15 (Millennium Ecosystem Assessment, 2005). Ecosystem services concepts have been adopted as tools for conservation and have become a very popular in recent years. Often it is possible to show (in financial terms) what value an ecosystem has to society in its ability to reduce flood risk, absorb carbon, provide clean water, provide food and so on. This can therefore help policy-makers invest in ecosystem protection where they can see it is more cost effective to do so for society as a whole rather than pay for the costs of damage caused by a degraded ecosystem (more flooding, poor water quality and so

on). In 2011, countries committed to the Convention of Biological Diversity signed a new plan until 2020, there are currently 196 parties signed up to the Convention. This included a commitment to stop biodiversity loss and ensure healthy ecosystems providing services for people. The EU in response to this set up a project mapping and assessing ecosystems and their services across Europe and committed themselves to the, ‘better protection of ecosystems and the services they provide’ (European Commission, 2011). Despite this adoption, there are difficult issues about how to measure conservation success using ecosystem services. Mace (2014) pointed out that contradictory messages

BOX 11.4 ➤

295

Chapter 11 Ecosystem processes

➤ Regulating Benefits water, erosion, air quality, climate, disease

Cultural Non-material benefits spiritual, aesthetic, recreation, knowledge, social

ECOSYSTEM SERVICES

Provisioning Products food, fuel, fibre, pharmaceuticals, genetic resources

Supporting Services photosynthesis, nutrient cycling, water cycling, soil formation

Figure 11.15 Categories of ecosystem services.

may occur. High biodiversity is accepted to be important for the functioning of natural ecosystems, yet ecosystem service provisioning food for the world’s population have usually resulted in monocultures. This may result in confusing goals for policy. In addition to problems with measuring the success of conservation based around

ecosystem services, Carpenter et al. (2009) suggested that the ecosystems services concept is too utilitarian and there should be a move away from what nature can provide for people towards people and nature together, a move towards a socioecological system. Glaves and Egan (2010) stated that ecosystem disservices should

also been considered not just services. In agriculture these disservices include crop pests, such as herbivores and pathogens, and weed competition, both of which decrease productivity and potentially lead to crop failure. Disservices may also include the hosting of pathogens and pests and restricting human mobility.

BOX 11.4

Reflective questions ➤ What species are particularly susceptible to overharvesting?

➤ What is an edge effect? ➤ How can alien species affect ecosystem processes?

disturbance and biotic interactions. A conservation plan will not succeed in the long term unless ecosystem processes are considered in conjunction with other measures (Bennett et al., 2009). The methods for managing different ecosystem processes will vary depending on what is being conserved. A conservation plan for a peatland or a rare insect will need to be very different because their response to changes in ecosystem processes will vary greatly. Early conservation approaches typically involved creating a reserve to prevent human intervention. Even if these reserves were planned with knowledge of edge effects on fragments it is may be that this conservation effort would not be successful in the long term. Some fens, for example, 296

if left, would through succession, develop into scrub and then woodland. If conservation of the fen is important for the high biodiversity of plant and animal species, then management may sometimes be needed to halt succession. Climax communities often have lower biodiversity than intermediate communities, as intermediate communities have a mixture of pioneer and climax vegetation and a greater variety of habitats for animals. If high biodiversity is the aim of the conservation effort then an understanding of succession in that community is needed. Some ecosystem processes are altering because of large-scale external forcing. Climate change will result in changes in species composition in many areas: some species will be lost but others will replace them. The connectedness of reserves or intact habitat becomes crucial to enable migration and dispersal to new areas. Planning of suitable habitat corridors is therefore essential to mitigate climate change effects. In addition, the consideration of island biogeography theory and the understanding that ‘islands’ closer together or closer to the ‘mainland’ are likely to have higher rates of immigration is important to consider during conservation planning.

Further reading

11.6 Summary

environmental conditions change. Sinks and sources of carbon have been particularly keenly studied.

Ecosystem processes include how energy flows through an

Exactly how energy and nutrients move through an ecosystem

ecosystem, the biogeochemical cycling within the system, the

will be determined by biological interactions between species

community dynamics and responses to disturbance events.

called symbioses. Symbioses can be positive, where species ben-

In the majority of ecosystems, primary producers use solar

efit from the interaction, or negative, and can strongly influence

energy to convert CO2 and water into carbon sugars through

ecosystem processes. Mycorrhizal symbioses will, for example,

photosynthesis. Where conditions for photosynthesis are good

increase the availability of phosphate and nitrogen to plants,

(high temperature, light and water availability) the net primary

increasing plant growth, and net primary productivity, thereby

productivity of an ecosystem will be higher and the ecosystem

increasing the number of organisms the ecosystem can support.

therefore has the potential to support more organisms. The

If competition for resources is too strong then species will be

energy produced through photosynthesis moves through the

eliminated from an area.

trophic levels of the ecosystem food web as organisms in turn

Ecosystems are not static in time, but fluctuate in response

are eaten. As energy dissipates as heat at each level, follow-

to internal and external environmental conditions. These may be

ing the laws of thermodynamics, less energy is available to the

predictable changes over an annual or inter-annual cycle or may

next trophic level. This limits food webs to around five trophic

be responses to one off disturbance events. Many ecosystems

levels, although, it can be slightly higher in aquatic ecosystems.

have evolved with disturbances as part of their natural cycle and

Organic detritus is broken down by detritivores and in this

disturbance regimes can help maintain high biodiversity within

way nutrients are made available again in the soil for primary

an ecosystem. Severe disturbance events, such as lava flow,

producers to take up. Some elements and molecules will cycle

fires or landslides may lead to succession. Through this process

locally within the ecosystem. Others will enter the global bio-

ecosystems reform slowly until a climax community is reached or

geochemical cycles leaving the particular ecosystem in water,

another disturbance event occurs.

dust or as gas. Some ecosystems will be a net store of certain

Human actions are impacting significantly on ecosystem pro-

elements and are known as sinks, and others will be a net emit-

cesses. As humans are reliant for survival on the services provided

ter of them and so be a source of that element. To understand

by ecosystems it is vital that the processes underpinning them are

global climate change and to make predictions about the future,

fully understood. There is an increasing realization that working with

ecologists are trying to determine which ecosystems are sinks

ecosystem processes and enhancing them, even in the human-dom-

and sources of different elements and how this changes as

inated urban ecosystem, can bring advantages to human well-being.

Further reading Begon, M., Howarth, R.W. and Townsend C.R. (2014) Essentials of ecology, 4th Edition. Wiley, Chichester. A clear undergraduate introduction to ecology. Chapin, F.S. III, Matson, P.A. and Vitousek, P.M. (2011) Principles of terrestrial ecosystem ecology, 2nd edition. Springer, New York. An easily accessible undergraduate book providing detailed coverage of the main concepts in ecology. Cox B., Moore, P.D. and Ladle, R.J. (2016) Biogeography: an ecological and evolutionary approach, 9th edition. WileyBlackwell, Oxford.

A good undergraduate text introducing biogeography, covering both the background and more contemporary research in this area. Currie, W.S. (2011) Units of nature or processes across scales? The ecosystem concept at age 75. New Phytologist, 190, 21–34. A paper exploring the ecosystem concept and processes and the scales at which these should be considered. Mace, G.M., Norris, K. and Fitter, A.H. (2012) Biodiversity and ecosystem services: a multilayered relationship. Trends in Ecology and Evolution, 27, 19–26. A paper exploring the relationship between biodiversity and ecosystem services. There are clear explanations throughout on the concepts covered. 297

C HAPTER 12

Freshwater ecosystems Lee E. Brown School of Geography, University of Leeds

Learning objectives After reading this chapter you should be able to:

➤ understand some of the key scientific concepts underpinning the study of life in rivers and lakes

➤ recognize some of the major groups of freshwater plants and animals

➤ appreciate the major differences between flowing and still water ecosystems

➤ describe some of the ways in which freshwater ecosystems have been altered by human environmental change

12.1 Introduction Although rivers and lakes constitute only an estimated 0.01% of the world’s water resources and cover approximately 0.8% of the Earth’s surface, these habitats have a disproportionately high diversity of plants and animals with at least 6% (or 7100 000) of known species estimated to be found in freshwaters (Dudgeon et al., 2006). Over 10 000 fish species are known from freshwaters with some 90 000 species of invertebrates described, the richest

groups being the insects, crustaceans, molluscs and mites. Other well-studied groups of freshwater organisms include amphibians, mammals, birds, macrophytes (plants) and algae (e.g. diatoms, phytoplankton) (Figure 12.1). A diverse assemblage of bacteria, fungi, protozoa and rotifers is also found. However, knowledge of freshwater diversity remains incomplete and new species are identified every year. Organisms inhabiting freshwaters can be grouped according to their role within aquatic food webs. For example, producers or autotrophs are the plants and algae that synthesize biomass from inorganic compounds and light. Producers can be attached to surfaces such as rocks or other plants (e.g. filamentous algae or macrophytes), be rooted in loose sediments (macrophytes), or be free living in the water column (e.g. phytoplankton). Those species that exist within the aquatic ecosystem and directly provide energy to the aquatic food web are termed autochthonous producers. Heterotrophs are organisms that obtain organic matter by consuming autotrophs, other heterotrophs or detritus. Members of this group can be considered either as herbivores (consumers of attached algae, plants and phytoplankton), detritivores (consumers of dead organic matter) or predators (consumers of living heterotrophs), although some organisms

12.1  Introduction

(a)

(b)

(d)

(e)

(c)

(f)

(g)

(h)

(i)

(j)

Figure 12.1 Some examples of the diverse life forms associated with aquatic ecosystems: (a) insect larvae (blackfly); (b) juvenile fish (dolly varden);

(c) insect larva (cranefly); (d) amphibian (pyrenean brook newt); (e) adult aquatic insect (mayfly); (f) filamentous algae; (g) frog; (h) single-celled algae (diatoms) magnified 100x; (i) macrophytes (water lilies); ( j) fish (tench). (Source: photo (d) Mark Ledger, University of Birmingham; photo (h) Jubal Harshaw)

are omnivorous, feeding on a variety of resources. The diets of many aquatic heterotrophs are also subsidized with organic materials from adjacent terrestrial ecosystems (e.g. leaf litter from trees, terrestrial insects).

Resources originating externally to the river are termed allochthonous. Further grouping of freshwater organisms can be made based on where they spend the majority of their 299

Chapter 12 Freshwater ecosystems

existence. Benthic organisms (or the benthos) are those living on, in, or near the bed sediments of rivers or lakes. Nektonic organisms (or the nekton) are known collectively as organisms that can actively move around within the water column, in contrast to planktonic organisms that are suspended and passively float or drift around in the water column. A fourth group exists predominantly on, or just beneath, the surface of water bodies and these organisms are collectively termed the neuston. The focus of this chapter is river and lake freshwater ecosystems. Wetlands (areas where the water table reaches the surface and persists long enough to support aquatic plants) can also be considered as freshwater ecosystems (see Dobson and Frid, 2009) but these systems are not considered in this chapter (see Chapter 10). The study of any ecosystem unavoidably requires some knowledge of both the living organisms and their effective environment (see Chapter 11). Where necessary, this chapter makes reference to the aquatic environment to place understanding of the biota into relevant context, but the reader is directed to Chapters 15 to 20 for more detailed information on hillslope processes, sediments, catchment hydrology, fluvial geomorphology and solutes.

12.2 Running waters: rivers and streams The terms stream and river are often used interchangeably when referring to flowing waters, because in reality there is no obvious distinction and the latter term is typically used when describing a ‘larger’ running watercourse. Some authors use the term lotic to describe running water systems of any size. The term ‘river’ is used for consistency in this chapter. Rivers are extremely diverse in their geomorphological form and physicochemical characteristics, which in turn influences the remarkable diversity of organisms that we find in flowing water ecosystems. Describing the characteristics of an individual river ecosystem can be difficult owing to this diversity of characteristics. However, it is helpful to learn about river ecosystems based upon: (i) their hierarchical organization, which considers the nested scales at which components of river ecosystems can be observed; (ii) the interactive pathways, or spatial dimensions, of interest, including upstream–downstream changes, land–water interfaces and surface–subterranean water mixing zones; and (iii) the temporal scale over which observations are made. Human influences have had a major effect on river ecosystem structure and functioning. In many environments, such changes are obvious, but even the most 300

remote and apparently pristine river environments will have undergone some degree of alteration due to the effects of atmospheric pollution (Moss, 2017).

12.2.1 River ecosystem geomorphological units A widely used spatial framework used to aid understanding of nested river ecosystem units is that of hierarchical organization (Frissell et al., 1986). Spatial units include the whole catchment, river segments, river reaches, mesohabitats and patches (or microhabitats) (Figure 12.2). The environment varies considerably at all of these scales, with larger units exerting significant influence over those at smaller spatial scales. This variability is reflected in the plants and animals that inhabit these different spatial units. The most influential environmental variables at each spatial scale can be considered as ‘filters’, for which species must possess appropriate biological traits enabling them to disperse to, and exist in, that unit (see Box 12.1 for further details). For example, at the catchment scale geology and vegetation are generally important natural factors influencing which organisms are present. With respect to vegetation, dense forest cover is likely to restrict the presence of in-stream producers and their consumers but can be beneficial to detritivores. In individual river segments, water quality variables such as stream temperature, or the flood or low-flow disturbance regime, may be important determinants of species distribution. Closer examination of river segments reveals distinctly different reaches at the scale of tens of metres. For example, a river may flow through a relatively narrow, deep, slower section (e.g. Figure 12.2c) before entering a wider reach where the wetted channel is separated from the bordering riparian vegetation by deposits of sediment (e.g. Figure 12.2d). Consequently, there may be reach-scale differences in shading affecting the level of primary production, for example. Each reach is typically composed of smaller units known as mesohabitats, including deeper, flatter, slower-flowing pools with lower dissolved oxygen and finer sediment accumulations compared with shallow, steeper, faster-flowing riffle sections (Table 12.1). Within mesohabitats there are clear differences in habitat at the patch (microhabitat) scale, which can play important roles in the temporal dynamics of stream communities (Townsend, 1989), often by providing refugia from flow disturbances. Differences in flow velocity or bed sediments lead to small-scale variability in accumulations of detritus or algal growth, which in turn can influence the abundance of heterotrophs. Alternatively, small-scale differences in flow velocity may influence the abundance of aquatic

(a)

(b)

(d)

(c)

(e)

(f) Glide Riffle

Pool

Run (i)

(g)

(h)

Figure 12.2 The hierarchical organization of river ecosystems, showing features at nested spatial scales: (a) the catchment (1000 m + ) is composed of

(b) various river segments (100 m). Each segment is composed of (c and d) different reaches (10 m). Closer inspection at the reach scale reveals (e) different mesohabitats (1 m). Mesohabitats are themselves composed of patch-scale (0.1 m) components, for example (f) collections of leaf litter and algal tufts on an individual boulder, (g) sand–silt film over cobbles, (h) a thin algal covering on a partially exposed rock and (i) moss attached to a small rock. (Source: photo (e) Sandy Milner, University of Birmingham) 301

Chapter 12 Freshwater ecosystems

Table 12.1 Mesohabitat types and defining features as typically found in rivers with pool–riffle sequences

Mesohabitat

Features

Riffle

Sloping bed, shallow depth, poorly defined thalweg (i.e. line of greatest water depth)

Run

7 Depth but 6 slope compared with riffles. Thalweg often well defined

Pool

Deep, surface slope almost zero, often located at the outside of meander bends

Glide

Located immediately downstream of pools, where the stream bed slopes to become progressively shallower

(a)

(b)

invertebrates depending on, for example, their ability to adhere to the sediment or their morphology (e.g. whether streamlined or not). It is obvious from these examples that river habitat can be extremely varied. Therefore, there are major differences in both the abundance of individual species as well as the composition of the stream biotic community depending on the level of examination across these nested spatial scales. The variability that can be observed across even the smallest spatial scales is important for scientists to consider when studying river ecosystems or attempting to manage degraded systems. Multiple patch-scale samples will be necessary to characterize the local species pool of just one mesohabitat. River ecologists often have to accept some level of sampling error as a consequence of the exceptional level of habitat variability, although careful sampling can ensure comparable data sets for different river sites. For example, if the aim was to identify changes along an entire river segment, a typical approach for sampling aquatic invertebrates would be to focus on riffle mesohabitats at several points along the upstream–downstream gradient. Sampling may then be undertaken using a device such as a Surber sampler (Figure 12.3a), or a composite sample may be collected from several patches over a fixed time period by taking a kick sample (Figure 12.3b). Figure 12.3 Collection of river benthic macroinvertebrates. (a) A Surber

12.2.2 Spatial variability of river ecosystems The concept of hierarchical organization in river ecosystems recognizes that stream habitats, and the biota they support, are strongly influenced by the surrounding catchment (Hynes, 1975). When considering the importance of these spatial perspectives on river ecosystem pattern and processes, it is also important to acknowledge three interactive spatial pathways, or dimensions (Ward, 1989): 302

net. The operator places the quadrat onto the bed of the stream and uses a hand to disturb a small patch of stream bed in a fixed area (in this case 0.1 m2). Water is directed into the net by the orange side panels. Typically between five and ten samples are collected from each stream reach using this method. (b) A kick net. The operator places the net onto the bed of the stream and then disturbs the sediments immediately upstream of the net with one foot. The operator typically moves to a different part of the stream reach or mesohabitat every 30 seconds, sampling for a total of three minutes per site. Nets for both sampling methods typically have mesh size of between 250 mm and 1 mm. (Source: (a) Jonathan Carrivick, University of Leeds)

12.2  Running waters: rivers and streams

BIOLOGICAL TRAITS AND ENVIRONMENTAL FILTERS Biological traits are variants of a phenotypic character (e.g. size, shape, life-history attributes, dispersal mechanism and diet) that may be inherited or environmentally determined (or both). Biological traits have been used to classify organisms since the early twentieth century when the Saprobian system was developed to assess the extent of organic pollution of rivers based on species’ oxygen requirements (see Statzner and Bêche, 2010). Cummins (1973) later developed a classification of river invertebrates based on their dominant mode of feeding, so-called functional feeding groups (FFGs). The concept of FFGs was subsequently adopted by Vannote et al. (1980), forming the theoretical foundations of the hugely influential river continuum concept. More recently, freshwater ecologists have developed methods based on characterizing multiple biological or ecological traits of species, which can potentially be used to better understand how assemblages vary across environmental gradients compared with more traditional taxonomic approaches. Traits can also be used to generate estimates of functional diversity (i.e. the diversity of roles played by organisms in ecosystems) in addition to standard measurements of taxonomic diversity. These approaches can then be used to test theoretical ideas about how communities assemble or break apart (Brown and Milner, 2012). While predicting the abundance of species as a function of environmental features has been a long-standing goal in river ecology, the complexities associated with spatial habitat variability and frequent disturbance episodes (particularly in rivers) mean this has proven difficult to achieve (Poff, 1997). However, freshwater habitats with similar environmental features should theoretically have species

assemblages with comparable biological trait attributes, even for two water bodies in different regions of the world, because traits respond to environmental selection regimes regardless of biogeographic boundaries (e.g. Townsend and Hildrew, 1994). Freshwater species can be grouped on the basis of numerous traits broadly categorized as follows: (i) life history, (ii) mobility, (iii) morphology and (iv) ecology (Table 12.2). Individual traits can be related to habitat gradients; for example, species inhabiting frequently disturbed habitats will often possess traits that confer resilience (i.e. an ability to recover quickly from the disturbance) such as early age at reproduction, short reproductive cycles (multivoltine), high adult mobility and high fecundity.

Natural ‘filters’ Immigration, extinctions, climate, geology

Alternatively, taxa may possess resistance (ability to withstand the disturbance) traits such as clinging habit, streamlined or flattened body shapes or life-cycle forms such as egg stages or diapause and life stages outside the river. A knowledge of species traits allows researchers to consider the selective environmental forces operating at different spatial scales (region to patch), each of which acts as a ‘filter’ reducing the total species pool of a region to the different assemblages found in habitat patches (Figure 12.4). For example, consider a regional species pool containing 100 different species. Natural factors such as barriers to dispersal (e.g. waterfalls, prevailing wind direction) may restrict 20 species from establishing in a given catchment

REGIONAL SPECIES POOL

Anthropogenic ‘filters’

Catchment

Climate change, acidification

Water quality, thermal regime, vegetation

River

Dams, land use

Shading, substratum, predators

Reach

Road construction, mine drainage, gravel mining, industrial discharges

Moss, algae, detritus, interstices, flow velocity

Patch

Fishing, local tree cutting

PATCH SCALE SPECIES POOL Figure 12.4 Examples of some natural and anthropogenic ‘filters’ at hierarchical spatial scales, which act upon species traits. Filters operating at the catchment scale restrict the occurrence or abundance of traits at smaller spatial scales, and so on. As river ecosystems are dynamic, the ways in which the environmental filters restrict the species pool can change in space and time. (Source: modified from Malmqvist, 2002)

BOX 12.1 ➤

303

Chapter 12 Freshwater ecosystems

➤ Table 12.2 Some indicative traits of invertebrates and fish

Trait category

Indicative biological traits

Life history

Voltinism (e.g. number of generations per year) Maximum age (e.g. fish) Development (e.g. seasonal, non-seasonal for insects) Synchronization of adult insect emergence (e.g. none, poor, good) Adult life span (e.g. days, weeks, months) Fish reproductive strategy (migratory, broadcaster, simple nester, complex nester–guarder and bearer) Resistance forms (e.g. eggs, diapauses) Fecundity (number of eggs laid) Egg type (single or multiple batches)

Mobility

Adult insect dispersal distance Adult insect flying strength Travel distance (drift, crawling, swimming, flight) Swimming ability (strong, weak, none) Locomotion (cruisers, accelerators, maneuverers, benthic high-velocity huggers, benthic low-velocity creepers)

Morphology

Shape (streamlined or not for invertebrates; torpedo, arrow, disc, arched, teardrop, elongate for fish) Morphometric characteristics of fish (e.g. body length/depth, height of caudal fin, head width) Mode of respiration (gills, tegument, plastron, spiracle) Size at maturity (body length or biomass)

Ecology

Habitat (e.g. erosional, depositional, sediment types, mesohabitat (fish)) Thermal preference (cold, cool, warm)

because they lack the necessary traits for dispersal to that catchment. Within that catchment, if we sampled an individual river in its entirety then we might find only 40 of the 80 species, perhaps due to a cool thermal regime selecting against species preferring warm water. Within an individual reach, only 20 of the 40 species may be able to colonize because shading by riparian vegetation means there is too little algal production to support grazing invertebrates. Finally, an individual habitat patch may then have only 10 of the remaining 20 species because the high flow velocity selects against species without streamlined bodies. In addition to natural factors, researchers also have to consider the effect of habitat filters imposed by the actions of human interference which are widespread in river systems, and which themselves operate across different spatial scales (Figure 12.4). The concept of environmental filtering is based on the idea that the species pool at each spatial scale is filtered by the environment. It is important to realize that the actual species pool at each scale is also influenced by dispersal constraints, such that some species may be absent even though the habitat is suitable for their existence. Biotic interactions should also be taken into account, because a given species may be absent due to a lack of food or the existence of a predator, even if the abiotic environment is suitable.

Invertebrate habit (e.g. burrower, clinger, swimmer) Fish stream size preference (expressed in m) Feeding behaviour (e.g. FFGs for invertebrates; herbivores, detritivores, planktivores, invertivores and carnivores for fish) Trophic classification (e.g. mouth position, teeth, gut size for fish) Water body type (river, lake, wetland, estuary) Elevation distribution Salinity tolerance Oxygen requirements pH tolerance

BOX 12.1

304

12.2  Running waters: rivers and streams

(i) the upstream–downstream continuum, or longitudinal dimension; (ii) exchanges of matter between the river channel and the riparian zone/floodplain, or the lateral dimension; and (iii) interactions between the channel and groundwater, or the vertical dimension.

12.2.2.1 Longitudinal dimension The longitudinal (upstream–downstream) dimension is the best studied of the three spatial dimensions in rivers. At the largest scale, river biota change markedly with distance downstream from the headwaters linked to changes in physical properties such as increased river size (width/ depth), lower gradient, downstream fining of river bed sediments and a typically warmer, less variable thermal regime. One of the most influential research ideas in the history of river ecology is the river continuum concept (RCC; Vannote et al., 1980), which generalized the changes in basal resource supply, the functioning of stream macroinvertebrates, with reference to their functional feeding groups (FFGs; Table 12.3) and fish communities that are typical in temperate forested river systems from headwater streams downstream to larger rivers (Figure 12.5). In headwater or upland reaches, the river continuum concept considers rivers to be net heterotrophic

(i.e. respiration 7 production). Heavy shading by trees limits light and thus instream primary production. Coarse particulate organic matter (CPOM; predominantly leaf litter but also small pieces of wood 7 1 mm) from deciduous trees and plants acts as a main source of energy for aquatic food webs. Leaf litter is colonized by microbes and fungi making it more palatable for aquatic invertebrates. The physical action of the river’s current and movement of sediment can also break leaf litter into smaller pieces. This organic matter is consumed by invertebrates classified into a functional feeding group called shredders, which includes organisms such as freshwater shrimp (Gammaridae), water louse (Asellidae), some stoneflies (e.g. Leuctridae) and some caddis fly larvae (e.g. Limnephilidae). The processes of shredding and instream physical breakdown break the coarse material into fine particulate organic matter (FPOM; 6 1 mm). FPOM easily washes downstream, thus serving as an upstream–downstream connection in the aquatic food web, hence the use of the term ‘continuum’. Fine particulates are considered to be a particularly important food source in the mid reaches of rivers. FPOM can be collected from the water column by organisms such as blackfly larvae (Simuliidae) which possess filter-feeding adaptations (collector–filterers) or gathered from deposits on the

Table 12.3 Functional feeding groups (FFGs), their feeding mechanisms, food sources (FPOM/CPOM 5 fine/coarse particulate organic matter, respectively) and typical size range of particles ingested

FFG

Feeding mechanism

Main food source

Particle size range (mm)

Collector–filterer

Suspension feeders – filtering particles from the water column

FPOM, detritus, algae, bacteria, fungi

0.01–1.0

Collector–gatherer

Deposit feeders – collecting deposited particles of FPOM or sediment

FPOM, detritus, algae, bacteria, fungi

0.05–1.0

Grazer

Graze surfaces of rocks, plants and wood

Producers such as attached algae, and associated detritus, bacteria and fungi

0.01–1.0

Predator

Capture prey and either engulf whole or pierce and ingest body fluids

Living animals

7 0.5

Shredder

Chewing

CPOM or live plant tissue

7 1.0

Parasite

Microparasites absorb nutrients often living within host cells

Producers and consumers

Not applicable

Macroparasites are visible with the naked eye and spend only part of their life cycle within a host digesting host tissue or gut nutrients. Endoparasites attach to and absorb nutrients from host gut or tissues. Ectoparasites feed on host blood or tissues (e.g. gills) (Source: modified from Merritt et al., 2008)

305

Chapter 12 Freshwater ecosystems

River width (m)

Predators

Coarse particulate organic matter

Grazers

Microbes 1–2

Trout

Shredders Grayling Collectors

Periphyton 4–6

UPLAND REACHES

OM FP Grazers

Microbes

Barbel Periphyton Collector s

FP

Perch

OM

50–75

Predators

MIDDLE REACHES

Shredders

Vascular plants

Bream

Predators

Tench

Phytoplankton

Microbes

Collectors Zooplankton

500+

LOWLAND REACHES

Figure 12.5 Diagrammatical representation of the river continuum concept, showing the proposed relationship between river size and the downstream

changes in structural and functional attributes of the river ecosystem. (Source: after Vannote et al., 1980)

riverbed by collector–gatherers. However, the increasing width of the river channel means that shading of the river by riparian vegetation is largely restricted to the margins. 306

Therefore, more light can reach the stream allowing for an increase in primary production and a shift to net production. Abundant algal growth supports a greater abundance

12.2  Running waters: rivers and streams

of grazers or scrapers (e.g. snails, some mayflies) which directly consume producers growing on surfaces. In the deeper, wide, slow-flowing lowland reaches of rivers, light may be unable to penetrate to the river bed due to the high level of suspended sediment and organic material transported from the upper reaches of the river system. Therefore, primary production falls and these reaches will be net heterotrophic. FPOM washed from upstream is a dominant food source, and the community will be almost exclusively composed of collectors. Throughout, the whole river, shredders, collectors and grazers are themselves consumed by predators such as invertebrates (e.g. some stoneflies; Perlidae), fish and amphibians. There is also a major microbial component to aquatic food webs throughout a river’s length, fuelled largely by dissolved organic carbon (DOC) sourced from the decomposition of organic matter in soils and the stream (Meyer, 1994). Carbon and other nutrients in river water can be absorbed and assimilated by plants and microbes or adsorbed to sediments. When these nutrients are eventually released back to the water, some may be recycled immediately into the biota but most move some distance downstream with the flow. The combined action of nutrient cycling and downstream movement can be conceptualized as a spiral. This idea forms the basis of another important piece of work that is built around the longitudinal dimension of rivers, the nutrient spiralling concept (Newbold et al., 1982). It should be realized that some organisms cannot be easily placed into functional feeding groups as diet can vary with life stage and body size, while some species are omnivorous, consuming a variety of different food resources. It is also important to appreciate that the RCC describes only an idealized sequence of ecosystem change along undisturbed, temperate forested rivers. The downstream sequence of changes is not ‘continual’ and can be interrupted by tributaries which deliver water, sediment, food sources and organisms, and alter the geomorphology. Riparian forest clearings also allow reaches of high primary production in otherwise net heterotrophic sections of river. The RCC cannot be applied to rivers which naturally lack abundant forest cover, such as those in Arctic or alpine regions (see Box 12.2, for example), and it is not considered applicable to streams in New Zealand due to an absence of shredders, because many streams are unstable due to the frequent overriding influence of rainfall and snowmelt-induced flood disturbances, and because many streams lie above the treeline (Winterbourn et al., 1981). Additionally, the natural characteristics of many rivers and their catchments have been altered by a long history

of human modification (e.g. deforestation, dam building, channelization; see Section 12.2.4) meaning that the downstream changes predicted by the RCC no longer apply.

12.2.2.2 Lateral dimension As the RCC suggests, river-dwelling organisms and the adjacent terrestrial ecosystem (or riparian zone) can be linked through the consumption of leaf litter by shredders. However, the diet of aquatic predators such as fish and some invertebrates can also be supported by inputs of terrestrial invertebrates (e.g. ants, beetles, caterpillars) which fall into the water and are washed downstream. In this sense it can be said that the terrestrial ecosystem feeds, or subsidizes, the aquatic ecosystem (Nakano et al., 1999). Inputs of terrestrial invertebrates to rivers vary seasonally, with peaks typically occurring from spring to early autumn in temperate zones. Some studies have shown that the diet of river-dwelling fish can be made up of 750% terrestrial invertebrates (Baxter et al., 2005). The flow of resources and energy between the terrestrial and aquatic systems is not, however, a one-way process. Recent research indicates that there are subsidies from the river back to the terrestrial ecosystem (Baxter et al., 2005). Aquatic insects develop in the stream as larvae (nymphs) but towards the end of their life cycle they emerge as adult flies with a terrestrial aerial stage. The number of flies emerging from rivers varies widely but in some studies 7150 000 per m2 have been measured. The emergence of adult aquatic insects is typically concentrated into a small part of the year in temperate regions (peaking in early summer) whereas in tropical areas it may occur year round. These insects provide food for many different terrestrial organisms including spiders, bats, birds, amphibians and beetles. High abundances of these predatory organisms can be found along some river channels at times of peak aquatic insect emergence. Terrestrial animals can also benefit from fish populations in rivers. In the Pacific north-west region of North America, for example, the annual salmon runs are often characterized by thousands of fish moving upstream in each river to their spawning grounds. During and after this migration, the salmon die leading to accumulations of carcasses in the stream, entangled in bankside vegetation and washed up on river banks. These salmon provide abundant food resources for insects, scavenging birds and mammals including bears. Bears also actively predate live fish, as the many TV documentaries attest. Bear populations can be up to 80 times denser in areas where salmon are abundant compared with fishless rivers 307

Chapter 12 Freshwater ecosystems

(Gende et al., 2002). Nutrients from salmon carcasses may also fertilize terrestrial vegetation such that growth may be increased significantly in areas where bears deposit carcass remains and excrete waste.

12.2.2.3 Vertical dimension Rivers were traditionally viewed as components of the landscape where water was flowing over the surface. However, river scientists now understand that the spatial extent of rivers does not cease at the upper surface of the river bed. At small scales (up to metres), river water flows within the openings of river bed sediments (called interstices). These spaces are important habitats for invertebrate and fish eggs and larvae, offering refuge from swift-flowing surface currents and some predatory organisms. The slower flow of water through the interstices, compared with the water column, is important for nutrient cycling and organic matter respiration by microbial organisms because it results in greater sediment–water contact times. At larger spatial scales (up to kilometres) river ecosystems can extend downwards and outwards where subsurface and river bank sediments are permeable and surface water interacts with subsurface groundwater (Ward et al., 2002). The boundary between the river water and groundwater, known as the hyporheic zone, can have steep physical, chemical and biological gradients and is considered to be particularly important for nutrient cycling, a permanent habitat for some species and as a refuge from disturbance for organisms that typically inhabit the upper layers of the stream bed.

12.2.3 Temporal variability of river ecosystems In addition to the three spatial dimensions discussed above, a fourth dimension, time, exerts significant influence on river ecosystems (Ward, 1989). It is useful to consider temporal river ecosystem dynamics either as those that are relatively regular and predictable allowing the biota to adapt or persist, or those that are discrete disturbances, defined as ‘an event in time that is characterized by a frequency, intensity and magnitude that is outside a predictable range and that disrupts ecosystem, community or population structure’ (Resh et al., 1988, p. 433). Predictable changes in river ecosystems occur over diurnal timescales, where there can be, for example, significant changes in water temperature due to solar radiation receipt, causing fish to migrate temporarily to cooler reaches. In mountain rivers fed by snow and ice there are usually diurnal discharge variations due to melt 308

(see Chapter 18), which can influence the behaviour of macroinvertebrates, causing them to move down into bed sediments to avoid higher flow velocities, or to migrate downstream as drifting individuals. At longer timescales there are seasonal changes in the composition of river ecosystems due to life-cycle progression and migrations to avoid/take advantage of wet or dry seasons. At interannual timescales, researchers have found evidence that the abundance and diversity of fish and invertebrates respond to climatic cycles associated with phenomena such as the El Niño Southern Oscillation and the North Atlantic Oscillation (Hurrell et al., 2003; see also Chapters 6 and 7). Long-term changes have also been documented in some river ecosystems as a consequence of climate warming and the retreat of glaciers (Box 12.2). Less predictable or stochastic disturbances in river ecosystems are typically associated with floods (the most common form of disturbance in rivers) or droughts, although major changes to river ecosystems can also result from disturbances associated with freezing, landslides, high winds depositing large numbers of trees into rivers, or even anthropogenic pollution events. High-flow events may be devastating for the biota of small, steep upland streams owing to large sections of river being affected by sediment erosion and deposition. In lowland rivers, floods can be beneficial because of the potential for floodwater to expand across the floodplain and transfer energy and nutrients between the terrestrial and aquatic systems, whilst providing slower-flowing refuge areas for mobile biota (Junk et al., 1989). During droughts, the contraction of wetted area can lead to river biota becoming concentrated in pools and short reaches, increasing the risk of predation. Extreme climate events (e.g. high rainfall, heat waves and prolonged droughts) are expected to occur with higher frequency and magnitude and/or timing outside of their historic range in coming decades due to climate change, and more research is needed to understand the effects and allow river managers to plan for these changes (Jones et al., 2013)

12.2.4 Human alterations to river ecosystems Humans have severely changed river ecosystems through hydrological (e.g. water abstraction and dams), geomorphological (e.g. channelization, culverts, dams and gravel extraction), water quality (e.g. acidification, organic pollution, nutrients and thermal pollution) and biological alterations (e.g. introduced species; see Vörösmarty et al., 2010). These anthropogenic alterations span the range of river ecosystem spatial scales (see Box 12.1 and

12.2  Running waters: rivers and streams

RIVER ECOSYSTEM RESPONSE TO GLACIER RETREAT Glaciers are distributed worldwide, not only at high latitudes but at lower latitudes where they are located in mountainous areas (e.g. the Himalayas and the equatorial Andes). Glacier melt contributes significantly to river flow and water resources across the globe, and rivers with glacial meltwater inputs provide habitat for fisheries and a number of rare and endemic aquatic invertebrate species. Glacier-fed rivers are considered to be among the

most dynamic on seasonal to inter-annual scales because of strong interconnections between atmospheric forcing, snowpacks, glacier mass balance (see Chapter 23), stream flow, water quality and fluvial geomorphology and river biota (Milner et al., 2009). Glacier retreat is frequently attributed to a warmer climate and will lead to major shifts in water sourcing of Arctic and alpine rivers, with glacier and snowmelt reductions and changes in river hydrological and geomorphological dynamics. These changes are likely to have

significant, widespread consequences for plants and animals of alpine stream ecosystems which are strongly influenced by river channel stability and water temperature, two variables determined by the amount of meltwater contribution and valley geomorphology (Brown et al., 2003). The non-biting midge larvae (Chironomidae) genus Diamesa typically dominates European glacier-fed rivers where maximum water temperature is 62 °C and river channel stability is low (Figure 12.6). Further downstream from glacier margins, river channels become

Glacier

Diamesinae Tmax625C Low channel stability

Tipulidae

Tmax645C

Baetidae

Heptageniidae Tmax665C

Rhyacophilidae

Simuliidae

Tmax685C

Taeniopterygidae

Tmax6105C High channel stability

Limnephilidae

Figure 12.6 Diagram outlining the likely first appearance of macroinvertebrate taxa along a longitudinal continuum from the

ice margin for European glacier-fed rivers. (Source: based on data from Milner et al., 2001)

BOX 12.2 ➤

309

Chapter 12 Freshwater ecosystems

➤ more stable and water temperature increases allowing mayfly, stonefly and caddisfly larvae to become increasingly dominant along with blackfly (Simuliidae) and other groups of chironomid larvae. In Glacier Bay, south-east Alaska, glacial retreat has been occurring since around 1750 (the end of the Little Ice Age), opening up vast areas of deglaciated terrain, and creating hundreds of kilometres of new rivers that subsequently undergo colonization and primary succession by biotic communities. Stream ecosystem response to the loss of glacial ice masses has been studied in detail since 1978 at Wolf Point Creek. Progressive increases in stream temperature, riparian vegetation cover and instream habitat complexity over time have been accompanied by a diverse group of

macroinvertebrate colonizers (Milner et al., 2008). In 1978, when catchment glacial cover was ∼70%, just five taxa (all Chironomidae of the subfamilies Diamesinae and Orthocladiinae) were found. The first mayflies and stoneflies were found at ∼50960% glacier cover in 1986, the first non-insect taxa (Oligochaete worms) at ∼30% in 1992, and Dytiscidae beetles and Corixidae (pond skaters) in 2000 and 2003, respectively, after ice masses had almost vanished. More recently, these systems have been impacted by extreme rainfall events in 2005 and 2014, which have led to major reorganization of the ecological communities (Milner et al., 2013). These kinds of extreme events are expected to impact rivers more frequently in future (Jones et al., 2013).

Further loss of snow and ice masses in the future will alter the spatial and temporal dynamics of river basin runoff through changes in the relative proportion of snowmelt, glacier melt, hillslope and groundwater contributions. If the climate continues to warm, glacial runoff may increase initially but eventually it will decrease in the long term as the ice disappears. Seasonal discharge patterns will therefore shift, with earlier and higher spring peaks and lower summer flows, and diurnal meltwater cycles may be reduced with day-to-day variability increasing with a greater disposition to flooding. These hydrological changes will alter the habitat within which freshwater organisms exist, necessitating adaptation or potentially causing species losses (Jacobsen et al., 2012).

BOX 12.2

Mason, 2002). At the catchment scale and more widely, climate change is expected to drive changes in the thermal regime of freshwaters (Woodward et al., 2010b). Warming increases metabolic rates of ectothermic organisms (i.e. those reliant on the environment to maintain their own body temperature), and above certain limits can induce stress on physiological systems, leading to the loss of species. Thermal regime changes can also occur from afforestation or deforestation, as well as the release of heated effluent from factories and power stations into streams and rivers. Warmer waters hold less dissolved oxygen and this will consequently impact respiration. Climate change effects on river ecosystems may also be indirect, with changes in precipitation leading to more frequent and severe droughts in some areas but more severe and unpredictable flooding in others. Many organisms have adapted their life cycles and morphological traits to the natural flow regime over long timescales. Changes to the magnitude, timing, frequency, predictability or duration of extreme flow events are therefore likely to induce major changes in river ecosystems (Lytle and Poff, 2004). Therefore, increasing numbers of reservoirs are being used to generate artificial floods in attempts to mimic the natural cycles of floods and lower flows (Gillespie et al., 2015). At the scale of individual river systems, major alterations can occur following land-use changes, in particular 310

the conversion of forest to grassland or arable farmland and the urbanization of catchments. Agricultural development leads to increased diffuse nutrient fluxes to rivers (see Chapter 20), potentially increasing primary production which in turn may lead to changes in assemblages of invertebrates and fish (Allan, 2004). Nutrient enrichment can alter rates of leaf litter breakdown by microbes and fungi, leading to changes in the metabolic balance of freshwaters. Sediment eroded from bare agricultural fields can smother river bed sediments, reducing algal production and, by filling in the voids between larger sediments, affecting the spawning habitats of fish. Globally, regulation of rivers by flow control structures (typically dams) has caused major negative impacts on aquatic biodiversity by changing river flow regimes, altering habitat and impeding the migration of organisms. In the United States alone, there are an estimated 2.5 million structures controlling river flow, and only 2% of rivers are considered to have no impacts from flow control structures (Lytle and Poff, 2004). There is a long history of humans discharging waste products to rivers from industrial processes and sewage treatment works via point source pipes or culverts (Figure 12.7). Sewage works were historically associated with inputs of organically rich effluent which led to severe oxygen decreases in affected river reaches downstream of the outlets (Mason, 2002). The severe lack of oxygen

12.2  Running waters: rivers and streams

(a)

(b)

(c)

(d)

Figure 12.7 Examples of point source discharge of sewer effluent to rivers.

caused by organic enrichment typically leads to losses of most invertebrates and fish with the exception of a few tolerant groups such as worms and some midge larvae. However, in the last two decades there have been major improvements in river water quality (particularly in the UK) linked to investment in enhanced sewage treatment technologies. Some problems still remain, though, owing to the inadequacies of sewage treatment processes in removing the active compounds contained in human pharmaceuticals, recreational drugs and personal care products such as toothpaste, shower gels and shampoo (Hughes et al., 2013). Research in the last couple of decades has, for example, found that male roach (Rutilus rutilus) have

developed female characteristics in many UK rivers due to steroidal estrogens and chemicals that mimic estrogens reaching the river in sewage effluent (Tyler and Jobling, 2008). The ecological effects of the many hundreds of other synthetic chemicals released into rivers remain to be studied. While many rivers have been physically or chemically altered by humans, there are also biological changes caused by the intentional or accidental introduction of ‘exotic’ species. The rainbow trout (Oncorhynchus mykiss), for example, originates from North America, but has been spread intentionally across the world for the purposes of sport angling and as a food source. When 311

Chapter 12 Freshwater ecosystems

individuals escape they can compete for food and habitat with other species of fish, often to the detriment of the native species (Allan and Castillo, 2007). Similarly, introduction of the brown trout (Salmo trutta) to New Zealand has led to dramatic declines in populations of native galaxiids (McIntosh et al., 2010). Another example of biological change in rivers concerns the American signal crayfish (Pacifastacus leniusculus), which was brought to the UK because it is much larger than the native white-clawed crayfish (Austropotamobius pallipes), and was therefore more appealing as a species to farm for food supply. While the signal crayfish was initially introduced to selected fish farms, some individuals escaped into river systems and over time these have bred and expanded their range. The signal crayfish are better competitors for food and habitat, and they carry a fungus which causes mortality among the white-clawed crayfish, meaning they are at risk of being lost from the few rivers where they still exist (Dunn et al., 2009). Chapter 11 contains more on alien species.

Reflective questions ➤ What are the key features controlling the spatial and temporal variability of river ecosystems?

➤ In what ways have the activities of humans altered the spatial dimensions of river ecosystems?

➤ How might the modification of a river by damming the headwaters alter the upstream–downstream continuum

although some may be more isolated in the landscape and fed by rain, groundwater seepage/springs or inundation during times of flooding (Dobson and Frid, 2009). Similar to rivers, the study of lakes cannot be undertaken without considering the effects that humans have had on many of these systems. Describing the characteristics of an individual lake ecosystem can be difficult owing to the diversity of morphological and physicochemical characteristics. However, it is helpful to learn about lake ecosystems based upon: (i) their geomorphological origin, and (ii) the tendency or otherwise to stratify, either thermally or chemically. Within individual lakes it is then possible to identify (iii) spatially discrete habitat zones. Each of these classification systems is detailed below.

12.3.1 Classification of lake ecosystems 12.3.1.1 Geomorphology and processes of formation Seventy-six types of lakes were classified by Hutchinson (1957) and the list has been extended to over 100 lake types in the intervening period. However, 10 distinct groups can be recognized (Table 12.4), with the most common being formed by glacial activity (Figure 12.8). The geomorphological characteristics of lakes and their surrounding landscape are particularly diverse. These characteristics control the nature of a lake’s drainage, nutrient inputs and residence time, which in turn have strong influences on physical, chemical and biological characteristics (Wetzel, 2001).

of food sources and macroinvertebrate functional feeding groups?

12.3 Still waters: lakes and ponds Lakes and ponds are distinct bodies of ‘still’ water that form in geomorphological settings with outflows small relative to the size of the water body. These water bodies are not completely still, with currents driven by wind, thermal currents and the slow movement of water from the inflow to outflow. However, they have relatively longer residence times than rivers, and some authors use the term lentic rather than referring interchangeably to still waters, ponds or lakes. The term ‘lake’ is used for consistency in this chapter because most of the examples refer to larger bodies of water. Lakes and ponds are typically fed by rivers,

312

12.3.1.2 Stratification of lake ecosystems Some lakes, such as those that are shallow and strongly influenced by the actions of the wind, are well mixed. In contrast, many lakes become subject to stratification at certain times of the year, when water that is not readily mixed forms layers with distinct physical and chemical properties (in particular, differences in water temperature). The absorption of solar radiation from the Sun decreases exponentially with water depth (Moss, 1998) and, as a result, there should theoretically be an associated temperature decrease. However, the effect of wind at the lake surface is to mix the water and, if this mixing is insufficient to reach the bed of the lake (i.e. because it is too deep relative to the mixing depth), then a warmer upper layer of water (termed the epilimnion) can form. Where distinct layers occur and temperatures drop rapidly the ‘boundary’

12.3  Still waters: lakes and ponds

Table 12.4 Groups of lakes, with a brief description of processes by which they were formed

Group

Typical processes of formation

Example

Tectonic basins

Depressions formed by movement of the Earth’s crust that infill with water. The major types are caused by faulting of the crust

Lake Baikal, eastern Siberia. Covers 31 500 km2 with a maximum depth of 1620 m

Volcanic

Depressions and cavities in cooling magma may collect water and form lakes. Crater lakes are often found occupying inactive volcanoes. Small depressions where magma has been ejected and overlying material has collapsed. Caldera lakes form where the rook of partially emptied magmatic chamber collapse. Lava flows may create basins or block existing rivers

Lake Toba, North Sumatra; 100 km long * 30 km wide, maximum depth of 505 m

Landslides

Sudden movements of unconsolidated material form new depressions or block existing rivers. Many lakes formed in this way are transitory, lasting for only a few weeks or months because the dams are usually susceptible to erosion and failure

Hunza River, north-west Pakistan, formed in January 2010 following a massive landslide

Glacial

These are the most numerous type of lake. Gradual erosional and depositional activities of glacial ice create depressions. With the retreat of the last Pleistocene glaciers (see Chapter 4) many small lakes were formed. Lakes can also form in, on or under ice masses, or behind terminal moraines in river valleys where glaciers still exist

Windermere, Lake District, England. Main fish species include trout, char, pike and perch. Much of the pioneering research into lake ecology was undertaken here by the Freshwater Biological Association

Solution lakes

Slow chemical weathering of soluble rock creates depressions. Commonly found on limestone terrain. Typically very circular and conical

Many lakes in Florida, USA, lie on limestone and have formed by dissolution

River activity

Plunge pool lakes – excavated at the foot of waterfalls along previous river channels

Grand Coulee plunge pool system, Washington State, USA

Lateral lakes form where mainstem rivers deposit large amounts of sediment overbank, blocking tributaries

Lake Chicot, Arkansas, USA, is the largest oxbow lake in North America and was originally part of the Mississippi River system

Also lakes in the rift valley of East Africa

Several types of floodplain lakes form where rivers inundate the riparian zone Oxbow lakes form where river meanders become isolated from the main channel (see Chapter 19) Wind-formed

Form in arid regions where wind-blown sediments form dunes that block rivers

Common in dry regions such as inner Australia, South Africa and arid regions of the USA

Deflation basins form where sediments are blown away and depressions form Shoreline activity

Longshore currents can deposit marine sediments across the mouth of bays, separating the bay from the sea (see Chapter 22)

Biological origin

Lakes formed by the damming action of plant growth or detritus build up. Can also include beaver dams (see Figure 5.16)

Human

Reservoirs formed behind dams

Slapton Ley, England

(Source: after Wetzel, 2001)

is referred to as the thermocline. Lakes can also stratify in winter when ice cover develops at the surface, resulting in a warming with depth and a reverse thermocline close to the surface. The classification system proposed by Hutchinson (1957) is still used today to describe differences in lake stratification. Lakes that never stratify are termed amictic;

these lakes are characteristically found at high latitudes or altitudes where surface ice cover persists year-round and prevents wind from mixing the water. Meromictic lakes mix incompletely and form where lakes are either very deep, preventing complete turnover, or where inflows are denser than lake water owing to high solute concentrations, causing them to sink to the lake bottom. A third group of

313

(a)

(b) (b)

(c)

(d)

Figure 12.8 Four contrasting glacially formed lakes: (a) Latnajaura, near Abisko, Sweden, a lake formed by glacial scour, and which is frozen for up

to 10 months per year; (b) a small cirque lake in south-east Alaska, in the depression formed at the head of a former glacier; (c) Lake Wakatipu, near Queenstown, New Zealand, a moraine dammed lake; and (d) kettle lakes, formed through the melting of ice blocks deposited in unconsolidated sediment by retreating glaciers. 314

12.3  Still waters: lakes and ponds

(a)

stratified except for when irregular storms induce mixing. During stratification, the upper layer is typically referred to as the epilimnion and the lower layer the hypolimnion. The zone characterized by the thermocline (zone of rapid temperature change) is the metalimnion. It is important to bear in mind that these general patterns may be altered by, for example, local variations in climate, lake geomorphology and water movement.

Depth

Temp

(b)

(c)

12.3.1.3 Nutrient status Lakes are often described on the basis of their nutrient status with two common descriptions being oligotrophic and eutrophic, although there are various other trophic states (e.g. dystrophic, mesotrophic and hypertrophic). Various methods exist for determining the trophic status of lakes, and although there are no universally accepted definitions, some quantitative differences have been proposed (Table 12.5). These include taking measurements of total phosphorus concentration, phytoplankton biomass or chlorophyll-a concentration in a water sample, primary production rate measured as the uptake of carbon, and water transparency measured using a device called a Secchi disc (Figure 12.10).

(d)

Figure 12.9 Annual cycle of stratification for a cold temperate dimicitic

lake. Circular arrows denote zones of mixing. Panels to the right show the temperature changes that occur with increasing distance from the surface. (a) In spring, there is complete mixing, whereas (b) in summer a thermocline develops. (c) The thermocline breaks down in autumn and the lake becomes thoroughly mixed, until (d) in winter stratification can occur for a second time due to surface ice cover resulting in a reverse thermocline at the surface. (Source: modified from Dobson and Frid, 2009)

12.3.2 Spatial variability of lake ecosystems lakes are holomictic, and these mix completely. Holomictic lakes can be subdivided into: (i) monomictic lakes which have a single season of mixing, or overturn (e.g. warm, temperate lakes, or some tropical lakes that have large temporal changes in depth); (ii) dimictic lakes which stratify in the summer and winter (e.g. cold temperate lakes, Figure 12.9); (iii) polymictic lakes, such as warm tropical lakes which may stratify as often as each day, albeit slightly; and (iv) oligomictic lakes, which may be continually

12.3.2.1 Edges and bottom Lake habitats can be considered within a simple framework (Figure 12.11). The littoral zone is the part of a lake nearest the shore. It is the zone of both shallow and deep lakes where light can penetrate to the bed allowing the growth of diverse macrophytes (plants) and benthic algae assemblages. Primary producers can dominate the energy base of aquatic food webs in these areas. In small and

Table 12.5 Approximate values for four parameters used to categorize the trophic status of lakes

Total phosphorus (Mg L −1)

Maximum chlorophyll-a (Mg L−1)

Primary production (mg C m −2 day −1)

Maximum Secchi depth (m)

6 4

6 2.5

6 30

7 6

Oligotrophic

4–10

2.5–8

30–100

3–6

Mesotrophic

10–35

8–25

100–300

1.5–3

Eutrophic

35–100

25–75

300–3000

0.7–1.5

7 100

7 75

7 3000

6 0.7

Ultra-oligotrophic

Hypertrophic (Source: modified from Dobson and Frid, 2009)

315

Chapter 12 Freshwater ecosystems

shallow lakes, the littoral zone may constitute a major habitat due to its areal dominance. The edges and bottom of lakes are considered to be relatively productive and diverse perhaps because of the light penetration, combined with abundant sediment nutrient stores and the spatial heterogeneity of these habitats (Moss, 2017). The shallow edges and bottom in the littoral zone have an assortment of sediment sizes supporting a diverse assemblage of organisms. Waves act to regularly disturb the lake bed and create a mosaic of sediments (ranging from silts and mud in calmer waters, to bare rock and gravel where wave action is intense) which in turn determine the composition of biotic communities. Wave disturbance is a major factor influencing the distribution of benthic organisms in the littoral zone, with wave-impacted rocky shorelines only covered with attached algae, bacteria, limpets and occasional caddisfly or mayfly larvae. More sheltered areas of the shoreline may have gravel deposits but this moves readily and therefore has only a small permanent community. Where wave action is less considerable, finer sand deposits can build up and be colonized by aquatic plants and diverse communities of bacteria, protozoa or epipelic (living on, or in, fine sediment) algae. There may also be an array of emergent, submerged and floating plants, fish, invertebrates, and both herbivorous and piscivorous birds and mammals (Moss, 1998). The calm waters around these plants can be particularly beneficial for the larval development of nuisance species such as mosquitoes (Box 12.3).

Figure 12.10 Measurement of water transparency using a Secchi disc.

The disc is 20 cm in diameter with quarters painted black and white. The operator lowers the disc into the water until it is no longer visible. The Secchi depth is then recorded from the attached tape measure. (Source: Mel Bickerton)

High Littoral r plant

highe Limit of

ation

coloniz

s

munitie

ted com

omina phyte d

Euphoti

Profundal

Profund al benth os in silt s

and mud s

Light availability

c depth

d bryo

Algal an

ion

ypolimn

vel of h

Upper le

Low/none hen deo

cking w

s la Bentho

ted xygena

Figure 12.11 A general scheme of littoral and profundal habitats in a freshwater lake. (Source: modified from Moss, 1998)

316

12.3  Still waters: lakes and ponds

MOSQUITOES AND INFECTIOUS DISEASES Mosquitoes are familiar as the small flying insects which can deliver painful bites. The name mosquito does not refer to one individual species; in fact, there are more than 3500 known species of mosquito from around the world. Many species are found only in tropical and subtropical regions but some have adapted to life in temperate and subarctic regions. While these insects spend the majority of their life cycle as adult flies (between 4 and 8 weeks), three of their four life-cycle stages (egg, larvae and pupa) are completely reliant on freshwater (with the exception of a few species which can develop in salt marshes). Adult females lay their eggs in still waters such as ponds, lakes and wetlands and also water butts, buckets, or water-filled hollows of plants. These eggs develop into larvae and subsequently pupae over a period of up to two weeks depending on the environmental temperature and food availability. In the larval stage (Figure 12.12a), the mosquito has a well-developed head with

mouth brushes that are used for feeding on phytoplankton and suspended bacteria. The larva’s body is segmented and lacking legs but has a distinctive eighth abdominal segment called a spiracle, through which it breathes. To breathe, it must hang beneath the water surface with the spiracle piercing the surface film. When disturbed, the larvae swim to depth by undulating the body or using hairs around the mouth for propulsion. Following the larval stage, each individual forms a pupa which also breathes from the water surface but it does not feed and is far less active than the larvae. After a few days the adult fly emerges from the pupal cocoon. During the adult stage of the life cycle (Figure 12.12b) both male and female mosquitoes feed on plant nectar. However, the female flies of most mosquito species also require a blood meal which provides the protein and iron required for the development of egg masses. When these blood meals are obtained from humans, the mosquito can deliver a painful bite as its proboscis pierces the skin and injects anti-coagulant saliva to prevent localized blood clotting while it feeds. The bite itself

typically causes a red itchy spot in most people that subsides quickly. However, mosquitoes pose major health hazards in many tropical and subtropical areas because they are vectors for several lifethreatening diseases. The most commonly known mosquitoborne disease is malaria which is spread to humans by mosquitoes of the genus Anopheles. Malaria is caused when humans become infected by parasitic protists of the genus Plasmodium which are carried by the mosquito. These parasites have a complex life cycle, partly in the mosquito and partly in vertebrate animals. When a mosquito bites an infected person, malarial parasites can pass into the mosquito with the blood. These parasites then develop in the mosquito over a period of about one week until they are subsequently injected into another victim with the mosquito’s anti-coagulant saliva. The Plasmodium parasites then develop in the human liver over a period of at least two weeks before they start to multiply and malarial symptoms arise. Symptoms of malaria can include severe fever and headache as well as hallucinations, coma

(b) (a)

Figure 12.12 Mosquitos: (a) larvae; (b) adult female mosquito feeding on blood with the distinctive proboscis used to pierce the flesh of its prey. (Source: (a) Svetoslav Radkov, (b) anat chant, Shutterstock.com)

BOX 12.3 ➤

317

Chapter 12 Freshwater ecosystems

➤ and even death if left untreated. While a variety of anti-malarial drugs are available these are not accessible to many of the world’s poorest people, particularly in developing countries. Consequently there are thought to be more than a million deaths per year across the world due to

malaria. Additionally, some species of mosquito carry a parasitic worm that causes elephantiasis (severe swelling of different body parts). The mosquito species Aedes aegypti spreads the viral diseases yellow fever, dengue fever and Chikungunya, whilst other mosquitoes spread Ross

River fever, West Nile virus and Japanese encephalitis. An estimated 700 million people worldwide are affected by mosquito-transmitted diseases each year, and the scale of the problem is expected to increase, with northward shifts of the disease projected due to climate change.

BOX 12.3

The zone where light penetrates the water column is known as the euphotic zone. Below this is the profundal zone and here the lack of light prevents photosynthesis. Thus, any organisms inhabiting this area permanently are dependent on detritus (dead organic matter) as a food source, either supplied from the overlying water column or washed from the littoral zone. The profundal zone typically supports a lower diversity of organisms than the littoral zone, and is often characterized by simple communities of bacteria (including Cyanobacteria or blue–green algae), protozoa, invertebrates and fish (Moss, 1998). Often the zone below the hypolimnion will become deoxygenated, restricting the biotic community to anaerobic bacteria and protozoa.

moving boats to filter larger volumes of water, or which can be lowered from the side of a vessel and then slowly pulled upwards. Rotifers, protozoa and crustaceans such as the cladoceran Daphnia form the major groups of freshwater zooplankton, an assortment of organisms which comprises heterotrophs feeding on phytoplankton, suspended organic particles and sometimes other smaller

12.3.2.2 Open water The region of open water above the profundal and out from the littoral is called the pelagic zone. The open-water habitat of lakes contains a complex assortment of organisms, although in reality many of these are invisible to the naked eye except for where they form immense colonies, or blooms. The planktonic organisms that are primary producers are collectively known as phytoplankton, and range in size from 1 mm (picoplankton), 65 mm (ultraplankton), 5960 mm (nanoplankton) and 760 mm (net plankton). This group includes a hugely diverse array of bacteria, green, yellow–green and golden-yellow algae, diatoms, cryptophytes, euglenoids and dinoflagellates (Moss, 1998). The growth of phytoplankton is strongly influenced by light and dissolved nutrients such as nitrogen and phosphorus. Phytoplankton can be sampled by collecting water samples (usually 1 L) from the water surface, from depth using a pump and hose, or by deploying sealed bottles which are submerged to the required depth before being opened remotely. Alternative methods include the use of plankton nets (Figure 12.13), which can be towed behind 318

Figure 12.13 Sampling of lake plankton from a small boat. The fine mesh

net is lowered into the water column to the desired depth then pulled up to the surface to sample a column of water. (Source: Mel Bickerton)

12.3  Still waters: lakes and ponds

zooplankton. Other less common constituents of the zooplankton include freshwater jellyfish, some flatworms and mites (Moss, 1998). Some groups of zooplankton such as the Cladocera are much more active than the phytoplankton, being able to move actively through the water column, form shoals and undertake migrations through the water column either in response to diurnal cycles or to avoid predators. Larger, faster-moving zooplankton groups include shrimp-like mysids (up to 2 cm in length) which are predators of other zooplankton. Sampling zooplankton can be tricky due to their spatial aggregation and ability to move actively. Plankton nets are the favoured method because bottle samples of lake water are unlikely to capture many individuals. A familiar third group of open-water-dwelling organisms, with major socio-economic importance, is the fishes. Over 10 000 species, or ∼40% of the world’s known fishes, are from freshwaters (Dudgeon et al., 2006). Most freshwater fish species are found in the tropics, perhaps as result of shorter generation times in warmer waters permitting faster evolution, coupled with the long history of permanent water bodies in these regions (Moss, 1998). The fishes have particularly diverse morphological, behavioural and ecological traits, meaning that they inhabit a range of lake habitats as well as the open water, and they have considerable dietary differences with some being herbivorous or detritivorous, whilst others are predatory. Many fish are

ANTHROPOGENIC EUTROPHICATION While eutrophic lakes (high nutrients, high productivity) can occur naturally, the term eutrophication is more synonymous with the problem of anthropogenic eutrophication. Anthropogenic eutrophication is caused by an increase in nutrient concentrations (nitrogen and phosphorus) in lakes, usually from treated sewage effluent, raw sewage (from sewer overflows, farm runoff or fish farm effluent), and by leaching of fertilizers from arable farmland in the surrounding catchment

omnivorous (feeding on both plants and animals) and their diet changes as they develop from larvae to fry to adults. Fish have different spawning needs and thermal tolerances but being highly mobile they are able to move between different areas of lakes.

12.3.3 Human influences on lake ecosystems Lakes suffer from many of the same problems as rivers (see Section 12.2.4 and Mason, 2002), particularly those related to hydrological, water quality and biological alterations. The hydrological behaviour of many lakes throughout the world has been modified by human abstractions for water supply and irrigation. Some lakes have been modified morphologically by the construction of dams to raise the water level (e.g. Malham Tarn, UK) which alters the configuration of the shoreline and littoral zone. Diffuse pollutants from the wider catchment, and point source pollutants entering either directly through wastewater pipes or indirectly via inflowing streams and rivers, are particularly major problems. In many lakes around the world, nutrients (nitrogen and phosphorus in particular) contained in diffuse runoff from surrounding agricultural land as well as point source discharges have led to major changes in the trophic status of lakes, leading to the widespread problem of eutrophication (see Box 12.4 and Moss, 2017). Some lakes are used as direct disposal routes for sewage effluent

(Moss, 1998). The accumulation of phosphorus and nitrogen in previously oligotrophic (low-nutrient, low-productivity) lakes can cause a dramatic shift along the nutrient and productivity spectrum towards a eutrophic state. The problem was particularly prevalent in the 1960s to 1970s when the water quality and fisheries of many lakes across Europe and North America deteriorated as a consequence of massive algal blooms associated with eutrophication. Eutrophication is now increasing in countries such as China which are undergoing major economic development.

The problem of eutrophication is recognizable by increased growth of aquatic plants and phytoplankton. Occasionally these blooms contain species of blue– green algae that produce substances toxic to mammals, in particular affecting cattle using the lake margins as a source of drinking water or dogs swimming in the water. Algal blooms also clog filters where water is extracted from lakes for human consumption, increasing the cost of treatment and sometimes altering the taste of the water (Moss, 1998). In extreme cases the high densities of algae can lead to deoxygenation of lakes because

BOX 12.4 ➤

319

Chapter 12 Freshwater ecosystems

➤ night-time algal respiration, coupled with the decomposition of detritus originating from the increased phytoplankton bloom, consumes dissolved oxygen from the water column more quickly than it can be replaced by diffusion from the atmosphere. Consequently, there can be major losses of fishes and other aquatic organisms during these deoxygenation episodes. If the hypolimnion becomes deoxygenated, sulfides can be released from bed sediments causing further water quality problems for water supply companies and aquatic ecosystems. Solving the problem of eutrophication can involve treating the symptoms, by either physically removing aquatic plants and phytoplankton or altering the mixing regime of lakes. However, these interventions can be costly in the long term and do nothing to address the root cause of the problem. Laboratory studies carried out in the 1960s and 1970s suggested that phosphorus, nitrogen and carbon were all responsible for eutrophication. However, experimental studies in lakes showed that the problem could be controlled successfully by restricting the input of phosphorus alone. In a series of experiments, scientists from the Canadian Fisheries and Marine Service manipulated the nutrient content of lakes in north-western Ontario. Figure 12.14 shows an image of Lake 226, which has two basins of similar size separated by a narrow constriction (Schindler, 1974). The constriction provided an opportunity to physically separate the basins using a barrier. Both basins were treated with nitrogen and carbon throughout the year but only the north-east basin received additions of phosphorus. The north-east basin quickly developed a major bloom of the blue–green alga Anabaena spiroides whilst the phytoplankton

Figure 12.14 Lake 226 on 4 September 1973. The yellow line is the top of a vinyl curtain rein-

forced with nylon, which was sealed into the lake bed sediments and fastened to the lakeside bedrock. The lower part of the lake (north-east basin) received additions of phosphorus and quickly developed a bloom of blue–green algae as can be seen from the light green colour of the water. (Source: Fisheries & Oceans, Canada: E. Debruyn)

community of the adjacent basin did not differ from the pre-treatment period. The findings of this and other experiments led to a realization that the problems of eutrophication could be mitigated by making efforts to: reduce phosphorus loads in sewage treatment effluent discharges

(either by phosphate stripping or by reducing phosphorus in household and industrial detergents), control the application of phosphorus to farmland, reduce livestock densities in vulnerable lake catchments, or manipulate the aquatic food web (Scheffer et al., 1993).

BOX 12.4

which can cause problems related to nutrient enrichment, organic pollution and the introduction of pharmaceuticals. For example, Windermere, England’s largest natural lake, receives effluent from two sewage treatment works 320

and while these remove a large component of the potential pollutant and nutrient load, some pollutants remain in the effluent. Additionally, releases of untreated waste are common from overflow pipes during heavy rain.

12.4  Summary

Figure 12.15 Typical lake fish cages. (Source: Mikhail Olykainen

Shutterstock.com)

One of the major ways in which lakes provide goods and services to humans is through fisheries. In many regions of the world, lake fisheries provide fish that comprise the major dietary component. However, a typical problem is overfishing, which in turn often leads to the restocking of fish to replace those which have been over-exploited. In some places commercial fish farms are common on lakes, with fish being reared in large cages (Figure 12.15). Inevitably, some fish escape and these can be particularly problematic if they are non-native species because they often out-compete native species. Other invasive species that have caused major problems in lakes are zebra mussels (Dreissena polymorpha). These bivalve molluscs originate from the Caspian Sea which borders

12.4 Summary

northern Iran and southern Russia, but they were spread to western Europe in the nineteenth century, North America in the late 1980s, and to other parts of the world, most likely in the ballast waters of cargo ships. Since being introduced to North America they have spread quickly across the Great Lakes and into Canada. Dispersal is rapid because high numbers of larval offspring are produced by each adult zebra mussel (e.g. Johnson and Carlton, 1996), and these are distributed by lake currents. Adult mussels are also transported between lakes attached to boats, and their unrestricted spread has caused major problems with the fouling of vessels, docks and blocking of water supply pipes. Their mode of feeding by filtering detritus and phytoplankton from the water column has also led to increased clarity of lakes, meaning increased sunlight penetration and algal growth with knock-on effects throughout the lake food web.

Reflective questions ➤ Can you draw the typical zones of a lake ecosystem? ➤ Consider a lake that you have seen recently. By what process is it likely to have been formed and what classification might it be accorded based on the stratification process described by Hutchinson (1957)?

➤ How have the activities of humans altered lake ecosystems?

over timescales ranging from days to years. The river continuum concept has been a very influential theory over the past three

This chapter has provided an introduction to some of the diverse

decades and describes the aquatic zones within river systems

habitats and major biotic groups of plants and animals that

from headwater streams downstream to larger rivers. Connec-

can be found in rivers and lakes, the differences between these

tivity is provided by flowing water carrying, for example, fine

two types of freshwater system and some of the alterations

particulate organic matter or nutrients as part of the freshwater

imposed by human activities. River systems can be examined at

nutrient spiral. Some of the ways in which freshwater ecosystems

different scales and the chapter has introduced the concept of

have been altered by human environmental change include the

longitudinal, lateral and vertical spatial dimensions as well as

introduction of alien, invasive species, changes to chemical and

some of the important changes that occur in river ecosystems

thermal characteristics through pollution, modification to flow



321

Chapter 12 Freshwater ecosystems

➤ regimes through river and lake control structures (e.g. dams),

are particularly diverse ecosystems, and the ideas discussed

and climate change impacts (e.g. changes to glacial meltwater

within this chapter for the most part provide general frame-

release, floods or drought). Lake ecosystem classification sys-

works, rather than a blueprint for how individual river or lake

tems including geomorphological, stratification and nutrient

systems can be expected to work. The reader is encouraged to

status have been outlined. An individual lake ecosystem can be

consult the recommended further reading, key research papers,

described by a simple conceptual framework of zones depend-

references and weblinks on the Companion Website to build up

ent on physical and chemical characteristics which interact with

a more detailed understanding and knowledge of freshwater

biota. However, it is important to remember that rivers and lakes

ecosystems.

Further reading

Scheffer, M., Hosper, S.H., Meijer, M.L., Moss, B. and Jeppesen, E. (1993) Alternative equilibria in shallow lakes. Trends in Ecology and Evolution, 8, 275–279. Steady states and thresholds of change in lake systems.

Allan, D.J. and Castillo, M.M. (2007) Stream ecology: Structure and function of running waters. Springer, Dordrecht, Netherlands. This is a very thorough text providing a detailed review of research studies on the subject of stream ecology. It is suitable for more advanced undergraduate study but also has introductory information on stream hydrology, water chemistry and human modifications of river systems.

Vannote, R.L., Minshall, G.W., Cummins, K.W., Sedell, J.R. and Cushing, C.E. (1980) The river continuum concept. Canadian Journal of Fisheries and Aquatic Sciences, 37, 130–137. This is the paper that proposed the highly influential concept in river ecosystem studies.

Moss, B. (2017) Ecology of freshwaters: Earth’s bloodstream. Wiley-Blackwell, Chichester. A comprehensive book covering lakes and wetlands but including some chapters on streams and rivers. The book provides a detailed overview and is aimed at students who wish to gain an integrated view of freshwaters.

Ward, J.V., Tockner, K., Arscott, D.B. and Claret, C. (2002) Riverine landscape diversity. Freshwater Biology, 47, 517–540. An excellent review of landscape features in river corridors, landscape evolution, ecological succession, connectivity and biodiversity.

322

C HAPTER 13

Vegetation and environmental change John Grace School of GeoSciences, University of Edinburgh

Learning objectives After reading this chapter you should be able to:

➤ understand the way vegetation responds to climatological variables, and appreciate some of the underlying mechanisms of this response

➤ outline the evidence that shows how temperature and water supply have a leading role in determining the global patterns in vegetation

➤ appreciate how researchers are using models to predict future vegetation patterns

➤ discuss how human activities interact with climatic impacts, in both ‘good’ and ‘bad’ ways

➤ understand the relationship between climate models, vegetation models and observations, and realize that much remains to be understood about the impact of climate on vegetation and vice versa

13.1 Introduction Vegetation responds to a large number of factors in the physical environment, such as temperature, the available solar energy, and the supply of nutrients and water via the soil. Humans have been discovering the nature and extent of this response for a very long time, at least since biblical times when people first began to grow crops. The impact of drought, for example, is woven into the history of tribes

and whole civilizations. In modern times, the study of the relationship between plant life and the climate falls within several scientific disciplines. Plant physiology is concerned with understanding the functioning of plants, and this includes the response of plants to their environment and the acquisition of resources by plants. From plant physiology we learn how species have different environmental requirements as a result of differing structural and biochemical make-up. Agronomy, horticulture and silviculture are applications of plant physiology in the service of humankind to provide field crops, garden plants and wood products. Ecology, on the other hand, looks more broadly at vegetation, and has a focus on plant distribution; here we learn that plant distribution is not only constrained by climate and soil, but also influenced by factors such as competition, herbivory, fire and disturbance (Chapters 10 and 11). Palaeoecology is about ancient distributions (see Chapters 4 and 5), often inferred from deposits of pollen in a few places where material is well preserved, such as peat bogs and lake sediments. Every year, the surface of the peat or the bed of a lake is added to by a ‘rain’ of pollen grains. Cores can be extracted from the material, and slices can be acid-digested to remove organic debris leaving pollen grains. The temporal patterns (with some assumptions) can be correlated with changes in temperature. Likewise, patterns of growth over long periods can also be measured in some organisms, such as corals and trees. Dendrochronology is the study of climate patterns from the annual growth

Chapter 13 Vegetation and environmental change

rings of trees, and the related dendroecology is the study of tree rings to investigate ecological processes. In old trees these growth records extend over hundreds of years, and by joining together the records of past generations of trees, for example using trees found preserved in bogs or the timbers of ancient buildings, it is possible to obtain a record over thousands of years. Finally, phenology is the study of the annual variations in the timing of key events in the life cycle of plants, such as the date when specific tree species open their leaves, the date when flowers appear, or the date of autumn colouring when leaves lose their chlorophyll and photosynthesis ceases. There are long-term records of these phenological events, and they can be related to trends in the climate. In this chapter, we will draw upon work from all these disciplines, noting at the same time that to synthesize knowledge over a range of disciplines often requires some kind of mathematical model that incorporates knowledge and understanding, and that can be run from historical climate data or from data generated by climate models. In fact, we use predictions of what the climate may be in the future to estimate how the vegetation may change in the future. We may also note in passing that there is an inevitable relation between climate and the native vegetation, as discussed in the classic work by Wladimir Köppen, a German climatologist, in a 1931 paper (see Chapter 8), and Holdridge, an ecologist, in 1947. We emphasize ‘native’ vegetation, because much of the world’s vegetation is affected by humans and transformed or removed so that the land can be used for agriculture and forestry. It turns out that for the native vegetation, the most important variables determining distribution on a global basis are average annual and monthly temperatures, and precipitation (see Figure 10.7 in Chapter 10). Here, the land cover is represented as ‘biomes’, each biome occupying a particular region of climate space. From such relationships we see how a warmer world might shift particular locations from one biome to another.

13.2 Fundamentals of how plants respond to climatic variations 13.2.1 Light About 50% of the dry weight of plants is carbon, which has been accumulated through the process of photosynthesis (from the Greek phos = light and synthesis = combination) summarized as follows:

Green leaves achieve this by capturing photons (energy as light from the Sun) with a set of pigments, of which the most important are the chlorophylls (green pigments); then they use the captured energy to drive a series of chemical reactions that result in CO2 being absorbed from the atmosphere and converted into simple sugars, such as glucose. Subsequently, glucose is made into storage compounds, the most common being starch, and structural compounds such as cellulose (of which cell walls of leaves and roots are made) and lignin (a component of the cell walls of woody stems). From this discussion, it is clear that green plants need light, and that the more light they have, the more growth can be expected. It is worth pointing out that the process of photosynthesis has been going on for nearly 4 billion years, and that the by-product of photosynthesis, oxygen, enables all aerobic life including our own. From a biochemical perspective, different types of photosynthesis are recognized. The majority of plant species are found to fix CO2 into 3-carbon compounds, triose phosphates. However, some other species have a different enzyme system and they make a 4-carbon compound instead, oxaloacetic acid. The former condition is known as C3 photosynthesis and the latter as C4 photosynthesis. C4 plants are relatively recent. We see them in the fossil record only 20–25 million years ago and they spread remarkably only 8 million years ago. Most of them are grasses, some are sedges, a few are herbs and shrubs, and very few of them are trees. When they evolved from C3 ancestors, fire was on the increase because of a drying climate, and browsing increased with the evolution and spread of large mammals in the Oligocene and Miocene. Only a few per cent of species are currently C4 but they can utilize high levels of solar radiation more effectively and may account for as much as 30% of global photosynthesis, mostly in the tropics and mainly in savannas (see Chapter 10). There are many differences between C3 and C4 but here we need note only a few of them. C4 plants utilize water more efficiently and so they tolerate periods without rain. C4 photosynthesis works best in warm climates, and so we can reasonably expect C4 plants to be favoured in the future, warmer climate (Figures 13.1 and 13.2). They are among the faster growing crops of the world, including maize, sorghum, millet and sugarcane. They dominate some of the most productive natural ecosystems in the world, from the floodplains in Brazil

6CO2(gas) + 12H2O(liquid) + photons S C6H12O6(aqueous) + 6O2(gas) + 6H2O(liquid) carbon dioxide + water + light energy S glucose + oxygen + water 324

(13.1)

13.2  Fundamentals of how plants respond to climatic variations

0–30% C4

30–60% C4

60–100% C4

Figure 13.1 Grasslands of the world, showing the percentage of the grasses that are C4. (Source: from Ehleringer et al., 2005)

(b)

(a)

(c)

50

CO2 uptake (omol m-2 s-1)

40

C4 grass C4 grass

C4 grass 30

20

10

C3 grass

C3 grass

C3 grass

C3 tree

C3 tree

C3 tree

0 500 1500 1000 Light (omol photons m-2 s-1)

2000

10

20 30 40 Temperature (5C)

50

Days without rain (arbitrary units)

Figure 13.2 Environmental responses of photosynthetic CO2 uptake by leaves of three hypothetical species: a C4 grass, a C3 grass and a C3 tree.

(a) Under different amounts of light, the C4 grass has higher rates of CO2 uptake and is able to continue to exploit the incident solar energy even in bright sunlight; C3 grasses and herbs often have a higher rate of photosynthesis than C3 trees. (b) The performance of the three species at high temperature; we see how the C4 grass has a higher optimum temperature for photosynthesis. (c) How the hypothetical species might respond to a period with no rain: C4 species have a higher water use efficiency than C3 species (they lose only 250–300 g of water for every gram of CO2 fixed, while C3 lose 400–500 g water per g CO2); grasses have near-surface roots and they die down in drought but trees often have deep roots and can exploit stored water more effectively (grasses will regrow leaves when the rain comes).

where we find a C4 aquatic grass Echinochloa polystachya to the lakes in Africa where mono-specific stands of Cyperus papyrus occur, the sedge used by ancient civilizations to make ‘papyrus’. They are widespread as the C4 grasses

in tropical savannas, where they co-exist with C3 trees. A third type of photosynthesis, first discovered in the plant family Crassulaceae, is associated with leaves that are fleshy and accumulate malic acid during the night. This 325

Chapter 13 Vegetation and environmental change

Figure 13.3 Cyperus papyrus plants on a lake – an example of a tropical C4 sedge. (Source: C. Jansuebsri/Shutterstock.com)

condition is termed crassulacean acid metabolism (CAM). In contrast to all other plants, CAM plants keep their stomata closed in the day and hence avoid water loss; but they open them at night using a different enzyme system to capture CO2 and form malic acid. Then, during the day the malic acid is broken down to yield CO2 which is fixed into glucose in the same way as we find in C3 plants. As well as members of the Crassulaceae, CAM is found in cacti, the pineapple family (Bromeliaceae) and orchids.

13.2.2 Water Inside the plant cell, water is the medium in which all the important biochemistry takes place. For land plants water is additionally important as they are mechanically supported by the water pressure inside them. Take away water from a plant and it becomes flaccid (it wilts). So, rain (strictly, the supply of water) is an important variable in controlling the distribution of plants, and determining their growth rates. Only a few plants can survive desiccation. These are the so-called resurrection plants that can disassemble the photosynthetic machinery when drought comes and then reassemble it later when water is abundant. Almost all land plants benefit from an extra supply of water, but species vary hugely in how well they can tolerate dry periods. Some tolerate drought by shedding leaves or by having special characteristics such as a thick and waterproof coating (cuticle) on their leaf surfaces, or by having 326

small thick leaves and special organs to store water. Plants with adaptations to dry conditions are called xerophytes, and are found especially in deserts. Changes in water supply are expected as a component of climate change: as we saw in Chapter 7 we expect some regions to become wetter and others to be drier; we may therefore expect vegetation to change in some places and patterns of food production to alter. For example, it has been predicted that the Amazon rainforest will be replaced by a more xerophytic vegetation within 100 years as El Niño events become more extreme and possibly more frequent (Cramer et al., 2001).

13.2.3 Temperature The biochemical reactions involved in photosynthesis and growth all require warmth. Most plants photosynthesize and grow best between 10 and 30 °C, although there are important variations. Some organisms can even thrive in the extreme conditions of hot springs and others live in the coldest places on Earth. But these extremophiles are mostly bacteria, not vascular plants. Much of the planet’s surface is too cold for many plant species, and the long winters of the boreal zone limit photosynthesis and prevent cell division. As climate warms we expect northern and mountain regions to become greener, to photosynthesize more rapidly and to grow faster. Indeed, there is evidence from historical photography and satellite imagery that this is already happening (Fraser et al., 2014). We expect that in

13.3  Observational studies: how we know for sure that vegetation responds to a changing climate

the north especially, there will be widespread changes in the distribution of plants, as the length of the ‘growing season’ increases. For example some of the boreal forests in the southern parts of the taiga biome may die back (browning of the forest) to be replaced by grasslands. There is evidence that this shift has already begun (Buermann et al., 2014).

are acting as a carbon sink. So far, there is just one experiment on mature trees, and a few on trees enclosed in large chambers. Consequently we conclude that the extent of the stimulation of photosynthesis by CO2 is one of the unknowns in global models of vegetation.

13.2.5 Other climatic variables 13.2.4 Carbon dioxide concentration The atmospheric concentration of carbon dioxide (CO2) has risen from 270 parts per million (ppm) before industrialization to over 390 ppm today. As photosynthesis involves the diffusion of CO2 molecules from the atmosphere through the stomatal pores to the active sites inside the leaf where photosynthetic fixation occurs, it is to be expected that any increase in the external concentration of CO2 will increase photosynthesis. This is because, according to Fick’s law, the rate of diffusion increases in proportion to the concentration gradient. The increase in photosynthesis might also increase the growth rate and final yield of the plants. In fact, commercial growers have often used high CO2 concentrations in greenhouses to speed up the production of crops like tomato and lettuce. The increase in photosynthesis may not be as clear as expected, however, because plants raised at high CO2 develop fewer stomata and their stomata become somewhat closed at high CO2. Moreover, the products of photosynthesis may not be readily translated into growth and yield. For example, there may be a shortage of a vital nutrient such as nitrogen, or the temperature may be too low for cell division to occur. The extent to which these expectations about the effect of high CO2 are true can be tested using a type of experimentation called FACE (Free Air enriChment Experiment). In FACE, a set of supply tubes surrounds a small circular field in which crops, small trees or other plants are being raised. Then, CO2 is fed through the tubes to enrich the atmosphere. Sensors measure the concentration and adjust the supply rate. Usually the experiment is aimed at doubling the CO2 concentration, to try to predict what may happen in a future high CO2 world. The results of such experiments show that the increase in photosynthesis, productivity and yield is rather smaller than people have imagined. A doubling of external CO2 will almost double the diffusion gradient, so in a simple physical system we might expect this to translate into a near doubling of CO2 uptake. However, in a review of experimental data, Leakey et al. (2009) reported a mean stimulation in biological production of 17% and in yield of only 16%. This type of experimentation cannot easily be done on major biomes such as the tropical and boreal forest, which collectively

Other climatic variables are important too, especially on a local basis. Photosynthesis responds strongly to changes in humidity, as leaves tend to close their stomata in dry air. Wind is an important factor also. It plays a direct role in some regions of the world, where it may limit the extent to which forest or woodland can develop. In all environments wind is important because it determines the relation between the climate and the microclimate at the surface of leaves. In this regard, the influence of wind on heat transfer between leaves and the atmosphere determines surface temperatures, and plants that are sheltered are generally a few degrees warmer by day than those that are exposed in the same location. Wind may also convey adverse materials, including salt spray and pollutants.

Reflective questions ➤ What are the key climate variables determining vegetation distribution and growth rates?

➤ Should we take into account extremes of climatological variables or do we work merely with mean values?

➤ Is the temperature of a leaf the same as the temperature of the air?

13.3 Observational studies: how we know for sure that vegetation responds to a changing climate In this section, we review examples of field observations that show how natural vegetation frequently changes as a result of climate change.

13.3.1 The forest/savanna boundary in southern Amazonia The Amazon rainforest occupies about 6 million km2 and holds around one-third of the global biodiversity. However, the world’s most famous rainforest may not have been as 327

Chapter 13 Vegetation and environmental change

extensive just a few thousand years ago. Records obtained from lake sediments in Bolivia in the southern part of Amazonia suggest that the boundary with the savanna may have undergone substantial changes over the past few thousands of years (Mayle et al., 2000). The results are presented as a pollen diagram (Figure 13.4), in which we see the fluctuations of some major groups of plants found as pollen from a 3 m core taken from the sediment at the bottom of a deep lake. In such studies, the age of the samples is determined by the 14C dating (see Box 5.1 in Chapter 5) of the strata from which the samples have been taken. At present, the lake is surrounded by rainforest and the pollen ‘rain’ into the lake is dominated by

ca ce Ce ae cr o M pi au a Cu riti ra a/ te M Po lla au ac am riti ea e el e ric la an a Iso et es

rti /U ea e

ic

or ac M

Or ga n

0

co

yr ( 14 C e Ag

St

ra t

ig

ra p

hy

BP

)

nt en

t

pollen of the families Moracaceae and Urticaceae, which contain predominantly trees. There is also a signal from Cecropia, a tree genus. Grass pollen from members of the Poaceae is relatively uncommon, and so are Mauritia and Mauritiella, both palms of wet places. However, just 4000 years ago the situation was very different. The site was evidently much drier. We can tell the water level in the lake must have been much lower because we see pollen of Isoetes (quillwort), a plant that lives in the shallows at lake margins (Figure 13.4). Most significantly, there is a strong component of grass pollen and relatively little tree pollen, showing this area to have been savanna at the time. There is also charcoal, an indication of dry

20

1650 ; 40 40

2740 ; 50

Depth (cm)

3220 ; 40 60

7980 ; 70

80

9380 ; 70

100

9590 ; 70

120

9720 ; 60 9820 ; 70 11030 ; 80 38 600 ; 1000 46 000 ; 2900

140 160 180

41 400 ; 1500 200

43 900 ; 1900

220 240

43 700 ; 2000 260 280

750 850

0

Organic mud

20 0

20 40

Sand

0

0

0

Clay

0

20 40 0 20 40 60 80

Clay/sand

Figure 13.4 Vegetational history of southern Amazonia seen in the pollen record. (Source: Mayle et al., 2000, from F.E. Mayle, R. Burbridge & T.J. Killeen,

Millennial-scale dynamics of Southern Amazonian rain forests, Science, 290: 2291–2294, 2000. Reprinted with permission from AAAS.) 328

13.3  Observational studies: how we know for sure that vegetation responds to a changing climate

conditions and human presence. The conclusion, that rainforest in the area is quite recent, and preceded by grassland vegetation, can be supported by other evidence. Mayle et al. (2000) point to a regional increase in rainfall at the time tree pollen began to be common: it is known that the water level in Lake Titicaca (located in the Andes on the border of Peru and Bolivia) rose after 3200 BP, reaching modern levels by 2100 BP. This change in climate was not anthropogenic, although the savannas were probably occupied and fire was used as a tool, but rather the consequence of a southward shift in the intertropical convergence zone (see Chapters 6 and 8), itself triggered by Milankovitch forcing (see Chapter 4). A few similar studies have been conducted at other tropical and subtropical regions. For example, the Sahara Desert was formerly much wetter and vegetated, with a considerable human population (see Chapter 5). The work is relevant to ideas of how tropical vegetation will change in the future. Some climate models predict substantial drying in Amazonia and elsewhere, and it is thought likely that part of the region will revert to savanna. Today, the changes might well be accelerated by humans who are inclined to use fire for clearing. When the forest is dry there can be large-scale destruction of forests by fire as we have seen in many parts of the world in recent years. This removal of forest by fire is one of the mechanisms by which savanna replaces forest.

13.3.2 The northern tree line Trees are excluded from cold places. In most older texts it is claimed that trees are absent from sites wherever the temperature of the warmest month is less than 10 °C, following a line of thinking by Köppen. This is not entirely true: for some Andean sites the tree line coincides with a maximum summer temperature of only 6 °C. However, there is a consensus that summer temperatures are important and that they must be sufficient for adequate rates of photosynthesis and cell division. The consequence of this is the existence of a phytogeographical boundary where trees give way to dwarf shrubs, found at northern latitudes and on mountains all over the world. Around the boundary, known as the tree line, the trees grow slowly and are often stunted with contorted stems. The German word to describe the woodland composed of these trees is krummholz (crooked wood). Climate warming has so far been more rapid in the extreme northern regions than elsewhere, and so it is in the north that we might expect to see a sign of warming in the vegetation, for example, increases in the rates of growth, and northerly advances in the distribution of trees. In fact, observations from several regions of the world support this

view. In some places it is possible to obtain photographic evidence of shifts in vegetation, either by repeatedly revisiting the same place, comparing aerial or satellite photographs, or simply using photographs taken by ecologists and naturalists on the ground. Higgins et al. (2016) showed using satellite data globally that biomes with tall vegetation and minimal vegetation activity in the cold season shifted to higher productivity biome states between 1981 and 2012. Frost and Epstein (2014) studied shrub and tree advance in Siberian tundra. At their 11 study regions, spread across Arctic Siberia, ranging in size from 34 to 78 km2 each with more than 40 years of data, they found shrub and tree expansion had occurred since the 1960s in both lowland and upland areas (Figure 13.5). Some spatial differences (e.g. some decreases in tree cover at some locations) were related to disturbance due to permafrost melt and more active floodplains (Figure 13.5). Similar changes in tree line species have been recorded in other parts of the world too including sub-Arctic spruce–tundra in northern Québec, Canada (Gamache and Payette, 2004) and Pinus sylvestris populations in Sweden where there have been strong increases in growth and reproduction and a decrease in winter damage (Kullman, 2005).

13.3.3 Upward march of vegetation in mountains Even away from high latitudes observations have shown an upwards expansion of shrubs and trees. One such case is the Montseny Mountains of Catalonia in north-east Spain (Peñuelas and Boada, 2003), where ground-based images were available from 1945. In these mountains the beech forest has moved up the slope by 70 m at the highest altitudes. At medium altitudes the existing beech forest is becoming degraded, partially defoliated and is not regenerating itself as it formerly did. Nearby, heathland is being replaced by the more drought and heat-tolerant holm oak. Changes like this have been reported in Japan (Nakashizuka et al., 2016) and from the European Alps as well. There, the existence of many small alpine species is threatened with local extinction as their particular habitat declines in area as a result of global warming (Pauli et al., 2007). This general phenomenon, first recognized three decades ago, has prompted the establishment of long-term sample plots, which are revisited every few years for enumeration of the plants. The project has established a global network of sites in mountain regions, known as GLORIA (Global Observation Research Initiative in Alpine Environments, www.gloria.ac.at). They have shown, for example, from 867 vegetation samples above the treeline from 60 summit sites in all major European mountain systems, that ongoing climate change is gradually transforming mountain 329

Chapter 13 Vegetation and environmental change

22.0

Normalized expansion rate (% decade–1)

10 8 6 4 2 0 Kh O Tn Tz D L U P V

Kh O Tn Tz D H L U Ko P

Kh O Tn Tz H Ko P V

Kh O Tn Tz D H Ko P V

–2 –4

Upland Inactive floodplain

–6

Lowland Active floodplain –10.4 –13.4

Figure 13.5 Net change in shrub and tree cover at northern Siberian tundra ecotones since the 1960s across four landscape types. As the period of

data record is not the same for all regions, the percent change in cover is normalised to express the decadal rate of canopy cover change as a percentage of total canopy cover present at the beginning of the study period for each landscape. The letters refer to the names of the regions studied. (Source: Frost and Epstein, 2014).

plant communities with a resultant compression of cold mountain habitats (Gottfried et al., 2012)

13.3.4 Changes in the timing of flowering The recording of the annual cycle of growth and development of plants has interested amateur naturalists and professional meteorologists for over 100 years (Jeffree, 1960;

Sparks et al., 2000). In some countries, ‘phenological gardens’ have been established to monitor such things as the date of bud-break every spring, the date of first flowering and the date on which trees shed their leaves. For example, there is a large network of phonological sites across Europe (see http://tinyurl.com/j3pdtnr) (e.g. Figure 13.6). The data sets generated by these gardens are a rich archive of information, linking biology and meteorology. In some

Figure 13.6 Johnstown Castle gardens in County Wexford, Ireland with some of the trees and shrubs in flower. The dates of plant development (e.g. first

flowering) are recorded each year for the various species in the garden contributing to the Irish and European phenology network databases. (Source: Captblack76, Shutterstock.com) 330

13.3  Observational studies: how we know for sure that vegetation responds to a changing climate

cases these records are very long ones, broken only by interruptions by war or failure of funding. In other cases, they are the result of one person’s painstaking efforts over a lifetime. One of the best examples is provided by the naturalist Richard Fitter, who recorded the data of first flowering for a set of 557 wild plant species over a period of 47 years in south-central England (Fitter et al., 1995; Fitter and Fitter, 2002). These authors found a very strong dependency of first flowering date on temperature in the previous months. Only 24 out of the 243 species selected showed no dependency. Menzel et al. (2006) completed an analysis of more than 100 000 time series of phenological stages across Europe. They found that nearly all the phenological events, and also farmers’ activities, were related to temperature (Figure 13.7). In Figure 13.7, the recordings represent the date of onset of the event. A negative correlation coefficient therefore means that warming is

associated with an earlier date of onset. Nearly all phenological events occur earlier as a result of warming in the spring. Overall, the authors calculate that as a result of warming, spring in Europe is arriving 2.5 days earlier per decade. A few events have positive correlations with temperature, meaning that warming is associated with a later occurrence of the event. For example, warm summers are associated with later leaf colouring in the autumn. We may safely conclude that, at least for temperate perennial plants, the life cycle is to a large extent set by the temperature, within certain limits. The cycle of growth and development may be advanced or retarded, and the growth period shortened or prolonged, according to the temperature pattern of a particular year. From such data, it is possible to model what might happen with a few degrees of warming. To go further and to predict the ecological consequences is somewhat more

(a) 0.80

Correlation with temperature

0.60 0.40 0.20 -0.20

Farmers’ activities Flowering Leaf unfolding agriculture

Ear formation–agriculture Ripeness–agriculture Fruit ripening–wild plants

-0.40

Leaf unfolding wild plants

Leaf colouring

-0.60

Shooting

0

-0.80 -1.00 -1.20 30

60

90

120 150 180 210 240 Mean date (day of year)

270

300

330

(b) Regression with temperature (days 5C-1)

4 2 0 Corylus avellana F Tussilago farfara F Aesculus hippocastanum F Syringa vulgaris F

-2 -4

Robinia pseudoacacia F Taraxacum officinalis F Betula pendula LU Betula pendula LC

-6 -8 30

60

90

120

150

180

210

240

270

300

330

Mean date (day of year) Figure 13.7 The relationship between phenological events and temperature measured on different days of the year: (a) the correlation coefficient

(negative coefficient means high temperatures on that day of the year makes the event happen earlier); (b) the sensitivity of the bud opening to temperature on that day of the year for a range of six trees and two perennial herbs. For example, a regression of –4 on day 120 means that one degree of warming stimulates early bud-opening by 4 days. F = flowering, LU = leaf unfolding, LC = leaf colouring. (Source: from Menzel et al., 2006) 331

Chapter 13 Vegetation and environmental change

difficult. If insect-pollinated plants flower early, for example, they will not be fertilized unless the relevant insect pollinators are available at the same time. Also, if flowering is too early there might be some increased risk of damage by frost. Clearly, this work has enormous economic relevance and it assists growers to plan for the future and also to evaluate risks. Today, satellite sensors can be brought to bear on the problem of relating climate change to the cycles of plant life, especially in relation to the greening up of the land in spring. However, satellites look at huge swathes of landscape and it is usually hard to disaggregate the signal of natural vegetation from the frequent changes in agricultural practice. So, there remains a role for amateur phenologists, and interest in phenological gardens is currently as high as it has ever been.

Reflective question ➤ What evidence is there that vegetation is responding to climate change?

13.4 Models for prediction In the preceding sections, we saw something of how plants respond to climatic variables. There are substantial differences between species: how long they live, their growth rates and physiology, and this in general is why different species occupy different regions of the land surface. In principle, it would be possible to determine a set of parameters for each species, and then to use a model to determine where the species could possibly occur within the world’s biomes. The task would be large, as even closely related species respond differently to environmental variables, and in any case there are important other influences, to which we have already alluded, such as competition between species and the tendency for dominant species to modify the conditions for other species (Chapter 11). It would not be practical to screen the characteristics of all extant species of vascular plants, as there are about 400 000 of them. The size of this task is one of the bottlenecks in the development of predictive models. The research community is still discussing the best way to link knowledge of species to knowledge of biomes; however, there is a consensus that some coarse classification of species is required so as to capture the essence of what makes one vegetation type different from another. Ideally, 332

this description should have a ‘meaning’ across several disciplines, so that remotely sensed data on reflectance of the land cover can be related to what ecologists can discern as a more or less homogeneous entity. To this end, and after a number of international conferences on the subject, the idea of plant functional types (PFTs) emerged. A PFT is a group of plants with similar traits and which are similar in their association with environmental variables. Each PFT is defined by a variety of optical, morphological and physiological parameters. At present there is no real consensus beyond this, but in Table 13.1 we show one set of PFT definitions from Woodward et al. (2004). From this starting point we can make mathematical models of how vegetation responds to a changing climate. Most models keep track of the flow of carbon as well as the PFTs and biomes. The important processes which need to be modelled in that case are as follows. 1. The rate of photosynthesis, and its dependency on environmental factors and the supply of nutrients. The rate of photosynthesis on an ecosystem scale is termed gross primary productivity (GPP); it is the CO2 uptake in the daytime period, adjusted by an estimate of how much carbon is simultaneously lost by plant respiration. 2. Allocation patterns. The allocation of the carbon acquired in photosynthesis to different plant parts (leaf, stem and root) is required and so is the respiration rate required to build complex molecules such as proteins and lignin from the simpler ones like glucose. If we know how this process of allocation depends on environment, and we also know the rate of photosynthesis, then we can compute the growth rate. On an ecosystem scale this is known as net primary productivity (NPP), related to GPP as follows: GPP = NPP + Ra

(13.2)

where Ra is the respiration that the plant expends, known as autotrophic respiration. There are some established theoretical relationships that help us to find Ra. For example, the synthesis of cellulose requires the energy obtained from the breakdown of glucose as well as a supply of glucose molecules, so the formation of 1 g cellulose is associated with the release of a specific quantity of respiratory CO2. These are ‘classical’ issues in plant physiology, discussed by Amthor (2000). 3. Birth, death and phenology. Plant parts are formed and they are shed, often on an annual cycle. Plants as a whole are ‘born’ every time a seed germinates and they eventually die. Some are annuals (programmed to die after a few weeks or months), others are biennials (the life cycle

13.4  Models for prediction

Table 13.1 A plant functional type classification adopted by Woodward et al. (2004)

Classification

Comment

Evergreen needleleaf forests

Lands dominated by trees with a canopy cover of more than 60% and height exceeding 2 m. Almost all trees remain green all year. Canopy is never without green foliage

Evergreen broadleaf forests

Lands dominated by trees with a canopy cover of more than 60% and height exceeding 2 m. Almost all trees remain green all year. Canopy is never without green foliage

Deciduous needleleaf forests

Lands dominated by trees with a canopy cover of more than 60% and height exceeding 2 m. Consists of seasonal needleleaf tree communities with an annual cycle of leaf-on and leaf-off periods

Deciduous broadleaf forests

Lands dominated by trees with a canopy cover of more than 60% and height exceeding 2 m. Consists of seasonal broadleaf tree communities with an annual cycle of leaf-on and leaf-off periods

Mixed forests

Lands dominated by trees with a canopy cover of more than 60% and height exceeding 2 m. Consists of tree communities with interspersed mixtures or mosaics of the other four forest cover types. None of the forest types exceeds 60% of the landscape

Closed shrublands

Lands with woody vegetation less than 2 m tall and with shrub canopy cover more than 60%. The shrub foliage can be either evergreen or deciduous

Open shrublands

Lands with woody vegetation less than 2 m tall and with shrub canopy cover between 10 and 60%. The shrub foliage can be either evergreen or deciduous

Woody savannas

Lands with herbaceous and other understorey systems, and with forest canopy cover of between 30 and 60%. The forest cover height exceeds 2 m

Savannas

Lands with herbaceous and other understorey systems, and with forest canopy cover between 10 and 30%. The forest cover height exceeds 2 m

Grassland

Lands with herbaceous types of cover. Tree and shrub cover is less than 10%

(Source: Woodward et al., 2004)

takes two years, sometimes a little longer) and perennials (most grasses and all trees are good examples). Generally, in global models we have to provide some ‘rules’ for each PFT. As in all models, we have to ignore the exceptions and abide by simple rules. For conifer trees, for example, leaves are retained for several years whereas for deciduous trees they are shed on an annual cycle. The resulting ‘litter’ from shedding and death is incorporated into the soil, and decomposed by microbes. The microbial respiration gives rise to further efflux of CO2 from the ecosystem, Rh, known as heterotrophic respiration. The resulting net flux of CO2 between the ecosystem and the environment is the net ecosystem production (NEP), related to the previous terms as: GPP = NEP + Ra + Rh

(13.3)

In a world with a constant environment, all the GPP would be respired by plants or microbes and so NEP would be zero. In the real world there are additional losses of carbon caused by natural and human influences. To take these into account we would need to insert a new carbon loss, the disturbance flux Rd. 4. Population-level processes. Models need to represent the process of plant succession, whereby species

colonize the land and develop towards an equilibrium or near-equilibrium state. To do this, the attributes of each PFT are needed, and some further rules are required to determine in what circumstances one species succeeds another. Models are run using a sequence of climatological data, in time steps that vary from months to minutes, according to the scheme of Figure 13.8. How good are such models? They can be tested in several ways: (i) by examining whether the model calculations can reproduce today’s vegetation by running the models from historical climate data; (ii) by investigating whether they produce the carbon fluxes that are measured in field studies; and (iii) by comparing them with other models, developed in other laboratories more of less independently. An important comparison of six models was reported by Cramer et al. (2001). They computed the ‘expected’ vegetation on the basis of a world divided into regions of 3.75° longitude and 2.5° latitude, given the climatic scenario. In this scenario of climate change, temperature increases by 4.5 °C over the next 100 years, and the CO2 concentration rises from 380 to 800 ppm by the end of the century. Although the six models do not all agree in 333

Chapter 13 Vegetation and environmental change

Atmosphere CO2

CO2

Leaves

Stem

CO2 CO2

CO2 CH4 Root CO2 CH4 Soil

River

Ocean

Figure 13.8 Structure of a very basic model of the carbon flows associated with photosynthesis, respiration, growth and storage of carbon. ‘Losses’ of

CO2 from leaves, stems and roots (autotrophic respiration) are associated with maintenance and growth of the plant tissues; losses from the soil are either from microbial respiration (heterotrophic respiration) or in the drainage water. Such a model might be configured for an individual plant or for a set of PFTs in an ecosystem. More advanced models include flows of water and nutrients, and their interaction with carbon flows.

detail, and although they do not faithfully reproduce the current vegetation, there is a general consensus on how vegetation change will occur (Figure 13.9). The principal changes are:

t t t

Transitions from forest to savanna. Such transitions are predicted to occur in the Amazon, the central part of the American continent and in South-East Asia. Evergreen forest will replace ‘grassland’ in parts of North America and north Europe. Parts of the Mediterranean grasslands will become savannas.

As for the carbon balance, the models produce an interesting trend, and there is some agreement between models. The NPP is stimulated by warmer conditions (felt especially in the cold northern regions) and also by the elevated CO2 (Figure 13.10). However, the heterotrophic respiration is increased by warming to an even larger extent, and the consequence is a large rise in global respiration over several decades. It is interesting to put these numbers into perspective as follows. The heterotrophic respiration from all the microbes (heterotrophic respiration of all terrestrial ecosystems) is currently around 50–60 gigatonnes of carbon per year (Gt C yr-1), which completely dwarfs the fossil fuel emissions of about 6.5 Gt C yr-1.

334

Large though Rh is, it is more than offset by photosynthetic production, so the NEP is positive. In other words the terrestrial ecosystems are collectively a ‘sink’ for carbon. In fact, the current terrestrial sink strength of 1–2 Gt C yr-1 is predicted to rise for a few decades before taking a downward turn as the effect of temperature on respiration is increasingly felt. There is a substantial predicted downturn in NEP so that by 2100 the sink has diminished and, in one of the models, has turned into a source. The most significant weakness of these models is that they are not coupled to a model of the climate system, so the CO2 effect of a vegetational source or sink on the temperature is not apparent. In the ‘real world’ the atmospheric CO2 and therefore the global warming rate would be influenced by the vegetation. Such feedbacks may well be important. In recent years, simple vegetation models have, however, been coupled to global circulation models (GCMs) in an attempt to capture the essential feedbacks. Here, we refer to a synthesis study (Friedlingstein et al., 2006) in which 11 models were compared. The results are quite variable and the models do not all show the same general trend (Figure 13.11). They clearly differ in their sensitivity to climate change. The most sensitive result of all is shown by the Hadley Centre Model where the vegetation becomes a progressively weakening carbon sink in the

13.4  Models for prediction

(a) 100

HYBRID IBIS LPJ SDGVM TRIFFID VECODE

90 80

NPP (1015 g Cyr-1)

70 60 50

2000

40 (b) 100

Rh (1015 g Cyr-1)

90 80 70

2100

60 50 40 (c) 10

NEP (1015 g Cyr-1)

5

2100 + 100 Desert

Savanna

Mixed forest

Grassland

Deciduous forest

Evergreen

0

Figure 13.9 The predicted distribution of biomes as a result of climate

change and elevated CO2 according to Cramer et al. (2001). The models stopped simulating climate and CO2 change in 2100. Therefore, changes from the middle to bottom panel are related to vegetation dynamics as the system equilibrates. See text for further interpretation. (Source: from Cramer et al., 2001)

next few decades and then moves into carbon deficit rather in the same way as the uncoupled models discussed in the previous paragraph. For the other models in the Friedlingstein study, there is a weakening of the sink except in one case where the sink continues to intensify. Further work is clearly needed before we can make further estimates of the future based on models. One of the more difficult aspects of predicting the future from models is that humans interact with climate change and this interaction is hard to predict. In truly managed ecosystems, the economically relevant outputs

-5

1900

1950

2000

2050

2100

2100+50 2100+100

Figure 13.10 Predicting the carbon balance of terrestrial vegetation for

the next century by means of six models: (a) the net primary productivity (NPP); (b) the heterotrophic respiration (Rh microbes and animals); (c) the overall carbon flux made by subtracting Rh from NPP. Each line is a different model, as shown in (a). (Source: from Cramer et al., 2001)

are the ones that are recognized, monitored and increased largely as a result of applying science. In modern UK agriculture, for example, the agriculture has become more intensive since the Second World War, and has enjoyed substantial government subsidies, yet a recent study showed widespread recent carbon losses from the soil in England and Wales, thus contributing inadvertently to global warming.

335

Chapter 13 Vegetation and environmental change

13.5 The complex interaction between human activities and climate change

1000

Atmospheric CO2 (ppm)

900

13.5.1 Does atmospheric pollution sometimes benefit plants?

800 700 600 500 400 300 12.0 10.0

Land uptake (GtCyr-1)

8.0 6.0 4.0 2.0 0.0 -2.0 -4.0 -6.0

1850

1900

1950

2000

2050

2100

Year Figure 13.11 Carbon uptake from the atmosphere by terrestrial

ecosystems (units: Gt = gigatonnes 5109 tonnes 5 1 billion tonnes). The calculations are made after coupling several GCMs to simple models of the terrestrial biosphere. Each line is the result from a different model. The current terrestrial carbon ‘sink’ is somewhere between 0 and 4 Gt of carbon per year. Some models predict that the land sink will become a source. (Source: from Friedlingstein et al., 2006)

Reflective questions

As a result of human activities, notably agriculture and the driving of motor vehicles, much more nitrogen in a chemically active form (ammonium and nitrate especially) is deposited to the land surface than hitherto (Box 13.1). If there is too much N-deposition, ecosystems may show signs of ‘nitrogen saturation’, a condition whereby the land surface may ‘leak’ nitrogen to the drainage water and give off nitrous oxide, another greenhouse gas. If, on the other hand, the deposition rate is below a certain threshold level, there may be a fertilizer effect for many ecosystems, especially those that are otherwise N-deficient. This would be seen as a stimulation of photosynthesis and possibly an increase in growth rates and a strengthening of the carbon sink. Indeed, many model calculations of the impact of environmental change on global vegetation contain a term to allow for this. The N-effect and climate change interact in ways that are not understood very well, as highlighted by Magnani et al. (2007). Data on the C-fluxes over forests in Europe and North America were collected, and attempts were made to relate the GPP and Rh to temperature. In fact, both showed a remarkable linear relation with temperature; but the NEP (i.e. GPP - Reco, where Reco is the total ecosystem respiration) is rather a weak function of temperature (Figure 13.12). However, NEP is strongly related to the deposition of anthropogenic nitrogen from the atmosphere. Hence, we may suggest that pollution of the atmosphere with nitrogen (principally from vehicles, agricultural systems) is at a level that stimulates production. This is of course a somewhat controversial claim, as most people link N-deposition to nitrogen saturation of ecosystems or, worse, with the production of acid rain that has deleterious impacts on forests.

➤ Models are only as good as the understanding built into them. What are the gaps in our understanding of vegetation–climate models?

➤ How well have the vegetation models been tested? ➤ What is a plant functional type? ➤ What important processes need to be modelled in PFT, biome and carbon flow vegetation models?

336

13.5.2 How does fire interact with climate change? Many predictions regarding climate change suggest that some areas of the tropics will become dryer as a result of an increased frequency and harshness of El Niño events. As we saw above, this increase in drought is one of the causes of the expected conversion from forest to savanna.

13.5  The complex interaction between human activities and climate change

(a)

(b) 20 y = 0.52x + 6.97 R 2 = 0.92

15

Average GPP (t Cha-1 yr-1)

Average Reco(t Cha-1 yr-1)

20

10

5

0 -10

10 0 Mean annual temperature (5C)

15

10

5

0 -10

20

(c)

20

(d) R 2 = 0.41

Average NEP (t C ha-1 yr-1)

Average NEP (t C ha-1 yr-1)

0 10 Mean annual temperature (5C)

6

6

4

2

0 -10

y = 0.71x + 7.86 R 2 = 0.92

0 10 Mean annual temperature (5C)

20

R 2 = 0.97 4

2

0

0

3 6 9 N wet deposition (kgNha-1 yr-1 )

12

Figure 13.12 Measured carbon fluxes over 18 forests in Europe and North America: (a) ecosystem respiration, Reco; (b) gross primary productivity,

GPP; (c) and (d) the overall carbon flux as directly measured as NEP, showing a carbon sink of between 0 and 5 t C ha - 1 yr - 1. Note that the overall carbon flux NEP is a fairly weak function of temperature but a strong function of the deposition of nitrogen from the atmosphere. (Source: Magnani et al., 2007, reprinted by permission of Macmillan Publishers Ltd: NATURE, F. Magnani, M. Mencuccini, M. Borghetti, P. Berbigier, F. Berninger, S. Delzon, A. Grelle, P. Hari, P.G. Jarvis, P. Kolari, A.S. Kowalski, H. Lankreijer, B.E. Law, A. Lindroth, D. Loustau, G. Manca, J.B. Moncrieff, M. Rayment, V. Tedeschi, R. Valentini, J. Grace, The human footprint in the carbon cycle of temperate and boreal forests, vol. 447: 848–850. Copyright 2007)

THE NITROGEN CYCLE AND ANTHROPOGENIC PERTURBATIONS Nitrogen is an important constituent of proteins and nucleic acids, and so is essential for life. Nitrogen exists primarily as an unreactive gas in the atmosphere as dinitrogen, N2. It constitutes 79% of the air we breathe. Very important reactive forms of nitrogen also exist as gases and ions. The gases are ammonia (NH3) and the oxides NO, NO2

and N2O, all present in trace concentrations. The main ions, found in soils and water, are ammonium NH4+ , nitrate NO3 and nitrite NO2 - ions (see Chapter 17). Two natural processes convert nitrogen to reactive forms that can be taken up by the roots of plants: lightning and biological nitrogen fixation (BNF). Very small amounts of N2 are reacted with O2 during lightning, to form the gas nitric oxide, NO, which eventually reaches the ground as nitrate. BNF is quantitatively

more important than lightning as an agent of nitrogen fixation: bacteria living in the soil fix N2 to make the reactive forms ammonium NH4+ , and nitrate NO3 - which can be used by plants. Some of these nitrogen-fixing bacteria are free-living, but others form symbiotic relationships with plants, especially those of the pea family Leguminosae. Many members of this family are used in agricultural systems as a ‘free’ source of nitrogen fertilizer (examples are

BOX 13.1 ➤

337

Chapter 13 Vegetation and environmental change

➤ Nitrogen in atmosphere (N2)

Plants

Assimilation Denitrifying bacteria

Nitrogen-fixing bacteria in root nodules of legumes

Nitrates (NO3-) Decomposers (aerobic and anaerobic bacteria and fungi) Ammonification

Nitrification

Ammonium (NH4+) Nitrogen-fixing soil bacteria

Nitrifying bacteria

Nitrites (NO2-) Nitrifying bacteria

Figure 13.13 The nitrogen cycle.

clover, lucerne, groundnuts, soybeans, alfalfa and lupins). Such plants have root nodules containing populations of the nitrogen-fixing bacteria. As a consequence of the free-living nitrogenfixers and the symbiotic nitrogen-fixers, 130–330 Gt N yr - 1 are made available in the soil solution as ammonium NH4+ and nitrate NO3 - and can thus be taken up by plants to make protein and other biochemical constituents. Herbivorous animals obtain their protein by consuming plant material in prodigious quantities. Dead plants and animals decompose in the soil, and some of the nitrogen is acted upon by denitrifying bacteria. Nitrogen is thereby returned to the atmosphere as N2 or N2O. The process is thus cyclic (Figure 13.13). Humans perturb the nitrogen cycle in fundamental ways. The industrial

fixation of N2 was developed in 1909 by Haber and Bosch, and provides a supply of nitrogen fertilizer, estimated to be 78 Gt N yr - 1. This is applied to the soil, but much of it is released to the atmosphere as nitrous oxide N2O, a greenhouse gas. Some of it enters drainage water, and causes excessive growth of algae in streams and rivers. There are other anthropogenic sources of nitrogen. The internal combustion engine and some other fuel-burning devices are responsible for emissions of oxides of nitrogen to the atmosphere, through the combination of atmospheric oxygen and nitrogen inside the combustion chamber. Moreover, cultivation and disturbance of the land results in emissions of nitrous oxide. Finally, animal rearing is associated with the emission of NH3, produced by the decomposition of urine and faeces.

The consequence of increased formation of NH3 and oxides of nitrogen is an enhanced rate of ammonium and nitrate deposition to land and waters. Nitrogen deposition rates are now much higher in populous regions of the world than they were in pre-industrial times (Figure 13.14). This deposition occurs as dry deposition of the gases themselves, and also as nitric acid (HNO3), a contributor to acid rain. The growth of plants is generally limited by the availability of nitrogen in the soil, and so the enhanced deposition may be increasing plant growth, and contributing to a widespread increase in the rate at which trees grow. On the other hand, in some areas, the imbalance in nutrients in acid rain may cause damage to forests.

BOX 13.1 ➤

338

13.5  The complex interaction between human activities and climate change

➤ 905N

(a)

605N

305N

05

305S

605S

905S 1805

1205W

605W

0 905N

605E

05

2

10

20

50

1205E

1805

1205E

1805

250

(b)

605N

305N

EQ

305S

605S

905S 1805

1205W

605W

0

605E

05

2

10

20

50

250

Figure 13.14 Deposition of reactive nitrogen to the Earth’s surface (in mmol N m - 2 yr - 1): (a) pre-industrial levels; (b) current atmospheric deposition. Note the enhanced deposition in densely populated areas. (Source: after Galloway et al., 1995)

BOX 13.1

However, as humans encroach upon the rainforest, some researchers believe the effect will be amplified by the creation of forest edges in a fragmented forest, and the use of fire (Laurance and Williamson, 2001). Dense forests have a microclimate characterized, for example, by high

humidities and daytime temperatures that are usually lower than those measured at the top of the canopy. At forest edges the situation is different, with free horizontal ventilation and mixing of canopy air with air from outside. When drought occurs, relatively dry air penetrates the 339

Chapter 13 Vegetation and environmental change

Deforestation

Forest fragmentation

Less evapotranspiration

Pastures + regrowth

Logging

Greater fire vulnerability More smoke

More fires

Less rainfall + higher surface temperatures

ENSO droughts ? Global warming

Figure 13.15 The Laurance–Williamson model of how droughts and human impacts combine to degrade and destroy rainforests.

(Source: Laurance and Williamson, 2001)

canopy, and reductions in humidity and increases in plant mortality have been measured at up to 100 m from the canopy’s edge. Humans light fires, and these fires are likely to ignite more easily and spread more rapidly in the dry conditions of the forest edge. Hence, the forest is damaged and possibly destroyed at a faster rate to what would occur in the absence of humans. The processes involved are quite complex and collectively amount to a positive feedback (Figure 13.15). In the Laurance–Williamson model, deforestation causes less evaporation, which in turn leads to less rainfall and hence droughts are exacerbated. Logging can also be important as it thins the canopy and increases vulnerability to combustion.

Reflective questions ➤ How might atmospheric pollution benefit plants? ➤ How much N-deposition is required before N-saturation occurs? (Research the literature and see what you can find.)

340

13.6 Loss of biodiversity Biodiversity is defined as the number and variety of species in ecological systems, at local, regional and global scales (see Chapter 10). There is concern that both human activities and climate change are causing a decline in the number of species. Given global warming, the importance of climate change is likely to become progressively more important, as pointed out in the recent report by the IPCC (IPCC, 2014). There are at least two reasons why we may expect biodiversity to alter under climate change. In the first place, the pattern of land use will change as the demand for agricultural land and biofuel increases. Thus, species-rich habitats such as rainforest and savanna in the tropics will decline and at some critical point the species they contain will be lost. The second, and more subtle, reason for loss of biodiversity is that species differ greatly in their sensitivity to warming and to water supply. Thus, particular species in an ecosystem may migrate, leaving a functional void in their original location. In the new ecosystems that they come to occupy, they are ‘invasive species’ and may be strong competitors, ousting some of the more delicate species. Thus, ecosystems which are more or less in a

13.7  Agriculture and food security

species equilibrium may be destabilized with unknown consequences. Although it is impossible to know the total number of species in the world (only 1.5 million are known but many are yet undiscovered), extinctions themselves are generally well documented. Since 1600, a minimum of 490 plant and 580 animal species have become extinct. Some groups, such as mammals and birds, have suffered more than others. In geological time there have of course been catastrophic mass extinctions. The natural or background extinction rate can be estimated from the fossil record. For example, in mammals the background rate is about one in 400 years, but this is much lower than the observed rate. The number of species that are threatened far exceeds our capacity to protect those species, and so conservationists have identified ‘hot spots’ where conservation effort and resources should be greatest (Myers et al., 2000). It is particularly important to protect areas with a high degree of endemism (an endemic species is one restricted to a particular region). When this exercise was carried out there were some surprises, as shown in Box 13.2. For example, the natural vegetation of the tropical Andes heads the list. Its vegetation has been reduced to 25% of its original extent, yet it contains 6.7% of all plant species in the world and 5.7% of all vertebrate animals.

HOT SPOTS AND CLIMATE CHANGE Myers et al. (2000) reported that 44% of all species of vascular plants are confined to 25 hot spots constituting only 1.4% of the land surface. The authors urged these to be singled out for the attention of conservationists, to attempt to protect them. The Myers et al. (2000) map of hot spots is reproduced here, along with projections of temperatures from the IPCC (2007b) 4th Assessment Report (Figure 13.16). Figure 13.16b shows the warming predicted from the A2 scenario (see Chapter 7). Note: all hot spots will be warmer.

Reflective question ➤ Why is biodiversity important?

13.7 Agriculture and food security Native vegetation has largely been replaced by agriculture. Some 40% of the land surface is used to produce food for 7.3 billion people, which is around the size of a soccer pitch per person (8200 m2, to be exact). But the human population of the planet is set to increase, albeit at a lower rate than in the past. The land available for agriculture cannot increase much, except at the expense of regions which are currently of high conservation value. Moreover, the increasing use of biofuels will take out significant areas of land which are now under agriculture. How then, will climate change impact upon the increasing need for food production? The question is not a simple one, as the answer depends to a large extent on location. Northern lands will benefit from warming because the length of the growing season will increase, and so agricultural production in the boreal zone is likely to increase. However, areas like the Mediterranean are expected to become drier and warmer,

Over the next century there will be further extinctions, and climate change will be an important driver, along with land-use change. Everyone agrees that species should be protected, and that natural environments have an inherent value. Indeed, the idea that humans are the guardians of nature is deeply embedded in many religions. There is also much folklore that evokes the conservation ethic. The author’s grandmother used to say: if you wish to live and thrive let a spider run alive. This appears to be a reference to a ‘keystone species’ (see Chapter 11), as

spiders are voracious predators, required to control populations of small flying insects, which often carry disease. When keystones are removed the ecosystem is in trouble. At the moment there is a plethora of international agreements to protect species and habitats and many individuals subscribe to organizations such as Friends of the Earth or Greenpeace. The challenge for nature conservation is to protect natural ecosystems but traditional conservation and protection against human encroachment is clearly not sufficient: it does not protect species and ecosystems from the impact of human-made climate change.

BOX 13.2 ➤

341

Chapter 13 Vegetation and environmental change

➤ (a)

Caucasus

Philippines

Mediterranean Basin

California Floristic Province

South-Central China

Caribbean

Mesoamerica

Brazil’s Cerrado

Choco/ Darién/ Western Ecuador

Polynesia/ Micronesia

Indo-Burma

Eastern Arc and Coastal Forests of Tanzania/Kenya

Polynesia/ Micronesia Western Ghats and Sri Lanka

W. African Forests

Tropical Andes

Sundaland Brazil’s Atlantic Forest

Central Chile

Wallacea New Caledonia

Succulent Karoo

Southwest Australia

Madagascar Cape Floristic Province

New Zealand

(b)

2030

0

0.5

2100

1

1.5

2

2.5

3

3.5 4 (5C)

4.5

5

5.5

6

6.5

7

7.5

Figure 13.16 (a) Twenty-five biodiversity hot spots identified where 44% of all species of vascular plants and 35% of all species in

four invertebrate groups are confined to the hot spots comprising only 1.4% of the land surface of the Earth. (b) Projected temperature increases for 2030 and 2100 from model scenario A2. (Source: (a) from Myers et al., 2000, reprinted by permission of Macmillan Publishers Ltd: NATURE, Norman Myers, Russell A. Mittermeier, Christina G. Mittermeier, Gustaro A.B. da Fonseca & Jennifer Kent, Biodiversity hotspots for conservation priorities, vol. 403: 853–858: Copyright 2000; (b) from IPCC, 2007a)

BOX 13.2

and agriculture will need to adapt over several decades. Plants everywhere are expected to benefit from rising CO2, but this benefit may be offset by other factors. Porter et al. (2014), writing for the IPCC, considered that the damaging effects of elevated tropospheric ozone will offset CO2

342

fertilization, and reminded us that weeds too will benefit from CO2 fertilization. Moreover, crop production systems, and consequently food prices, are very sensitive to extreme events. As these increase we may expect shortages of certain foods and instability in the human food chain.

Further reading

Although there is little scope for increasing the area of agriculture, its efficiency in most parts of the world can increase. Organic farmers have claimed that one person can obtain a full (vegetarian) diet from 100 m2 of land, rather than the 8200 m2 which prevails currently.

13.8 Summary

Reflective question ➤ How might climate change affect food production in the area of the world where you live?

carbon cycle and the climate system. Specifically, regions of the world such as the humid tropics, which are now believed to be

Plant growth is influenced by climatological variables, especially

a carbon sink, may become a source; conversely, cold northern

light, temperature and moisture. For natural vegetation it is pre-

regions will become warmer and therefore more favourable for

dominantly temperature and moisture that determine the type of

the growth of plants and especially for trees. They may become

land cover on a global scale. Both temperature and moisture are

a sink for carbon. Predictions are, however, based on state-of-

changing and they are expected to change rapidly in the next 100

the-art model calculations, and there are still many uncertainties.

years. Observations in the field, and use of satellite data over the

One of the largest unknowns in the system is the behaviour of

last few decades, support the general view that the vegetation is

humans. Their behaviour determines the rate of global warming

changing in response to global climate change. Over the next cen-

and the nature of the land cover, and also modifies the response

tury, it is likely that substantial changes in vegetation will result,

of vegetation on a global scale through the use of agents such as

and that these are likely to be so important as to impact upon the

cultivation, fertilizers and fire.

Further reading Cleland, E.E., Chuine, I., Menzel, A. et al. (2007) Shifting plant phenology in response to global change. Trends in Ecology and Evolution, 22, 357–365. Interesting review of data and approaches from different fields. Leakey, A.D.B., Ainsworth, E.A., Bernacchi, C.J., Rogers, A., Long, S.P. and Ort, D.R. (2009) Elevated CO2 effects on plant carbon, nitrogen and water relations: six important lessons from FACE. Journal of Experimental Botany, 60, 2859–2876. Useful review of experimentation. Lovejoy, T.E. and Hannah, L. (2005) Climate change and biodiversity. Yale University Press, New Haven, CT. A good overview book.

Malhi, Y. and Phillips, O. (2005) Tropical forests and global atmospheric change. Oxford University Press, Oxford. A research-level enquiry into one of the most important issues raised in this chapter. Porter, J.R., L. Xie, A.J. Challinor, K. Cochrane, S.M. Howden, M.M. Iqbal, D.B. Lobell, and M.I. Travasso (2014) Food security and food production systems. In: Climate change 2014: Impacts, adaptation, and vulnerability. Part A: Global and sectoral aspects. Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Field, C.B., V.R. Barros, D.J. Dokken, K.J. Mach, M.D. Mastrandrea, T.E. Bilir, M. Chatterjee, K.L. Ebi, Y.O. Estrada, R.C. Genova, B. Girma, E.S. Kissel, A.N. Levy, S. MacCracken, P.R. Mastrandrea, and L.L. White (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, pp. 485–533. The IPCC’s scientific overview of climate change and agriculture.

343

PA RT V

Geomorphology and hydrology Figure PV While tectonics drives global landform development, the actions of water, ice, gravity and wind combine with biotic and chemical processes to alter those landforms and to create new landforms at large and small scales. Understanding the interactions between these processes is the key to geomorphology.

Part contents ➤ Chapter 14: Weathering ➤ Chapter 15: Slope processes and landform evolution ➤ Chapter 16: Sediments and sedimentation ➤ Chapter 17: Soils ➤ Chapter 18: Catchment hydrology ➤ Chapter 19: Fluvial geomorphology and river management ➤ Chapter 20: Solutes and water quality ➤ Chapter 21: Drylands ➤ Chapter 22: Coasts ➤ Chapter 23: Glaciers and ice sheets ➤ Chapter 24: Permafrost and periglaciation

Part V Geomorphology and hydrology

Scope This part of the textbook deals with those processes that mould and shape landscapes. Geomorphologists are interested in trying to ascertain the processes and mechanisms that interact with, and create, the form of the landscapes we study. Often, while most processes may be operating on all land environments, the dominating processes depend on the regional climate characteristics or tectonic context. Tectonic processes were dealt with in Part II of this textbook and so Part V focuses on other geomorphological themes and how the processes (including tectonic ones) come together in different places at different times to produce distinctive Earth features. On a rudimentary level, it is the Earth’s attempt to create some kind of balance of inputs and outputs of energy that leads to the internal reorganization of sediment and materials, resulting in the changing shape of landscapes. A common theme in Part V is one of ‘budgets’ in which inputs and outputs of water, ice, sediment or energy are calculated and used as a framework for studying dynamic system adjustments. It is through uplift, erosion and sedimentation that landscape forms are maintained. Erosion and sedimentation are dependent upon the physical agent exchanging the energy in the system, which is most commonly a transporting fluid (water, wind or ice), although gravity also plays a significant role. Although erosion and sedimentation are contrasting processes, they are inextricably linked and the same sediments can be reworked many times. The chapters in this part of the book highlight that it is often the relative abundance of the various processes, and the balance between the process rates, that determines the nature of the landforms in different environments. The first seven chapters of this part (Chapters 14–20) detail processes that can occur on any landscape on Earth whereas the remaining four chapters describe the processes and nature of particular Earth environments (Chapters 21–24). The processes of weathering are considered in Chapter 14 and the discussion includes consideration of weathering under different environmental conditions including in the urban environment. Weathered materials are transported by erosion. Chapter 15 considers erosion processes and deals with the interactions between the shape of the landscape and erosion processes. Chapter 15 deals

346

with issues of form–process relationships in relation to slope profiles and landscape evolution. However, the landform itself is not simply a product of geomorphological or hydrological processes: the processes themselves will be influenced by the landform. In addition, landscapes may incorporate a legacy from the operation of geomorphological processes under past climatic regimes such as dryland and post-glacial environments. Therefore contemporary processes and landscapes are likely to be linked to historic processes and landscapes and there may be time lags in landscape response. Change through time in hillslope form is covered in Chapter 15 ranging from slow creep processes to fast, hazardous landslides. Sediments transported by erosion processes can also form landscapes elsewhere and the nature of sediments and sedimentary landforms is covered by Chapter 16. The products of weathering and sediment movement often help to form soils. The soil acts as the interface between the atmosphere, lithosphere, hydrosphere and biosphere in which we exist. It performs a wide range of essential functions that sustain life. The properties of soils are determined by the processes that maintain the inputs and outputs of material through the soil system. Soil can be degraded, lost and improved by natural or human activities. An appreciation of these issues is imperative if we want to preserve our soil in a healthy state. Chapter 17 attempts to provide such a basis. This part of the book also deals with the nature of water (especially Chapters 18, 19, 20 and 22), solutes (Chapter 20) and ice movements (especially Chapters 23 and 24) in a range of environments including drylands (Chapter 21), spectacular coastal environments (Chapter 22), glacial landscapes (Chapter 23), and periglacial environments (Chapter 24). It describes motions of energy and mass on the hillslope, in the floodplain, within channels dominated by water and ice, and in coastal zones. The focus is an understanding of processes so that as physical geographers we can explain different landforms and how they are likely to change in the future. We also seek to understand hazards such as landslides, flooding, sea-level rise, changes in river channels, water pollution, coastal erosion and subsidence caused by permafrost melt. In each chapter in Part V, special consideration is given to the influence of humans on landscape processes and how understanding of the processes can result in improved management strategies.

C HAPTER 14

Weathering Bernard J. Smith Late of School of Geography, Archaeology and Palaeoecology, Queen’s University Belfast

Learning objectives After reading this chapter you should be able to:

➤ describe the main weathering processes ➤ explain the role of water in weathering ➤ understand the importance of multiple weathering processes operating in combination or sequence

➤ describe the climatological and geological controls on weathering at different scales

➤ outline the main factors associated with building stone weathering

14.1 Introduction This chapter will deal with how geomaterials (rocks and regoliths) react when exposed to weather. Studies of weathering are studies of change or ‘metamorphism’, in which materials that were typically formed under conditions of high temperature and/or pressure within the Earth’s crust interact with the highly variable and often rapidly fluctuating low-temperature, low-pressure and chemically complex conditions at and near its surface. In most cases this involves a progressive simplification of the mineralogy and

chemistry of geomaterials and a decrease in their structural integrity. This chapter will examine one of the most crucial Earth surface processes. Without weathering there would be no soils (Chapter 17), and thus no plants and no people. Without weathering there would also be a dramatic reduction in global erosion and deposition (Chapters 15 and 16). Weathering directly provides the solute load of rivers (Chapter 20) and is an essential precursor that unlocks the transport of sediment by all but the most energetic of geomorphological processes (Chapter 16). The changes involved in weathering are often individually very small, gradual and to a large extent hidden from view either inside rock or within and beneath a blanket of soil. The physical measurement of weathering effects is challenging, time consuming and requires long-term observation using specialist equipment before any noticeable change can be recorded. This is not to say, however, that over geological time the gradual changes brought about by weathering cannot be just as effective in shaping the Earth at the landscape scale as, for example, large floods which might be dramatic, but infrequent. What weathering lacks in terms of magnitude it more than makes up for in terms of persistence and pervasiveness. While there are some processes that act

Chapter 14 Weathering

in isolation, it is much more common to find weathering processes operating in combination or in sequence to the extent that some processes may act as essential precursors to the operation of others. This chapter explores these processes and the outcomes of these processes on the materials around us.

14.2 Environmental and material controls on weathering The controls on weathering processes represent a competition between the environment and the material. It is often considered that the environment is the main driving force of weathering (Figure 14.1). Hence there is often a desire to classify weathering on the basis of climate into, for example, categories such as tropical and periglacial. However, the consequences of the exposure of geomaterials to different environmental conditions are strongly influenced by material properties. Where these properties are particularly distinctive they may overtake environmental influences in terms of their significance. Hence distinctive features and landscapes can be recognized that are associated with the weathering characteristics and controls exerted by, for example, granitic rocks and limestones (e.g. see Box 14.5). There are no weathering processes that can be uniquely ascribed to a particular climatic zone. Hence a weathering classification based on climate alone may be problematic. For example, in the case of ‘tropical weathering’, we are not referring to a set of unique processes found only in these regions, but rather to particular combinations and intensities of generic weathering processes related to a specific climatic regime. This in turn results in distinctive weathering products and, in limited cases, supposedly definitive landforms. Such ‘climatic landforms’ are, however, not always what they appear to be. For example, granite tors that are found in many textbooks on tropical environments are in fact a feature of most climatic regimes

Process

Environment Materials

Figure 14.1 The weathering system.

348

Form

(Figure 14.2). Their origins may have followed different pathways, but their final appearance often differs only in terms of the detailed morphology of the individual granite blocks. This convergence of shape or form is partly in response to a strong, but largely passive, lithological control on weathering conditioned by the characteristic mineralogy of granites. However, it also owes much to active structural controls that are a response to the release of physical stresses contained within the rock that in turn result in the formation of distinctive patterns of joints and fractures. The shape of the surface is important since there are multiple feedbacks in the weathering system (Figure 14.1). As weathering progresses it often develops complex surface morphologies such as variously sized hollows. This topography can then influence how processes operate. This might, for example, be through the modification of surface temperature by the creation of shade, by allowing water to accumulate, by allowing salt to accumulate in hollows protected from rainwash on vertical surfaces and by creating micro-environments in which organisms can flourish. These are positive feedbacks that encourage further growth of the hollow. The atmosphere is an important environmental control on weathering. It provides oxygen required for many weathering processes. Carbon dioxide (CO2) is also very important because when it mixes with water (e.g. in rainfall) it forms carbonic acid which dissolves rocks such as limestones. Weathering processes, which characteristically operate at the scale of individual grains and crystals, are controlled on a day-to-day basis by micro climates at the rock–atmosphere interface and within the pores of fresh rock and weathered regolith and the soils above them (see Chapter 17). At the micro-scale, it is temperature and moisture that mainly control patterns and rates of weathering. In the case of temperature, seasonal and diurnal cycles together with erratic variations in surface temperature driven by, for example, variations in cloud cover and wind speed, interact to create patterns of temperature change over time. These parameters are particularly significant for a wide range of near-surface physical weathering processes, both directly through influencing responses such as thermal expansion of minerals and indirectly through their influence on variables such as relative humidity and moisture uptake, movement and evaporation. Overall, however, the moisture balance within these materials is driven primarily by the availability of precipitation across different spatial and temporal scales. ‘Deep wetting’ of materials is, for example, most likely to reflect seasonal rainfall patterns, whereas

14.2  Environmental and material controls on weathering

(a)

(b)

Figure 14.2 (a) A granite tor in the Guadarrama Mountains near Madrid, Spain composed of joint bounded boulders. (b) Section in a road cutting near

to the tor in (a) showing an ‘incipient’ tor shaped by differential subsurface weathering controlled by the joint system within the granite. If the weathered regolith were to be removed by erosion the tor would then be revealed.

near-surface wetting and drying within a few centimetres of the surface is more likely to be ‘event related’ and controlled by factors such as the amount, duration, intensity and frequency of individual storms. Included in these factors is wind direction and velocity, which studies have shown to be crucial in determining the efficiency of driven rain in the wetting of rock (Ashurst and Dimes, 1998). Scientists are still trying to understand how individual wetting events contribute over time to longer-term, deeper wetting and drying and what role some processes such as direct condensation of water onto rock play in weathering. The most complex reactions between geomaterials and the atmosphere are most likely to occur in the course of urban stone decay. This is primarily the result of the addition to the atmosphere of gases such as sulfur dioxide (SO2) and various oxides of nitrogen (NOx) from the burning of fossil fuels. Through a combination of photochemical oxidation and reaction with moisture in the atmosphere this results in the formation of a chemical cocktail that includes naturally occurring carbonic acid as well as sulfuric and nitric acids and which falls as ‘acid rain’. Total ‘acid deposition’ is, however, more complex than this as acids can be produced directly through the reaction of gases with moist stone surfaces (dry deposition) or precipitated onto surfaces from fog and condensation (occult deposition). While such deposition is, like naturally occurring carbonic acid in the atmosphere, very effective in dissolving carbonate rocks such as limestone, it plays a much more significant role when it reacts with building materials to produce soluble salts. These are responsible for widespread damage to stone through a

variety of mechanisms associated with the process of salt weathering (Goudie and Viles, 1997). The most potent of these salts is generally considered to be gypsum (calcium sulfate, CaSO4 # 2H2O), formed most commonly by the reaction of sulfuric acid with stone containing calcium carbonate, although it can be deposited directly onto building stone where gases have reacted in the atmosphere with carbonate-rich combustion particles (particulate deposition). In this way, and through reactions with carbonate mortars, it is possible for all stone types to be ‘contaminated’ with complex salts, even if they themselves contain no carbonate fraction. Because of this the urban environment not only ranks alongside deserts and coasts as an environment particularly prone to salt weathering, but is for much of the world’s population the environment in which they are most likely to encounter readily identifiable evidence of weathering. In fact, it could be argued that the built environment represents a well-constructed ‘weathering experiment’, in which friendly architects simultaneously expose different combinations of stone types to the same environmental conditions, conveniently align structures to permit evaluation and comparison of meteorological effects and often provide the researcher with a considerably greater knowledge of the physical characteristics and stress histories of the stone than is normally available for natural rock outcrops. The following sections explore the salient points raised above by trying, as far as is possible, to link weathering processes to the environmental and material factors that control them and the features and products they produce. 349

Chapter 14 Weathering

Reflective question ➤ Why is it inappropriate to classify weathering only on the basis of regional climate?

14.3 Weathering and the role of water Alteration of geomaterials occurs in two realms. There are complex changes that take place relatively deep within the Earth’s crust, generally in the absence of oxygen, in response to reactions with heated and chemically complex gases and liquids. Often referred to as hydrothermal alteration, these processes are enormously important in, for example, the mobilization and concentration of specific elements or weathering products into zones that can be then economically mined or quarried. This includes extensive deposits of the clay mineral kaolinite formed through the hydrothermal alteration of granites in areas such as the south-west of Britain, and which form the basis of that region’s ‘china clay’ industry. In this chapter, however, we are primarily concerned with weathering that takes place at or near the Earth’s surface under mainly sub-aerial conditions where they are in close contact with the atmosphere. In these surface environments the most important control on chemical alteration is the presence of water. This is both as a potential reagent and as the principal means by which the products of weathering are removed and weathering allowed to continue. The precise role of water, especially in chemical alteration, depends on a number of key properties. Water molecules are surprisingly complex. They consist of one oxygen and two hydrogen atoms, but exist as a covalent bond in which electrons are shared between atoms. In water, this bond is uneven and the oxygen atom attracts electrons more strongly than the hydrogen ones. This creates polar molecules with an asymmetric charge distribution and partial negative and positive charges at the ends of each molecule. It is this polarity that allows water to separate other polar solute molecules and explains why water is such an effective solvent (see Chapter 20). On its own, water can undergo self-ionization, in which one of the bonds of the liquid water molecules breaks to create positively charged aqueous hydrogen ions or protons (H+) and negatively charged hydroxyl ions (OH - ). In pure water the free hydrogen ions react with other water molecules to produce H3O+. At a normal temperature (25 °C) the concentration of free hydrogen 350

ions is extremely low at only one-tenth of a millionth of a gram per litre of water. To bring the figures into a more manageable range, the concentration is expressed as pH with a neutral water solution of pH 7 (see Chapter 17 for an explanation of pH). Any solution with a greater concentration of hydrogen ions is acidic and has a lower pH value, while those with a lower concentration have a higher value and are considered to be alkaline. In turn, any substance which, when added to the water, increases the proton concentration is therefore said to be acidic or a proton donor. In contrast, substances that are proton acceptors are said to be alkaline and are referred to as bases. As an example, adding a base such as sodium hydroxide to water results in the dissociation of an aqueous sodium ion (Na+) and an aqueous hydroxyl ion (OH - ), which will then combine with any free hydrogen ions. In nature, perhaps the most common proton donor is carbonic acid, formed by the reaction of atmospheric CO2 with water. When this dissolves in water it dissociates to form hydrogen and bicarbonate ions: H2O + CO2 S H2CO3 S H+ + HCO3 -

(14.1)

The importance of proton donation for weathering is that these protons can react with minerals such as many commonly occurring aluminium silicates (e.g. feldspars) and replace positively charged metal cations (e.g. Na, K+ and Ca2+ ) in their crystal lattices through a process known as hydrolysis. This leaves behind weathering products made up mostly of ‘hydrated aluminium silicate minerals’ or clays. In addition, pH is a key control on the solubility of substances, especially weathering products such as metal oxides and hydroxides. The importance of pH in influencing relative solubility is indicated in Figure 14.3. A further key property of water that controls chemical reactions and determines solubility is its redox potential, or Eh. This is the ability of a chemical species to bring about an oxidation or a reduction reaction. Eh is explained in Chapter 20. In terms of rock weathering the most common oxidation or reduction reactions are those associated with iron oxides. Iron commonly exists in two valent forms, either as oxidized, trivalent ferric iron (Fe3+ ) or divalent, reduced ferrous iron (Fe2+). Ferrous iron is therefore typically found under saturated conditions and most commonly in its hydrated form as minerals such as limonite (FeO(OH) # nH2O) and goethite (FeO(OH)) that give soils such as gleys (see Chapter 17) their distinctive yellowish green colour. When exposed to the atmosphere these minerals oxidize (rust), most typically into the mineral haematite (Fe2O3) that is responsible for the reddish

Al2O3

MgO

FeO

Al2O3

TiO2

8

Fe2O3

14.3  Weathering and the role of water

Solubility (millimoles/litre)

6

4

SiO2 2

0

0

2

4

6

8

10

pH Normal range of soil pH Extreme range of soil pH Figure 14.3 Solubility of metal oxides plotted against pH for some

common products of chemical weathering. (Source: modified from Douglas, 1979)

colour of many iron-rich soils. Another important distinction between these two valent forms is their relative solubility. From Figure 14.3 it can be seen that the reduced form of iron is relatively soluble under typical Earth surface conditions, so that it can, for example, be mobilized and concentrated under saturated conditions. If it is allowed to oxidize, however, it will become fixed under all but the most acid of conditions. When associated with repeated wetting and drying this switch from ferrous to ferric states can lead to the progressive concentration of iron as, for example, nodules within a soil profile or more uniformly in distinct soil horizons. The difference in solubility of iron in relation to its oxidized or reduced states shows how solubility in natural environments is dependent on both pH and Eh, which are, in turn, influenced by temperature. The relationships between pH and Eh are typically shown in the form of ‘Pourbaix diagrams’ for individual elements to establish the range of conditions under which mobilization and fixation are likely to occur. The phase diagram for iron in water is shown in Figure 20.1 (see Chapter 20), indicating its various states in relation to combinations of Eh and pH. As well as the chemical properties of water that are significant for weathering, water also possesses a range of physical properties that strongly influence both weathering environments and the processes that operate within them.

For example, the strong hydrogen bonds in water molecules give water a high cohesiveness and consequently a high surface tension, adhesion and capillarity (see Chapters 17 and 18). Within large pores the majority of the water is held by a combination of surface tension and capillarity, but very close to particle surfaces the water is bonded in its ionic state by much stronger forces of electrochemical adsorption. This creates a very thin film of water bound very strongly to the underlying material. The different states at which water is held exert a major influence on processes such as the evaporation of water held within porous materials and explains why initial drying may be quite rapid, whereas thorough drying requires considerably more time and energy. In nature, it is unlikely that all adsorbed moisture will be completely removed and so even in dry deserts adsorbed water is found around solid particles. The same tension and electrochemical forces act to attract moisture into dry materials by exerting a ‘suction potential’. Finer pores exert a greater suction and attract and hold moisture and any dissolved substances for longer. Therefore porosity and the pore structure have a considerable influence on many weathering processes. Adsorption and capillary attraction are not the only ways in which water can be held and it can also be absorbed into the structure of a material. The so-called mineral hydration occurs when hydroxyl and hydrogen ions are absorbed into a substance, leaving the non-water component chemically unchanged. The most common reactions of this type involve salts and results in the creation of new minerals or hydrates. The reaction is reversible and on drying the salt reverts to its original form. The other area where water absorption is important is in clay minerals (see Box 17.3 in Chapter 17), and there is an important group of these minerals known as smectites that can absorb water into their crystal lattice. These swelling clays, of which montmorillonite is the best known, form a major soil group, but when found within rocks the potential swelling and contraction associated with wetting and drying could be a contributory factor in their eventual physical breakdown. Indeed wetting and drying, or slaking, is recognized by many as a weathering process in its own right, to which certain rock types such as basalts are particularly prone because of the presence of secondary minerals that hydrate and dehydrate. Finally, water also has a unique property of having a ‘triple point’ that allows it to freely change between liquid, gaseous and solid states under normal Earth surface conditions. This facilitates not only wetting and drying, but also the best known of all volume changes, the expansion which occurs when water becomes ice. 351

Chapter 14 Weathering

14.4 Chemical weathering The processes by which geomaterials weather are normally grouped into three categories: physical, chemical and biological. In reality all three interact with each other and it is equally valid for the physical and chemical consequences of biological activity to be split and considered under the other two headings. Here we cover biological processes under the two main headings. Most weathering scientists consider there to be five main chemical processes that act individually and in combination to alter geomaterials, all of which involve water. These are described below.

14.4.1 Hydrolysis Hydrolysis is the chemical reaction between a mineral and water that is slightly acidic. Aluminium silicates, of which the best known are the feldspars, typically combine a metal with silica, aluminium and oxygen atoms. Within the feldspars these range from calcium feldspar or anorthite (CaAl2Si2O8) and sodium feldspar or albite (NaAlSi3O8), both of which belong to the plagioclase group, to potassium feldspar or orthoclase (KAlSi3O8). Other commonly occurring minerals which are weathered include groups of metal silicates, most notably ferromagnesian minerals containing iron and magnesium such as olivine ((Mg,Fe)2SiO4), and phyllosilicate or sheet minerals such as

CARBONATION EQUATIONS The carbon dissolution process is important in the formation of karst landforms (see Box 14.5). It is also referred to as incongruent dissolution, on the basis that any clay minerals found are derived from impurities within the original limestone rather than the alteration of the rock itself. The precise sequence is: CO2 + H2O S H2CO3

(14.3)

H2CO3 S H+ + HCO3 -

(14.4)

biotite (K(Mg,Fe)3(AlSi3O10)(OH)2) that are found particularly in igneous rocks. During the course of hydrolysis the metal cations are selectively replaced within the crystal lattice by a process known as congruent dissolution. Typically the first to go are the most mobile, monovalent cations such as sodium and potassium, followed by divalent ones such as calcium and magnesium and finally by the least mobile polyvalent ions such as iron and aluminium. Through this process, these minerals are progressively altered into clay minerals that are themselves phyllosilicates, together with soluble oxides that are usually lost from the system. A typical, simplified, reaction is that for orthoclase: 2KAlSi3O8 + 2(H+OH - ) S Al2Sl2O5(OH4) + K2O + 4SiO2

(14.2)

or orthoclase feldspar + ionized water gives kaolinite + soluble potassium oxide + soluble silica.

14.4.2 Carbonation Pure water does not exist in nature, mainly because rainwater reacts with CO2 in the atmosphere to form a weak solution of carbonic acid (H2CO3) and it is this acidified water that reacts with minerals through carbonation and hydrolysis. The best-known example of this sequence of reactions is that involving the calcium carbonate that comprises most limestones, which produces soluble calcium bicarbonate that is then lost in solution (Box 14.1).

Ca2+ + CO3 2 - + H+ + HCO3 - S Ca2+ + 2(HCO3 - )

(14.5)

or calcium carbonate + carbonic acid gives soluble calcium bicarbonate. Carbonation is not restricted to carbonate minerals. Carbonation and hydrolysis are also effective in the weathering of silicate minerals such as orthoclase: 2KAlSi3O8 + 2H2CO3 + 9H2O S Al2Si2O5(OH)4 + 4H4SiO4 + 2K+ + 2HCO3 -

(14.6)

or orthoclase + carbonic acid gives kaolinite + soluble silicic acid + soluble potassium bicarbonate. Carbonation is also effective for ferromagnesian minerals such as olivine: Mg2SiO4 + 4CO2 + 4H2O S 2Mg2+ + 4HCO3 - + H4SiO4

(14.7)

or olivene + carbon dioxide + water gives soluble magnesium carbonate +soluble silicic acid.

BOX 14.1

352

14.4  Chemical weathering

Box  14.1 shows that the carbonation process is highly dependent on the continued presence of CO2 in solution, which in turn is initially dependent upon its concentration in the atmosphere. Beneath a soil cover, however, the relative concentration (partial pressure) of CO2 in soil air, and also cave air, is typically greater than that in the open atmosphere (see Chapter 17, Section 17.2.4). This can lead to a higher content of dissolved CO2 in soil moisture and helps to explain how limestone is often dissolved more rapidly beneath a soil cover than when exposed to the atmosphere. It also helps explain how limestone can be precipitated from solution (i.e. a solid form produced from a dissolved form) as water flows out from under the ground and CO2 is lost to the atmosphere. An interesting anomaly arises from the partial pressure of CO2 being inversely related to temperature. It falls by approximately 50% between 0 and 20°C. This has led some scientists to suggest that karst solution (see Section 14.7 below) should, unlike almost all other chemical weathering, be most rapid in areas with cold climates. This ignores, however, a number of key factors, especially in relation to the elevated level of CO2 within soils that derives primarily from biological processes that would be inhibited under cold conditions. It also fails to consider the relative aridity of cold regions or the effective aridity associated with the prolonged freezing of what water is available.

14.4.3 Solution Water is a highly effective solvent. There are many naturally occurring minerals that are highly soluble in water. The degree of solubility is highly dependent on factors such as the pH and Eh of the water solvent. One important factor is that the rate of dissolution can be rapidly curtailed if there is no water flow and the water that is static above a surface becomes saturated with the solute.

14.4.4 Oxidation and reduction Oxidation and reduction have already been dealt with in Section 14.3.1, where they are seen to be the controlling factors in the transformation of substances through either the addition or removal of oxygen. In transforming substances, oxidation and reduction also exert considerable control on the mobility of substances in solution.

14.4.5 Biologically related chemical weathering There are numerous ways in which organisms act to initiate and enhance rock breakdown and alteration. The most obvious chemical process is related to the large quantities

of CO2 produced by the respiration of microorganisms and the breakdown (humification) of organic matter within soils, both of which considerably enhance the carbonation process. Certain bacteria may also produce oxygen, which can lead to the formation of inorganic acids. In particular, oxygen from bacteria is central to the nitrification process in soils in which ammonium is converted into nitrate, with the release of two hydrogen ions (protons) leading to the acidification of the soil. As well as inorganic acids, the biodegradation of organic matter produces organic acids, which react directly with and alter the mineral constituents of soils. These are typically complex mixtures of acids containing carboxyl and phenolate groups, the most common groupings being dark brown humic acids that are soluble only above pH2 and yellower fulvic acids that are soluble in water under all commonly occurring pH conditions. One of the most important roles within soils of organic acids is that they act as effective chelating agents. Chelation is the process by which metal ions, in particular, are kept in solution and prevented from precipitating within, for example, plant roots. Chelation is, however, also the process by which these cations can be complexed and removed from the surfaces of fine mineral soil particles and colloids by ion exchange, in which they are typically replaced by hydrogen ions. Such ‘cation exchange’ (see Chapter 17) is a major route by which plants obtain nutrients, but also a mechanism by which mineral soils are weathered and, importantly, progressively acidified. Organic acids are not the only available chelating agents and microorganisms, especially bacteria, secrete a range of ‘siderophores’ or ‘iron-carrying’ compounds which are equally effective in chelating a range of other metals including aluminium, copper, zinc and manganese. In a weathering context, siderophores are especially important in scavenging iron from minerals in wet environments where it can then be removed in solution. As well as their role in weathering and soil formation, there is growing interest in the role that organisms, particularly bacteria, fungi and algae, play in the surface weathering of rock (e.g. Allsopp et al., 2004). When examined microscopically, all exposed rock surfaces on Earth are found to be colonized by microorganisms. Bacteria, in particular, are able to survive extreme environments. Some of these organisms also penetrate pores and cracks and live successfully, especially where light is still able to penetrate through translucent grains and crystals. These microorganisms do not make a major contribution to landscape evolution, but they can be locally important and result in a range of micro-scale phenomena that are often diagnostic of certain environmental conditions. There are three main types of bacteria based on where they 353

Chapter 14 Weathering

obtain their nutrients. Heterotrophic bacteria obtain nutrients from organic compounds, autotrophic from inorganic compounds and mixotrophic from both. In addition to these there is an important sub-group of chemolithotrophic bacteria, which are capable of oxidizing metals from minerals with the aid of an enzyme. These have been used extensively on a commercial basis to leach metals such as copper from ores. They have also been linked to the extraction of manganese, which can then be precipitated on rock surfaces as a black crust or ‘rock varnish’ especially in arid and semi-arid environments where a soil cover is absent and there is limited surface water. A further grouping of bacteria is cyanobacteria or blue-green algae, which obtain their energy through photosynthesis. Colonial forms of these bacteria can often be found living in depressions on rock surfaces and they are commonly associated with locally concentrated weathering and erosion. On sandstones, for example, it has been suggested that photosynthesis by endolithic bacteria can result in more alkaline conditions which could be responsible for the weakening of inter granular cements through the dissolution of silica leading to grain loss. Conversely, it is also proposed that some cyanobacteria are able to mobilize manganese from superficial dust on rock surfaces, which, when precipitated, provides an alternative mechanism of rock varnish formation (Dorn, 1998). Lichens (Figure 14.4) are not an individual plant, but are interdependent communities of fungi and

oxygen-producing algae. The presence of lichens on the surface of a rock is often seen as an indicator of stability and age. However, they are also linked to chemical and physical weathering processes. Some lichens may be considered to be bioprotectors shielding the surface from the actions of other processes and others may be considered more likely to promote weathering (Lisci et al., 2003). Lichens can excrete humic and other organic acids (so-called lichen acids) that are thought to promote both hydrolysis and chelation. At the same time, lichens also produce oxalic acid which, on limestones, can produce a coating of relatively insoluble calcium oxalate that could protect the underlying surface from carbonate dissolution. In the past, builders were known to have coated limestone surfaces with a variety of organic substances, including urine and eggs, to create a surface oxalate crust or patina of ‘case-hardened’ limestone.

14.4.6 Products of chemical weathering Although the underlying chemistry of weathering is relatively well researched and understood, there is often uncertainty in how weathering has operated at a site or what might happen in the future. This is because of complex combinations of materials and varied environmental conditions. As a starting point the following section attempts to introduce a degree of order by establishing some broad categories of weathering products derived from primary rock-forming minerals. Broadly speaking, these come under four headings as shown in Figure 14.5.

14.4.6.1 Resistates Resistates are unaltered primary minerals that are effectively immune to a specific process. At any one stage during the weathering of a rock it might include minerals that survive for longer because they are more resistant to alteration than others, even though in the fullness of time and with continued weathering they would also disappear. The most commonly referred to of these minerals is quartz (Figure 14.6), which survives as recognizable crystals within the weathered residues of rocks such as granite, long after the other minerals, typically feldspars and micas, have turned to clay.

14.4.6.2 Secondary minerals Figure 14.4 Lichens growing on the surface of granite in

the Namib Desert. When they dry and lift from the surface they can ‘pluck’ away loosened grains as dust. 354

Secondary minerals are the weathering products of primary

minerals. The most common alteration is of silicate and aluminium silicate minerals into clay minerals through

14.4  Chemical weathering

Primary minerals

Chemical alteration

Resistates

Solutions

Unaltered primary minerals

Colloidal ionic

Gels

Flocculation

Aluminosilicate Fe3+ oxyhydroxide

Ageing dehydration

Ion exchange

Secondary minerals 2:1 Clays: illites, chlorites, vermiculites, smectites, interstratified clays Loss of cations, silica 1:1 Clays: kaolinites, halloysites Loss of silica Al oxyhydroxides Fe3+ oxyhydroxides

Lost from system

Figure 14.5 Processes and products of chemical weathering. (Source: from Fookes et al., 1988)

Figure 14.6 Large quartz crystals. (Source: Stefano Cavoretto

Shutterstock.com)

hydrolysis. The precise composition of the clay mineral depends on the initial mineral and the intensity and/or duration of weathering. The most complex clay minerals (and therefore those found in less intensely weathered environments) are so-called 2 : 1 lattice clays made up of two sheets of corner-sharing SiO4 tetrahedra, sandwiching a sheet of AlO4 octahedra. Different groups of 2 : 1 clays can be distinguished by the presence of, for example, interlayer potassium (illite) or the substitution of magnesium, iron, nickel and manganese within the sheets (chlorite). These additions give the clays a rigid, non-swelling structure. The sheets in other 2 : 1 groups such as the smectites

are less rigidly attached and allow water to penetrate and the minerals to swell. Continued or more intense weathering can, however, lead to the loss of one of the silica sheets (desilicification) of 2 : 1 clays, and the formation of a 1 : 1 lattice clay, the best known of which is kaolinite. Under extreme weathering conditions the second silica sheet may also be lost, to leave only aluminium or iron oxyhydroxides, such as the aluminium hydroxide, gibbsite (Al(OH)3), which is one of the main constituents in the aluminium ore bauxite. The fact that bauxite is commonly associated with highly weathered tropical soils has encouraged researchers to establish links between certain clay mineral assemblages and different climatic regimes and different intensities of weathering (Bland and Rolls, 1998). However, there are likely to be significant variations in weathering intensity related to factors such as local drainage conditions.

14.4.6.3 Solutions During the course of chemical weathering it is normal for metal cations, in particular, to be oxidized or transformed into salts and lost in solution. This is normally in the form of ionic solutions (e.g. NaCl(s) S Na(aq)+ + Cl(aq) - ), but they can also be in the form of colloidal solutions which contain extremely fine particles between 1 and 1000 nanometres in diameter (a nanometer is a thousandth of a millionth of a metre). Typically, these solutes are lost to the weathered material, although they may be precipitated 355

Chapter 14 Weathering

nearby if environmental conditions (e.g. pH, Eh, temperature or the partial pressure of a gas) change, and they are the source of most terrestrial geochemical sediments.

14.4.6.4 Gels The final category of weathering products is gels. These are solid, jelly-like substances ranging from soft and malleable to hard and tough. In a weathering context, perhaps the best studied are silica gels that can be formed by the evaporation of aqueous silicate solutions in surface and shallow subsurface waters, themselves the product of, for example, the hydrolysis of primary minerals.

14.5 Physical weathering Physical weathering is the physical breakdown of material though either the application of an external stress or the release of an accumulated internal stress, without the involvement of any chemical alteration. In this way, material can be reduced in strength and size to a point where it becomes susceptible to movement by a process of erosion. Where there is an existing relative topography, material can be lost directly through gravitational collapse (see Chapter 15). In addition, the creation of fracture networks creates pathways for moisture penetration and biological colonization that facilitate chemical weathering, as well as greatly increasing the available surface area over which chemical reactions can occur. Fracturing can be brought about through a number of different mechanisms, but all of them are ultimately linked to some form of expansion or contraction. This volume change is either of the material itself or of an alien material created within it or absorbed during the process of weathering.

14.5.1 Dilatation – pressure release The most common form of dilatation is that associated with the release of pressure when intrusive igneous rocks (see Chapter 2) such as granite are exposed through the erosion and removal of the overlying rock into which they were intruded. At this point, the stresses accumulated during cooling at high pressure are released and result in the physical expansion of the rock. At the margins of an intrusion this expansion is greatest towards the surface. Therefore, there are internal shear stresses parallel to the surface. This in turn results in characteristic curvilinear ‘sheet joints’ parallel to the curved surface of the cooling granite intrusion or batholith (Figure 14.7). 356

Figure 14.7 Curved sheet jointing generated by dilatation in

granite, Yosemite National Park, California.

Figure 14.8 Half Dome, Yosemite National Park, California, demonstrating

the structural control of curvilinear sheet joints on the shape of the dome prior to its truncation by glacial erosion that exploited vertical joint sets within the batholith. (Source: Topseller / Shutterstock.com)

Such joints are common in newly exposed intrusions and help to give the characteristic dome shape to many granite hills (Figure 14.8). They are also particularly useful in the quarrying of granite as they provide an initial set of joints that can then be exploited to break the stone into manageable blocks. The fact that they are ‘pressure release’ features is demonstrated by so-called A-Tent structures. These are formed where an isolated sheet of granite arches upwards from the surface of the

14.5  Physical weathering

Figure 14.9 An A-Tent structure on a granite batholith, South Australia. The granite slabs are larger than the hole from which they sprang demonstrating

the expansion associated with pressure release.

granite mass (Figure 14.9). Measurement of the sheets always shows them to be longer than the hole from which they came, indicating not only their original compressive loading, but also their expansion once this pressure is removed. Deeper within the intrusion the stresses are resolved into sets of joints at right angles to each other that define large-scale rectangular blocks. These joints provide the pathways along which moisture and ultimately chemical weathering can penetrate (Figure 14.2b). The density of these joints is highly variable, depending on the structure of the intrusion, and on a large scale can result in the so-called compartmentalization of the landscape as areas with a high joint density are weathered and eroded more rapidly than those with a low joint density. Such processes are often cited as central to the formation of isolated granitic hills characterized by widely spaced joint systems. As well as the development of joints and fractures spaced over metres or tens of metres, pressure release is also operates at the scale of millimetres and less. In this way it can lead to the development of dense networks of microfractures around and through individual crystals as minerals with different physical characteristics expand at different rates following pressure release. Depending on the minerals present, these fractures represent a secondary porosity along which moisture and chemical alteration can penetrate leading to the formation of a clay-rich or ‘argillaceous’ weathered regolith (Figure 14.10a). Alternatively, if the minerals present are less susceptible to chemical weathering a sandier or ‘arenaceous’ regolith can develop in which the original minerals and structures of the rock are retained, it becomes so fragmented that it

(a)

(b) Figure 14.10 Granite weathering through microfracturing: (a) weathered

granite in which all of the feldspars have been weathered by hydrolysis into white kaolinite; (b) an arenaceous regolith or grus in Montana, USA. (Source: (a) BORTEL Pavel - Pavelmidi / Shutterstock.com (b) Dr. Marli Miller, VISUALS UNLIMITED / Science Photo Library) 357

Chapter 14 Weathering

can be dug out by hand (Figure 14.10b). In granites these sandy regoliths are sometimes referred to as ‘grus’ after the term coined by quarry workers in the south-west of England, where spreads of sandy material derived from the disaggregation of arenaceous regoliths are a common feature.

14.5.2 Thermoclasty Temperature change always works in conjunction with other weathering factors. Even in the driest desert moisture plays a role in weathering. Hence understanding the role of temperature in weathering has been tricky. To get over these problems researchers have attempted to isolate thermal effects under laboratory conditions. One of the earliest of these experiments was carried out by Griggs (1936), who repeatedly heated a block of granite in front of an electric fire and allowed it to cool in a dry cellar. The upshot of the experiment was that despite heating the block many times and far beyond any temperature to be found in nature, he observed no physical change until he gave up and started to cool the block by pouring water on it. Since this experiment, laboratory tests have become much more sophisticated and now employ, for example, programmable climatic cabinets that can mimic temperature and humidity conditions from a wide range of climates. However, these experiments have had comparatively little success in inducing breakdown that is uniquely attributable to thermoclasty. In designing these experiments, researchers normally choose to study one of two potential thermal processes. The first of these is what Yatsu (1988) described as ‘thermal shock’. This is where temperature change is so rapid that the stresses generated by expansion or contraction of the rock cannot be accommodated by the required deformation. The most obvious example of thermal shock is that experienced during a fire, and there are many observations of cracked boulders after the passage of bush and forest fires. It seems reasonable to suppose that over time thermal shock could play an important role in the breakdown of debris in, for example, Mediterranean and savanna environments prone to such fires. Other than this special case, it is difficult to imagine how natural cycles of heating and cooling could generate a sufficient shock to fracture fresh rock. Recently, however, there has been interest in the potential role of short-term, near-surface reductions in rock temperature triggered by, for example, the spread of cloud cover over a surface previously exposed to bright sunlight (Smith, 2009). This is particularly the case in polar and high mountain environments where high rock 358

surface temperatures can occur at the same time as low air temperatures, resulting in a very rapid fall in surface temperature when clouds or shade appear. It is most likely, however, that the mechanical stresses associated with any such temperature change will be restricted to a very shallow, near-surface zone. Adjacent crystals of different minerals heat up and expand at different rates in response to differences of thermal characteristics such as albedo, thermal conductivity, heat storage capacity and coefficient of thermal expansion as well as in some cases the orientations of their crystal lattices along which expansion can be concentrated. This, in turn, sets up stresses at crystal boundaries that can be translated into stresses across adjacent crystals. Thermal shock effects are now considered to operate at small scales, as compared with earlier ideas about the splitting of large boulders. It is also recognized that the forces generated are most effective where the expansion is constrained in some way by surrounding crystals or rock (because they have something to push against). This constraint effect has increasing practical significance as architects use more and more stone as thin claddings on buildings and are faced with replacing bowed and cracked panels put in place with inadequate expansion joints. The alternative to thermal shock as a mechanism for rock breakdown is that of ‘fatigue failure’. This is where breakdown occurs through the repeated application of a low-magnitude stress, which on its own is insufficient to initiate a fracture, but cumulatively can weaken the rock to the point where it becomes susceptible to even a low-magnitude stress. This is in much the same way as if you were asked to snap the handle of a metal spoon: you would not attempt it in one go, but instead would bend it backwards and forwards many times until it eventually broke. Fatigue failure is seen by many as a more viable contributory mechanism to rock weathering not least because there is no need to invoke extreme and abnormal temperatures. However, fatigue failure is difficult to test in the laboratory or in nature as it is extremely difficult to isolate and test the effect. In the laboratory, for example, many weathering tests seek to accelerate the process under study, but any attempt either to speed up the heating and cooling process or increase the temperature range would invalidate any link with the fatigue effect. Any such tests would therefore have to run in real time and are thus completely impractical. In nature, meanwhile, the fact that fatigue effects only manifest themselves over very long time periods makes it difficult to untangle them from the contributions of other weathering mechanisms.

14.5  Physical weathering

14.5.3 Freeze–thaw (frost weathering) Of all weathering processes, freeze–thaw is the one that the majority of people can most easily comprehend, in part because of its clear link to an easily recognizable environmental condition (water freezing), but also because it is perceived to involve a relatively straightforward mechanism – water expanding on freezing. However, a number of researchers question some of the long-held assumptions about its effectiveness. For example, there has been an assumed relationship between freeze–thaw and the production of angular rock fragments. In reality, however, the conditions required for freeze–thaw to generate significant stresses within a stone are much more complicated and rare than a simple drop in temperature below 0°C and the freezing of water already in the pores of a stone. For example, the maximum expansion associated with the transition of water into ice occurs at - 22°C. If this point is actually reached, the pressure generated, theoretically 207 MPa, is more than enough to fracture most rocks. As temperatures drop, however, a number of factors combine to inhibit freezing and also to relieve any pressure generated by ice that has formed within pores. For example, when ice begins to form within pores the pressure on the remaining water increases. This lowers the freezing point and slows down further freezing. The presence of impurities in the water such as salts will also lower the freezing point, and in an unsaturated, open-textured stone the pressures generated by freezing and expansion might be partly absorbed by the compression of air within the pores and feasibly by displacement and possible expulsion of water via any exposed surface. For this reason the rock should ideally be at or near saturation and frozen from all sides if freeze–thaw is to be most pronounced. The process is accentuated by rapid freezing and so is considered to be most effective at or near the rock surface, leading to the preferential release of individual grains and fine rock fragments. If the expulsion of water is particularly rapid or the surrounding porosity relatively low, the pressure of the displaced water may itself lead to breakdown of the rock by a process known as ‘hydrofracture’. Ice segregation is the process whereby water is gradually attracted towards areas of ice formation by a suction gradient established in adjacent pores as the ice forms (see Chapter 24). In this way, lenses of ice are created that can theoretically continue to grow against a constraining pressure. Such phenomena have been recognized for many years in soils and result in the fracturing and heaving of these and other fine-grained, weak materials, but recent experimental studies have suggested that ice segregation

and growth might also be capable of fracturing stronger rocks, especially where pre-existing microfractures can be exploited and extended. Unlike volumetric expansion, the conditions required for ice segregation are not especially exceptional and ice can continue to grow under conditions of relatively slow freezing associated with sustained low temperatures. Experimental and field observations also suggest that disruption associated with ice lens formation could be a much more deep-seated process than volumetric expansion, especially because it can be enhanced by prolonged freezing leading to ‘slow ice separation’ at depth. This could in turn lead to fracturing at greater depths within rock formations and potentially the production and release of larger and bigger quantities of rock debris. Debris produced by freeze–thaw seems to be both fine and coarse. Microgélivation is considered to occur where ice crystallizes within pores and microfractures at the grain and crystal scale, and is responsible for the creation of fine debris. It is also considered by some to be responsible for the rounding of larger debris and rock surfaces. This contrasts with macrogélivation, which is principally the exploitation of fractures and potential lines of weakness within larger rock masses to produce coarse, angular debris or clasts (Figure 14.11).

14.5.4 Salt weathering Traditionally salt weathering has been seen as something that is important in coastal and desert environments where salt is available in abundance, but largely irrelevant outside of these areas. However, it is possibly the most important cause of rock breakdown in built environments. Building owners across the world will testify, irrespective of climate, to the effectiveness of salt, from sources such as

Figure 14.11 Frost weathering resulting in coarsely cracked granite rock,

Iceland. (Source: MyImages – Micha Shutterstock.com) 359

rising groundwater, in weathering stone. Salt weathering also occurs in polar and mountain environments that are themselves often arid. Unlike freeze–thaw, salt weathering is not constrained to the precise coming together of a limited and very specific set of environmental parameters for its operation. Salt weathering is a physical weathering process associated with the growth or expansion of salt crystals in pores and fractures. Of course, the processes driving this expansion are predominantly chemical and relate either directly to crystal growth out of solution or to the absorption of moisture (hydration) by salts that have previously crystallized within the rock. Theoretically, crystal growth alone could result in rock breakdown as, provided that the solution remains supersaturated and a thin liquid film is maintained at the boundary between the crystal tip and the adjoining pore or crack wall, crystals will continue to grow against any confining pressure. The stresses generated within an individual pore are unlikely to be sufficient to cause failure as they act on too small an area (Smith and McGreevy, 1988). Instead, it is the ability of salts to spread and accumulate throughout a zone within the rock that allows them to generate sufficient stress to cause failure by interacting with larger structural flaws that ultimately control rock strength. The exception to this might be at the rock surface, where individual grains and crystals are only partially confined. The effectiveness of salt crystallization in causing breakdown is also closely linked to the nature of the pores and it has been suggested that the controlling factor is the presence of micropores (less than 5 millionths of a metre in diameter) that have the suction potential to absorb moisture and are readily bridged by salt crystals. Such a process is more difficult to accomplish in larger pores. The growth of clearly defined, pore-bridging crystals may not be essential for salt weathering as hydration could be favoured where pores become completely filled with microcrystalline salt. The same might be said of differential thermal expansion as a possible weathering mechanism, where salts are present with coefficients of thermal expansion greater than that of the enclosing rock (Figure 14.12). For further details on salt weathering processes and hazards, see Goudie and Viles (1997). As with other processes linked to physical breakdown, the importance of salt weathering lies not so much in the shock effect of one-off crystal growth, but in the fatigue effect associated with regular and frequent volume changes linked to repeated solution and recrystallization, hydration and dehydration, and possibly differential thermal expansion and contraction. These changes are largely controlled by alternating heating and cooling and wetting and drying 360

% Volumetric expansion from 205C to indicated temperature

Chapter 14 Weathering

NaCl 5

KCl

4 NaNO3 3 BaSO4 2

Granite 1 CaCO3 (Calcite) 0

0

100

200

300

400

Temperature (5C) Figure 14.12 The thermal expansion of salts linked to salt weathering,

compared with that of granite. (Source: from Cooke and Smalley, 1968)

that may in turn be linked to natural diurnal cycles, but may also be event driven in relation to individual storms. It is in this context that hydration and dehydration take on special significance, in that, depending on the salt, these processes can occur when specific combinations of temperature and relative humidity are crossed without the presence of liquid water. For some salts these thresholds could potentially be crossed several times during the course of a day as temperature and relative humidity rise and fall. The hydration pressures generated can be considerable, and theoretically could exceed the tensile strengths of many common rocks. The relative significance of crystallization, hydration and differential thermal expansion is still unclear but it is most likely that all three mechanisms work together. It should be stressed, however, that not all salts hydrate or dehydrate under commonly experienced environmental conditions (e.g. sodium chloride), and many commonly occurring salts (e.g. calcium sulfate) can have a low solubility that makes it difficult for them to penetrate far into the rock. This is where having a mix of salts can be important. The presence of sodium chloride is known to increase the solubility of calcium sulfate, and together they are considerably more effective in causing decay. A mix of salts could also increase the number of hydration and dehydration cycles that occur within a rock. Some salts such as sodium chloride can also absorb moisture from the atmosphere (deliquesce) and promote the operation of other

14.5  Physical weathering

chemical weathering processes. Salinity is also known to enhance the solubility of certain elements, most notably silica. While this is unlikely to play an important role in dissolving the rock mass as a whole, it could be significant in some sandstones where crystalline quartz grains (with a generally very low solubility) are held together by amorphous silica cements with a somewhat higher solubility. In this situation there may be preferential removal of the cement enhancing the susceptibility of the stone to more classic salt weathering, as well as other mechanical weathering processes. For urban stone decay, one of the diagnostic traits of salt weathering is so-called sanding, whereby running one’s hand over, say, a sandstone block removes a layer of loose material, which would otherwise collect as the small piles of sand that are often seen along the bottoms of walls (Figure 14.13a). Larger grains may form from salt weathering of other rock types (e.g. granite shown in Figure 14.13b). This process occurring on stone is referred to as granular disaggregation. It is generally agreed that this disaggregation is associated with a near-surface or surface accumulation of salt (efflorescence) and both laboratory experimentation and field observations from innumerable buildings have linked this to the slow evaporation of salts brought to the surface in solution, a process that is enhanced if the salt itself has a high solubility. The complementary process to surface salt accumulation is thought to occur when rock surfaces are rapidly heated, especially in conjunction with a surface air flow. Under these conditions, rapid evaporation dries out the immediate subsurface zone more rapidly than moisture can be drawn out from the interior. In this way the water connectivity with the surface is broken and the only way that moisture can be lost is as vapour, causing salt to crystallize at depth. This concentration is clearly enhanced if the salt itself has a relatively low solubility. Repeated many times, this process allows salt to accumulate in a subsurface zone, close to a frequent wetting depth, and eventually results in contour scaling, in which a complete surface layer up to several centimetres in thickness can first blister before falling away (Figure 14.14). This can sometimes reveal a subsurface accumulation of crystallized salt. Box 14.2 discusses an alternative mechanism for salt weathering. Granular disaggregation and contour scaling cannot be claimed as uniquely salt weathering in origin, being the possible end points of a range of linked processes, including thermal expansion and contraction that can establish stresses either between grains or between the rock surface and its interior. However, a more definitive indicator of the operation of salt weathering is the

(a)

(b) Figure 14.13 Granular disintegration: (a) small pile of sand accumulating

at the bottom of a wall near a road in an urban area after disintegration of the stones above; (b) granite, arid southern Namibia.

Figure 14.14 Surface scaling of granite, arid southern Namibia.

361

Chapter 14 Weathering

SALT WEATHERING BY ION DIFFUSION Recently attention has been drawn to ion diffusion as an alternative mode of salt weathering (e.g. McCabe et al., 2013). In this process salts can migrate through stone to concentrate in certain areas. This is where the individual ions (e.g. Na + and Cl - ) migrate through a saturated rock

from a zone of high concentration to one with a low concentration without any flow of the solution itself. The process could be important in areas where rock is saturated for long periods by rising groundwater, or possibly during long periods of wet-season saturation that could penetrate deep into a rock. One specialist area of weathering where this process is especially important is that of concrete, where chloride

diffusion and subsequent corrosion of iron reinforcing rods is a major source of structural failure as the iron expands. Therefore new ways of designing concrete have to be developed including making layers of concrete with different mixtures for the same building structure to ensure that chloride diffusion cannot simply occur from the outside inwards towards iron reinforcements (Zeng and Song, 2013).

BOX 14.2

occurrence of cavernous weathering at a variety of scales indicative of the tendency for salt weathering to create hollows protected from rainwash, in which salts are then preferentially retained. The hollows also establish their own microclimates that are typically cooler and moister than adjacent exposed surfaces, which in turn enhance the direct precipitation of moisture, its retention and the deeper absorption of salt in solution. They thus create a positive feedback that promotes further weathering and the growth of the cavern. Two scales of hollow are generally recognized: honeycombs of the order of a few millimetres to a few centimetres wide and deep (Figure 14.15); and tafoni that can be measured in tens of centimetres to metres (Figure 14.16). The location of these hollows may reflect an initial, localized variation in weathering susceptibility across a rock surface, perhaps related to porosity, grain size, mineralogy or degree of cementation (Turkington, 1998). Sometimes these structural controls are easily recognized as lines of hollows that pick out a particular horizon within a rock, but in other cases the controls may be very subtle and the distribution may appear to be random. As honeycombs grow and begin to intersect each other to cover a surface they can appear to take on a strange degree of uniformity. In other situations it is possible to identify a degree of environmental control so that in hot desert environments, for example, basal tafoni can be found in cliff-foot locations where groundwater seepage is concentrated.

362

Figure 14.15 Honeycomb weathering of a carbonate cemented quartz

sandstone in a salt-rich coastal environment, Yehliu, north-east Taiwan. Weathering of the less consolidated sand beneath the carbonate-rich nodule has produced a ‘pedestal rock’, which is also a common feature of salt weathering in many rocks.

14.5  Physical weathering

Figure 14.16 Cavernous hollows or tafoni in sandstone, in Writing-On-Stone Provincial Park, Alberta, Canada. (Source: National Geographic Creative /

Alamy Stock Photo)

Cavernous hollows of all types are particularly common in coastal and desert locations but their widespread occurrence in urban environments suggests that it is the presence of salt, rather than any specific environmental control, that is the key to their formation. It should also be remembered that salt weathering is not the only process that favours warm, humid and sheltered environments and the formation of hollows is also associated with various forms of biological weathering in which, for example, bacterial and algal communities create their own microenvironments in otherwise harsh locations.

14.5.5 Biologically related physical weathering The most visual link between biological processes and the physical breakdown of rocks is when the growth of tree roots is seen to split and prise apart rocks (Figure 14.17). As in Figure 14.17, the penetration of the root is greatly aided by the pre-existing presence of a network of joints and a degree of prior granular disaggregation. At a smaller scale, it has been suggested that processes such as grain release can be facilitated through the generation of tensile stresses as fungi grow along micro-cracks. It is also proposed that the expansion of both fungi and lichens as they rehydrate after a dry period can exert significant stresses through exertion of ‘turgor’ pressure (the pressure of water against cell walls). Some authors have suggested that the volume expansion of some species could be close to 4000% on rehydration.

Figure 14.17 The growth of a tree root within a sandstone outcrop,

enlarging the joints, Valaam Island, Russia. (Source: Denis Iamshchikov) Shutterstock.com

363

Chapter 14 Weathering

Reflective questions ➤ Why is water important for weathering? ➤ What are the key chemical weathering processes? ➤ What are the key physical weathering processes?

14.6 Climatic controls on weathering Most chemical reaction rates double with every 10°C rise in temperature and so the mean annual temperature of a region will be an important control on weathering rates. Of course such reactions are also likely to be influenced by the availability of water and the nature of any biological activity, linked in turn to other factors such as available CO2. However, for processes such as freeze–thaw the links with mean annual temperature become much weaker and there are many other more meaningful environmental parameters that control its operation and effectiveness. These include absolute temperature below zero and the rate of freezing in relation to the number and duration of frost events and the depth of frost penetration. Thus it is possible to envisage constraints on freeze–thaw activity controlled not just by conditions that are too warm to freeze, but also too cold to thaw and too dry for ice to form. For moisture it is not necessarily the absolute amount of rainfall that is significant but the balance between rainfall and potential evapotranspiration which provides a better measure of moisture availability. It is also the case that rainfall distribution can exert a considerable influence on patterns of weathering. In savanna environments, for example, alternate wetting and drying linked to the rise and fall of the water table might favour cycles of leaching and precipitation of certain minerals within soils (especially iron) and ultimately the formation of duricrusts such as ferricretes or laterites. Continually wet conditions under rainforests, together with the rapid recycling of organic matter, could favour continuous leaching of soils and weathered regoliths (see Chapter 17). Such assumptions apply specifically to patterns of deep wetting and associated deep weathering. In terms of weathering closer to a rock surface, weathering is much more likely to be related to the characteristics (duration, frequency, intensity and total amount) of individual rainfall events linked to patterns of evaporation immediately following them. There is no minimum annual rainfall below which weathering does 364

not occur. This is because many weathering processes, such as salt weathering and microbial activity, require very little water for their operation and they can be adept at obtaining this water directly from the atmosphere or from groundwater. In his famous diagram, Strakhov (1967) attempted a broad linkage between climate, weathering depth and the nature of weathering products (see Figure 17.11 in Chapter 17). As a generalization this is undoubtedly justified, but it masks a raft of other factors. The underlying assumption is that increased intensity of weathering is associated with high temperatures and abundant all-yearround rainfall. However, global climates do not conform to a perfect zonal model, and it is common, even within the tropics, to find a range of climatic types at any one latitude, including climates associated with, for example, mountain ranges (Chapter 9). Within each of these zones it is also the case that weathering processes and depths are highly dependent on rock type. Within the humid tropics this most probably means ancient, basement rocks in stable shield areas that are likely to be granitic in character and have a propensity for deep weathering because of their combined physical and mineralogical characteristics. Under these circumstances it might also be the case that profiles are deep, not only because of the intensity of processes such as hydrolysis, but because the profiles are old and the rate of surface erosion under a protective forest cover is relatively slow. That is, we can substitute stability and longevity for weathering intensity. There are many areas of the humid tropics in steep, tectonically active, weathering-resistant rocks that have relatively thin soils. Conversely, it is also possible to find deep weathering profiles outside of the tropics. In many cases this reflects a long-term climatic inheritance. For example, pre-Quaternary deep weathering profiles have been found in areas such as Scotland and Scandinavia and this has questioned the power of cold-based ice sheets in particular to reshape the landscape (Migoń and Bergstrom, 2001; Olvmo et al., 2005). Like Strakhov (1967), others have attempted the same climate zoning exercise using the occurrence of specific clay minerals. This is based on the assumption that weathering intensity increases with increased temperature and rainfall and that increased weathering intensity is in turn reflected in the progressive simplification of clay minerals from 2 : 1 lattice clays, to 1 : 1 clays, to iron and aluminium oxides and hydroxides. While such sequences do occur it is also true that specific weathering processes (e.g. the carbonation of orthoclase) are capable of bypassing this sequence to directly produce clays such as kaolinite. Likewise, recent

14.6  Climatic controls on weathering

Percentage clay in soil

100

Gibbsite (bauxitic minerals) Kaolinite Smectite 0

0

2000

4500

Annual rainfall (mm) Figure 14.18 Influence of climate on clay mineralogy under a perennially wet climate in Hawaii in which high rainfall and rapid leaching cause the prefer-

ential removal of silica and metal cations. (Source: after Sherman, 1952)

studies from countries such as Thailand (Hermann et al., 2007) have suggested that gibbsite, which is usually thought of as the end product of weathering under wet tropical conditions, can appear as an early and direct transformation from micas and feldspars in rocks such as granite and gneiss without any intermediates. The key variable in this case appears to be freely draining soil conditions and high leaching rates, a control previously emphasized in a

DEEP WEATHERING IN RIO Despite the widespread assumption that the humid tropics are underlain by a uniformly thick blanket of deeply rotted rock reduced mainly to simple clay minerals, research has emphasized that there is considerable local variability. A study of deep weathering around Rio de Janeiro by Power and Smith (1994) showed that soils on well-drained crests, subject to enhanced cation and silicate removal, were dominated by gibbsite and kaolinite, whereas complex clays such as montmorillonite were more common in poorly drained foot slopes and valley floor locations. A study

study of basalt weathering from Hawaii (Figure 14.18). Sherman’s (1952) study on Hawaii is frequently used to demonstrate climatic controls on weathering and clay formation, but given that it is a study of one small island it could more accurately be thought of as a demonstration of variability within one broad climatic zone based upon local conditions. Such variability has also been demonstrated in many other studies from the tropics (Box 14.3).

in the same area by Smith and Sanchez (1992) also demonstrated how, under the same climatic conditions and with the same climatic history, subtle variations in mineralogy could drastically affect weathering response. Granitic rocks containing predominantly potassium (orthoclase feldspars) weathered to produce a deep, but predominantly sandy, arenaceous regolith. Adjacent rocks rich in calcium and sodium feldspars (plagioclase), which are more prone to chemical alteration, produced equally deep, but clay-rich, argillaceous regoliths (Figure 14.19). One important consequence of this in terms of potential slope hazard is that

slopes underlain by the clay-rich regoliths are much more likely to fail catastrophically through mudflows and slides (Figure 14.20), whilst the arenaceous profiles are more likely to experience erosion by surface wash. Active erosion also means that, particularly on steep slopes, deep weathering profiles are replaced either by thin soils that are prone to ‘slip’ off the underlying bedrock after prolonged rainfall, or by large boulders held in place by surrounding finer debris (Figure 14.19). These boulders are themselves prone to rolling downslope, with great destructive potential, if the supporting material is washed away.

BOX 14.3 ➤

365

Chapter 14 Weathering

➤ (b)

(a)

Forest/scrub Shallow organic horizon Yellow bleached horizon Red, clay-rich horizon, kaolinite clays formed in situ

Ú30 m

Deeply weathered bedrock, high clay content, most feldspars altered to kaolinite or gibbsite. Quartz remains as angular, coarse sand grains

Deep, clay-rich weathering profiles with red-yellow podzolic soils underlie much of the countryside in the north of the area

Forest/scrub Shallow organic horizon Yellow bleached horizon Red, clay-rich horizon, kaolinite clays formed in situ

Ú30 m

Occasional corestones

Deeply weathered bedrock, very high sand content of quartz and unweathered feldspars, clays often 610% mainly kaolinite. Still has appearance of bedrock but can be dug out Corestones rare

Weathering front Crystalline bedrock

Weathering front Crystalline bedrock

Deep, sand-rich profiles with less well-developed red-yellow podzolic soils. Common on structurally deformed augen gneiss in south-west of area

Figure 14.19 Schematic diagrams of (a) clay-rich and (b) sandy deep weathering profiles on granitic rocks in the Rio de Janeiro region of Brazil.

Sandy profiles tend to form on rocks comparatively rich in potassium feldspars.

Figure 14.20 Slope failure at Niteroi, Brazil in May 2010 resulting from the saturation of deeply weathered,

clay-rich weathered granite regolith.

BOX 14.3

It is important that global models of climatic controls on weathering are recognized for what they are, namely generalizations, and only the broadest of starting points to any understanding of weathering. Thus, while they indicate that the occurrence of certain processes may be favoured in particular climatic zones this is no guarantee 366

of their exclusivity or their universal effectiveness. The danger comes if one seeks to employ the relationships they imply to the interpretation of weathering and weathering products at the landform scale. At this scale it is essential that attention is focused appropriately on the scale at which the processes in question operate.

14.7  Geological controls on weathering

Reflective questions ➤ What are the key climatic controls on weathering? ➤ Why are global generalizations of weathering depth or clay mineral form in relation to climate zone problematic?

14.7 Geological controls on weathering The composition and structure of the rock, combined with its age and exposure to weathering are important geological controls on weathering. Researchers have produced lists of different rocks and their relative susceptibility to weathering. The most widely quoted of these is based on the Bowen reaction series (Figure 14.21a), created in the early twentieth century, which ranked common minerals found in igneous rocks in terms of their temperature of

formation. This ranking was later picked up by Goldich (1938) who, on the basis that the Earth’s surface is a lowtemperature environment, used the listing as an indication of relative weathering susceptibility under near-surface conditions, with the most stable and therefore resistant at the bottom of the list (Figure 14.21b). In reality, patterns and rates of decay will vary considerably in response to localized variations in the weathering environment and there is no guarantee that the stability sequence could not be reversed under certain circumstances. The techniques described in Box 14.4 can assess susceptibility to weathering rather than providing any insight to the actual nature of rock breakdown. The physical breakdown of rocks tends to result from the application of three main stress types applied either generally across a rock mass or at specific points right down to the microscale. The first of these is compressive stress, in which an axial ‘pushing force’ is applied to a material that crushes it once the compressive strength of the material is exceeded.

Early crystallization Ferromagnesian minerals

Non-ferromagnesian minerals (feldspar) Least stable Calcium-rich

Olivene

Pyroxene

High temperature (14005C)

Sodium-rich

Orthoclase (potassium feldspar)

(a)

Calcium plagioclase

Augite

Mixed plagioclase

Hornblende

Sodium plagioclase

Plagioclase

Amphibole

Biotite mica

Olivene

Biotite

Muscovite mica

Potassium feldspar (orthoclase)

Quartz

Muscovite

Low temperature (8005C)

Quartz

Late crystallization

Most stable

(b)

Figure 14.21 (a) The Bowen reaction series of igneous minerals and (b) the weathering series of Goldich (1938).

367

Chapter 14 Weathering

MEASURING ROCK HARDNESS There are a variety of forces to which materials might be subject, to which the same material will have a varying degree of resistance. Included in these are scratch hardness, indentation hardness and rebound hardness. ‘Moh’s hardness scale’ (scratch hardness), is used by geologists as an aid to mineral identification (Table 14.1). It tests the ability of one mineral to scratch another mineral. It does not provide any great insight into weathering in itself, although it might give an indication of abrasion resistance during erosion. Rebound hardness has received some attention in recent years thanks to the use of the ‘Schmidt hammer’ (Figure 14.22). This was initially developed to test the hardness of concrete, and fires a spring-loaded metal bolt at the surface to measure its rebound. While of debatable use on rough and weathered surfaces, where it tends to compress the material rather than rebound, it has had some success in, for example, discriminating between rocks that are more or less susceptible to different types of karst formation (Day, 1981). Recently, the development of similar looking,

Table 14.1 Moh’s scratch hardness scale

Scratch hardness

Example mineral

1

Talc

2

Gypsum

3

Calcite

4

Fluorite

5

Apatite

Figure 14.22 A Schmidt hammer in use. The hammer punches the rock

6

Orthoclase feldspar

with a metal core and the rebound distance of the core can be measured.

7

Quartz

8

Topaz

9

Corundum

10

Diamond

but more sensitive rebound instruments, such as the ‘equotip portable hardness tester’ has offered greater potential for the characterization of weathered material in a non-destructive manner (see Viles et al., 2010).

BOX 14.4

368

14.7  Geological controls on weathering

This contrasts with tensile stress, in which an axial ‘pulling force’ is applied, which results in the distortion of the material, followed by its rupture or fracture once the tensile strength is exceeded. Between these two are shear stresses, in which offset forces induce a sliding failure parallel to the direction of the force once shear strength is exceeded. Resistance to these stresses is by no means uniform and the same material will invariably respond differently and more or less readily to the different stress types. The nature of this response can be shown in the form of a stress–strain curve in which the applied stress is plotted against the distortion of the material. Figure 14.23 shows a representative stress–strain curve for material under tensile stress, in which strain is expressed as the ratio of the increase in length of the sample compared to its original length. From the curve it can be seen that there is an initially linear increase in strain, during which the material behaves elastically, and will return to its original dimensions once the stress is removed. The gradient of this straight line is known as the Young’s modulus for the material. Following this, the material may enter a phase of plastic deformation, having surpassed its elastic limit, which is characterized by non-reversible shape change. If the stress continues, eventually a point is reached at which the material fractures or ruptures. This final failure could be an eventual response to the gradual development of networks of micro fractures developed during repeated plastic deformation, and it is this process that lies at the heart of fatigue failure. Obviously, the precise nature of this curve will differ greatly between materials. Some materials

Brittle fracture when deformation is still elastic

o def tic s a Pl

tion rma

Ultimate failure point

Ela

sti c

de

fo rm a

tio

n

Stress

Elastic limit

Tensile stress

Strain L = original length

Gradient (rate of change) = Young’s modulus L Material under stress

l Tensile stress

l/L l = increase in length

Figure 14.23 Idealized stress–strain curve.

are, for example, more ‘ductile’ than others and have a greater ability to deform under an applied stress, although this property is in turn affected by other factors such as temperature and a material typically becomes more ductile when heated. In contrast, other materials (e.g. glass) may be brittle in character and have little if any capacity for deformation, and instead will tend to fracture under the application of a well-directed stress. Therein, however, lie a number of contradictions, in that brittle materials can often be considered as hard and are particularly resistant to, for example, abrasion forces. Likewise there is often confusion over the difference between weak and strong, in that it is often perceived that materials that bend are weak, and that those that do not are strong. Many materials that experience significant elastic and plastic deformation are in fact capable of surviving stresses that would fracture more brittle materials. In the context of weathering, one further complication is that the nature of materials can be altered by their surroundings. Thus, when a rock in the form of a pebble is unconstrained it can expand and contract elastically in response to, say, heating and cooling and absorb the stresses generated. In contrast, the same material, when constrained within a larger rock mass, cannot absorb stress through expansion and will behave inelastically, making it more susceptible to the development of, for example, micro fractures and ultimately brittle fracture. The chemical and physical properties described above can be broadly thought of as lithological in character, relating as they do to intrinsic properties within the material. As we will see in the following section dealing with stone in the urban environment, however, by the time that most geomaterials are subject to weathering they will have experienced complex stress and deformation histories. A common result is the compartmentalization of the rock mass into structural units ranging in dimension from sub millimetres to spacings measured in kilometres. These are defined by joints and fractures, some of which are ‘real’ and others that are ‘incipient’ and only revealed when exploited by subsequent weathering. As indicated earlier in this chapter, the three-dimensional distribution of these structural properties exerts major controls on both the operation of weathering processes and the morphology of subsequent landforms. Nowhere is this combination of lithological and structural control better illustrated than in the case of the hard, non-porous, well-jointed limestones that form the distinctive karst topographies of the world (Box 14.5).

369

Chapter 14 Weathering

KARST LANDFORMS Where rocks are highly soluble, so that most of their mass can be removed in solution, a set of distinctive, or ‘karst’, landforms may be formed. Karst occurs most commonly in limestone, but may also occur in other soluble rocks, such as gypsum. Some comparable forms can also be generated through melting of ice to form ‘thermokarst’. Where the rocks are strong, water flow and solution is concentrated along joints and other porous areas, enlarging them in a complex network of passages, many of them underground, although generally with some connections with the surface (Figure 14.24). These passages may evolve over very long periods, and some become enlarged into caves some of which can be hundreds of metres in extent and form networks over many kilometres. Joint passages evolve most rapidly where there is most water flow, so that they tend to be formed mainly above the water table, although this can alter over time through climate or base-level change. This also means that caves frequently develop along the lines of stream valleys, because the streams have the largest and most continuous water flow. Streams, in this way, can progressively bring about their own demise, as more and more of the water sinks into the bed and follows underground passages, which may only emerge at the boundaries of the soluble rock mass. Other factors that increase the rate of solution are the presence of soil organic matter and cold temperatures, because CO2 is more soluble in the cold. The combination of these factors with the flow means that karst development is most rapid in humid tropical areas, and slowest in hot deserts.

Figure 14.24 Karst cavern system, Planina cave, Postojna, Slovenia. Inside

this large karst Cave is a confluence of two underground rivers, Pivka and Rak Rivers. (Source: Ralf Siemieniec Shutterstock.com)

Figure 14.25 Towering karst pinnacles covered in trees, rural Guilin,

Guangxi, China. (Source: Sean Pavone Shutterstock.com)

Rapid karst development lowers the hills until they intersect the water table, so that they eventually form towers rising out of an almost level plain (Figure 14.25). In temperate areas, karst usually develops more slowly, so that cave systems are formed beneath a surface topography of hills and valleys. Where the landscape has

BOX 14.5 ➤

370

14.8  Urban stone decay and lessons for rock weathering

➤ been glaciated, as in northern Britain, the rock is, in many places, initially smoothed by glacial abrasion. The action of solution and joint enlargement only gradually destroys the initial smooth surface, leaving a very characteristic surface of clints and grikes (Figure 14.26). In unglaciated landscapes, no visible trace of the original surface is left, and a landscape of pinnacles, bounded by joint planes, is equally characteristic (Figure 14.25).

Figure 14.26 Clints (higher rock) and grikes (gaps between) formed in

limestone in the Yorkshire Dales, northern England.

BOX 14.5

Reflective questions ➤ What are the main geological controls on weathering? ➤ Under what main weathering controls do karst landforms dominate?

14.8 Urban stone decay and lessons for rock weathering Placing stone in a building does not immunize it from natural weathering processes (Smith et al., 2008). Outside of the dissolution of limestone by acid rain, the principal cause of urban stone decay is salt weathering. Salts can derive either directly in the form of pollution particles deposited on stone, or indirectly through reactions between different forms of acid deposition and stones and mortars. This applies especially to limestones in sulfur-rich atmospheres and leads to the formation of calcium sulfate in the form of gypsum. In areas of elevated atmospheric pollution, and where this reaction takes place faster than the gypsum can be dissolved and washed away by rain, this can result in a uniform crust across

complete facades, turned black by the inclusion of soot and other particles (Figure 14.27). Where levels of pollution are lower or rainfall amounts higher, such black gypsum crusts are largely restricted to areas sheltered from rainwash, where they form primarily as a result of dry gaseous and occult deposition. While it is the presence of such crusts that drives many owners to expensively clean their buildings, the potential longer-term significance lies in the possibility that the gypsum can be washed into the stone, sometimes during the course of cleaning. There it can join other salts (derived from atmospheric pollution, rising groundwater, marine aerosols, road salt and other sources) to cause breakdown through, for example, repeated crystallization and dissolution, hydration and dehydation and differential thermal expansion. This can result in many of the patterns of decay, such as scaling, flaking, granular disaggregation and honeycombing, that we have come to associate with salt weathering of natural rock outcrops (Figure 14.28). Because of the financial costs involved in construction, repair and replacement of stone in buildings there has been considerable research into the causes of urban stone decay in recent years, but those involved face the same difficulties as rock weathering researchers in terms of convincing building owners of the differences between symptoms and causes, explaining the underlying

371

Chapter 14 Weathering

Figure 14.27 Black encrusted façade on an Asian temple. The lower and central part has been cleaned. Note that the cleaning may not stop further

decay and may in many cases increase stone decay rates. (Source: Asian Images/Shutterstock.com)

(a)

(b)

Figure 14.28 Honeycombing produced by salt weathering of sandstone used as: (a) building stone for Lindisfarne Priory and (b) a gravestone at Tyne-

mouth Priory, both on the north-east coast of England. 372

14.8  Urban stone decay and lessons for rock weathering

complexities of the processes operating and demonstrating the importance of accurate diagnosis for effective conservation. Such a communication process is made even more difficult by the widespread public belief that stone buildings should last forever, and that if they do not it is because of some human failing rather than a natural process of change. What follows therefore are some suggestions as to what building owners and architects might usefully understand about stone decay.

14.8.1 Stone decay is multifactorial Urban stone decay derives from the superimposition of anthropogenic factors on top of those associated with natural change. There is a wider variety of salt types found within urban environments that can exploit an increased range of environmental thresholds in terms of processes such as hydration and dehydration. This also includes the consequences of the construction process itself, in which, for example, constraining stone within hard impermeable mortars (especially Portland cement) increases the likelihood of fracture development as the stone attempts to expand against this restraint in response to heating and cooling. At the same time, forcing water to drain through the stone rather than through the surrounding mortar can promote salt weathering that further exploits the micro-fracture network. The end result of this is a phenomenon known as boxworking in which stones disappear to leave only the hard mortar as an empty framework (Figure 14.29).

Figure 14.29 Boxworking resulting from the rapid weathering and erosion

of soft oolitic limestone constrained within a hard cement mortar, Oxford, England.

14.8.2 Rates of stone decay are unpredictable Research into rates of urban stone decay, which go back to the nineteenth century and the work of the famous Scottish geologist Archibald Gieke on the erosion of dated gravestones, has typically envisaged a gradual rate of change. This is partly because of an emphasis on gradual loss of material in solution, but also because of a desire to justify the extrapolation of long-term rates of decay from short-term observations. There are, however, numerous problems with this approach. Environmental conditions are never stable, especially in urban environments where patterns, types and levels of pollution are known to have varied greatly in historical and recent times. It is also the case that many building stones, such as quartz sandstones, are prone to decay in a very non-linear fashion. Typically they can show no surface evidence of decay for many years, especially if this is also associated with case hardening by, for example, near-surface iron cementation. Then they can lose a complete outer layer almost instantaneously through the falling away of a contour scale. If the newly exposed subsurface layer has been sufficiently weakened, perhaps by the outward migration of iron cements, and if enough salt has penetrated deeply into the stone, a set of positive feedbacks could be initiated that lead to rapid decay and complete loss of the stone block. Such a process is possibly encouraged by the fact that as the stone retreats it creates in front of it a humid, shaded environment in which more salt can accumulate. Conversely, if there is only a shallow subsurface zone of weakening and limited salt penetration below the scale, its loss may be followed by only limited granular disaggregation before a negative feedback is initiated and the newly exposed surface stabilizes. Stabilization can also be brought about if, for example, a black gypsum crust is allowed to develop following scaling. In fact, recent studies of clastic limestones in the still highly polluted city of Budapest and the once heavily polluted city of Oxford have identified multiple crust formation, where stone has gone through many cycles of scaling and stabilization. Interestingly, there is some evidence that a decrease in atmospheric sulfur and particulates (which help to catalyze the rapid formation of gypsum) resulting from clean air legislation could retard this ‘scabbing over’ and might accelerate the decay of some intrinsically weak stones in the longer term (Smith et al., 2010a). This process of episodic decay is illustrated in Figure 14.30, which also highlights how effects such as scaling and rapid surface loss may also be triggered by human interference such as aggressive surface cleaning. What this also illustrates is the danger and uncertainties associated

373

Chapter 14 Weathering

Threshold 1

Threshold 2

Depth of stone

Scaling triggered by extreme event, e.g. severe frost Rapid

Salt accumulation and biological growth

Breaching of iron-rich crust

disaggregation of exposed surface Formation of iron-rich outer skin

Alternative 1 Hollow develops at site of scale, salts accumulate leading to rapid disaggregation, multiple flaking and tafoni development

Threshold 3

Threshold point

Stone cleaning, removal of black crust and surface layer

Negative feedbacks operate

Rapid disaggregation of iron-depleted sublayer Growth of gypsum black crust-salt accumulation

Alternative 2 Loss of cement and weakened zone below ironrich skin promotes tafoni development as in 1

Positive feedbacks operate

Rapid disaggregation of weakened sub–crust zone

Contour scaling Regeneration of black crust

Alternative 3 Exposure of leached zone rich in salts promotes tafoni development as in 1

Rapid disaggregation and multiple flaking of weakened stone leading to catastrophic decay

Time Figure 14.30 Schematic diagram of potential decay sequence of a quartz sandstone used in construction in a polluted environment. (Source: from

Smith, 2003)

with any attempt to precisely predict the long-term performance of stone, especially based on one-off observations or only short-term monitoring

14.8.3 Decay is spatially variable Casual observation of almost any ancient stone building or structure will reveal stones and groups of stones that have experienced markedly different patterns and rates of weathering and erosion (Figure 14.31). Given that individual walls are likely to be constructed of stone from the same source that was quarried at the same time, the seemingly random distribution of decay shows how subtle differences in mineralogy and structure can determine susceptibility to different processes and/or relative resistance to the same one. Alternatively, on some walls there does appear to be a degree of spatial organization of decay. This could possibly be a response to environmental factors, of which the most obvious is decay near to ground level in response to salts derived from groundwater. One way of examining this spatial dimension is to study the degree of connectivity exhibited by particular types of decay across a wall. This can be done by assigning a principal decay type to each block, and counting the number of sides that are adjacent to blocks demonstrating the same type of decay. When this was done for quartz sandstones in a polluted area of Belfast, this showed that stones experiencing scaling and rapid retreat typically had a low level of connectivity, mainly zeros and ones (Smith 374

and Viles, 2006). This could suggest that the factors controlling susceptibility to this type of mechanical decay are mainly related to differences in the physical characteristics of individual blocks. In contrast, soiling by black gypsum and biological crusts were associated with a higher average score for connectivity, suggesting that their formation is controlled to a greater extent by localized environmental factors that override material differences between blocks.

Figure 14.31 Weathering of quartz sandstone at Bamburgh Castle, north-

east England illustrating the variable response of the same stone type based upon subtle differences in porosity, bedding, cementation and other physical and chemical characteristics.

14.9  Summary

14.8.4 Stress history is important Following the successful implementation of clean air legislation across much of western Europe, many building owners believed that it was safe to clean buildings without the fear that they would rapidly resoil and safe in the knowledge that stone decay would effectively cease. It came as something of a surprise that despite much cleaner air many buildings continued to decay. In general this has been put down to a ‘memory effect’, but it is also a tacit recognition that natural rock weathering processes also have a role to play in urban stone decay. During the course of its lifetime, building stone, or any natural rock or rock outcrop for that matter, will experience a range of environmental conditions and be subject to a variety of forces. For building stone this could include a number of pre-emplacement effects such as dilatation and associated micro fracturing during quarrying, as well as cutting and shaping before it is placed in a building. Post-emplacement, it could be subject to many years of environmental cycling and physical stressing through heating and cooling, wetting and drying, and freezing and thawing as well as potential chemical alteration and loading with a range of pollutants, including complex salts, under changing environmental conditions. These factors could enhance its susceptibility to future decay irrespective of whether atmospheric pollutants continue to be available. The importance of these ‘complex histories’ has been demonstrated in a recent study of coastal medieval sandstone churches in Ireland and Scotland. Over their lifetime these buildings have been subject to many changes, including the application and removal of plaster renders

14.9 Summary

that have loaded the stone with lime, often multiple fires and a period of enhanced freeze–thaw activity associated with the Little Ice Age of the sixteenth to nineteenth centuries, which itself followed the Medieval Warm Period (see Chapter 5). These changes are in addition to previous campaigns of conservation including repointing. In an experiment to understand the effects of these different factors, samples of fresh sandstone were given complex stress histories by subjecting them in the laboratory to different combinations of freeze–thaw cycles, addition and removal of lime plaster and heating in a wood fire, after which they were artificially salt weathered to replicate the ongoing stresses associated with their coastal location (McCabe et al., 2007). The upshot of this was that different combinations produced very different patterns of weight loss during the course of the salt weathering simulation, perhaps the most interesting being the pattern of loss associated with stones that had been heated by fire. In comparison with other stones, weight loss was initially slow, perhaps because the heating made the stones water repellant for a period, but at a later stage the stones started to rapidly fracture and decay. This possibly reflected the exploitation of a deeper-seated network of microfractures initiated during their rapid heating and cooling.

Reflective question ➤ What key lessons for rock weathering have we learned from studying building stone decay?

that drive the weathering of geomaterials. These include chemical weathering processes of carbonation, solution, oxidation

Weathering is an essential precursor to almost all other geomor-

and reduction and the enhancement of these processes by

phological processes. It is all around us and is happening all the

biological action. Physical weathering processes include dila-

time. It is pervasive, persistent and can operate under normal

tion, thermoclasty, freeze–thaw, salt weathering and increased

environmental circumstances. The shapes of features produced

stresses caused by biological activity. In the above processes

by weathering have a feedback effect on weathering processes

water plays an important role as it is an effective solvent and it

by influencing micro-environmental conditions. This chapter has

has important physical properties related to surface tension and

introduced the key chemical, physical and biological processes

adsorption.



375

Chapter 14 Weathering

➤ Rarely do weathering processes operate singly. Weathering

millimetres and in minutes rather than years or even days. Like-

is the result of the interaction of two or more processes operat-

wise, if salt and ice weather rock by exploiting pores and micro

ing together or in sequence and these combinations result in

fractures the first thing we need to understand is what goes on in

the particular features and landforms that we see around us. All

these spaces and what controls it. An implication of such a small-

geomaterials carry within them a stress history, even so-called

scale approach is that eventually we have to upscale our under-

fresh rock will have embedded ‘memories’ of the conditions under

standing to explain the formation of landforms and landscapes

which it was formed, as well as any subsequent tectonics and the

over much longer timescales.

processes responsible for its exposure at the Earth’s surface. This

For urban stone decay it is important to remember that many

stress history, especially where it is complicated by, for example,

factors contribute to decay and so eradicating one will not neces-

numerous environmental changes over a long period of exposure,

sarily prevent decay. Often the rates of stone decay are hard to

can have a major influence on subsequent rates and patterns of

predict and can be spatially highly variable. Stress inheritance is

weathering.

especially relevant in situations where decay is episodic in nature

Despite an historical tendency to classify weathering pro-

and where it is the long-term accumulation of stress and mate-

cesses, forms and products on the basis of simplified large-scale

rial weakening that typically brings rock close to the threshold

environmental controls, such as meteorological measures of

beyond which it is subject to rapid and often catastrophic decay.

mean annual rainfall and mean annual temperature, the most

As rock nears this strength–stress threshold it becomes more sus-

important environmental controls on weathering processes tend

ceptible to damage from exceptional environmental conditions

to act at the meso- and micro-scales at the interface between

and processes that did not necessarily contribute to its previous

materials and the environment. Therefore, to understand weath-

weakening. Hence the common coincidence of weathering with

ering it is important to focus investigations on the temporal and

extreme events has led to an over-emphasis of the effects of, for

spatial scales over which the responsible processes operate.

example, extreme high and low temperatures to the detriment

Thus, if breakdown is through granular disaggregation one should

and neglect of more mundane conditions and regular environ-

focus initially on understanding the behaviour of individual grains

mental cycles that lay the foundations for much of the change

in response to environmental changes that operate over a few

and decay that constitutes weathering.

Further reading Bland, W. and Rolls, D. (1998) Weathering: An introduction to the scientific principles. Arnold, London. This is an excellent book for the reader to explore the chemistry of weathering in greater detail. Ford, D.C. and Williams, P. (2007) Karst hydrogeology and geomorphology. John Wiley & Sons, Chichester. An excellent overview of karst processes and associated landforms. Goudie, A.S. and Viles, H.A. (1997) Salt weathering hazards. John Wiley & Sons, Chichester.

376

An in-depth study of how salt weathering has turned out to be one of the most pervasive weathering processes that has many implications not just for natural landscapes, but particularly for the durability of materials in the built environment. Hall, K. and Thorn, C. (2011) The historical legacy of spatial scales in freeze–thaw weathering: Misrepresentation and resulting misdirection. Geomorphology, 130, 83–90. This paper challenges traditional assumptions about freeze–thaw. Yatsu, E. (1988) The nature of weathering: An introduction. Sozosha, Tokyo. One of the most detailed accounts of rock weathering.

C HAPTER 15

Slope processes and landform evolution Mike Kirkby School of Geography, University of Leeds

Learning objectives After reading this chapter you should be able to:

➤ describe differences in slope profiles ➤ describe the key processes of sediment and solute transport across slopes

➤ explain how the balance and rate of processes influence the evolving form of hillslope profiles

➤ use slope profiles to interpret the processes shaping them

15.1 Introduction Over 90% of landscapes that are currently not glaciated consist of hillslopes, and the remainder consist of river channels and their floodplains. Although hillslopes are not generally the most active part of the landscape, they provide almost all of the material which eventually leaves a river catchment through the more active channelways. The processes by which parent material is broken down by weathering (see Chapter 14) and carried to the streams are therefore vital to an understanding of how the catchment works as a geomorphological machine. The weathered layers of debris on hillslopes (the regolith or the

‘critical zone’) is also the raw material from which soils are developed (see Chapter 17). Land management such as agricultural practice strongly affects the rate and types of hillslope processes. The way in which farmland is managed can dramatically influence whether soil erosion remains at an acceptable level, or is increased to a rate which leads to long-term and often irreversible degradation of the soil. Terrestrial landscapes are dominated by erosion, and the material removed is ultimately transported to the oceans where it takes part in the continual slow recycling of the Earth’s crust as tectonic plates spread apart and collide to form new mountains (Chapter 2). Geomorphological processes form an essential part of this crustal recycling which periodically renews the surface of our planet in episodes of orogeny, erosion and isostatic response. Water helps to break up rocks as part of the process of weathering, and drives sediment transport processes that carry soil materials down to the ocean, progressively eroding the land. This chapter reviews the various hillslope sediment processes, the factors which influence their rates and the ways in which the processes in an area influence the form of the hillslopes. The balance between process rates has a very strong influence on the form of both the landscape and its soils, and plays a large part in the distinctive appearance of landscapes in different climatic regions of the Earth. Weathering and erosion

Chapter 15 Slope processes and landform evolution

provide the raw material which rivers transport through their valleys to the oceans. However, changes in hillslope erosion rate may not be matched by similar changes in river transporting capacity, resulting in either floodplain aggradation and widening, or valley incision.

15.2 Slope profiles The appearance of hillslopes around the world varies considerably (Allen, 1997). Some of these differences are related to the vegetation and soils on the hillslopes, but there are also important differences in the form of hillslope profiles. Most profiles are convex near the divide (top of the slope) and concave near their base (Figure 15.1). The most significant differences between them are in their total length, gradient and convexity.

(a)

15.2.1 Slope length Slope profiles are generally surveyed along a straight line (on a map) from the divide, following a line of steepest descent, to the nearest point at their base (Figure 15.1), which is usually in a river or floodplain, or, for coastal slope profiles, directly into the sea. The length of inland slope profiles is inversely linked to the drainage density (the total length of channels per unit area) (see Chapter 19), which varies, depending on gradients, climate and rock type. In temperate areas such as Britain, slope profiles are typically almost a kilometre in length, whereas some semi-arid areas have slopes that are less than 10 m long. Drainage density therefore generally tends to be low (1–5 km km - 2) in humid areas such as north-west Europe,

Average slope gradient

zones (Yorkshire, England, with stone walls demarcating field boundaries) and (b) semi-arid badlands (Badlands National Park, South Dakota, USA). (Source: (b) Robert Cicchetti / Shutterstock.com)

Profile line

Slope profile Vertical distance

Figure 15.2 Comparison of typical slope lengths in: (a) humid temperate

and higher (10–500 km km - 2) in semi-arid areas, particularly where the soils are developed from impermeable rocks such as shales, clays or marls, on which intensely dissected badlands may occur (Figure 15.2).

Divide

15.2.2 Slope steepness Convexity Surveyed profile down a line of steepest slope in a catchment Concavity

Horizontal distance Total slope length Figure 15.1 A typical hillslope profile.

378

(b)

Slope profiles also differ in their average gradients, from steep cliffs to gentle slopes with almost imperceptible gradients. Figure 15.3 shows the four possible elements of a profile, consisting of an upper convexity, a free face, a straight slope of almost constant gradient and a basal concavity. Not all of these features are found in every slope profile, and some may be repeated more than once within a profile. For example, many profiles do not have a free face at all, but grade smoothly from convexity to concavity, as in Figure 15.1, whereas many steep profiles have a

15.2  Slope profiles

number of free faces, each with a straight slope and/or basal concavity below it.

t

t

t

t

Upper convexity: from the divide, there is generally a convex slope of increasing gradient and with slopes from level (0°) to a maximum of up to 35° (although usually less). Free face: below the upper convexity there may be a cliff of free face, usually consisting of bedrock at a slope of up to 70° (although exceptionally steeper and/or locally overhanging). The free face may be somewhat stepped if there are rock layers of different resistances. Straight slope: there may also be a straight section of almost uniform gradient. Below a free face this usually consists of a scree or talus slope, at 30° to 40°, with a surface of loose stones. In some cases the straight slope is cut into bedrock, which is usually visible in patches beneath a thin layer of loose stones, and is called a boulder-controlled slope. Where the rock is layered, the sequence of free face and talus slope may be repeated several times. Straight slopes at 10° to 25°, in fine materials, are also commonly found in sands and clays, although not associated with a free face, but merging directly into the upper convexity. Basal concavity: at the foot of the slope there is usually a basal concavity which leads down towards the valley-bottom river and floodplain. This is usually in sand or finer materials.

Between these slope elements, there may be breaks in slope (BiS in Figure 15.3), within which the slope gradient changes relatively abruptly. There are commonly breaks in slope at the top of a free face, and between the free face and the straight slope below it. In arid and semi-arid areas, there is usually another break in slope between the straight slope and the basal concavity, although this is not normally

found in humid areas, and is associated with the different balance of processes. Comparing hillslopes of different overall steepness, the proportion of different slope elements changes, and the free face and/or straight slope are commonly absent on lower-gradient slopes. These differences are sketched in Figure 15.4, and this represents one of the very many possible evolutionary sequences over time. On even steeper slopes, the basal concavity and/ or the straight slope may also disappear, so that the cliff plunges directly into a river as a gorge. The slope steepness reflects the relationships between the hillslope and conditions at the slope base. Where there is a river that is cutting rapidly downwards at the slope base, then slopes are inevitably steeper than where the river is stable. The most rapid downcutting is usually associated with tectonic uplift, since the river is generally able to cut down almost as fast as the land is uplifted, while the upper part of the hillslopes is initially little affected. Steep slopes are also associated with coastal areas, owing to undercutting by wave action (Chapter 22). Coastlines are some of the commonest locations for good free face development. Cliffs are also common in formerly glaciated areas, where slopes have been steepened by glacial erosion, and post-glacial processes have not yet (after 10 000–15 000 years) had time to eliminate them. Over time, and in tectonically stable areas, slope processes progressively erode the landscape to produce more and more gentle slopes. The gentlest slopes are therefore found in tectonically stable shield areas, such as West Africa, central Australia and northern Canada and Eurasia. Where rocks are more easily eroded, slopes flatten more quickly, so that, in any area, the gentlest slopes are usually found on clays and shales. However, there are exceptions which include badland areas where the dissection is severe, promoting steep gully sides.

Upper convexity BiS

Upper convexities

Free face

Free faces BiS

Straight slope BiS? Straight slopes Basal concavity

Lower concavities Figure 15.3 Elements of a slope profile: upper convexity, free face,

straight slope and basal concavity. BiS is break in slope.

Figure 15.4 Slope elements on steep and gentle gradient hillslopes.

379

Chapter 15 Slope processes and landform evolution

‘skyline’ profile

‘steepest descent’ profiles

Figure 15.5 Cliffs and gullies near Tabernas, SE Spain. Visually the skyline profile follows the divides between the concave gully thalwegs. However, true

slope profiles are from each local divide into the nearest channel, following a line of steepest descent.

15.2.3 Slope convexity The amount of the hillside that is convex in profile is usually expressed as a proportion of the total slope length. Hillslopes vary from almost complete convexity (100% convex) to a narrow convexity and a much broader concavity, although most slopes are mainly convex. Where slopes appear to have long concavities, as in the ‘skyline’ profile shown in Figure 15.5, the profiles seen from a distance are not the lines of steepest slope, but a line or a series of lines along the divides between channels. These divides follow the concave profiles of the streams between them, whereas the true ‘steepest descent’ slope profiles run from the divides into the nearest stream and are convex, or convex and straight. The proportion of convexity is related to both the balance between slope processes and the relationship between the slope and the river at its base. Where the river is cutting down rapidly, or undercutting the base of the hillside, then slopes not only are steeper but also tend to be more convex. Similarly, where the streams are aggrading, more of the slope profile tends to become concave.

Reflective questions ➤ Compare slope profiles you have seen or studied in different areas: how do they differ in the proportions of each slope component (Figures 15.3 and 15.4) – or are they all the same?

➤ Where might you typically expect to find the longest and the shortest average slopes?

380

15.3 Hillslope transport processes Slope processes are of two very broad types: (i) weathering and (ii) transport of the regolith. Within each of these types, there are a number of separate processes, which may be classified by their particular mechanisms into groups (Table 15.1), although many of these processes occur in combination. Most slope processes are greatly assisted by the presence of water, which helps chemical reactions, makes masses slide more easily, carries debris as it flows and supports the growth of plants and animals. For both weathering and transport, the processes can conveniently be distinguished as chemical, physical and biological. Landscapes evolve over time in response to the internal redistribution of sediment, usually with some net removal of material to rivers or the ocean. The way in which landscapes and slopes evolve depends on their initial form, the slope processes operating and the boundary conditions which determine where and how much sediment is removed (Allen, 1997). These relationships between process and form are discussed later in this chapter. Weathering is dealt with in Chapter 14 and so the subsequent section deals with transport processes. Where there is a plentiful supply of material, and the process which moves it can only move a limited amount for a short distance, the rate of transport is limited by the transporting capacity of the process, which is defined as the maximum amount of material which the process can carry (Kirkby, 1971). A transport process of this kind, such as rainsplash (see below), is described as transport limited. Some other processes are limited, not by their capacity

15.3  Hillslope transport processes

Table 15.1 Classification of the most important hillslope processes Surface

Chemical

Weathering processes

Transport processes

Mineral weathering

Leaching Ionic diffusion

Physical

Freeze–thaw

Type (S/T)

Mass movements:

Salt weathering

Landslides

S

Thermal shattering

Debris avalanches

S

Debris flows

S

Soil creep

T

Gelifluction

T

Particle movements:

Biological

Animal digestion Root growth

Rockfall

S

Through-wash

T

Rainsplash

T

Rainflow

T

Rillwash

T

Biological mixing T (often included within soil creep)

Types: T = transport@limited; S = supply@limited removal (see text).

to transport, but by the supply of suitable material to transport, and are described as supply limited. For example, rockfall (see below) from a cliff has a very large potential capacity to carry material, but is limited, fortunately, by a shortage in supply of freshly weathered material. There is not always a clear distinction between transport- and supply-limited processes, but it is an important distinction which has a substantial impact on the way in which hillslopes evolve over geological time periods. Landscapes that are dominated by transport-limited removal are generally covered by a good layer of soil and vegetation, and slope gradients tend to decline through time. Landscapes where removal of material is mainly supply limited, however, tend to have sparse vegetation, thin soils and steep slopes which tend to remain steep throughout most of their evolution (Carson and Kirkby, 2009).

15.3.1 Chemical transport processes (solution) The process of solution is closely linked to chemical weathering, as rock and water interact. The chemical reactions that are altering rock material in place are, at the same time, releasing the lost material dissolved in the water. Rocks that weather rapidly therefore lose material

Evapotranspiration (clean water) Overland flow (containing concentrated solute)

Regolith

Bedrock

T

Rainfall

Solution of regolith material

Percolation (with solutes) Recharge to groundwater at depth

Subsurface flow in weathered rock (with solutes)

Shallow subsurface flow (with solutes)

Figure 15.6 Cycling of solutes in soil water and runoff.

in solution rapidly too. In Figure 15.6, rain falls on the soil, where it picks up solutes from the regolith, in proportion to the concentration of each constituent in the regolith, and its solubility. As water in the rock gradually weathers the bedrock, solutes also diffuse out of the upper layers of rock into the soil water. Some water is lost by percolation, containing solutes carried, for example, into limestone cave systems. Some water is lost to evapotranspiration, and this carries little or no solutes, so that the remaining overland flow and subsurface flow runoff have an increased concentration of solutes. This concentration effect is only marked in relatively arid climates, where the evapotranspiration is high. In extreme cases, some of the soluble material reaches its maximum saturated concentration, and any further concentration leads to redeposition of the dissolved material near the surface. This occurs most commonly for calcium, which is often found to form crusts of calcrete near the surface in arid and semi-arid areas. The concentration of solutes is therefore generally highest in dry climates, but the total amounts removed in solution are much less than in more humid areas. Where a flow of water contains dissolved material, the rate at which the solutes are carried away or advected is determined by their concentration in the runoff water. This is called advective solution or leaching, and is very effective at removing solutes from regolith near to the surface, both in runoff and in percolating waters. Once material is leached out, it generally travels far downstream, and its rate is supply limited. Leaching is not very effective, however, in removing material from the bedrock–regolith boundary because little water usually flows across this boundary. However, there is a rapid change in solute concentration at this boundary as water remains in contact with material of different composition (see Chapter 20). Close to the bedrock, there is a high concentration of solute ions in the water. Further above the bedrock, there is a lower concentration in the slightly more weathered regolith. 381

Chapter 15 Slope processes and landform evolution

This difference in concentration results in a net upward movement of ions, which means that there will be a movement of solutes away from the bedrock towards the regolith, even though no water is moving. This is because of ionic diffusion. Ions move about randomly over short distances and Figure 15.7 compares the number of downward movements from the upper area of low concentration and the number of upward movements from the lower area of high concentration. Even though the movements are random, Figure 15.7 illustrates how random movements in all directions cause a net diffusion of material from areas of higher ion concentration to areas of lower concentration. Around the regolith–bedrock interface the concentration of ions is higher near the unweathered bedrock than in the partially weathered regolith, driving a net upward movement of ions, carrying solutes away from the bedrock, even though little water is moving. In this case the solute load depends not on the flow of water but on the differences in concentration between the layers. This movement of solutes by ionic diffusion is not as fast as by leaching where there is appreciable water moving, but can be very important in the early stages of rock weathering, when little water is able to flow through the almost intact rock (Yatsu, 1988). Because material moves only a short distance by ionic diffusion, and is not limited by the supply of suitable material, solution by ionic diffusion is a transport-limited process. Leaching is generally the most important process in carrying solutes down the slope and into the rivers. Both leaching and ionic diffusion, however, play an important role in moving solutes vertically. One particularly significant role of vertical leaching is in carrying plant nutrients

Low ion concentration

High ion concentration

Figure 15.7 Net transport resulting from random diffusion. Because the

concentration is greater nearer the bottom of this figure than at the top then there will be more upward movements than downward movements and so the net transport will be upwards. 382

down to the roots from decaying leaf litter deposited on the surface, and so completing the nutrient cycle of the vegetation.

15.3.2 Physical transport processes When material is physically transported down a hillslope, it may travel as a mass or as independent particles. In a mass movement, a block of rock or soil moves as a single unit, although there may be some relative movement within the block. The movement of the block is mainly determined by the forces on the block as a whole, and the individual rock or soil fragments within the block are in close contact, so that they are moved together, almost irrespective of the properties of the individual constituent grains. The alternative to a mass movement is a particle movement, in which grains move one, or a few, at a time, and do not significantly interact with one another as they move. For a particle movement, forces act on each particle separately, and they move selectively, mainly depending on their sizes, but also on other factors such as shape and density (Selby, 1993). Some processes can behave in either way, according to the size of an individual event. For example, in small rockfalls there is little interaction between the few blocks coming down the cliff face, but larger blocks may break up into fragments which interact as they fall, giving them some of the characteristics of a mass movement. Both mass movements and particle movements can occur at a range of rates. In general, however, movements driven by large flows of water tend to be more rapid than drier movements. The more rapid movements also tend to carry material farther, and so tend to be supply limited, whereas slower movements tend to be transport limited.

15.3.2.1 Force and resistance Movement of material is decided by a balance of forces, some of which promote movement and some of which resist movement. For mass movements, the forces act on the block of material which is about to move, and for particle movement on each individual particle. The main forces promoting movement are those of gravity and water detachment. On a slope, there is always a component of the weight of the material that tends to pull it downslope, and this applies equally to particle and mass movements. Flowing water can detach fragments of rock or soil if it passes over them rapidly. It can do this in three ways: (i) flow of water over the surface picks up material from the surface; (ii) raindrops strike at the surface at up to 10 m s - 1 and jets of water from their impact can return at

15.3  Hillslope transport processes

almost equal speed; and (iii) in a deposit which has enough fine-grained material in it, water can also permeate the entire deposit, and convert it into a mixture of water and sediment which moves as a thick slurry. This process is called fluidization, and is able to carry large masses of material in debris flows (see below). Friction and cohesion provide resistance to movement. Box 15.1 provides details of such resistance processes and the balance of forces operating on hillsides. In simple terms, material begins to move when the forces promoting its movement become larger than the resistances holding it back. The ratio of these forces is known as the safety factor: Safety factor (SF) =

If SF 7 1 then movement will not occur; if SF … 1 then movement will begin. As soon as movement begins, the resisting forces on the material usually decrease, as the moving material detaches itself from the bonds which originally held it in place. The moving material therefore accelerates at first. Material slows down again only when the promoting forces also decrease. This usually happens where the material comes down to lower gradients where the downslope component of the gravity force becomes less, or where the water flow carrying the material spreads out and moves more slowly, or drains out into the ground or the surrounding area.

sum of forces resisting movement sum of forces promoting movement (15.1)

RESISTANCE AND THE BALANCE OF FORCES Resistance to movement is mainly due to friction and, to a lesser extent, cohesion. When a particle, or a block of material, rests on another, a component of its weight (together with the weight of any other material on top of it) provides the ‘into-slope’ force shown in Figure 15.8. For given materials in contact, the frictional

m = tan f

Downslope component of weight = p/6 r g d 3 sin b

velo

city

=v

Into-slope component of weight, N = p/6 r g d 3 cos b

Fluid entrainment force for grain with friction factor f: E = p/8 f rw d 2 v 2 Water

Slope

Weight, for particle of diameter d and density r: W = p/6 r g d 3

Slope angle b

Figure 15.8 Forces acting on a particle resting on a slope in overland

flow.

(15.2)

In Figure 15.8, where there is no water entrainment (E = 0), the downslope

Upthrust due to submerged weight in water of density rw: U = p/6 rw g d 3 cos b

Frictional resistance, F = (N - U) tan f = p/6 (r - rw) g d 3 cos b tan f Flow

resistance that can be exerted is a fixed proportion of the into-slope force. This is called the coefficient of friction, m. Another way to express the coefficient of friction is as the angle of friction, f. For a dry slope, the angle of friction is related to the coefficient of friction by the equation:

component of the weight (mg sin b) exactly balances the frictional resistance (F = N tan f = mg cos b tan f) when the slope angle, b, reaches the angle of friction, f. The angle of friction is therefore very easy to measure experimentally (Figure 15.9), and, for coarse material, is approximately equal to the angle of repose found in natural scree slopes, of 30–35°. In many kinds of partly weathered bedrock, the material consists of roughly rectangular blocks, separated by joints or bedding planes. The possibility of sliding parallel to the surface along the zig-zag Revolving drum

Level Angle of repose Scree surface Figure 15.9 A transparent drum is slowly

turned so that the angle of repose of a scree material can be measured.

BOX 15.1 ➤

383

Chapter 15 Slope processes and landform evolution

➤ line indicated in Figure 15.10 is very much hindered by the interlocking of the two surfaces, and can begin only when the surfaces are lifted apart, or dilated in the direction shown by the arrow, at an angle to the slide direction, so that free sliding only takes place along the surface after the sliding mass has lifted clear of the main cliff. Under these conditions, which are typical of most cliffs, the effective angle of friction, 𝚽, is increased by the dilation angle, so that Φ = f + u

(15.3)

It is for this reason that cliffs are commonly able to stand at angles of 75° or more, made up of an angle of friction of 35° plus a dilation angle of 40° or more. The frictional resistance to sliding of a block of material is very strongly affected

Dilation angle, u

by the water within the material. If the material is saturated with water, then part of its weight is carried by the water, following Archimedes’ principle that the upthrust is equal to the weight of water displaced. On a hillslope, the surface of the flowing water is parallel to the surface, and so this upthrust, or weight relief, affects only the into-slope component of the weight and not the downslope component. The effect of saturating the material is therefore to reduce the frictional resistance, which is proportional to the into-slope force. This is a very important factor, and is able to reduce the frictional force to roughly half its value under dry conditions. This means that a slope that becomes saturated from time to time can be stable only at slopes of about half of the angle of friction. In some materials, especially unweathered rocks and some clays, there is still some resistance to movement even when there is little or no overburden weight. This residual resistance is called the cohesion of the material, and the total cohesive force is equal to the cohesion value for the material multiplied by the area of effective contact. Cohesion is thought to develop in materials that have been consolidated at depth. Hence clays formed close to the surface, such as some tills, have little cohesion, while older clays, particularly

those consolidated over geological time periods, have substantial cohesion. When these compressed clays are brought up to the surface, however, weathering along fissures in the clay gradually reduces the cohesion, so that, over a period of 50–100 years, the cohesion becomes very small again. Material strength is made up of friction and cohesion. Both of these are usually expressed as a stress, or force per unit area, and measured in megapascals (equal to millions of newtons per square metre) (see also Box 23.4 in Chapter 23). The total strength, s, exerted to prevent sliding is expressed as a stress, or force per unit area: s = c + s tan f

(15.4)

where c is the cohesion, s is the normal force (N) per unit area and f is the angle of friction. The angle of friction and the cohesion are essentially properties of the material. Some typical values are summarized in Table 15.2. The general pattern in this table is that materials with a lot of clay minerals (clay, till and shale) have lower angles of friction than others, and heavily consolidated materials (limestone, granite, sandstone and shale) have higher cohesion than unconsolidated materials which have been formed closer to the surface (sands, gravels, clay, till and chalk).

Table 15.2 Typical values for soil strength parameters in soil and rock materials

Figure 15.10 The stable slope angle for a cliff with interlocking joint blocks and/or bedding planes. Blocks are only released from the slope when they are able to slide out at an angle to the slope surface, the ‘dilation angle’ shown, which may be 40° or more.

Sand/ gravel Clay Cohesion (MPa or million N m - 2) Angle of friction (degrees)

0

Glacial till Chalk

0–10 0–10 10–20

30–40 5–20 15–35 30–35

Limestone Granite

Sandstone Shale

100–400

100–400

10–30

10–40

30–35

30–35

30–35

10–25

BOX 15.1

15.3.2.2 Rapid mass movements There are many names for different types of rapid and slow mass movements. In rapid mass movements, the crucial distinction is between slides, in which the moving mass 384

essentially moves as a block, and flows, in which different parts of the mass move over each other with differential movement or shear. Figure 15.11 shows the difference between velocity profiles for a slide and flow. It is usually

15.3  Hillslope transport processes

Table 15.3 Rapid mass movements classified by water content

(a) Slide

Surface

Water content

Density (kg m − 3) Types of flow

Sediment forms

More solids than water

2600

Slides

Back-scar and toe

1900

Debris flows

Thixotropic forms

1700

Debris avalanche

Marginal levees and lobes

1000

Fluvial sediment transport

Mid-channel bars

Uniform velocity

More water than solids

Sharply defined slide surface (b) Flow

Surface

Varying velocity with low rate of shear near surface Rapid basal shear but no sharp slide surface Figure 15.11 Velocity profiles of mass movement in idealized slides and

flows. Shear is the differential movement between layers. In a slide, shear is concentrated at the slide surface. In a flow, shear is spread throughout the moving mass.

found that flows occur in masses with more water mixed into the moving mass, in proportion to the amount of regolith or rock material. In a slide, water is often very important in reducing the frictional resistance and allowing movement to begin, but there is little water within the moving mass. In a flow, there is usually almost at least as much water as solids, and sometimes many times more. Water and regolith materials can be mixed together in almost any proportions if they are moving fast enough, although coarse materials (sand, gravel and boulders) can only remain suspended through the mixture in the fastest and largest flows. Table 15.3 shows a classification of mass movements based on their water content, and the resulting type of flow. In a slide, the form of the original block can usually still be seen, particularly at the upslope end, where the back-scar is usually clear. The slide mass may show multiple scars and cracks where it has moved. The downstream end, or toe, shows much more severe deformation producing a hummocky topography where the mass has

advanced over the previous surface. Slides may be more or less planar when there are lines of weakness which follow geological structures or are parallel to the ground surface. Cliffs with lines of weakness parallel to the face often fail in this way, creating slab failures in which a flake of rock collapses completely or partially (Figure 15.12a). Sometimes the flake only partly separates and begins to lean progressively outwards until it fails by toppling. Many low-angle (10–20°) clay slopes also fail in planar slides, along surfaces near the base of the weathered regolith (Figure 15.12b). Planar slides are long (in the downslope direction) relative to their depth (measured into the slope), with length : depth ratios of 10 : 1 to 20 : 1 (Skempton and DeLory, 1957). Slides also occur deep within the mass of a slope. In tills or consolidated clays, these rotational slides (Figure 15.12c) move on surfaces 5–10 m deep, but in strong rock, the slides may be at depths of 50–250 m, in proportion to the much greater cohesive strength of the rock compared to the clay (Table 15.3). Their length : depth ratios are typically 3 : 1 to 6 : 1. The largest slides, in (b) Back-scar (a) Rotated former surfaces

Length:depth = 10:1–20:1 Collapsed toe

(c) Back-scar

Length/depth = 3:1–6:1

Collapsed toe

Figure 15.12 Types of rapid mass movement: (a) slab failure on a steep

gradient; (b) low-gradient planar slide, length : depth = 10 : 1 to 20 : 1; (c) steep rotational failure, length : depth = 3 : 1 to 6 : 1. Red lines show general path of moving mass. 385

Chapter 15 Slope processes and landform evolution

Figure 15.13 The prehistoric Saidmarreh landslide, south-west Iran.

Material from the 15 km wide source area was spread over 64 km2, over riding the ridge on the other side of the valley. (Source: Geology.com, using Landsat Geo cover data provided by NASA)

rock, may therefore move entire mountain sides, and may be very destructive, such as the Hope and Turtle Mountain slides in the Canadian Rockies, the prehistoric Saidmarreh slide in south-west Iran (Figure 15.13). Where the slope material moves in a wetter mixture, the mass movement becomes more like a flow. In a pure

flow, such as a river, debris is lifted into the flow by turbulent eddies and settles to the bed again under its own submerged weight. Material is deposited as a series of bars within channels. As the concentration of solids is increased, many of the moving grains strike other grains before reaching the bottom, and are supported on a bouncing layer of grain-to-grain collisions. The lift provided by these collisions, called the dispersive grain stress, becomes more important than turbulence in a hyperconcentrated flow, and is maintained by the power of the flow. Along the edges of the flow, the water drains out sideways, leaving a levee of material, while the centre of the flow continues downslope. When the flow eventually comes to a stop on gentler slopes, the levees are joined by a loop of debris around the front edge of the flow. This kind of flow is relatively common on steep mountain slopes, and is generally known as a debris avalanche. However, some lower-level debris avalanches have had a significant impact. The destructive 2014 debris avalanche in northwest Washington shown in Figure 15.14 covered about 1.3 km2 and caused substantial loss of life and damage and dammed the river causing flooding. In many formerly glaciated valleys, debris avalanches come down the steep side slopes, and their lobes spread over the gentler foot slopes, each flow following the lowest path available over the lobes of previous flows. Creation of the mixture which behaves as a hyper-concentrated flow is thought to occur through small slides on gully sides which fall into a rapidly flowing stream of water down the centre of the gully. Where there are sufficient fine-grained materials (silt and clay) in the mixture, then a true debris flow can occur,

Figure 15.14 Large debris avalanche known as the ‘Oso landslide’, north-west Washington, March 2014. The event involved about 16 million tonnes of

material and caused the loss of 43 lives in the community of Steelhead Haven. Flooding can be seen in the image due to the damming of the river by the sediment from the event. (Source: Mark Reid, USGS) 386

15.3  Hillslope transport processes

with a still higher ratio of solids to water. Drainage at the edges of the flow is less evident, and grains are supported because they sink only slowly in the mixture, which is both dense and viscous (sticky), so that the flow is much less turbulent and even large rocks sink very slowly. The flow may still move at dangerously high speeds, but also tends to stop suddenly, as drainage lowers the water content to a critical level, and the whole mass suddenly sets into a rigid mass. As water drains out, the viscosity rises sharply as grains are no longer separated by flowing water, but collide with one another and coagulate. This behaviour, in which the viscosity decreases as the rate of shear (relative movement) increases, is called thixotropic. This creates more and more rapidly moving masses once the movement has been triggered, and very rapid solidification as the movement slows down, through drainage and/or on lower gradients. Debris flows can be triggered by landslides during intense storms, particularly in semi-arid areas where there is plenty of loose material and in steep mountain areas worldwide. After volcanic eruptions, freshly deposited ash commonly provides a plentiful supply of

loose, low-density material, which is particularly prone to debris flows, known as lahars. Flows may also be triggered by subsurface seepage of water into the bottom of an unstable human-made deposit, as occurred in the 1966 Aberfan slide in South Wales, where springs saturated a low-density mine spoil heap and coal waste flowed into a school and houses, causing many casualties. In steeplands, whole areas may be affected by swarms of mass movements. These are usually triggered by rainfall as it wets the soil and reduces the safety factor, but the underlying cause may be exceptional storms, for example linked to hurricanes or typhoons, that lead to unprecedented levels of soil saturation. Swarms of mass movement can also be caused by land-use change, for example where logging removes the trees that help to transpire soil water and add to the cohesion of the soil with their roots. Earthquakes may also be responsible, shaking the ground, and so adding to the forces that promote movement and reducing the safety factor. Figure 15.15 shows swarms of landslides and debris flows triggered by the April 2015 Nepal earthquake.

Figure 15.15 Multiple mass movements following an earthquake in Nepal in April 2015. The image shows an area around 40 km2 near the border

between Nepal and China, with some example landslides highlighted by red circles. (Source: Jesse Allen, USGS/NASA) 387

Chapter 15 Slope processes and landform evolution

15.3.2.3 Slow mass movements The essential characteristic of slow mass movements is that they do not involve movement bounded by a discrete slip surface. Failures occur between individual soil aggregates, and not over the whole of an area. Movements are usually driven either by ‘heaves’ of expansion or contraction, or by apparently haphazard movements between aggregates. Heaves are usually caused by freezing and thawing of soil water, or by wetting and drying of the soil. Haphazard or apparently random movements are usually caused by biological activity which mixes the soil in all directions. In all cases, these movements do not cause any net movement when they occur on level ground, but on a slope the steady action of gravity causes more downhill than uphill movement, and there is a gradual transport of regolith material, at a rate that increases with gradient (Figure 15.16). One particular form of slow mass movement, which is intermediate between slow and fast behaviour and is driven by freeze–thaw activity, is gelifluction, which occurs in periglacial areas (see Chapter 24). Wetting or freezing of the regolith has been shown experimentally to produce an expansion that is almost perpendicular to the soil surface. As the soil then dries (after wetting) or thaws (after freezing), it sinks back closer to the vertical, under the influence of gravity on the more open texture of the expanded soil (Figure 15.16). The rate of movement is therefore thought to be roughly proportional to the slope gradient. At different times of year, the expansion penetrates to different depths, so that, totalling over the year, the lateral movement of material is greatest close to the surface, and dies away gradually with depth. Biological organisms (e.g. earthworms, rabbits and gophers) move material more or less randomly in all directions, but gravity again provides a bias which leads to net downhill movement on a slope at a rate roughly proportional to gradient.

S urf ace Expa cont nding a n ract ing l d ayer U na s ffect ed la yers

Net movement

Figure 15.16 Soil creep due to expansion and contraction in a sloping

regolith. The net movement is downslope. 388

The three main drivers for soil creep, wetting–drying, freeze–thaw and biological mixing, can all be of similar magnitudes, although one or other dominates in any particular site (Selby, 1993). For example, in temperate deciduous forests, biological mixing may dominate, whereas in upland areas, freeze–thaw is probably the most important. Most measurements suggest that rates of near-surface movement by soil creep are typically 1–5 mm per year, dying away to nothing at depths of 300 mm. The total sediment transport capacity, C, is usually estimated, from these measurements, as about: C = 10 * (tangent of the slope gradient) cm3 cm - 1 yr - 1 (15.5) Because soil creep moves the regolith material, and only operates when there is an ample soil cover, the process is considered to be transport limited, so that transport is always at the transporting capacity. An important anthropogenic process that behaves like an accelerated soil creep is tillage erosion, which is the result of ploughing, either up- and downslope or along the contour. Each time the soil is turned over, there is a substantial movement of soil. Up- and downhill ploughing produces a direct downhill component of movement as the turned soil settles back. Contour ploughing can move material either up or down, according to the direction in which the plough turns the soil. Contour ploughing in which the soil is turned downhill moves approximately 1000 times as much material as soil creep. Contour ploughing in both directions (soil turned uphill and then downhill or vice versa) or ploughing up- or downhill produces a smaller net movement, but the overall rate is still about 100 times greater than natural soil creep. Sediment transport is more rapid using modern heavy machinery than with primitive ploughs, but it is clear that tillage erosion may have been responsible for more soil movement in some areas during the past few centuries than natural soil creep during the whole of the last 10 000 years. The accumulated effect is often apparent from the build-up of soil behind old field boundaries and from the infilling of hollows within arable fields. Soil conservation measures seek to reduce the effects of tillage erosion and often involve having contoured buffers across large fields to trap sediment as shown in Figure 15.17.

15.3.2.4 Continuous creep For several days or weeks, both before and after landslides (Figure 15.18), the soil commonly moves slowly along the slide surface in a process which is intermediate

15.3  Hillslope transport processes

Figure 15.17 Buffer terraces to reduce the effects

of tillage erosion on slopes in Montgomery County, Iowa. (Source: Claus Meyer / Minden Pictures / Getty Images)

Rate of movement (mm yr-1)

10 000

1 000

100

10 SLIDE 1

0

1

2 Time (months)

3

between a landslide and soil creep. The rates may be characteristic of soil creep, but the movement is concentrated along a well-defined surface. In some clays, much of the movement on a hillside seems to consist of this continuous creep, which is usually shown up by occasional tears in the vegetation cover, and by evidence of more distinct landslides here and there (Figure 15.19). In this type of movement, the main movement is a downslope glide rather than an expansion and contraction, but some sites show a combination of normal soil creep and continuous creep, so that many combinations of heave and glide are possible.

Figure 15.18 Continuous soil creep before and after a slide. Figure 15.19 An earthflow in

southern Italy, where continuous creep is occurring over most of a field.

389

Chapter 15 Slope processes and landform evolution

15.3.3 Biological mixing Plant roots and burrowing by creatures of all sizes produce an overall biological mixing. Because there is more material in dense soil than in uncompacted soil, there is a net diffusion of material from denser to looser soil (Figure 15.20). Small organisms such as bacteria are very abundant, but generally move little material, as they are much smaller than the soil aggregates. Larger organisms such as rabbits may move large quantities of material, but are much less abundant than earthworms. The greatest total effect is generally due to creatures of moderate size, such as earthworms, termites and ants, which are just large enough to move the aggregates. The rates of mixing are much lower when the soil environment is not suitable for them due to waterlogging, cold, lack of nutrients and/ or lack of air (usually at depth). However, in some areas, larger rodents, such as gophers and mountain beavers, have the greatest impact. The regolith is normally loosely packed near the surface, and denser at depth, and this bulk density profile is a result of two processes, both strongly driven by biological mixing: (i) the balance between a net upward diffusion of mineral material by biological mixing, and its settlement under the action of gravity, at a rate which is greatest where the bulk density is least; (ii) the balance between the net downward mixing of low-density organic litter and its decomposition to carbon dioxide. The combination of these two processes gives a bulk density which is very low at the surface, and gradually increases with depth to a constant value at 0.2–2.0 m according to conditions, where the soil is deep enough (Figure 15.20). Within this layer of biological mixing, the mechanical mixing processes are Density (kg m-3) 1000

0

2000

Organic dominated

Depth (cm)

30

60

Mineral dominated

90

Figure 15.20 A bulk density profile, showing the combined effects of

increasing pore space and increasing organic content towards the surface. 390

generally much more rapid than any chemical processes, so that the mineral soil shows only very small chemical differences. If weathering goes to depths beyond the reach of biological mixing, there is a clear distinction between the homogenized mixed upper layers and an undisturbed saprolite in which the detailed rock structures are preserved intact within the weathered regolith.

15.3.4 Particle movements 15.3.4.1 Rockfall and screes Although cliffs may lose material in large slab failures, they more commonly lose smaller blocks, which fall as they are released from the cliff face by weathering along the joints and/or bedding planes around them (Terzaghi, 1962). These blocks often break into smaller fragments on impact, and the pieces bounce and roll down the scree slope at the foot of the cliff, with little or no interaction between them (Figure 15.21). The scree slope is itself constructed by accumulation of the falling blocks as the cliff retreats. As it retreats, the scree covers the base of the cliff, and protects the base from further loss. In this way a bedrock core is established within the scree. This can occasionally be seen in road-cuts, or where the loose scree is quarried away. The blocks falling onto a scree have a range of sizes, and the scree slope acts as a dynamic sieve which sorts the stones as they bounce, roll and slide down its length. Each time a block makes contact with the scree surface, it may come to a stop, or it may continue downhill. A small block landing on a surface of coarser blocks can readily be trapped between the blocks and stop, whereas a large block tends to slide over the gaps between smaller blocks (Figures 15.21b and 15.22). Small blocks therefore tend to stop near the top of the scree slope, and larger blocks go farther down, creating a slight down-scree coarsening of the grain size, which is maintained as the scree continues to accumulate. The broken blocks will, in time, also weather away. In arid areas, the boulders are often covered in a tough, dark desert varnish, which is produced as the interior of the boulder weathers. When the varnish is broken, perhaps by an impact, the weathered interior of the boulder breaks down into sand, which is easily washed off the steep scree slope (Melton, 1965). The way in which the scree slope develops depends on the ratio between the rates of these two stages of weathering: (i) from intact bedrock to boulders and (ii) from boulders to sand. Where the second stage is very slow, the cliff is gradually buried in its own detritus as the scree extends farther and farther up the cliff (Figure 15.23a). This is the normal pattern

15.3  Hillslope transport processes

(a)

Figure 15.21 Two scree slopes. (a) Boulder-

controlled slope in Nevada where cliff and scree are dynamically retreating. (b) Static accumulation on a scree in Yorkshire showing mixed sediment sizes including large blocks that have slid over smaller blocks. (Source: (a) Antony Blundell, University of Leeds)

(b)

(a) Small block lodging between larger blocks on surface

Scree surface

(b) Rockfall

Large block sliding over smaller blocks on surface No weathering of scree blocks

Figure 15.22 Relative roughness for movement of stones which are

smaller or larger than the scree surface.

Rockfall

Weathering of scree blocks to sand which is washed away

Figure 15.23 Static and dynamic cliff and scree evolution: (a) static accumulation of scree with burial of cliff; (b) continued dynamic retreat of cliff and boulder-controlled slope.

391

Chapter 15 Slope processes and landform evolution

observed on cliffs in formerly glaciated areas, such as Britain (Fisher, 1866). Beneath such a scree there is usually a convex parabolic rock core. Where the second stage is faster than the first, however, material is removed from the scree as quickly as new material is added to it by rockfall. Here scree is only a thin veneer on a bedrock slope at the angle of repose of the scree material (30–40°), which is called a boulder-controlled slope (Bryan, 1922). The landform, consisting of a cliff and boulder-controlled slope (Figures 15.21a and 15.23b), retreats across the landscape, maintaining an almost constant ratio of total to scree heights (equal to the weathering ratio described above). Such forms are also familiar from the American southwest (e.g. Monument Valley), where cliff and scree have retreated until outliers are separated by a broad desert plain. Just below the base of the cliff, the bedrock core can often be seen, preserved at the angle of repose of the scree (Figures 15.21a and 15.24).

15.3.4.2 Wash processes Water is directly responsible for the other main processes of material transport as particles which move more or less independently. The least significant process is throughwash, in which regolith particles are moved through the

regolith. The pores between grains of equal size are much smaller than the grains themselves, so that grains can be washed through textural pores only if they are at least 10 times smaller than the grains they are passing through. Through-wash is significant, therefore, only in washing silt and clay out of clean sands (e.g. in the sandy layers of podzolic soils), and in washing clays into and through structural pores, such as cracks and root holes, which are often lined with clay skins which have been deposited there in this way. The more important wash processes take place at the surface. Material may be detached by two processes, raindrop impact and flow traction, and transported either by jumping through the air or in a flow of water. Combinations of these detachment and transport processes give rise to the three different processes: rainsplash, rainwash and rillwash, as indicated in Table 15.4. Raindrops detach material through the impact of drops on the surface. Drops can be as large as 6 mm in diameter, and fall through the air at a terminal velocity which is related to their size. For the largest drops, the terminal velocity is about 10 m s - 1, but they attain this only after falling through the air for about 10 m. If their fall is interrupted by hitting the vegetation, drops hit the ground at a much lower speed, and have much less effect on impact. As drops hit the surface, their impact creates a shock wave which dislodges grains of soil or small aggregates up to 10 mm in diameter and projects them into the air in all directions. The total rate of detachment increases rapidly with the energy or momentum of the raindrops, and thus with the rainfall intensity. As a working rule, the rate of detachment is roughly proportional to the square of the rainfall intensity. Where the raindrops fall into a layer of surface water which is more than about 6 mm thick, the impact of the drop on the soil surface is largely lost. Impact through thinner films can still detach aggregates into the water, and other detached grains jump into flowing water films, which then transport grains which they do not have the power to detach. If water is flowing with sufficient force, it exerts a force on the soil which is sufficient to overcome the frictional and cohesion resistance of soil particles (Figure 15.8). Table 15.4 Types of wash processes

Transportation style Detachment by

Figure 15.24 Bedrock, at the angle of repose of scree material, exposed

at the top of a boulder-controlled slope in Provence, southern France. 392

Through the air

In overland flow

Raindrop impact

Rainsplash

Not applicable

Overland flow traction

Rainwash

Rillwash

15.3  Hillslope transport processes

This can be expressed by the safety factor (Box 15.2). An important feature of all particle movements is that different grain sizes are carried selectively. For a surface of mixed grain sizes, the safety factor is generally determined by the average of the coarser grain sizes present, as small grains hide behind and are protected by larger grains, and cannot easily be dislodged on their own. The coarsest material may also be only partially submerged in a shallow flow, increasing its safety factor because the fluid entrainment force and the upthrust (Figure 15.8) are both reduced. The threshold is also influenced by the vegetation cover, which absorbs some of the flow power. A dense grass cover may, in practice, provide an extremely resistant surface that is vulnerable only where there is a bare patch, due, for example, to grazing pressure. At low flows, some fines can be detached from between coarser grains, but at higher flows, the whole surface begins to break up together (‘equal mobility’), as coarse material releases trapped fines. Once detached, fine grains generally travel farther, but, as coarse grains settle, they again begin to trap fines in the pockets they create. However, in general, travel distance in an event, and therefore the contribution to total sediment transport, decreases with increasing particle size. Transportation through the air, in a series of hops, is able to move material both up- and downslope, but there is a very strong downslope bias on slopes of more than about 5°. As a rough guide, the net rate of transportation (downhill minus uphill) increases linearly with slope gradient, and inversely with the grain size transported. The gross rates of material transport, for rainsplash, are generally similar to those for soil creep. Rainsplash, however, is strongly particle size selective, and operates only on the surface, whereas soil creep operates over a significant

BALANCE OF FORCES ON A STONE IN WATER For a grain of diameter d, fully submerged in a flow of depth r on slope s, the safety factor SF is: SF =

∇d rs

(15.6)

depth of soil, and carries material together as a coherent mass. Protection from raindrop impact, either by vegetation or by stones, strongly suppresses rainsplash by reducing the impact velocity of raindrops. Individual stones may be left capping miniature pillars of soil as shown in Figure 15.25, and microtopography, including tillage features, are gradually smoothed out as rainsplash redistributes material, eroding high points and filling depressions.

Figure 15.25 Stones protecting soil pillars from rainsplash. Columns of

soil are left intact with a stone sitting on top of the pillar. (Source: PhotoDisc: Alan & Sandy Carey)

where ∆ = (rs - rW)/rW which is the ratio of submerged grain density to water density and rs, rW are the grain and water densities, respectively. The ratio ∆ has a value of about 1.65 for mineral grains, but may be much lower for aggregates. If the safety factor falls below a critical value, then the driving

forces are greater than the resistance, and the particle may be detached by the flow traction. Corrections to this expression for the safety factor are needed to allow for the cohesion force between fine grains, which increases roughly as 1/d, and for the tendency for particles to roll on their own down very steep slopes which are

BOX 15.2 ➤

393

Chapter 15 Slope processes and landform evolution

➤ close to their angle of repose. With these corrections rs

(15.7)

where c is a constant, related to the cohesion, with a value of 1–10 mm2. The critical value of the safety factor is also influenced by the effect of turbulence in the flow, so that the critical value for entrainment is not 1.0, but lies between 10 and 20 in experimental practice. The resistance and therefore the safety factor has a minimum at around 1 mm for sand grains in water, and has the general form shown in Figure 15.26. The particular values of the safety factor will depend on a number of other variables, particularly slope gradient (s) and flow depth (r). Once the safety factor falls below the critical threshold, the total rate of detachment is proportional to the deficit below this threshold.

100

Relative resistance

SF =

∆d(1 - s/tan f) + c/d

10

1 0.1

Frictional resistance, greatest for coarse grains

1

10

100

Grain size (mm) Figure 15.26 Resistance to detachment by overland flow, showing effects of cohesion and grain friction. Safety factor 5 resisting forces/ driving forces.

Where the surface is not protected from raindrop impact, either by overhanging vegetation or by coarse gravel, the impact of raindrops on soil aggregates leads to crusting of the surface. As raindrops strike the surface, some water is forced into aggregates, compressing air inside them, causing them to explode in a process known as slaking, and breaking them down to their constituent grains and smaller aggregates (Figures 15.27a and b). According to the grain sizes involved, these are then washed into the pore spaces around intact aggregates, creating an impermeable seal, which changes as each raindrop strikes the surface. Where the soil is mainly silt sized, a structural crust is formed at the surface (Figure 15.27c). Where there is appreciable sand, or stable sand-sized aggregates, the crust forms below the surface (a sieving crust, Figure 15.27d). Often the fine broken-down material washes into depressions and is redeposited in a layered sedimentary crust (Figure 15.27e). All of these types of crust create a relatively impermeable surface which strengthens the surface and severely limits infiltration, so that runoff from subsequent storms is increased. Vegetation cover protects the soil from direct raindrop impact and so dramatically reduces this crust formation, and, as a result, there is a very strong relationship between vegetation and runoff generation. Plot 394

Cohesion, greatest for fine grains

BOX 15.2

experiments on silt soils in Mississippi, for example, have showed a 40-fold difference in total water flow between a bare crusted field (80% annual rainfall converted to runoff) and a densely vegetated plantation (2% annual rainfall converted to runoff). Once there is overland flow, material can be carried in the flow, and some material can move much further than through the air during rainsplash. The presence of

(a)

(c)

(d) (b)

(e)

Figure 15.27 Processes of soil crusting: (a) original soil aggregates;

(b) break-up of surface layer of aggregates into constituent grains under raindrop impact; (c) structural crust at surface; (d) sieving crust below the surface; and (e) sedimentary crust of in-washed and deposited fines.

15.3  Hillslope transport processes

overland flow provides a thin layer of water on the soil surface, generally distributed rather unevenly, following the microtopography. This layer of water attenuates the impact of raindrops, and significantly reduces detachment when it is deeper than the raindrop diameter (6 mm). In shallow flows, the combination of detachment by raindrop impact and transport by the flowing water is the most effective transport mechanism, and is known as rainflow. This process provides a significant fraction of the material carried into and along rills and larger channels. When and where the flow is deeper than 6 mm, raindrop detachment becomes ineffective, and detachment is related to the tractive stress of the flowing water. Sediment is detached when the downslope component of gravity and the fluid entrainment forces overcome frictional and any cohesive resistance in the soil (Figure 15.26). Detachment increases with discharge and gradient, and decreases with grain size except where cohesion is significant. Flows powerful enough to detach material generally suppress raindrop detachment, and detached material is also carried by the flow. This combination of processes is called rillwash, and is responsible for most of the erosion by running water in major storms. Much of the material exported from an eroding field is the direct product of enlarging these small rill channels during the storm, and almost all of the material detached by raindrop impact also eventually leaves the area through these channels. Combining the effects of these three wash processes which are active during storms under a sparse vegetation cover, much of the area is subject only to rainsplash (Figure 15.28), which feeds into areas, some spatially disconnected, with thin films of water where rainflow is dominant. These areas in turn provide sediment to the eroding channels where rillwash is actively detaching material and enlarging the channels. In larger storms, the areas of rillwash and rainflow increase, and become better connected to the channels. The runoff generated per unit area and the area contributing runoff to the outlet both increase, giving a greater than linear response of runoff to increased storm rainfall. Because sediment transport also increases more than linearly with discharge, the non-linearity of the relationship between rainfall and sediment load is even stronger, making the erosion pattern very sensitive to topography at scales from the catchment to individual soil clods. Many sparsely vegetated areas develop temporary rills and gullies which are channels formed during storms and destroyed by infilling between storms (Figure 15.29). In agricultural fields, infilling is generally through tillage, sometimes deliberately after each storm and otherwise

Rainsplash dominant Rainflow dominant Rillwash dominant Figure 15.28 Domains of wash processes in a semi-arid microcatchment.

Figure 15.29 Rill development on an exposed slope.

395

Chapter 15 Slope processes and landform evolution

396

Some rocks, for example coarse sandstones and granites, produce a discontinuous distribution of grain sizes in their weathering products, dominated by joint-block boulders of weathered rock and the sand grains which are produced as the boulders break down. On these bedrocks, desert slopes often show a sharp break in slope at the base of steep hillsides, between straight slopes close to the angle of rest and the basal concavity (Figure 15.3). If grain size is plotted against gradient for these slopes, the sharp break in slope represents missing gradients which correspond to the gap in the grain size distribution.

15.3.5 The balance between erosion processes Process rates are affected by topography, particularly slope gradient, and the collecting area for overland flow; by runoff generation and flow paths controlled by climate, soil type and land use; and by the properties of parent materials mediated by the regolith. Each of the processes discussed above in the sections on weathering and transport may be dominant under some circumstances, and in this section some qualitative comparisons are made between the rates of co-existing processes. Gradient is the strongest and most universal driver of hillslope processes, but processes respond to it very differently (Figure 15.30). Solution rates are only slightly affected by gradient, at least until slopes are so low that little water circulates through the regolith. Several processes, including soil creep, gelifluction (see Chapter 24) and wash processes, increase almost linearly with gradient, although the rate begins to increase more rapidly as they approach the angle of stability (Carson and Kirkby, 2009). Rapid mass movements, including landslides and many

Rapid mass movements

Soil creep, gelifluction, wash

Process rate

following the annual cultivation calendar. In uncultivated areas, natural processes of wetting and drying, or freezing and thawing, create a loose surface layer which accumulates downslope along the depressed rill lines, and gradually obliterates them. Rills are small channels, generally 5–10 cm deep, that are formed on a smooth hillside and are not associated with a depression. Over a series of storms the rills reform in different locations, and gradually lower the whole hillside more or less evenly. Ephemeral gullies form along shallow depressions, and tend to reform along the same line in each storm, enlarging and deepening the depression, while the infilling processes bring material from the sides and gradually widen the depression. In a particularly large storm, channels may form that are too large to be refilled before the next event. These channels then collect runoff in subsequent events, leading to further enlargement, and may become permanent additions to the channel network. As material is exported, undercutting of the surface layer can lead to further rapid growth of a linear or branching gully system, which disrupts agriculture and roads, and may be very difficult to restore. Selective transportation removes fine material from the soil, leaving behind coarser material that ‘armours’ the surface. As the surface is lowered by erosion, the armour layer consists of the coarsest fraction in the layer of soil that has been eroded, and so develops more and more over time. The coarse armour progressively begins to protect the soil by reducing detachment rates, increasing infiltration and providing an increased resistance to flow. All of these effects reduce the rate of erosion until some equilibrium is approached. In this equilibrium, local differences in sediment transport rate balance differences in armour grain size. One effect of this process, acting over a period, is to establish a relationship between surface grain size and gradient, with coarser material on steeper slopes. The effects of selective transportation are only evident where the regolith contains some coarse material. This usually consists of weathered bedrock, but may consist of fragments of calcrete or other indurated soil horizons. However, erosion of soils that contain little or no coarse material, for example deep loess deposits or some deeply weathered tropical soils, cannot produce an armour layer, so may continue unchecked to great depths, often allowing the formation of extensive gully systems. On the other hand, shallow, stony soils produced by weathering of the bedrock show an enhanced effect of armouring because erosion brings lower layers of the regolith to the surface, containing less and less fines, and the end point of erosion may be a rocky desert, with no prospect of recovery.

Solution

Gradient Figure 15.30 Schematic relationships between process rates and slope

gradient.

15.3  Hillslope transport processes

30

Low

Temperature (5C)

20 SE Spain

Rio de Janeiro

10 High

0

-10 0

500

1000

1500

Precipitation (mm) Figure 15.31 Rates of solution of igneous rocks in different climate

regimes near Rio de Janeiro, Brazil and south-east Spain.

2000

30 High

20 Rio de Janeiro Temperature (5C)

debris flows, only begin to move above a fairly sharply defined threshold gradient. Thus on the gentlest slopes, solution may be the dominant process, while the steepest slopes are generally dominated by rapid mass movements. However, the transition gradient from one process to another depends on other factors, such as the regolith materials and climate. The effect of climate is closely linked to the role of vegetation. Here only areas with near-natural vegetation are considered, and it should be remembered that cultivation, fire and/or grazing can greatly modify these relationships. A rainfall–temperature diagram can be used to sketch the range of conditions. However, conditions generally change through the year at any site, and processes may therefore show a seasonal pattern, in which the vegetation cover responds to monthly changes with some delay. Removal in solution is primarily associated with the amount of subsurface runoff, and is greatest where rainfall is high and temperatures are cool, but not frozen. The pattern of relative rates is sketched in Figure 15.31 on this basis, showing a maximum rate in wet temperate climates. Annual climate loops for southeast Spain (Almeria) and south-east Brazil (Rio), showing monthly mean values of precipitation and temperature for each month of the year, have been included for reference. Although deeply weathered soils are most widespread in humid tropical areas, this distribution mainly reflects the much longer time for development in tropical shield areas (10–100 million years) than in recently glaciated temperate areas (10–20 000 years), more than differences in rates of removal.

SE Spain 10 Low

0

High -10 0

500

1000 Precipitation (mm)

1500

2000

Figure 15.32 Rates of wash erosion for uncultivated land in different

climate regimes near Rio de Janeiro Brazil and south-east Spain.

Removal by wash shows a more complex pattern, with two regions of high erosion potential (Figure 15.32), both associated with sparse natural vegetation. One high is in areas that are too cold to support vegetation, but warm enough to have at least seasonal runoff. The second high is in semi-arid climates, where sparse vegetation is combined with intense rainfall. At a given temperature, there is an initial rise with rainfall as runoff increases while vegetation remains sparse. Beyond a maximum, erosion declines as the increase in vegetation cover more than compensates for the increase in rainfall. Although not shown in Figure 15.32, there is some evidence for an eventual gradual rise in erosion at very high rainfalls, under a closed forest canopy which can provide no additional protection. This pattern strongly reflects the relationship between climate and natural vegetation. The corresponding pattern for a fixed vegetation cover such as the extreme of a bare surface shows a steady increase with rainfall, almost irrespective of temperature except under permanently frozen conditions. The difference between these two patterns gives some measure of the effect of clearing natural vegetation for agriculture, and shows why forest clearance is particularly damaging in its erosional impact, both for wash erosion and, as noted above, for mass movements. Other processes show less interesting responses to climate. Despite the importance of moisture in driving rapid mass movements, storm events occur in most climatic regimes and are, with earthquakes, the most important triggers of landslides and debris flows. Mass movements 397

Chapter 15 Slope processes and landform evolution

therefore show relatively large increases with increasing rainfall and for increasingly frost-prone climates. The contrast between processes in different climates therefore shows several features, although with many exceptions according to local conditions and histories:

t t t

on steep slopes, rapid mass movements are generally dominant; on very low slopes, solution is generally dominant; on moderate slopes, wash processes are favoured under semi-arid conditions and solution processes under humid conditions.

Hillslopes dominated by wash processes typically show sparse vegetation, thin stony soils, often with a surface armour and poor soil development with little organic matter. Hillslopes dominated by solution show deep soil profiles with strong development of clay minerals, often encouraging mass movement. The soil is fine grained with high organic content and a dense vegetation cover. These soil and vegetation characteristics are also mirrored by differences in hillslope form.

Reflective questions ➤ Can you summarize and explain the response of each hillslope sediment transport process to slope gradient?

➤ What processes would you expect to be dominant on gentle slopes in North Africa and New England?

➤ How would you expect the surface and subsurface appearance of soils undergoing soil creep and rillwash to differ? Consider their texture, sorting, organic matter content and horizon development.

➤ Can you compare the processes you would expect to find in (a) tectonically active mountains and (b) humid tropical shield areas?

➤ How do wash processes depend on hillslope hydrological processes?

responds to these factors in a distinctive manner, and therefore shows a characteristic distribution over an area, or down the length of a particular slope profile. These differences gradually change the form of the hillslope profile and, if the climate and tectonic regime remain reasonably uniform, lead to a consistent relationship between profile form and the dominant processes acting. This section explores these relationships in order to understand the principles that link process and form. Some simple models are examined that make use of these principles and show how they may be used to interpret real landscapes.

15.4.1 Concepts The history of hillslopes is primarily one of erosion, and land masses would eventually become rather flat plains close to sea level (after about 10–100 million years) if there were no tectonic uplift. However, few, even cratonic, areas ever reach this stage, partly because few areas are absolutely stable relative to sea level, and because erosion is substantially (∼75%) compensated by isostatic uplift. More commonly, areas reach an approximate balance, or equilibrium, between erosion and tectonic uplift and such landforms are relatively easy to analyze and understand. The most useful single concept in understanding how hillslopes evolve is the principle of mass balance. When sediment is transported, the loss from the source area exactly balances the addition to the receiving area. In a few landscapes, a sequence of landforms can be seen which represent either the linear progress of a process, or different process rates along a climatic gradient, and it is possible to substitute space for time (ergodic method), and interpret the spatial sequence as an evolutionary sequence over time. Where such simplifying assumptions can be made, even approximately, landscapes can be most readily interpreted. Often, however, quantitative models are required to understand how process and landform are related to each other.

15.4.1.1 Mass balance

➤ How do climate and vegetation influence the rate and dominance of slope processes acting?

The most general statement of mass balance is the storage equation: Input - output = net increase in storage

15.4 Evolution of hillslope profiles Hillslope processes move material around the landscape, primarily, for each climatic zone, in response to gradient and hydrological conditions. Each process discussed above 398

(15.8)

This expression can be applied to the mass of any identifiable component of a hydrological or geomorphological system. The component may be water, total Earth materials, a chemical element or compound, a sediment fraction defined by grain size or source rock, or a population of

15.4  Evolution of hillslope profiles

tracers (e.g. radioactive or painted pebbles). Budgeting may be done in absolute terms, or with reference to a chosen fixed datum. For example, Earth materials may be budgeted with reference to sea level as a datum, and water may be budgeted as deficit or surplus relative to saturation. Finally the physical space for which a budget is calculated can be whatever is most convenient: it may, for example, be for a one-dimensional balance of vertical fluxes at a point; for a channel reach; for a particular catchment; or for the whole Earth surface. What is important is that inputs and outputs take full account of gains and losses for the component system of interest, and include all transfers of mass across the boundaries of the defined system. Examples of mass balance approaches are given in Box 15.3.

15.4.1.2 Equilibrium and other simple landforms Although there is a complex interplay between landforms and processes, some understanding of how processes shape landforms is gained by considering simple landscapes in which there is an approximate balance between the rates of processes, and the shape of the hillslope is either constant (Hack, 1960) or evolving in a simple way.

A strict equilibrium can generally be achieved only when tectonic uplift is exactly equal to the rate of downcutting at every point in the landscape. In practice, both tectonic uplift (in earthquakes) and erosion (in major storm events) are episodic, so that equilibrium can only be considered by taking long-term averages, and most real landscapes depart even more substantially from a true equilibrium (see Chapter 1). Nevertheless, the concept of equilibrium provides a powerful tool, as in many other branches of science, for simplifying the analysis of a complex system, and offers important insights into the relationship between the set of processes acting and a corresponding characteristic form for the hillslope profile. Three types of situations are of particular value in approximating to recognized types of landscape behaviour, and in simplifying the relationship between form and process. The first is the constant downcutting form, in which uplift exactly balances vertical downcutting. Such forms are found in areas of strong tectonic uplift, in which slope gradients steepen until slopes and rivers carry away the sediment as fast as uplift raises new material. The second is parallel retreat, in which a hillslope profile migrates laterally across the landscape as it erodes. Landscapes of

MASS BALANCE One example of a mass balance is to consider total sediment for a floodplain reach (Figure 15.33). Using the terms in the figure to expand the basic storage equation: (Inputs from upstream + hillslope inputs) - output downstream = net increase in (in@channel and floodplain) storage

Input from upstream

Hil ls inp lope uts

Floodplain storage

In-channel storage

(15.9)

By measuring or estimating the terms in this expression, an estimate can be made of whether the floodplain is aggrading or degrading, and a succession of mass balances can provide an estimate of what is happening in an entire catchment, and how long sediment spends in different

Boundary of physical system considered

do Out wn pu str t ea m

Figure 15.33 Components of a valley floor sediment budget. The physical system is bounded by the edges of the floodplain and the top and bottom of the reach.

BOX 15.3 ➤

399

Chapter 15 Slope processes and landform evolution

➤ parts of the catchment. This residence time is calculated as follows: Residence time =

volume in storage average annual flux (15.10)

In this case the volume is the total volume of floodplain alluvium, and the average flux is the mean of the input and output rates. Estimates have shown that residence times are longest (10 000 years) within the hillslope soil layers, least in small channels (10 years), and gradually increase downstream (100–1000 years) (Dietrich and Dunne, 1978). The sediment budget can be further subdivided, and in this example it may be relevant to separate the budget into grain size fractions. In this case, an additional input for each separate size class is the breakdown from coarser sizes, and an additional output is the breakdown into finer material. For the hillslope, it is often convenient to break the length of the slope profile into equal sections (Figure 15.34), and examine the sediment budget for each section. For the section of interest, representing one particular store, the storage equation is Sediment in from upslope - sediment out to downslope = increase in section storage

(SIN - SOUT)∆t = ∆Z ∆x or

(SIN - SOUT) ∆Z = (15.13) ∆t ∆x where SIN and SOUT are the rates of sediment transport (per unit contour width). The left-hand side of equation (15.13) represents the rate of change of elevation over time, or the rate of aggradation (negative if erosion). The right-hand side represents the current rate of change of sediment transport with position. Thus the storage equation converts the spatial pattern of erosion (the right-hand side) to a forecast for the rate of change over time (erosion or deposition on the left-hand side). This is the basis for modelling hillslope evolution

0S 0z + 0t 0x

(15.14)

where S is the sediment transport. The change in sign from equation (15.13) is due to the convention that the change in S is taken in the sense of increasing downslope (i.e. as SOUT - SIN).

¢z = change in surface level

Section of interest

Next section downslope

SIN = sediment transported from upslope

(15.11)

SOUT = sediment transported to downslope Figure 15.34 A sediment budget for a section of a slope profile.

this general type are found in some semi-arid areas, where steep boulder-controlled slopes and cliffs are maintained in near-horizontal sedimentary rocks during the retreat of escarpments across distances of several kilometres (Figure 15.21a). The third is slope decline, in which the landscape profiles remain the same in form, but become increasingly muted in their vertical relief, and eventually decline to a horizontal plain. These forms are described 400

over time. Each short-term forecast of erosion or deposition changes the form of the hillslope. Each change of the hillslope form changes the rate of sediment transport as it responds to the topography. If the distances ∆z, ∆x and the time ∆t tend to zero in an appropriate way, the equation becomes the partial differential equation:

¢x = horizontal length of section

New surface Old surface Base of soil

Next section upslope

For a section of length ∆x over a short time period ∆t during which the surface elevation is increased by ∆z and there is no addition of, for example, wind-blown material, this storage equation can be put into symbols in the form:

(15.12)

BOX 15.3

in areas of long tectonic stability, and more commonly described for humid than for arid areas. Because of the relatively low density of continental rocks and the flexibility of tectonic plates, the unloading of the mantle by erosion is partially or completely compensated by isostatic uplift, which replaces about 70% of the loss by erosion, and is spread over an area which depends on the rigidity of the slowly flexing plate.

15.4  Evolution of hillslope profiles

Where erosion is occurring in a sequential fashion, in the migration of a meander cutting into the valley wall, or where a spit grows along the coast progressively to protect cliffs behind from erosion (Chapter 22), then the spatial sequence of visible profiles represents a sequence of passive slope recession over different periods since undercutting was active. Figure 15.35 shows an example of where this space–time substitution may be applicable. As the broad meander bend has migrated downstream from Y to X, the sequence of slope profiles from X to Y can be interpreted as an evolutionary sequence showing the retreat of a cliff by rockfall, with development of an angle of repose slope below it. Although it is often possible to draw slope profiles within an area and arrange them into such a sequence, it is not generally appropriate to do so except in very particular circumstances such as those shown in Figure 15.35. More generally, the profiles form a spatial set representing the differences in process rates among the range of topographic (area and slope gradient) settings found within a catchment, and should not be reinterpreted as a time sequence.

15.4.2 Models The discussion of slope evolution can only be carried forward through the application of mathematical and numerical models (Culling, 1963; Kirkby, 1971; Kirkby et al., 1993). The cornerstone of these models is the mass balance equation for a section of a slope, which converts the spatial pattern of sediment transport into a forecast for local rates of erosion or deposition. To create a model formally requires three other types of information: (i) the functional relationship between topography and sediment transport rates; (ii) the initial form of the profile at some

Narrow gorge

appropriate starting time; and (iii) the boundary conditions, which define how the model slope interacts with the rest of the landscape. In each case, simple assumptions can be used to show some aspects of slope development, but more complex conditions may be needed to match the evolutionary history of a particular hillside. The relationship between process rate and topography can be developed by summing the rates across the frequency distribution of storm or other events, linking the long-term rates to detailed process mechanics. However, in most slope models, this summation is taken for granted, and results are quoted directly for the long-term average rates. The driving variable of discharge is represented by its topographic surrogate which is often the distance from the divide (or the collecting area in a three-dimensional landscape). Furthermore, assumptions are commonly made about whether removal is transport limited or supply limited, and the discussion here will focus on the simpler transport-limited case. With these assumptions, there have been many attempts to express sediment transport as an algebraic function of distance from the divide, x, and local (tangent) gradient, g. Not all processes can be readily put in this form, and for landslides and other rapid mass movements they have only very limited validity, but the expressions in Table 15.5 provide a useful and relevant basis for comparing form and process. In the literature, there is some range of exponents which appear to give acceptable results, and these values should be regarded only as indicative. Creep, rainsplash and gelifluction are all driven primarily by slope gradient, operate even on gentle slopes and are not driven by flow processes. They are generally thought to be linearly dependent on gradient over the full range. Rainflow is similarly driven by a uniform detachment, but with material carried by flow, which therefore

(a)

Table 15.5 Indicative long-term sediment transport functions, assuming transport-limited removal

Sloping valley-side

Y

X Cliff and narrow scree

Cliff and broad scree

100 m

Tributary (b) X

Process

Sediment transport function

Creep, rainsplash, gelifluction

∼g

Rainflow

∼x

Rillwash

∼x2g2

Landslides

∼(g(g - g0)

Solution

(1 - g/gT) ∼x

Plateau

Y

Figure 15.35 The Rio de Aguas Gorge above Turre, south-east Spain,

showing space–time substitution: (a) sketch map; (b) schematic sequence of cliff–scree sections from X to Y.

, valid for g0 6 g 6 gT

x = distance from divide; g = local tangent gradient; g0, gT are constants. 401

Chapter 15 Slope processes and landform evolution

increases with distance from the divide. Rillwash depends on detachment by the power of the flowing water, and hence depends strongly on both gradient and distance (Schumm, 1956, 1964; Dunne and Aubry, 1986). Landslides only occur above a threshold gradient, g0, and the distance moved by material increases strongly as the angle of repose, gT, is approached. These two threshold slopes determine the rather complicated form of its dependence on gradient alone (Scheidegger, 1973). Solution is usually described by a constant rate of denudation, with material accumulating linearly with the collecting area. Boundary conditions describe the spatial relationship of the profile with the remainder of the landscape. If the profile follows the line of steepest descent, there are no exchanges of material with neighbouring profiles, and the important boundaries are at the top and bottom of the profile. It is normally convenient to take the top of the profile as the divide, and this is defined by no sediment crossing this line, or by considering the profile on the other side of the divide to be a mirror image.

g = T/A x

(a)

Gradient, g

If the slope is eroding (and uplifting) at a constant rate T, then the sediment transported past a point at distance x from the divide must be exactly Tx as this is the area between the new and old surface levels. The form of the equilibrium slope profile can then be derived directly from the sediment transport relationships in Table 15.5. For example, for soil creep, the sediment transport can be expressed both as Tx and through the process relationship as proportional to g, say equal to Ag for a suitable constant A. Putting these two expressions equal to each other, Tx = Ag, or the gradient g = Tx/A. In other words, the equilibrium slope is a convex parabolic shape in which gradient increases steadily and linearly downslope. The slope profile can be either expressed as a relationship between gradient and distance, or

recalculated as the parabolic slope profile (Figure 15.36).

Distance from divide, x z = z0 - –12 T /A x 2

(b)

Distance from divide, x z0

Elevation, z

MODELLING EQUILIBRIUM FORM OF HILLSLOPE PROFILES

The lower boundary condition usually describes the connection of the profile with the stream or floodplain at its foot. In reality there are interesting and complex interactions at this point, but, for simplicity, it is often adequate to assume that the stream is a passive agent, removing all the sediment delivered to it at an unchanging position. Another simple alternative is to assume that the stream is downcutting at a steady rate. With these tools it is possible to create a numerical model for the progress of slope evolution for a given process or combination of processes. However, a good qualitative idea of how slope form responds to process can be obtained by analyzing the constant downcutting equilibrium form, in which erosion exactly balances tectonic uplift at every point. For this and other equilibrium assumptions, the unvarying slope form obtained is independent of the initial form of the slope profile, greatly simplifying the range of possible outcomes. Box 15.4 gives an example of how this is done.

Figure 15.36 Constant downcutting equilibrium profile for soil creep: (a) gradient against distance; (b) elevation against distance.

The same procedure can be followed to work out the profiles associated with each of the separate processes in Table 15.5. Thus, for rillwash, Tx = Bx2g2 for some constant B, which can be re-expressed as g = (T/Bx). This expression shows that the gradient decreases steadily downslope, so that the profile is concave throughout. This procedure can also be applied for landslides, but gives an indeterminate result for solution or rainflow. In practice, processes generally occur together, and creep or rainsplash are generally the most important processes near to the divide. The same procedure can be applied to a combination of processes, as is shown below for a combination of creep and solution, and for rainsplash and rillwash. Constant downcutting form for creep plus solution:

t creep: S = Ag for constant A; t solution: S = Cx for constant C; BOX 15.4 ➤

402

15.4  Evolution of hillslope profiles



t in equilibrium with uplift at rate

200

T: Tx = Ag = Cx;

160

t rearranging, g = (T = C)/Ax, which is Constant downcutting form for rainsplash and rillwash:

t rainsplash: S = Ag for constant A; t rillwash: S = Bx2g2 for constant B; t in equilibrium with uplift at rate T: Tx = Ag + Bx2g2. When this is rearranged with g on the left-hand side the equation becomes a quadratic equation: - A + 2(A2 + 4TBx3) 2Bx2

(15.15)

Example results are shown in Figure 15.37 for a combination of rainsplash and rillwash keeping the process rates (A and B) constant and varying the rate of uplift (equal to denudation). The slope profiles developed are convexo-concave in profile

140 Elevation (m)

a uniform convex profile.

g =

1 m Myr-1 2 m Myr-1 5 m Myr-1 10 m Myr-1 20 m Myr-1 50 m Myr-1 Convexity

180

120 100 80 60 40 20 0

100

0

200

300

400

500

Horizontal distance (m) Figure 15.37 Constant down-cutting profiles for rainsplash plus rillwash under different

rates of uplift (in m per million years). The width of the convexity is greatest where uplift is slowest. The grey line indicates the transition from convex to concave slope profiles.

and are generally steeper under higher rates of uplift (and matching denudation), and the convexities tend to be narrower for the steeper slopes. The form of these slopes depends on the values of the process rate constants, which in turn depend on climate and soil controls, but these general conclusions stand. By changing the

relative rates of the rainsplash and rillwash transport, it can also be seen that as rillwash is increased (perhaps due to changed climate or land cover), the concavity becomes broader, and that the convex and concave sections of the slope correspond roughly to the zones where rainsplash and rillwash are respectively dominant.

BOX 15.4

In progressively less exposed situations, the angle of the cliff decreases. Inland, where there is little or no retreat, the same material forms low-angle slopes, with minimal impact of landslides. The model smoothes out the effect

100 80 Elevation (m)

A similar approach to that in Box 15.4 can be applied to the parallel retreat of hillslopes at a constant horizontal rate. This can be applied most fruitfully to the combination of creep and mass movements, to give the form of a steadily retreating hillslope. Figure 15.38 illustrates the types of profile generated in this way, with the same slope process rates, but with different rates of slope retreat. Each profile shows a convex section at the top, of gradually increasing gradient, associated with the dominance of creep processes. This convexity becomes sharper as the rate of retreat is increased. If retreat is sufficiently rapid ( 7 0.04 mm yr - 1 in this example), as in all the profiles drawn, the convexity continues until the threshold for landslides (g0 in Table 15.5: 22° or 40% in this example) is crossed. From that point, the slope gradient increases much more slowly downslope, and an increasing proportion of the material is carried by landslides. The profiles in Figure 15.38 can be visualized as a series of sea cliffs, with decreasing severity of wave attack going from left to right. The most aggressive wave attack produces the steepest cliffs, dominated by frequent mass movements.

mmyr-1 0.1 0.2 0.5 1 2 5 10

60 40 20 0

0

20

40

60

80

100

120

140

160

180

200

Horizontal distance (m) Figure 15.38 Modelled profiles for a 100 m high hillslope in equilibrium

with a constant rate of lateral retreat under creep and landslides. Legend shows rates of retreat in mm yr - 1. In this example, the threshold gradient for landslides is 40% (22°). 403

Chapter 15 Slope processes and landform evolution

showing no trace of the initial form. There is a strong similarity between the final forms in Figures 15.38 and 15.39, in both cases showing a hillslope dominated by creep processes. However, the effect of landslides in Figure 15.38 produces much more uniform gradient on the steeper slopes than under creep without slides in Figure 15.39. Figure 15.40 shows, for the same initial conditions, development under rainsplash and rillwash together. Although there is an initial rounding of the sharp plateau edge, the effect of rillwash becomes increasingly evident over time, with the development of a marked concavity in the lower part of the profile. As with creep, the final forms show no trace of the initial form, and appear to be declining smoothly towards a level plain. However, the combination

of individual landslides, which, in reality, create much less regular profiles, in which the form of individual slides shows up as a series of steps and this effect is greatest on the steepest slopes. Models can also be used to generate evolutionary sequences, for profile development over time from a given initial slope form (see the Companion Website for the book for simple slope models you can adjust). Figures 15.39 and 15.40 show two such sequences, both starting from an initial form of a plateau with a stream vertically incised into it, and then remaining stable in position. Figure 15.39 shows the uniform convexities associated with soil creep, rainsplash or gelifluction processes, and the eventual evolution to a uniform parabolic form 60

50

Elevation (m)

40

30 Initial 2000 yr 10 000 yr 50 000 yr 2 00 000 yr 10 00 000 yr

20

10

0

0

20

1000 yr 5000 yr 20 000 yr 1 00 000 yr 5 00 000 yr

40

60

80

100

120

140

160

180

200

Distance (m) Figure 15.39 Modelled slope evolution by soil creep.

60

50

Initial 500 yr 2000 yr

Elevation (m)

40

10 000 yr 50 000 yr

30

200 000 yr 500 000 yr

20

1 000 000 yr Concavity

10

0

0

20

40

60

80 100 120 Horizontal distance (m)

Figure 15.40 Modelled slope evolution by rainsplash and rillwash.

404

140

160

180

200

15.4  Evolution of hillslope profiles

of convexity and concavity is characteristic of the processes acting in each case, and the convex and concave areas roughly correspond to the areas where the rainsplash (or creep) and rillwash are respectively dominant. Thus models, both simple and complex, are able to make use of current understanding of process rates and mechanisms, and show that these processes are able to produce many of the features of observed landscapes. Three-dimensional models (Ahnert, 1976; Willgoose et al., 1991; Howard, 1994; Tucker et al., 2001) are also able to take account of the interactions between streams and hillslopes, which control the spacing or density of channels in the landscape. Some recent models are also able to incorporate other aspects of hillslope profiles, including the development of soils, armour layers and vegetation patterns (e.g. Vanwalleghem et al., 2013). With improved knowledge of the climatic drivers of process rates, and of past climates, there is also scope to understand how land landscapes have evolved and how we are currently modifying them through global changes in use and climate.

15.4.3 Interpreting landscape form The hillslopes themselves contain many clues to the processes that formed them, although, as erosional forms, they always tend to destroy rather than preserve the evidence of their formation. The gross form of the profile is generally related to the processes that formed it and partially reflects its history. Many authors once argued about whether this history reflected periods of erosion under tectonically stable conditions (Davis, 1954), the tectonic history (Penck, 1924) or a time-independent form (Hack, 1960). Convex profiles are generally associated with creep (Gilbert, 1909), rainsplash or gelifluction, but, as Figure 15.38 illustrates, rapid incision can lead to initial convexity almost irrespective of the process. Similarly, concave profiles are generally associated with rillwash or fluvial processes, yet rapid deposition can also lead to initial concavity irrespective of process. Mass movements generally lead to more or less rectilinear slopes of uniform gradient except in situations of exceptional activity. Thus landslides produce a landscape with uniform slopes close to and slightly above the threshold of sliding (g0 above), and scree slopes form at angles close to their angle of repose. However, very active areas, such as actively retreating coastal cliffs, are dominated by large rotational slides with a much more complex topography of large backtilted blocks and crumpled toe areas. Under less active erosion, as is generally found inland, the slides become

shallower and the slopes straighter, although often retaining a more or less hummocky topography associated with individual slide blocks. Smaller features are also important in interpreting process activity. The irregular hummocks that are the remnants of landslide back-scars and toe areas are one example of features that can survive for thousands of years in the landscape. Another important feature is the lines of accumulated sediment above contouring field boundaries, and equivalent erosion below them. These may be the result of deliberate terracing, but, in many cases, are accumulated over many centuries of agriculture by tillage erosion and/or wash processes. On a still finer scale, active wash processes can produce small terraces behind each clump of vegetation, and erosional steps a few centimetres high below plants or larger stones. Wash processes also sort and selectively transport surface stones, leading to a concentration of stones and some sorting of surface material. Generally erosional winnowing of fines gives rise to a pattern of downslope fining, while local patches deposition and the foot of talus slopes may show downslope coarsening as fines are trapped by coarser material. In some areas, stream head hollows give a good record of episodic erosional activity. They may fill with sediment from fast and slow mass movements over periods of thousands of years, and then empty catastrophically in a major event (Dietrich and Dunne, 1993). Similarly, large mass movements may bury former soil surfaces below their toe deposits. Such sites therefore offer some prospect of obtaining a stratigraphic record of slope history. Interpreting the form of real landscapes and understanding process mechanisms and rates are the two complementary halves of geomorphology, which need to be integrated within a broader view of environmental processes and Earth history. Two of the most exciting areas of current research are into the quantitative relationships between landforms and climate, and between landforms and tectonics. It is clear that different climatic regions have different assemblages of process and form. Humid areas are generally dominated by creep, mass movements and solution under a dense vegetation cover and well-developed soils. Hillslopes are usually mainly convex, typically with a low (1–5 km km - 2) drainage density. In contrast semi-arid areas have stony, shallow soils with sparse vegetation and surface armouring. These hillslopes evolve under supply-limited conditions, with dominant wash processes, low solution rates, concave slope profiles and high (10–100 km km - 2) drainage densities.

405

Chapter 15 Slope processes and landform evolution

Areas of rapid tectonic uplift also show distinctive slope morphologies, dominated by steep slopes and the mass movement processes which are most active on steep slopes. Largely irrespective of climate, soils are thin and stony, and processes are limited by the weathering of fresh bedrock to a state where it can undergo mass movements, and by removal processes in largely bedrock-floored steep mountain rivers. The rapid erosion fuelled by tectonics is further enhanced by isostatic uplift – as erosion removes mass from the land, the surrounding area gradually bulges upwards in response, as the flexible continental plate floats higher on the viscous mantle beneath. Erosion is most rapid along the river valleys, while the neighbouring peaks, which erode more slowly, consequently become even higher due to the isostatic uplift. This process may be partially limited by increased glacial erosion as the mountain peaks rise, creating the characteristic alpine landscapes of high mountain ranges worldwide.

15.5 Summary

Reflective questions ➤ How would you estimate the components of a small catchment sediment budget?

➤ Why are deep soils most commonly found in the tropics and subtropics?

➤ How well do familiar landscapes show a good relationship between current processes and current landforms?

➤ What are the three types of valuable situations when approximating recognized types of landscape behaviour and simplifying the relationship between form and process?

➤ What can modelling approaches tell us about hillslope evolution?

movements such as landslides and rockfall are rapid, while others such as creep are slow. Although the different processes are all

Slopes can be described by the length, steepness and convexity,

driven by water and gradient, they produce distinctive small-scale

features which result from the balance of processes operating in

features in the landscape (e.g. rills, soil pillars, crusts and screes).

different environments. The transport of material across slopes

Because processes depend differently on flow and gradient, they

can occur in dissolved form via leaching and ionic diffusion.

also create different and distinctive larger-scale hillslope forms,

Mechanical processes, usually aided by water, transport sediment

which can be analyzed through models. The shape of slopes can

downhill, in mass movements and surface wash. Some mass

be analyzed to infer the processes acting.

Further reading

Gregory, K.J. (2010) The Earth’s land surface. Sage, London. A very well-written guide to key geomorphological themes.

Allen, P.A. (1997) Earth surface processes. Blackwell Science, Oxford. An excellent book on hillslope processes. It is moderately demanding, and written from an earth science perspective.

Kirkby, M.J. (2011). Hydromorphology, erosion and sediment. In: McDonnell, J. (ed.), Benchmark volumes in hydrology. IAHS Publications, Wallingford. Another selection of classic papers in geomorphology in relation to hydrology.

Carson, M.A. and Kirkby, M.J. (2009) Hillslope form and process. Cambridge University Press, Cambridge. A classic from 1972 reprinted again in 2009.

Selby, M.J. (1993) Hillslope materials and processes, 2nd edition. Oxford University Press, Oxford. An excellent introductory text.

406

C HAPTER 16

Sediments and sedimentation Kevin G. Taylor School of Earth and Environmental Sciences, University of Manchester

Learning objectives After reading this chapter you should be able to:

➤ understand the origin, composition and classification of sediments

➤ appreciate the products and patterns of sedimentation in surface environments

➤ understand the response of sedimentation to environmental change

➤ be aware of the impacts of anthropogenic changes upon sediment systems

16.1 Introduction Sediments are an important component of the Earth’s surface and their movement and deposition are key processes in environmental systems. In conjunction with weathering and erosion (Chapters 14 and 15), sedimentation shapes the face of the Earth. Sedimentation creates floodplains, builds coastlines and fills the ocean basins. Sediments have a major impact on natural and engineered systems. In this chapter, we will introduce the concepts of sediments and sedimentation, describe the features that sedimentation produces and assess the impact that sedimentation has on the environment.

Sediments are collections of grains of pre-existing rocks, fragments of dead organisms or minerals precipitated directly from water at Earth surface temperatures. In Earth surface environments these grains accumulate (through the process of sedimentation) to form sediment packages and sedimentary successions. Sediment is a term restricted to loose, unconsolidated material (e.g. sand). If the sediment is buried and hardened, through the action of heat and pressure, to form a rock (a process that can take from tens to millions of years) it is then termed a sedimentary rock, and the suffix ‘-stone’ is added (e.g. sandstone). Sediment classification is based on a combination of the sediment’s constituents and the mode of origin of the sediment. The three major types of sediment are terrigenous clastic sediments (simply termed clastic sediments throughout this chapter), biological sediments and chemical sediments. Clastic sediments are composed of particles (clasts) derived from pre-existing rocks (which may be igneous, metamorphic or sedimentary in nature). These clasts are derived from the weathering and erosion of bedrock and soil material and comprise both primary and secondary minerals. The material that is formed as a result of this weathering is transported away from the site by water, wind or ice and will ultimately settle out and accumulate in a range of continental or marine environments. Biological sediments are derived from organic materials

Chapter 16 Sediments and sedimentation

and can be either the remains of dead organisms (e.g. shells and plants) or build-ups of framework-building organisms (e.g. coral reefs). Shell material and other skeletal fragments are commonly composed of calcium carbonate (CaCO3), in the form of either calcite or aragonite. Biological sediment composed of such material is often referred to as carbonate sediment. Sediments in which organic material (e.g. plant remains) is a significant component are commonly termed organic-rich biological sediments. For example, peat is an organic-rich biological sediment composed of dead plant material accumulated in wet environments. Chemical sediments are those that are produced by chemical processes and are formed predominantly as a result of precipitation of minerals directly from a water body. A good example of this is the precipitation of salt from evaporating seawater or an inland sea, as is happening in the Dead Sea today (Chapter 3). Sedimentation is the process by which sediment is deposited, leading to its accumulation. The most common cause of deposition is the settling out of sediment from a transporting fluid (water, wind or ice). As such, sedimentation is the opposite of erosion and transportation. While erosion removes material from a location, sedimentation leaves material behind. Although they are contrasting processes, sedimentation and transportation are closely interconnected processes. Many sediments can be sedimented and then reworked and transported many times prior to the material permanently accumulating in sediment packages. Sedimentation is therefore a dynamic process, which cannot be considered in isolation from erosion and transportation.

16.2.1 Classification of clastic sediments The most widely used scheme for classifying clastic sediments is based on the size of the clasts, or grains, within the sediment. Three major grain size classes can be recognized. Gravel refers to grains greater than 2 mm in size. Sand refers to grains less than 2 mm but greater than 63 mm (1/16th of a millimetre) in size. Mud refers to grains less than 63 mm in size. These three major classes can be subdivided further using the Udden–Wentworth grain size scale (Figure 16.1). The Udden–Wentworth grain size scale is the scheme most widely used to classify the grain size of sediments. Each grain size class within the Udden– Wentworth scale is a factor of two larger than the previous one, and is therefore a logarithmic scale (logarithmic to base 2). As can be seen from Figure 16.1, each of the three major classes is subdivided into smaller classes (e.g. very fine sand, fine sand, medium sand, coarse sand and very coarse sand). The Φ (pronounced ‘phi’) scale is a numerical representation of the Udden–Wentworth scale, based on the logarithmic nature of the scale. The Φ value = - log2 grain size (in mm). This scale is mathematically more convenient to use than fractions of millimetres. Note that a larger value of Φ represents a smaller grain size. Grain size on unconsolidated sediment is readily estimated by observing the sediment through a hand lens or a binocular microscope. A more accurate determination of grain size can be made by sieving. Sediment is shaken through a stack of sieves of reducing mesh size and the

mm

£

Class terms Boulders

16.2 Clastic sediments As stated above, clastic sediments are composed of grains of rock, weathered and eroded from pre-existing bedrock material. The dominant grains present within clastic sediments are quartz, feldspar, mica, clay minerals and iron oxides (Nichols, 2009). In sediments those minerals which are most resistant to weathering are concentrated relative to those that are less resistant. This is because more easily weathered material will be more likely to be dissolved into solution (see Chapters 14 and 20), remain in suspension or be transported well away from the source area. In general, sediment that accumulates near its bedrock source bears a greater resemblance, compositionally, to the bedrock material than does sediment deposited a long way from its source. Quartz dominates the composition of most clastic sediments as this mineral is most resistant to weathering and transport.

408

256 128 64 32 16 8 4 2 1 0.5 0.25 0.125 0.0625 0.0312 0.0156 0.0078 0.0039

-8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8

Cobbles

Pebbles

Granules

Sand

Very coarse Coarse Medium Fine Very fine

Silt

Coarse Medium Fine Very fine

Clay Figure 16.1 The Udden–Wentworth grain size classification scheme for

sediment grains. The Φ value = - log2 grain size (in mm).

16.2  Clastic sediments

percentage mass of sediment trapped at each sieve is weighed and calculated. Determination of grain size by sieving has the benefit of allowing statistical calculations to be made on the sediment (e.g. the mean and modal grain size). In consolidated sedimentary rocks grain size is commonly determined qualitatively by microscope examination.

16.2.2 Clastic sediment grain shape and texture A great deal of information about the origin, history, source and environment of deposition of a sediment can be inferred by studying textural properties of the grains within a sediment. The sorting of a sediment is a measure of the degree to which the grains in the sediment are clustered around one grain size (in other words, a measure of the spread, or standard deviation, of grain sizes in a sediment). Although this can be determined statistically by sieving analysis, sorting is more usually estimated visually through a microscope by comparison with sorting charts. In such a case, the terms poorly sorted, moderately sorted, well sorted and very well sorted are used. Although a number of factors control sorting within a sediment, in general the further a sediment has been transported from its source, the better the sorting of the grains. Grain roundness is another textural property that contains environmental information. Grains can be very angular, angular, subangular, sub-rounded, rounded and well rounded (Figure 16.2). Grains within sediments become rounded by continual abrasion during transport as a result of the impact of the sediment grains with each other. Wellrounded grains indicate that the sediment has undergone extended transport prior to deposition. Wind-blown dune sands in deserts are commonly very well rounded as a result of continual grain impacts during wind transport.

Rounded

Subrounded

Subangular

Angular

High sphericity

Low sphericity

Well rounded

Figure 16.2 Roundness and sphericity scale for sediment grains.

(Source: after Pettijohn et al., 1987)

Very angular

In contrast, sediment grains in scree slopes (see Chapter 15 and below) are commonly very angular as they have undergone only limited transport.

16.2.3 Sediment transport and sedimentation With the exception of in situ organic build-ups, such as reefs, and chemical precipitation of minerals directly from water, virtually all sediments are deposited after some element of transport. This is particularly true for clastic sediments. Sediment transport can take place as a result of gravity, but more commonly transport is by water, wind or ice. The density of the transport medium has a major control on the ability of the medium to carry sediment. The higher the density of the medium, the larger the grains that can be transported. Transport of sediment by gravity alone is only important on steep slopes and can be thought of as the first stage of erosion and transport of weathered material. Material may move down a slope, under the action of gravity, by a number of mechanisms, depending on the grain size and cohesiveness of the material, and the slope angle (see Chapter 15). Major mechanisms are rockfalls, landslides, soil creep and slumping (Chapter 15). In rockfalls, consolidated material falls and breaks up into a jumble of material at the base of a cliff or steep slope. In contrast, a landslide is where a large coherent mass of material moves down a slope undeformed. Slumping is similar to landsliding, but contains saturated slope material (pore spaces are full of water) which deforms upon movement. Rockfalls, landslides and slumps are rapid events. Soil creep is the very slow, imperceptible, movement of material down a slope (Chapter 15). Screes are accumulations of sediments that build up adjacent to mountain fronts, developed through the collection of loose sediment material removed from the mountain by gravity-driven sediment transport. Sediments fall onto the surface of the scree and move down the scree surface. They come to rest at the base of the scree where the slope shallows out. Scree slopes are composed mainly of poorly sorted, angular gravels, and exhibit only crude layering. Another form of sub-aerial gravity flow is a debris flow. A debris flow is a slurry-like flow containing both solid material and water but with a high ratio of solids to water. Debris flows are highly destructive and are common in all climate regimes. They are often started after heavy rainfall on debris-laden mountain slopes, but can also be initiated by earthquakes, volcanic eruptions and even forest fires. The sediments deposited by debris flows are generally poorly sorted with little, or no, internal stratification.

409

Chapter 16 Sediments and sedimentation

Water is by far the most common medium for sediment transport. Water moves as a result of flow in channels or as a result of currents generated by wind and tides. If water movement is fast enough it may carry sediment, and in many cases this transport can be for hundreds of kilometres before the sediment grains are deposited. Box 16.1 describes important principles associated with sediment transport and water. Transport of sediment by air (wind) can also be an important mechanism, but its effectiveness is limited by the low density of

SEDIMENT TRANSPORT BY WATER To understand transport of sediment by water we need to have a basic understanding of the nature of flowing water. Water can flow in one of two ways: laminar flow and turbulent flow. In laminar flow the water molecules all flow in the same direction, parallel to each other (Figure 16.3). As a result, almost no mixing of water takes place during laminar flow. Laminar flow is uncommon in surface waters, being restricted to low flow velocities and very shallow water. In turbulent flow, water molecules move in many directions, with an overall net flow in one direction. As a result water undergoing turbulent flow is well mixed. Laminar flow

Flow Turbulent flow

Flow Figure 16.3 Laminar and turbulent flow in

water.

air and as a result only small grains can be transported by the wind. Wind-blown sediments are important indicators of climate change in the past and may also have important impacts on climate itself (Lowe and Walker, 1997). Such issues are addressed in Box 16.2. Although ice is a solid material, it moves and deforms slowly in the form of glaciers and ice sheets and can therefore be thought of as a fluid. As a result it can transport large amounts of sediments slowly over relatively short distances (see Chapter 23; Leeder, 1999).

The Reynolds number (Re) is a dimensionless (it has no units) quantity that indicates the extent to which a flowing fluid is laminar or turbulent. The Reynolds number relates the velocity of a flow (u), the ratio between the density and viscosity of the fluid (v, the fluid kinematic viscosity) and the length of the pipe or channel through which the fluid is flowing (l). The Reynolds number (Re) = ul/v. It has been experimentally determined that when the Reynolds number is low (less than 500) laminar flow dominates, and when the Reynolds number is high (greater than 2000) turbulent flow dominates. The Reynolds number is applicable to both water and air, but the lower viscosity of air results in turbulent flow dominating at lower flow velocities than water. Sediment grains in water are transported by one of three processes (Figure 16.4). First, grains can be moved along the bed surface by rolling. Second, grains may bounce along the bed surface; this process is termed saltation. Third, material may be lifted off the bed surface and transported in suspension in the fluid, kept in suspension by turbulent flow in the fluid. Sediment transported by the first and second mechanism is termed bed load and that transported by the third process is called suspended load. Whether a sediment grain will be transported as bed load or suspended load will depend on the

size of the sediment grain and the velocity of the fluid flow. At low current velocities only small sediment grains (clays) will be transported in suspension. At higher velocities larger grains may be transported in suspension, but it is rare for grains larger than sand sized to be transported as suspended load. The Hjulström curve (Figure 16.5) shows the nature of flow velocity required to move sediment of different grain sizes in water. This graph shows both (a) Rolling Flow

(b) Saltation Flow

(c) Suspension Flow

Figure 16.4 The three mechanisms of sediment transport in flowing water: (a) rolling, (b) saltation and (c) suspension.

BOX 16.1 ➤

410

16.2  Clastic sediments

➤ Clay

Silt

VF S F S M S C S VC S Gran. Pebbles

1000

VF S = Very fine sand F S = Fine sand M S = Medium sand C S = Coarse sand VC S = Very coarse sand Gran. = Granular

Erosion and transport Ero

sio n

of

co

ns

100

oli

da te

dm

Ero

ud

Flow velocity (cm s-1)

p ns Tra

10 Erosion of unconsolidated mud

ed sb ta

an fs

ion sit po e D

o

b of

ed

Boulders

d loa

d loa

Deposition

Curves are approximate for flow depth of 1 m. The positions of the curves vary for different flow depths and different sediment characteristics.

dl po

sit i

on

of

su sp

en

de

1

el rav

dg

n da

oa d

Transport in suspension

or

n sio

Cobbles

0.1 0.001

De

the velocity of water required to keep a sediment clast in transport and the velocity required to move a stationary clast. Therefore, as a consequence it also shows the velocity below which sediment of a specific grain size will be sedimented. As can be clearly seen in Figure 16.5, as sediment grain size increases, a higher velocity is required both to keep the grains in transport and to move stationary grains. For fine sediment grain sizes only low flow velocities are required to keep sediment in transport. A consequence of this is that fine-grained sediments will accumulate only under very quiet water conditions. In contrast to this, fine-grained sediments require relatively more energy to be moved from a stationary position. This is due to the fact that clay grains, which dominate fine-grained sediments, are cohesive in nature and clump together. As a result of this, it takes greater flow velocity to transport stationary clay material than sandsized particles. This means that, although clay-sized particles are deposited only when current velocity effectively falls to zero, once deposited, clay size material is not easily eroded and transported again.

0.01

0.1

1

10

100

1000

Grain size (mm) Figure 16.5 The Hjulström diagram, illustrating the relationship between grain size and current velocity for sediment grain transport. The two curves show the energy required to keep sediment in transport (lower curve) and the energy required to transport grains from a stationary position. (Source: after Press and Siever, 1986)

BOX 16.1

WIND-BLOWN TRANSPORT OF SEDIMENT Wind-blown sediment transport is called aeolian transport (Figure 16.6). Although such transport can take place in many environments, it dominates in arid and semi-arid environments with little water. As air has a lower viscosity than water, higher flow velocities are needed to move sediment grains. At typical wind speeds, medium sand grains (up to 0.5 mm) are the largest

Figure 16.6 Sandstorm

in Dubai. (Source: Mehdi Photos Shutterstock.com)

BOX 16.2 ➤

411

Chapter 16 Sediments and sedimentation

➤ grains that can be transported (this contrasts with the pebbles and boulders that can be transported by water). As a result the deposits built up from wind-blown sediment grains are generally fine grained in nature. Transport distances for such material can be vast, with dust deposits transported thousands of kilometres from the Sahara of North Africa to the Atlantic Ocean (Figure 16.7). This material settles out on the sea floor, contributing to sedimentation in the oceanic environment. The amount of material removed is huge and for the Sahara approximately 2.6 * 108 tonnes per year. The extent of this dust transport has changed through time, a fact documented through climate and dust records preserved in ice cores and deepsea sediments. Dust transport from land surfaces was greater during the last glacial period than today. Thompson et al. (1995) documented increased dust transport during the Late Glacial Stage in ice cores from glaciers in the high Peruvian Andes. They concluded that atmospheric dust contents were up to 200 times as high as today as a result of increased aridity. As well as responding to changes in global climate, the transport of mineral dust through the atmosphere also has direct impacts upon climate and biological systems itself. Very fine-grained mineral dust is highly effective at scattering light and, therefore, may have a cooling effect on climate (see Chapter 4). Recently, it has been clearly documented that the element iron, present within wind-blown dust derived from deserts (Figure 16.7), acts as a nutrient in the surface layers of oceans and helps promote algal growth in the water column (so-called primary production) (Moore and Braucher, 2008).

This primary productivity results in the uptake of carbon dioxide (CO2) from the atmosphere and is, therefore, an important component of the global climate system. Such observations have led some researchers to suggest that the artificial addition of iron to the oceans could be one way to lower atmospheric CO2 levels (a process commonly termed iron fertilization). It has also been proposed that dust transported from the Sahara has had a detrimental impact on the development of Caribbean corals. It has been recognized that, since the late 1970s, fluxes of wind-blown dust to the Caribbean from the Sahara have increased and it has been suggested that this has led to environmental stress on Caribbean corals (Shinn et al., 2000).

Loess is a fine-grained (less than 50 μm) sedimentary deposit composed of grains of quartz, feldspar, carbonate and clay minerals transported by wind from arid land surfaces and deposited elsewhere, often thousands of kilometres from its source (Pye, 1987). Thick deposits of loess sediments are present in the Czech and Slovak Republics of central Europe, and in central China, in an area known as the Loess Plateau. It is believed that these loess deposits were formed during full glacial conditions. The loess was derived from winds blowing across arid glacial outwash plains. These loess deposits also contain soil layers which are interpreted to reflect times of wetter climatic conditions, with negligible loess sedimentation (Lowe and Walker, 1997).

Figure 16.7 Iron-rich wind-blown dust from the Sahara being delivered to the North Atlantic Ocean where it acts as a nutrient stimulating surface water primary productivity. (Source: image courtesy of SeaWiFS Project, NASA/Goddard Space Flight Center and ORBIMAGE)

BOX 16.2

16.2.4 Products of sedimentation – bedforms A bedform is a morphological feature formed when sediment and flow interact on the sediment bed surface. Ripples on a sandy beach and sand dunes in a desert are both examples of sediment bedforms. As sediment is 412

moved along the bed surface by the current, small irregularities on the bed surface influence the manner in which sediment is transported and deposited. Bedforms deposited under unidirectional flow (currents that are flowing in dominantly one direction such as river currents and wind)

16.2  Clastic sediments

are different from those formed under oscillatory flow (currents that oscillate backwards and forwards such as wave currents). Current ripples form under unidirectional flow and are small bedforms (up to 5 cm in height and 30 cm in wavelength) and form predominantly in sand-sized sediment. Current ripples are not symmetric but have a shallow stoss side and a steep lee side as shown in Figure 16.8a. Sediment is transported up the stoss side and avalanches down the lee side. As a result the ripple migrates in a downstream direction. In plan view (looking from above) the shape of current ripples can vary from straight to sinuous to linguoid (Figure 16.8b). This variation in ripple shape can be in response to water flow and water depth. Dune bedforms are larger than current ripples (up to 10 m high) but have similar cross-sections and form in a similar manner. Ripples are more predominant in silt to

medium sand-sized sediments, whereas dunes are more predominant in medium to coarse sands (Figure 16.9). A clear relationship has been documented between flow velocity and bedforms (King, 1991) (Figure 16.9). At low flow velocities, bedforms do not form. At greater flow velocities current ripples form within fine to medium sand-sized sediment, dune bedforms forming at higher flow velocity. At very high velocities bedforms do not form owing to the speed of sediment transport. This is known as the upper-plane bed stage. Wind-generated waves in water bodies (predominantly shallow marine settings, but sometimes also present in lakes) produce circular, oscillatory water motion (see Chapter 22). Beneath the surface of the water body, at the sediment surface, this motion is translated into horizontal oscillatory current movement (Figure 16.10). This motion sweeps grains away from a central zone and deposits

Wavelength

(a) Lee side

Stoss side

Crest Flow direction

Trough

Summit point

Brink point

Avalanching down lee Height

Point of flow separation

Crossstrata

Eddy in trough

Point of flow reattachment

(b) Straight-crested

Sinuous

Lunate

Catenary

Linguoid

Flow direction Figure 16.8 (a) Schematic diagram to illustrate the formation of a current ripple under unidirectional current flow. (b) Shape of current ripples in plan

view. The change from straight-crested to linguoid is governed by current strength and water depth. (Source: after Tucker, 1981) 413

Chapter 16 Sediments and sedimentation

Mean flow velocity (cm s-1)

100 80

Upper flat bed Subaqueous dunes

60 40 Ripples

20

0.6

0.8 1.0

0.4

0.2

0.08 0.1

0.06

0.04

No movement on flat bed

Sediment size (mm)

Silt

Very fine sand

Fine sand

Medium sand

Coarse sand

Very coarse sand

Figure 16.9 A bedform–flow diagram to illustrate the grain size and current velocity regimes under which sediment bedforms are present. (Source: after Nichols, 2009, and King, 1991)

particles as symmetrical ripples on the sediment bed. In cross-section and plan aspect, wave-formed ripples are symmetrical in shape and as such can easily be distinguished from current ripples formed under unidirectional flow conditions (Nichols, 2009).

Wind

Both sand ripples and sand dunes may be formed by wind-transported sediment. The most predominant environment for wind-produced bedform formation is that of the arid desert environments (Chapter 21), but localized sand dunes may also form in coastal environments as coastal sand dunes (Chapter 22). Arid zone aeolian (windblown) bedforms can be highly variable in size, ranging up to 600 m in wavelength and 100 m in height. When sediments are deposited from wind, the most common result is sand dunes (Figure 16.11). Under simple conditions of unidirectional wind patterns, simple dunes may form, with stoss and lee slopes. However, in many desert regions wind directions can change seasonally, which leads to more complicated dune bedforms. Unidirectional dunes can take two forms. Transverse dunes are linear features, with a shallow windward side and a steep lee slope. The internal structure is similar to dunes and ripples formed below water. Barchan dunes also have a shallow windward and steep lee side, but these are isolated dunes, with a characteristic crescent shape. These dunes are most commonly formed as sediment is moved across a hard substrate, such as a dried-up lake bed. Stellate (star-shaped) dunes form under conditions in which wind directions are variable, and with no particular direction prevailing. These dunes do not migrate and may be initiated at irregularities in the ground surface. Seif (or linear) dunes form where two distinct wind directions are present at approximately right angles to each other. Wind-formed ripples are commonly present on the surface of sand dunes. Chapter 21 provides further details on sand-dune forms and processes.

Waves Barchan dunes

Transverse dunes

Oscillation in water

Prevailing wind direction

Prevailing wind direction

Star dunes

Linear dunes Oscillatory motion becomes horizontal Sand grains swept into ripples

Figure 16.10 The formation of symmetrical wave ripples by oscillation of

a water body. (Source: after Nichols, 2009) 414

Prevailing wind directions

Prevailing wind directions

Figure 16.11 Aeolian bedforms in arid environments. (Source: after Nichols, 2009)

16.3  Biological sediments

Reflective questions ➤ What grain roundness would you expect for a beach sand? ➤ Looking at the Hjulström curve, what can you say about the transport of boulders within rivers and streams?

Ooid (62 mm)

Pisoid (72 mm)

Peloid (61 mm)

➤ How would you use ripples to distinguish between sediment deposited in a river and that deposited in a shallow marine wave-dominated setting?

16.3 Biological sediments

Oncoid (72 mm)

In areas where there is a minimal supply of clastic sediments, other components form the major contribution to sediments. The main component is the accumulation of dead organisms or the build-up of framework structures (bioherms such as coral reefs). Two main types of biological sediments can be recognized, carbonate sediments and organic-rich sediments. The major components in carbonate sediments are skeletal fragments of marine or freshwater organisms (e.g. brachiopods, molluscs, echinoids, corals, foraminifera and calcareous algae), bioherms (e.g. corals) and non-biological components (e.g. ooids and peloids) (Figures 16.12–16.14). Reefs are the result of framework building by organisms composed of calcium carbonate material (Masselink and Hughes, 2003). Organisms most characteristic of reef build-up are corals, but bryzoa, coralline algae and brachiopods have

Skeletal fragments

Bioherms

Molluscs (cephalopods, bivalves, gastropods)

Bryzoa

Brachiopods

Corals Cyanobacteria (stromatolites)

Echinoids Crinoids Corals Foraminifera Algae Red algae Green algae Yellow–green algae (sea grasses and lake algae)

Non-skeletal carbonate grains Ooids Pisoids Peloids (faecal pellets) Intraclasts Lime mud

Figure 16.12 Components of carbonate sediments.

Intraclasts

Figure 16.13 Non-skeletal components of carbonate sediments. (Source: after Nichols, 2009)

also produced bioherms in the geological past. Non-biological components form as the result of direct precipitation of calcium carbonate in the form of grains, although some of these grains may form via microbiological processes. Ooids and pisoids are concentrically coated grains whereas peloids are grains with little internal structure (Figure 16.13). Carbonate sediments are most abundant within warm tropical waters, as carbonate production is favoured in strong solar radiation, warm-water environments. Although traditionally viewed as forming in clear waters, free from terrestrial sediment inputs, reefs have also been reported from tropical shallow waters containing high levels of terrestrial sediment inputs (Figure 16.14b) (Perry et al., 2008). Carbonate sediments form predominantly in tropical and subtropical shallow marine environments, but calcium carbonate accumulation can also occur within temperate shallow marine environments and lakes. These sedimentary environments are sensitive to inputs of land-derived sediment and an example is discussed in Box 16.3. Sediments composed of shells of single-celled organisms (e.g. diatoms and foraminifera) also accumulate in deep-sea environments. Such sediment is commonly termed calcareous ooze, or siliceous ooze, depending on the composition of the shell material making up the sediment. When buried and lithified, carbonate sediments are known as limestones. Chalk, a pure white limestone formed throughout north-west Europe 80–65 million years ago (see Figure 2.4 in Chapter 2), is an example of a sedimentary rock composed of the remains of shells of a calcareous blue–green alga (Tucker and Wright, 1990). 415

Chapter 16 Sediments and sedimentation

(a)

(b)

Figure 16.14 Tropical carbonate sedimentary environments. (a) Assemblage from a clear, tropical reef setting at Eleuthera, Bahamas. Note the presence

of large coral colonies. All material in such environments becomes transported and deposited as sediment upon the death of the organisms. (b) A reef assemblage forming in a muddy, sediment-dominated reef setting, Queensland, Australia. (Source: Chris Perry)

DISCOVERY BAY, JAMAICA – TERRESTRIAL SEDIMENT IMPACTS UPON TROPICAL CARBONATE SEDIMENTS Carbonate sediments, composed of the mineral calcium carbonate, commonly accumulate in tropical shallow marine

environments. Much of this calcium carbonate is derived from the skeletal remains of organisms. This accumulation of calcium carbonate is an important sink for CO2 and so is, therefore, an important part of the global carbon cycle. Carbonate sediments accumulate mostly in clear waters, with minimal Discovery N Bay

(a)

inputs of land-derived sediment. In many parts of the world changes in land use, deforestation or climate change have led to increased inputs of land-derived sediments. One example where such sediments have had a major impact is that of Discovery Bay, north Jamaica (Figure 16.15). In this example, bauxite

(b)

Fe content (mg g-1)

el ann

cha

p ch 50 m

500 m

18500'N

Shi

5m

78500'W

Ship

20 m 10 m 5m

nne

l

JAMAICA

61000 1001-3000 3001-6000 6001-9000 9001-12 000 712 000

5m 10 m 20 m 30 m 40 m

Bauxite loading terminal

Figure 16.15 Discovery Bay, north Jamaica. (a) Water–depth map of the bay, also showing the location of the bauxite loading point.

(b) Distribution of iron within the bay, showing the high levels of bauxite accumulation within the bay.

BOX 16.3 ➤

416

16.4  Chemical sediments

➤ dust, an iron-rich, aluminium-rich material that is mined for aluminium production, has been discharged from a loading terminal into Discovery Bay, a semi-restricted embayment fronted by fringing reefs and dominated by the deposition of carbonate sediments. Since the 1960s these inputs

have led to the bay containing areas where surface sediment is composed of up to 35% bauxite sediment (Perry and Taylor, 2004). As well as significantly altering the composition of the surface sediments, this bauxite material has also led to markedly altered element cycling

within the sediment. Iron and phosphorus are released from the sediments into the overlying water column and greater calcium carbonate is buried, compared with sediments in areas of the bay that have not received bauxite material (Taylor et al., 2007).

BOX 16.3

Reflective question ➤ What types of landform are composed largely of biological sediments?

16.4 Chemical sediments

Figure 16.16 Peat formation occurs in areas of high rainfall and/or poor

drainage where waterlogging is common. In some places deposits of peat several metres thick can develop.

The most widely distributed type of organic-rich biological sediment is peat, which forms through the accumulation of dead plant material in wet conditions (Figure 16.16). The accumulation of peat is favoured in areas where rates of plant breakdown are low, which is most common in waterlogged stagnant conditions. Under such conditions, low oxygen, coupled with low pH, inhibits the breakdown of organic matter by bacteria and fungus. This leads to the accumulation of organic material. Peat has often been cut and dried for use as a burning fuel. If peat layers are buried beneath further sediment layers, water is squeezed from the material and volatiles (water vapour and CO2) are lost. This process leads to the formation of lignites and coals. Many of the large coal deposits of north-west Europe were deposited in freshwater swamps during the Carboniferous era, 350 to 300 million years ago.

The most important chemical sediment is that termed evaporative sediment. These sediments are formed as a result of minerals precipitated out of lake or seawater as waters are concentrated by evaporation. As seawater is evaporated the least soluble mineral precipitates out first. This is usually calcite (calcium carbonate, CaCO3) followed by gypsum (calcium sulfate, CaSO4) and halite (sodium chloride, NaCl) as the water becomes more saturated. If water becomes very concentrated a number of salts of potassium become precipitated (bittern salts). The most commonly encountered evaporite mineral is gypsum. Water concentrated to 19% of its original volume will precipitate gypsum. Halite is only deposited once the water has been reduced to less than 10% of its original volume and, therefore, is less common than gypsum. Many water bodies undergoing evaporation are periodically recharged by addition of water, either by rainfall or river water input, making it rare for such concentrated evaporation. The high solubility of NaCl also means that it is readily redissolved on exposure to water. Bittern salts are very rare and form only after complete evaporation of a standing body of water. Evaporative sediments are common in arid climates (see also Chapter 21) and may form in coastal settings or in standing bodies of water. In the case of coastal settings, 417

Chapter 16 Sediments and sedimentation

Uplift and exposure of rock

In situ weathering Transport

Sedimentation Continental Figure 16.17 Sabkha salt deposits among sand dunes in Qatar on the

margins of the Empty Quarter Desert. (Source: Paul Cowan Shutterstock.com)

the best documented examples are sabkhas (Figure 16.17). These are low-angle tidal flat environments (the coasts of the Persian Gulf being a classic example). Evaporation of groundwater draws in seawater, which upon evaporation precipitates gypsum. However, in these settings the water rarely becomes concentrated enough to precipitate halite. Thick deposits of halite have been formed in the past, such as those extracted in the Cheshire region of northern England. However, such thick deposits are not forming in the present day. To produce such thick accumulations of halite complete evaporation of large water bodies has been invoked, such as the Mediterranean Sea. However, this may become a phenomenon of the future if climate and land management change impact on evaporative processes.

16.5 Sedimentation in Earth surface environments Sedimentation takes place in a wide range of Earth surface environments. In general, sedimentary environments can be thought of as a traverse from upland mountainous environments, through lower-lying continental environments, shoreline and shallow marine settings, eventually to oceanic environments, and this is the structure adopted in the following section (Figure 16.18). This passage is particularly true in the case of clastic sediments where the transport of sediment by water, wind or ice moves sediments from continental environments to offshore environments. Given the scope of this chapter, only a brief summary can be given here of the characteristics of sedimentation in Earth surface environments. Other chapters of this book provide further details within specific environments. 418

Coastal and shallow marine

Oceanic

Figure 16.18 A schematic diagram to illustrate the pathways and major

sites of sedimentation of clastic sediments in surface environments. (Source: after Nichols, 2009)

Reflective question ➤ In which environments are chemical sediments more likely to be found?

16.5.1 Continental environments 16.5.1.1 River environments The term fluvial is generally used to describe river environments and processes. There are three major types of fluvial environment that can be recognized: braided (where the river consists of several shifting channels separated by islands, or bars); meandering (where singular sinuous channels are surrounded by low-lying floodplains); and anastomosing (where the river consists of stable, splitting and rejoining channels, thereby possessing elements of both braided and meandering form). For a detailed discussion of fluvial geomorphology see Chapter 19. Braided channels form most commonly where water flows over loose sand or gravel, often in mountainous or upland areas. A characteristic feature of braided rivers is the mobile nature of the channels and the intervening bars (Figure 16.19; see also Figure 19.4 in Chapter 19). The sediments deposited in braided rivers are most commonly composed of gravels and coarse sands. Fine-grained sediments may be deposited on bars, but they are not common. The predominant sedimentary bedforms present within braided river sediments are dune bedforms, formed as bars accrete and move forward within and between the channels.

16.5  Sedimentation in Earth surface environments

In contrast to braided rivers, meandering rivers form in low-lying areas, with a predominance of finer sediment. Meandering is a term used to describe the sinuous nature of the channels (Figure 16.19). Sediments deposited in meandering river environments are quite different from braided rivers, being predominantly finer grained than braided river sediments. Sediments deposited in the channels are coarser than those deposited in floodplains, as the water flowing within the channels transports the finer materials downstream. During periods of high stage (floods) water may overflow the channels and deposit suspended sediment upon the floodplain. As the sediment carried in suspension is silt and clay sized, the resulting sediment on floodplains is fine grained. Such fine-grained sediments are high in nutrients and as a result floodplains are fertile areas. Indeed, it is the flooding and deposition of fresh sediment that keeps the land fertile, which is why flood management and dam building can have a major negative impact on soil fertility in river areas. In addition to the floodplains, sediments accumulate on the inside of meanders, which leads to the formation of point bars (Figure 16.19). Current ripples, especially on the surface of point bars, are the most common sediment bedform in meandering river environments (Tucker, 2001).

16.5.1.2 Arid environments Desert environments are those regions of the Earth where potential evaporation exceeds rainfall (often the definition that potential evapotranspiration is more than twice the precipitation is used; see Chapter 21) and in these

environments standing bodies of water are rare. The major process of sediment transport and deposition is by wind action, although flood events can also lead to fluvial conditions prevailing under wet seasons. Three major forms of sedimentation take place in arid environments: sand seas, alluvial fans and playa lakes. Sand seas (also known as ergs) are areas of sand accumulations, and large sand seas are present in the Sahara, Namibia, south-western North America and western central Australia. Sand seas are not the same as deserts. Deserts are simply dryland environments and may or may not contain sand seas. These sand seas are composed of dune bedforms deposited from wind-blown sediments (see above). Sediment grains in these environments are typically well rounded as a result of grain–grain collisions during transport, and composed almost entirely of quartz. The red colour of desert sands is the result of a thin coating of iron oxide as a result of the oxidizing conditions in arid environments (see Chapter 21). Alluvial fans are cones of sediment that accumulate at mountain fronts. The major agent of sediment transport on alluvial fans is flowing water (which is generally present only during wet seasons). This flowing water spreads out and deposits sediment as it slows down in distant parts of the alluvial fan surface. The result is a semicircular fan Figure 16.20). Sediment deposits on alluvial fans are a mixture of gravels and coarse sands, with a general decrease in grain size away from the mountain front. Alluvial fans are particularly common in arid mountainous settings with well-developed fans present in Death Valley in California (e.g. Figure 16.21), and the Atlas Mountains of Morocco (Harvey, 1997). Meandering

Braided

Sand or gravel bar

Vegetated bar

Cross-stratification formed by downstream migration of bars, sand waves and dunes

Levee

Channel

Crevasse splay

Point bar

Floodplain

Lateral migration

Major erosional surface defining base of composite channel fill

Internal erosion surfaces

Fining-upwards of grain size within some channel fills

Epsilon cross-bedding/ lateral accretion surface

Figure 16.19 Sedimentation associated with braided and meandering river systems.

(Source: after Tucker, 1981) 419

Chapter 16 Sediments and sedimentation

Mountain Slope

16.5.1.3 Glacial environments

range Feeder channel Mountain front

Axial channel

Tributary channels

Fan surface

Ice plays a major role in sediment transport and sedimentation and gives rise to a number of characteristic landforms and sediment deposits. However, erosion and sedimentation through glacial activity are significantly different from those resulting from water and air. Detailed information on sediments in glacial systems can be found in Chapter 23.

16.5.2 Coastal and marine environments 16.5.2.1 Delta environments

Figure 16.20 Typical features of an alluvial fan. (Source: after Harvey,

1997)

Playa lakes are ephemeral (seasonal) bodies of water which accumulate during rainfall events. Water flowing into these lakes deposits a layer of silt and clay. As the lake dries up, evaporite minerals are precipitated (gypsum and halite), and desiccation cracks form within the sediment surface. If the water body dries up completely, a salt crust known as a salt pan is produced. At the next wet period further sediment is deposited followed by additional formation of evaporite minerals. This alternation of muds and evaporite minerals is characteristic of playa lakes.

There have been many definitions for deltas but a broad definition can be given as a discrete shoreline protuberance formed at a point where a river enters an ocean or other body of water. Deltas are sites where sediment supplied by the river is accumulating faster than it is being redistributed in the ocean by waves and tides. These environments vary depending on whether a delta is dominated by river processes (e.g. the Mississippi Delta), wave processes (e.g. the Rhone Delta) or tidal processes (e.g. the Ganges Delta). Further details of delta environments are provided in Chapter 22.

16.5.2.2 Estuaries and salt marshes An estuary is a semi-enclosed coastal water body where there is a mixture of river and seawater and where there is a mixture of fluvial and marine processes. At the present time estuaries are common as a result of the post-glacial rise in sea level drowning the mouths of rivers. Two major morphological elements are present in estuaries: tidal channels and tidal mudflats (Figure 16.22). Tidal channels

Figure 16.21 An alluvial fan in an arid mountainous region of Death Valley

Figure 16.22 An estuarine mudflat environment with a few salt-tolerant

National Park. (Source: Steve Carver, University of Leeds)

species on Lesbos island, Greece. (Source: Rudmer Zwerver Shutterstock.com)

420

16.5  Sedimentation in Earth surface environments

are major sites of sediment transport and hence consist of sand-sized sediment. Within tidal channels subaqueous dune bedforms and ripples commonly form. In estuaries both flood (incoming tide) and ebb (outflowing tide) currents can be present, but in general either the ebb or the flood current is strongest and this is reflected in the nature of the bedforms. Tidal mudflats are regions away from strong flood or ebb currents. Suspended sediment (silt and mud) is carried over the mudflats during high tide and deposited upon the mudflat as the tide turns and water velocity falls. The frequency of flooding depends on the height of the tide and the elevation of the mudflat. Mudflats gradually build upwards and therefore become flooded less frequently over time. If flood frequency is low, salt-tolerant plant species will colonize the mudflat, and these systems are termed salt marshes. In tropical environments, mangroves form in such environments as a result of similar processes (Figure 16.23). Chapter 22 provides more information on estuary and salt marsh systems.

16.5.2.3 Beaches, barriers and lagoons Between deltas and estuaries coastlines may be sites of sediment erosion or sediment deposition (see Chapter 22). Along coastlines that are sites of deposition, sedimentation may take place on beaches, lagoons or barriers. A beach is an area that is continuously impacted by waves. Sediment accumulating on beaches (which may be supplied by cliff erosion or by longshore drift) is continuously reworked and is characteristically well sorted and well rounded. On very shallow-sloped beaches, waves and wind may form ripples, but on steeper-sloped beaches low-angle sediment accumulation may be present, especially on wave-dominated beaches. A barrier island is a beach detached from the

main coast to form a ridge of sediment parallel to the coast (see Chapter 22). Such islands are most common along shorelines with a low tidal range and high wave energy. Sediment accumulates along the front of the island in a beach environment, whereas landward of the island, quiet conditions allow the accumulation of fine-grained material, either in salt marshes or lagoons. Lagoons are areas of low energy and are normally formed behind a barrier such as a barrier island. Along clastic sediment shorelines, muds and salt marshes develop, whereas in carbonate-dominated shorelines, fine-grained carbonate mud accumulates. Within tropical coastal settings in which clastic sediment input by rivers is minimal, the deposition of carbonate biological sediments can dominate. In such environments, sediment may be formed by the build-up of reef-building organisms, the accumulation of skeletal material and the inorganic precipitation of calcium carbonate. Reefs composed of corals accumulate in shallow, high-energy conditions and commonly act as barriers to shallow lagoon environments behind. Within these shallow lagoons, fine-grained sediment composed of precipitated calcium carbonate (lime mud) accumulates.

16.5.2.4 Shallow marine environments The nature of sediment deposited in shallow marine environments is governed by the strength of currents produced by tides and storms. Sands are the predominant sediment deposited, commonly as sand dunes up to 10 m in height. In wave- and storm-dominated environments (micro-tidal) sands are deposited as wave-rippled and symmetrical hummocks. In tidal-dominated shallow marine environments (macro-tidal) sand waves and sand ribbons form. Sand waves form under lower tidal flow and are aligned perpendicular to the direction of tidal flow. Sand ribbons form under higher tidal flows and are aligned parallel to tidal flow. Fine-grained sediment tends to be deposited in waters deep enough not to be affected by major storms and tides. At the present time, many shallow marine environments (e.g. the North Sea) are covered with a layer of coarse sand and pebbles. These deposits are relict from the time of much lower sea level during the last Ice Age. Finer-grained sediment is currently being trapped in estuaries.

16.5.2.5 Oceanic environments

Figure 16.23 Tropical coastal mangrove with tree roots that trap and sta-

bilize sediment. (Source: Nunnicha Supagrit / Alamy Stock Photo)

Oceanic environments include those environments from the edge of the continental shelf into the abyssal plains (see Chapter 3), and span a water depth from 100 m or less to over 8000 m in some of the deep-sea trenches. 421

Chapter 16 Sediments and sedimentation

Sedimentation in oceanic environments takes place via two major processes: turbidite currents and pelagic sedimentation. Turbidite currents are mixtures of sediment and water which, because of their increased density relative to seawater, flow down and along the bottom surface of the oceans. In this process, they transport sand and clay-sized sediment from shelf slopes to deeper oceanic environments, depositing sediment as a thin bed widely across the sea floor (Figure 16.24). Turbidite flows are commonly triggered by earthquake events, with one of the best documented examples being in the Grand Banks area of Newfoundland in 1929. The resulting turbidite flows broke transatlantic telephone cables on the seabed. Pelagic sedimentation is the slow background sedimentation of fine-grained material falling through the water column to the seabed. The best developed pelagic sediments accumulate in the deep sea where clastic sediment input from continents is minimal. Three types of pelagic sediments have been documented from deep-sea environments: brown clay, carbonaceous ooze and siliceous ooze. Brown clay is a sediment accumulation of fine-grained clay grains and glass fragments. The sediment is derived predominantly from wind-blown continental material, volcanic material and micrometeoric grains. Carbonaceous ooze

Figure 16.24 Sandstone beds deposited by turbidite currents, Tabernas,

southern Spain. 422

is an accumulation of calcite tests of microscopic organisms living in the water column (e.g. foraminifera and coccoliths), and is widely distributed on the ocean floor. Siliceous ooze is composed of the tests of microscopic organisms made of silica (e.g. radiolaria and diatoms) and has a localized distribution on the ocean floor.

Reflective questions ➤ What are the main differences in the dominant sediment processes between continental, coastal and oceanic environments?

➤ What type of bedforms do you find in rivers? ➤ What type of bedforms do you find in arid environments?

16.6 Response of sedimentation to environmental change Both natural and anthropogenic activities can have a major impact on the style and rate of sedimentation in surface environments. Therefore it is important that we can measure the rates of sedimentation on different environments and Box 16.4 provides some examples of how this can be done. The natural changes that impact most upon sedimentation are climate change and sealevel change. Climate change leads to changes in rainfall and vegetation which have major impacts upon sediment supply in the catchment, and thereby sedimentation in associated receiving water bodies. Sea-level change results in changes in the base level of sedimentary systems. The result is commonly either the marine inundation of coastal and continental environments, or the exposure of shallow coastal shelves. In general, these natural changes are slow and gradual, although there are numerous examples in the geological past where such changes have produced marked changes in sedimentation style. At the present time sea-level rise, associated with global climate warming, is having marked impacts on low-lying coastal systems (see Chapter 22). Of greater short-term impact and concern are the effects of anthropogenic activities on sedimentation. Such activities can be either direct, through the engineering of water bodies (e.g. dams and, reservoirs), or indirect, through changes in catchment characteristics (e.g. mining and, urbanization).

16.6  Response of sedimentation to environmental change

MEASUREMENT OF SEDIMENTATION RATES The rate of sedimentation is a measure of the thickness of sediment (normally measured in centimetres) that accumulates at a specific location over a specific amount of time. Generally, accumulation rates are quoted in centimetres or millimetres per year, but may also sometimes be quoted as grams per cm2 of sediment. However, in many environments sedimentation rate is generally very low, and so rates in centimetres per hundred years, or even per thousand years, are commonly quoted. Sedimentation rates can be measured for environmental systems using a range of techniques. Short-term measurements can be made by collecting sediment that accumulates in a sediment trap and measuring the amount of sediment deposited over a month or a year. Alternatively, in salt marsh or floodplain environments short-term sediment accumulation rates have been

measured by laying down grass mats on the marsh and measuring sedimentation upon the mats over durations of 1–10 years. Longer-term measurements of sedimentation rates can be made using radionuclide determination. Caesium-137 (137Cs) is a radioactive element that was released into the atmosphere by atomic weapons testing in the 1950s and by the Chernobyl nuclear power station incident in 1986 (Figure 16.25). By measuring the vertical location of these two concentration peaks

of 137Cs in the sediment, an estimate of annual sedimentation rate can be made. For longer-term sedimentation rate estimations, archaeological artefacts or 14C dating can be used to estimate sediment accumulation rates over hundreds to thousands of years (see Chapter 5). Under special circumstances, yearly layers of sediment (commonly called varves in lake sediments) can be recognized, allowing for accurate estimates of sedimentation rate in such cases (Figure 16.26).

160

Annual deposition of 137Cs (PBq)

140 120 100 80 60 40 20 0

1950

1960

1970

1980

1990

Year Figure 16.25 The inputs of radioactive 137Cs from the atmosphere to the

northern hemisphere since 1950. The peaks in the 1950s and 1960s as a result of atomic weapons testing, and in 1986 as a result of the Chernobyl incident, allow for estimates of sedimentation rate to be made for sediments over the past 50 years, assuming constant sedimentation rates. (Source: after Owens et al., 1996)

Figure 16.26 Annual sediment layers (varves) deposited in Quaternary Glacial Lake Riada, central Ireland. Core is approximately 25 cm in length. (Source: Cathy Delaney)

BOX 16.4 ➤

423

Chapter 16 Sediments and sedimentation

➤ Typical sedimentation rates in natural systems display a wide range of values. In general, sedimentation rates are low, being of the order of less than 1 cm yr - 1. Sedimentation rates are very low in oceanic environments (commonly less than 1 cm

per 1000 years), as a result of their great distance from sediment sources. Sedimentation rates in lakes, floodplains and coastal settings can be much higher (in the range of 0.1–10 cm yr - 1). It should be remembered, however, that sedimentation

is a dynamic process and rates of sediment accumulation may vary over timescales from daily to yearly. Longer-term changes in sedimentation rate will also result from natural and anthropogenic changes to the system.

BOX 16.4

16.6.1 Dams and reservoirs Dams and reservoirs have been constructed since early human history for regulation of water, but in the last 50 years the construction of major dams for water supply and hydroelectric generation has increased markedly. As well as having marked impacts upon water flow within catchments downstream of the dam, they also have a marked impact on sedimentation and sediment transport throughout the catchment. Although sedimentation impacts vary between catchments, it has been estimated, for example, that a typical water reservoir in upland United Kingdom loses 10% of its volume as a result of sedimentation over a 100 year lifetime (Figure 16.27). The two most significant impacts are the trapping of sediment behind the dam and the reduction in the sediment load of the river downstream. Sedimentation in lakes behind dams leads to less floodplain sedimentation downstream, which reduces nutrient supply and increases the potential for

Figure 16.27 An upland reservoir in the UK during drought conditions

showing layers of sediment that have accumulated within the reservoir. 424

erosion (clear-water erosion: see Box 19.4 in Chapter 19). Before 1930 the Colorado River, USA, carried up to 150 million tonnes of suspended sediment annually to its head in the Gulf of California. Since that time, a number of dams have been built on the Colorado. The Glen Canyon Dam, for example, was constructed as a sediment trap to prolong the life of Lake Mead behind the Hoover Dam which was completed in the 1930s. Sediment is currently being deposited and trapped behind these structures (e.g. in Lake Mead and Lake Powell) and is no longer discharged into the sea. Indeed, as a result of dams and water abstraction in southern California, water no longer enters the sea from the Colorado River. Box 16.5 discusses the world’s largest dam.

16.6.2 Mining The activity of mining economic deposits from the Earth can have major impacts on sedimentation in both river and coastal systems. Mining activity can take the form of subsurface mining for metals, or the open-cast mining of metals, coals and aggregate material. In all cases, the mining activity exposes large amounts of material to subaerial and fluvial erosion through the production of piles of spoil and waste material. This increase in erodibility leads to increased sediment loads in rivers and increased deposition of material in downstream environments. In addition to large sediment yields, the sediment that is deposited in downstream settings is commonly highly contaminated with metals, which has an impact on organisms living within those environments. A severe pollution event linked to the dispersion of metals associated with a tailings dam spill was the Aznalcóllar copper–silver–lead–zinc mine in Spain, 45 km west of Seville. On 25 April 1998 the tailings dam, which held the fine-grained metal-rich tailings waste from the mining activities, failed, releasing metal-rich sediment into the Agrio and Guadiamar Rivers. Approximately 4600 ha of floodplain land was flooded with an estimated 2 million

16.6  Response of sedimentation to environmental change

THE THREE GORGES DAM The largest dam-building project in the world was undertaken on the Yangtze River in China (Figure 16.28). The Three Gorges Dam wall is 2 km long and 100 m high and was completed in 2006. All hydroelectric plants on the dam were operational by 2012. The resultant reservoir stretches for 650 km upstream. As well as for hydroelectric generation, the Three Gorges Dam is designed to protect 10 million people downstream from devastating floods that have killed up to 300 000 people in the past 100 years. However, there is concern as to how long the reservoir will last, given that silt carried down by the Yangtze will sediment behind the dam, perhaps eventually filling the reservoir. On the Yellow River, also in China, a reservoir behind the Sanmenxia Dam filled with silt within four years of construction and had to be emptied, dredged and rebuilt. In the year 2000 the reservoir had less than half its original capacity (Chengrui and Dregne, 2001). The Yangtze carries 530 million tonnes of silt through the Three Gorges area each year. A similar, smaller dam on the Yangtze (the Gezhouba Dam), a test run for the Three Gorges, lost more than a third of

its capacity within 7 years of opening. To minimize this loss of capacity in the Three Gorges area, two approaches have been recommended. The first is to keep the reservoir levels low during high-flow

seasons to allow more transported sediment into the lower reaches of the river. The second is to increase tree cover in the catchment in order to decrease erosion of silt into the river.

Figure 16.28 The Three Gorges Dam, Yangtze River, China. This is a false colour satellite image

where red represents green vegetation, light blue is urban area or bare ground, and dark blue is water. (Source: NASA/GSFC/METI/ERSDAC/JAROS, and U.S./Japan ASTER Science Team)

BOX 16.5

m3 of metal-rich tailings, as well as over 5 million m3 of metal-rich acidic water. This was the worst recorded pollution event in Spanish history. The contaminated sediment was deposited along a length of river 40 km downstream from the mine, and 800 m wide. Similar toxic sediment inputs to river systems as a result of mining have happened recently in Bolivia (Rio Pilcomayo; Hudson-Edwards et al., 2001) and Romania (Tisa Basin; Macklin et al., 2003). The clean-up of affected rivers takes the form of mechanical excavation, and manual clearing of the deposited layer of tailings sediment. Recent research has shown that there are long-term impacts upon river systems as a

result of the deposition of these tailings’ sediment on the floodplain and the channel bed (Hudson-Edwards et al., 2005; Kraus and Wiegand, 2006).

16.6.3 Urbanization Urbanization has a marked effect on sediment sourcing and sedimentation, mainly through the anthropogenic nature of sedimentary material and the engineering of the land surfaces (Taylor and Owens, 2009). There are two main types of urban sediment: aquatic sediments in urban water bodies (e.g. canals and, docks), and as street sediments on road surfaces (‘street dust’). 425

Chapter 16 Sediments and sedimentation

Urbanization of sediment catchments has a number of effects on sedimentation in water bodies. Watercourses become engineered, commonly by channelization and culverting, and land surfaces are paved over. All this has the effect of increasing the rate and extent of sediment supply to receiving water bodies, while vegetation loss reduces the sediment storage capacity of the system. In addition to sediment yields, sediment quality also markedly decreases as a result of urbanization. Sediment can become contaminated by sewage, industrial pollution and vehicular pollution. As a result of this, sediment pollution is a problem within urban water bodies, and chemical reactions in the sediment can lead to the remobilization of this pollution and the generation of noxious methane gas (Taylor et al., 2003). Sediment accumulates on street surfaces as a result of industrial, vehicular and building activities and these sediments commonly contain high concentrations of lead and other metals (Robertson et al., 2003). These sediments have been implicated in respiratory diseases, as the finegrained fraction of this sediment can be resuspended in the atmosphere and inhaled by humans living and working in urban environments.

CONTAMINATED SEDIMENT MANAGEMENT AND TREATMENT, PORT OF HAMBURG, GERMANY The contamination of sediments with potentially toxic elements (e.g. metals, persistent organic pollutants) is increasingly being recognized as a major problem. In many cases, the presence of this contamination needs management and remediation. Such an example is the Port of Hamburg, Germany, which is a major economic shipping port on the River Elbe, approximately 100 km from the North Sea. Sediment is delivered to the port from upstream sources on the River Elbe and from tidal transport of sediment from the North Sea. Sedimentation rates are high,

16.6.4 Sediment management The field of sediment management has rapidly expanded in response to the environment pressures exerted upon sediment systems. As Box 16.6 indicates, many systems suffer from issues of sediment quality (e.g. pollution) and sediment quantity (e.g. sediment requiring dredging). Management of sediment in such cases may take the form of source control by cleaning up contaminated sediment sources such as mine sites, industry and wastewater treatment works, for example. Site-specific remediation, or clean-up, can take the form of dredging, sediment nourishment, sediment capping (to isolate contaminated sediments from overlying water) or in situ chemical or biological treatment.

Reflective questions ➤ Thinking of dams, mining and urbanization, which of one or two of these activities are most likely to: (a) increase rates of sedimentation; (b) change the type (or quality) of the sediment; and (c) decrease rates of sedimentation?

➤ Can you describe some methods of measuring sedimentation rates?

and sediment accumulation in the port significantly reduces water depth. As a result a dredging programme is needed to maintain water depths in the port to allow continued shipping access (Figure 16.29). Approximately 3 to 4 million m3 of sediment is dredged each year. Once the sediment has been dredged from the port, it cannot be dumped at sea as the sediment is contaminated and must be treated as a controlled waste. The River Elbe upstream of Hamburg flows through industrial and mining areas in Germany and the Czech Republic and for this reason the sediment in Hamburg Port is contaminated (Netzband et al., 2002), containing high levels of arsenic, mercury, chromium, lead and organic pollutants (polychlorinated

biphenyls, PCBs, and polyaromatic hydrocarbons, PAHs). The option taken to deal with this dredged contaminated sediment is to treat it in a specially built sediment treatment plant. The dredged material is passed through rotary screens and sieves to remove the coarser fraction of the sediment, leaving behind the fine sediment fraction ( 663 mm). This fine sediment fraction contains the majority of the contaminants as a result of the high capacity of silt and clay minerals to adsorb contaminants. The resulting coarser sediment fraction is lower in contamination and can be used for building material. The fine sediment fraction is dewatered and disposed of to a specially built landfill facility.

BOX 16.6 ➤

426

16.7  Summary

➤ By removing the coarse fraction from the dredged material the volume of material which requires landfill disposal is significantly reduced. This dredging of sediment from the port also acts as a pollutant filter to the North Sea, annually removing approximately 30% of metal contaminants from the River Elbe that would otherwise be discharged to the North Sea. This management of sediment represents a site-specific approach. A more sustainable approach for contaminated sediment management is one that considers sediment management on the river basin scale, identifying and minimizing contaminant inputs at source. It is increasingly being recognized that river basin scale approaches to sediment management are the most effective, from both economic and ecological viewpoints (Owens, 2005).

Figure 16.29 Dredging operations in Port of Hamburg.

(Source: Philip N. Owens)

BOX 16.6

16.7 Summary

reconstruction of past sedimentary environments through the recent and geological past. Both natural and anthropogenically

Sedimentation is a major process acting to shape the Earth’s

induced changes in environmental conditions impact upon sedi-

surface. Sediments are derived from fragments of pre-existing

mentation processes. These changes might be climatic, physical

rocks (clastic sediments), the remains of organisms (biological

or chemical in nature and such changes can be recognized. Of

sediments) and the direct precipitation of minerals from sea-

these changes, anthropogenic impacts (e.g. engineered struc-

water (chemical sediments). These different types of sediment

tures, mining and urbanization) have the most marked and rapid

have characteristic chemical and physical properties and occur

effects, both on sediment physics and sediment chemistry. It

in a range of characteristic settings. Sediments are deposited in

should be clear that a thorough understanding of sediments and

all surface environments, from continental settings to oceanic

sedimentation processes is required for physical geographers

settings. Within these environments distinct grain composition,

and environmental scientists to interpret the Earth’s surface

grain shape and sediment bedforms are preserved. This allows the

geomorphology.

427

Chapter 16 Sediments and sedimentation

Further reading

Lowe, J.J. and Walker, M.J.C. (1997) Reconstructing Quaternary environments. Longman, Harlow. This deals extensively with methods and examples of how Quaternary environments can be reconstructed from the rock record. As such, it has a large amount of information upon sedimentary successions and would give the student a good background to ancient sedimentary deposits. It also provides useful information on loess and wind-blown deposits. Moore, J.K. and Braucher, O. (2008) Sedimentary and mineral dust sources of dissolved iron to the world ocean. Biogeosciences, 5, 631–656. A paper on some of the material discussed in Box 16.2. Nichols, G. (2009) Sedimentology and stratigraphy, 2nd edition. Blackwell Science, Oxford. This book is designed for undergraduates studying sedimentology in the earth sciences. Although latter parts of this text are

428

based on interpreting sedimentary environments from the geological record, the first half of the book provides a good, clear overview of the major processes operating on sediments, the formation of bedforms and the sediments deposited in the full range of Earth surface environments. Perry, C.T. and Taylor, K.G. (eds) (2007) Environmental sedimentology. Blackwell Publishing, Oxford. This book provides an extensive introduction to the sedimentology of contemporary Earth surface environments and the impacts of climatic and environmental change upon these environments. A large range of terrestrial, coastal and marine environments are covered, and aspects of the biology, physics and chemistry of sediments are included, as well as good case examples of sediment pollution, management and remediation. Taylor, K.G. and Owens, P.N. (2009) Sediments in urban river basins: a review of sediment-contaminant dynamics in an environmental system conditioned by human activities. Journal of Soils and Sediments, 9, 281–303. A paper that brings together research knowledge on urban river sediment contaminants.

C HAPTER 17

Soils Pippa J. Chapman School of Geography, University of Leeds

Learning objectives After reading this chapter you should be able to:

➤ appreciate the wide range of functions soil perform ➤ describe the components that make up soil and understand how they affect the physical, chemical and biological properties of soil

➤ describe the processes of soil formation and the factors that control soil development, and understand how these lead to the

(water). Soil can be defined as a complex medium, consisting of inorganic materials (such as sand, silt and clay minerals), organic matter (living and dead), water and air, variously organized and subject to dynamic processes and interactions. As well as being a major component of the natural system, soil has a key role in the use and management of the environment by humans, where it performs a wide range of essential functions:

t

features seen in soil profiles

➤ compare and contrast the different methods of soil classification ➤ define the important physical soil properties of soil texture and soil structure

➤ explain important cation–clay mineral interactions and the pro-

t

cesses that control soil pH and relate these concepts to soil fertility

➤ understand the important functions soil biota perform and the factors that control their diversity and activity

t

➤ appreciate the environmental importance of soil and understand the threats posed to soil and soil processes by human activities

17.1 Introduction Soil is a major component of the Earth’s ecosystems and

forms at the interface of the atmosphere (air), lithosphere (rocks), biosphere (plants and animals) and hydrosphere

t t

It is a medium for plant growth, providing plants with support, essential nutrients, water and air. Plant life, in turn, supports animal life. Soil is therefore able to provide humans with food, fibre and fuel. This ‘production’ function is also referred to as the ‘forestry and agricultural’ function of the soil. It acts as a reservoir for water, influencing the quantity of water in our rivers, lakes and aquifers (see Chapter 18). It has a filtering and transforming role for materials added to the soil (see Chapter 20). Thus, soil is often able to protect the quality of our air and water against many forms of harmful substances (pollutants). It recycles dead plants and animals into nutrients needed by all living things. It stores huge amounts of carbon. It is estimated that there are 15 gigatonnes (15 thousand million) of carbon in the world’s soils; three times more than in all vegetation and forests.

Chapter 17 Soils

t t

t

It provides a habitat for organisms. A handful of soil may be home to billions of organisms, belonging to thousands of species. It provides raw materials such as clays, gravels, sands and minerals as well as fuels such as peat. It also provides a physical base for the foundations of buildings and roads. It helps to protect our cultural heritage. Soils preserve a diverse range of archaeological remains which are a vital resource for understanding anthropogenic history.

Soil is, therefore, essential to humans and the maintenance of the environment. Without soil, the biosphere in which we live could not function. As Doran and Parkin (1994) stated: ‘The thin layer of soil covering the Earth’s surface represents the difference between survival and extinction for most terrestrial life.’ Soil, however, is not an unlimited resource and can be lost, degraded or improved by natural processes and human activities. As most soils take thousands or even millions of years to form, they cannot be replaced if they are washed away or polluted. Understanding the nature and distribution of soils and the processes operating within soil is therefore essential if we want to preserve our soil in a healthy state and improve our understanding of ecosystem dynamics. This chapter begins by describing the components of soil before examining soil formation processes. These processes impact the physical and chemical properties of soil, which are also described. The types of organisms found in soil and the functions they perform are also covered in this chapter. The chapter concludes by examining a range of impacts of human activities on soils and soil processes.

17.2 The components of soil Soil comprises four major components: the inorganic or mineral fraction, organic matter, water and air. The relative proportion of these four components greatly influences the physical, chemical and biological properties of a soil. In a soil, the four components are mixed in a complex way. Figure 17.1 shows that approximately half the soil volume of a typical topsoil consists of solid material (inorganic and organic); the other half consists of voids or pore spaces between the solid particles. Most of the solid material is inorganic mineral matter. However, the influence of the organic component on soil properties is often far greater than its small proportion suggests. Most agricultural soils contain between 1 and 10% organic matter and are referred to as inorganic or mineral soils because of the low organic content. 430

Air 25% Solid material

Mineral 45%

Pore spaces Water 25%

Organic 5%

Figure 17.1 The components of soil. This diagram shows the composition

by volume of a typical topsoil. The dashed line between water and air indicates that the proportion of these two components varies with soil moisture. Water and air make up the pore spaces whereas the solid material is made up from mineral and organic matter.

Air and water fill the pore spaces between solid soil particles. The relative proportion of air and water fluctuates greatly and they are inversely related to each other. Following rainfall, water fills the pore spaces, expelling much of the air. As water gradually drains away or is used by plants, air refills the pores.

17.2.1 Mineral particles In most soils the mineral fraction predominates. Mineral particles are derived from the weathering of parent material. Weathering (Chapter 14) refers to the breakdown of rocks and minerals by the action of physical and chemical processes. The larger mineral particles, which include boulders, stones, gravel and coarse sands, are generally rock fragments, whereas smaller particles are usually made of a single mineral. There are two major types of mineral particles: primary minerals, which are minerals that have changed little since they were formed, such as quartz, feldspars and micas; and secondary minerals, such as clays and oxides of iron and aluminium, which are formed from the breakdown and chemical weathering of less resistant primary minerals . The composition and size of mineral particles have a great influence on the physical and chemical properties of a soil (see Sections 17.5 and 17.6).

17.2.2 Soil organic matter Soil organic matter can be divided into three categories: (i) decomposing residues of plant and animal debris referred to as litter; (ii) resistant organic matter known as humus; and (iii) living organisms and plant roots collectively referred to as the soil biomass. Fresh plant and animal litter

17.2  The components of soil

is progressively decomposed by soil microorganisms (see Section 17.7) to a more or less stable end product called humus which is resistant to further decomposition. During the breakdown of organic matter by microorganisms, plant nutrients are released, particularly carbon, nitrogen and phosphorus. This process is called mineralization. The balance between inputs of plant and animal materials and losses by decomposition determines the amount of organic matter in the soil. The organic matter content of soils varies greatly but it usually represents between 2 and 6% by volume, but even a small amount is important as it has such a large impact upon many of the major physical, chemical and biological properties of soil. In arid climates, where less vegetation grows, soil organic matter is low (0.5–2%), whereas in environments where decomposition processes are drastically slowed, such as in waterlogged conditions, a surface accumulation of only partially decomposed material builds up to form depths of several metres of peat, which typically contains more than 50% organic matter. Soil organic matter is a very important component of soils because it (i) is the main food for soil organisms (see Section 17.7); (ii) binds mineral particles together and therefore stabilizes the soil’s structure (see Section 17.5.3) and protects it from erosion; (iii) improves water holding capacity (see Section 17.2.3); (iv) improves porosity and aeration and therefore aids the growth of plants; and (v) is a major source of nutrients and therefore influences soil fertility.

17.2.3 Soil water Water is essential to the ecological functioning of soils. Plant and soil organisms depend on water to survive. Water is also a major driving force in soil formation as it is required for parent material weathering. All chemical weathering processes depend on the presence of water. Water together with soil air fills up the pore spaces between the mineral and organic components of the soil. Soil water, however, is not pure; it contains dissolved organic and inorganic substances and is known as the soil solution. When some compounds dissolve in the soil solution, the atoms become separated as ions. An ion is an atom, or group of atoms, bearing an electrical charge. For example, when table salt (sodium chloride), which has the chemical formula NaCl, dissolves in water the atoms separate and form ions (see Figure 17.2). The sodium ions have a single positive charge and are indicated by the symbol Na+, whereas chloride has a single negative charge as indicated by the symbol Cl - . Positively charged ions are

= Na+

-

Each water molecule d+H is polar

d O

H

d +H

= Cl-

O

O

H

H

H

O H

H H

H H

O

O H

H

H

O O

H

H

H

H

H

H

O

O

H

H O

O O

H Solid

H

H H Solution

Figure 17.2 The dissolution of table salt (NaCl) in water. Since water

(H2O) is a polar molecule (one end has a different charge from the other end), the negative end is attracted to positive ions and the positive end is attracted to negative ions. Table salt completely dissolves (dissociates) in water as the water molecules keep the sodium (Na+) and chloride (Cl - ) ions apart and stop them from reforming the solid salt. The figure shows how the water molecules are linked to the sodium and chloride ions.

referred to as cations and negatively charged ions as anions. In soil solution important cations and anions include those in Table 17.1. An important function of the soil solution is to ensure the continual supply of some of these cations and anions to the plant roots. Soil solution is also the main agent of translocation, carrying dissolved ions, including pollutant ions, and small particles through the soil to surface and groundwaters. Water is held in soil by the attraction of water molecules to each other and to soil particles. Water exists as one of three states in the soil (Figure 17.3). The amount held in each state changes over time, which affects the Table 17.1 Some important cations and anions in soil solution Cations H+ Na K

Hydrogen +

+

Anions Cl -

Chloride -

Sodium

NO3

Potassium

SO42 -

Ca2+

Calcium

HCO3

Mg2+

Magnesium

OH -

Al3+

Aluminium

Nitrate

-

Sulfate Bicarbonate Hydroxide

431

Chapter 17 Soils

30

Water (%)

24

Saturation

Field capacity

18

Available water (capillary water)

12 6

Figure 17.3 Soil water states. 0

432

Field capacity

Wilting point

Wilting point

amount of water available to plants and the potential movement of nutrients and pollutants within the soil. When all the soil pores are filled with water from rainfall, the soil is described as saturated (Figure 17.3 left box). The soil, however, does not stay in this state for very long as, under the action of gravity, water will start to drain out of the larger pores and is replaced by air. This water is called gravitational water and when all of it has drained away the soil is said to be at field capacity (Figure 17.3 middle box). The small pores retain water against the force of gravity: this water is known as capillary water and represents the majority of water that is available for plant uptake. This water remains in the soil because the combined attraction of (i) the water molecules to each other and (ii) water to the soil particles is greater than the gravitational force. Capillary water moves within the soil from zones of higher potential (wet areas) to lower potential (dry areas) (see Chapter 18). The most common movement is towards plant roots and the soil surface, where it is lost by evaporation and transpiration. The final type of water is hygroscopic water, which is held as a tight film around individual soil particles (right box in Figure 17.3). This water is unavailable to plants as the attraction between the water and the soil particles is greater than the ‘sucking power’ of plant roots. A soil in which all the water is hygroscopic will appear dry although some water still remains. The drier the soil, the harder the plant has to work to obtain the remaining water held in progressively smaller pores. Eventually there comes a point when plants cannot withdraw the tightly held water from the soil and this is known as the permanent wilting point (Figure 17.4). The water retained in the soil between the states of field capacity and the wilting point is known as the plant available water or available water. The amount of water held in each state is related to a number of factors including soil texture, soil structure (see Section 17.5.3) and soil organic matter content (see Section 17.2.2). Since soil water occurs as films around soil particles, if there are many small particles (i.e. claysized particles) in a soil it will hold more water due to the

(gravitational water)

Unavailable water (hygroscopic water) Sand Sandy Loam Silt Clay loam loam loam Particle size classes

Clay

Increasing % clay Figure 17.4 The general relationship between soil texture and soil water

availability.

greater soil surface area per unit volume. However, much of the water held in soils with a high proportion of claysized particles is unavailable to plants (i.e. hygroscopic water) because it is held in very small pores. The general relationship between soil texture and soil water availability is shown in Figure 17.4. Soil structure influences the nature and abundance of soil pores and soil permeability and therefore the rate at which water drains through the soil. For example, if a soil has a lot of well-connected pores and is very permeable, water will percolate rapidly through it. Organic matter increases the soil’s moisture holding capacity and indirectly affects water content through its influence on soil structure and total pore space (White, 1997).

17.2.4 Soil air Soil air occupies pores that are not filled with water. Soil animals, plant roots and most microorganisms use oxygen and release carbon dioxide (CO2) when they respire (breathe). In order to maintain biological activity, oxygen needs to move into the soil and CO2 must move out of the soil. This ventilation of the soil is known as aeration. Aeration is affected primarily by the pore size distribution, pore continuity, the soil water content and the rate of oxygen consumption by respiring organisms. As soil air is ‘compartmentalized’ by the presence of water and intervening soil particles, the composition of the soil air differs from that of atmospheric air. Generally it has a higher moisture content, higher CO2 concentrations and lower oxygen concentrations than the atmosphere (see Table 17.2). Carbon dioxide concentrations are often several hundred times higher than that in the atmosphere. However, the composition of soil air is constantly changing with marked diurnal and seasonal fluctuations.

17.3  Soil profile

Table 17.2 The composition (% by volume) of soil air relative to the open atmosphere. Soil air

Nitrogen (N2)

79.01

79.0

Oxygen (O2)

20.96

18.0–20.8

Carbon dioxide (CO2)

0.035

O Organic horizon — surface layer dominated by the accumulation of organic matter

0.15–1.0

These changes are often associated with the differences in biological activity between night and day or between summer and winter. There will be less respiration on a cold winter’s night in a temperate zone than on a warm summer’s day and so CO2 concentrations in soil air may be much lower in winter. The composition of soil air also varies considerably from place to place in the soil.

A Mineral horizon — mineral material mixed with decomposed organic matter. Dark colour due to the presence of organic matter

Solum

Atmosphere

E Mineral horizon — depleted in clay, iron and aluminium oxides. Eluvial horizon. Lighter in colour than A and B horizons B Mineral horizon — enriched in iron, aluminium and clay minerals. Illuvial horizon

C Unconsolidated material

Reflective questions ➤ What is the difference between primary and secondary

R Bedrock

minerals?

➤ Why does soil organic matter have such an important impact on the properties of soil?

➤ What factors affect the soil’s moisture holding capacity

Figure 17.5 A hypothetical mineral soil profile showing the relative positions of the major horizons that may be present in a well-drained soil in the temperate humid region. Not all the horizons described here are present in every soil profile, and the relative depths vary.

and why?

➤ Why is the concentration of CO2 in soil air higher than that in the air above the soil?

17.3 Soil profile Soils are described by the characteristics of their soil profile. This consists of a vertical section through the soil from the ground surface down to the parent material. It is made up of a series of distinctive horizontal layers known as soil horizons. This horizontal alignment is mainly due to the translocation of materials by the movement of water through the soil. The removal of solid or dissolved material from one horizon is called eluviation, while the deposition in another horizon is referred to as illuviation. The soil horizons are given letters according to their genesis (mode of formation) and their relative position in the profile. The major horizons are shown in Figure 17.5. Note that not all the horizons described here are present in every soil. The O horizon is a surface layer dominated

by the accumulation of fresh or partially decomposed organic matter. The A horizon can occur at or near the surface (beneath the O horizon) and contains a mixture of mineral and organic (mainly humus) material and is therefore usually darker than the horizons below. Beneath this occurs the E horizon or elluvial horizon. As the E horizon is a zone of depletion (e.g. of clay, organic matter and iron) it is usually a pale, ashy colour. E horizons are common in high-rainfall areas, especially in soils developed under forests. The underlying B horizon is often a zone of accumulation (e.g. of clay, iron, organic matter and carbonates) often referred to as the illuvial horizon. In some soils, the accumulation of iron oxides in the B horizon gives it a reddish colour. The A, E and B horizons are sometimes referred to as the solum (from the Latin for soil or land). It is in the solum that the soil-forming processes are active and that plant roots and animal life are largely confined. The B horizon usually grades into the C horizon, which largely comprises unconsolidated weathered parent material known as the regolith. Although the regolith is affected by physical and chemical processes it is little 433

Chapter 17 Soils

affected by biological activity and therefore not part of the soil solum. If unweathered rock exists below the C horizon it is called bedrock and is designated the R horizon. In some soil profiles, the soil horizons are very distinct in colour, with sharp boundaries, whereas in other soils the colour change between horizons may be very gradual, and the boundaries difficult to locate. However, colour is just one of the many physical, chemical and biological characteristics by which one horizon may differ from the horizon above or below it. The informal terms ‘topsoil’ and ‘subsoil’ are often used to describe soil. Topsoil refers to the upper portion of the soil (usually the A horizon or plough horizon) and is the part most important for plant growth. The subsoil refers to the part of the soil below the topsoil (plough depth) and usually relates to the B horizon.

Reflective questions ➤ Draw a diagram of the typical soil horizons and explain their importance in determining the properties of soil.

➤ What is the difference between eluviation and illuviation?

17.4 Soil formation processes 17.4.1 Pedogenesis The process of soil formation, called pedogenesis, takes place over hundreds and thousands of years. The soil is an open system, which allows input of materials to the soil, the loss of materials from the soil and internal transfers and reorganization of these materials within the system. Soil horizons develop as a result of a number of processes occurring within the soil, which can be classified into the following categories: additions, removals, mixing, translocations and transformations. The main addition of soil material comes from the parent material of the soil. Mineral particles are released from the parent material by weathering at the base of the soil, and contribute to the lower layers of the soil. Significant additions of material come from surface accumulation, particularly of organic matter. Additions also include solutes and particles carried by precipitation and the wind, energy from the Sun and gases from the atmosphere. The main losses (removals) from the soil occur through wind and water erosion and leaching. Leaching is the removal of soil material in solution and is most active 434

under conditions of high rainfall and rapid drainage. The percolating water carries soluble substances downwards through the soil profile, depositing some in lower layers but removing the most soluble entirely (see Chapter 20). Removals also include the loss of gases and uptake of solutes by plants. Mixing of organic and inorganic components is an important process that is carried out by soil animals, microbes and plant roots, freezing and thawing of water, and shrinking and swelling of the soil. Humans also cause physical mixing of the soil by ploughing. Chemical and biological processes can also transform soil components. Organic compounds decay and some minerals dissolve while others precipitate. These transformations result in the development of soil structure and a change in colour from that of the parent material. Translocation of material within the soil profile often occurs in response to gradients of water potential (e.g. suction) and chemical concentrations within soil pores. Suspended and dissolved substances may move up or down through the soil profile. The net result of these soil forming processes occurring over a long period of time is the formation of different soil horizons. However, the processes that dominate at a particular site are dependent on the environmental conditions at that site. In areas where rainfall exceeds evapotranspiration, net water movement is down through the soil (Figure 17.6a). The extent of leaching is often indicated by the acidity of the soil (Jarvis et al., 1984). In many freely draining soils, clay is carried from the upper horizons by percolating water to lower horizons and this is known as clay eluviation, or lessivage (Figure 17.6b). The clay is redeposited as skins or coats on the surfaces of aggregates or in pores and around stones. Soil horizons characterized by clay accumulation are described as argillic. Clay eluviation tends to produce a group of soils known as acid brown earths or luvisols (Figure 17.7). Podzolization may occur in soils where there is intense leaching and translocation of material (Figure 17.6c). Organic acids complex with iron and aluminium compounds that are transported downwards from the E horizon by percolating water and deposited in the B horizon. Podzolization occurs on freely drained sites under forests and heath plants and the end product of this process is a soil called a podzol (Figure 17.8), the characteristics of which are the presence of an organic layer, a leached E horizon and an accumulation of iron, aluminium and humic material in the B horizon. These soils are not very productive for agriculture because they are acidic and the free drainage results in leaching of fertilizers away from plant roots.

17.4  Soil formation processes

(a)

Leaching

cm 0

(b)

Clay eluviation

Loss of clay Fe, Al, humus

Clay

Losses N, Ca, Mg, Na and K

Some precipitation of calcium

Podzolization Accumulation of acid humus or peat

Depletion of bases

Losses N, Ca, Mg, Na and K

(c)

Accumulation of clay

Losses N, Ca, Mg, Na and K

Depletion of Fe and Al

Precipitation of Fe, Al, humus

80

Drainage water

Drainage water

(d)

Drainage water

(e) Gleying Intermittent aeration in presence of organic matter

Slowly permeable subsoil or fluctuating groundwater

Slow drainage

Mobilization and local precipitation of ferric iron

(f) Laterization

Evaporation

Intense weathering; loss of Si. Residual accumulation of Fe and Al

Losses Si, N, Ca, Mg, K and Na

Drainage water

May be formation of kaolinite

Figure 17.6 The movement of water

in the soil-forming processes of (a) leaching, (b) clay eluviation, (c) podzolization, (d) gleying, (e) laterization and (f) salinization. (Source: (a)–(d) Reproduced from National Soil Resources Institute, Bulletin 10, Jarvis et al. (1984) ‘Soils and their use in Northern England’, Soil Survey of England and Wales, Rothamsted Experimental Stations, Harpenden, Herts, Fig. 14, p. 47. © Cranfield University 1984. No part of this publication may be reproduced without the express permission of Cranfield University; (e) and (f) adapted from McRae, 1988)

Salinization Salt crust Deposition of soluble salts

Upward transmission of Na, Mg, K, Cl, SO4

Upward movement of saline groundwater

Figure 17.8 A podzol from the Upper Wye catchment, Wales. (Source: Figure 17.7 A brown earth soil. (Source: Davey Jones)

Chris Evans) 435

Chapter 17 Soils

In many locations waterlogging leads to the reduction, mobilization and removal or redeposition of iron compounds in the soil (Figure 17.6d). The reduction of ferric (Fe3+) to the more mobile, grey ferrous (Fe2+) iron compound (see Chapter 15) by microorganisms is known as gleying. The soil loses the brown/red colour of ferric oxide and becomes grey or bluish. Alternate phases of reduction and oxidation due to fluctuations in the water content result in soil having a mottled appearance with brown/red iron oxide spots or streaks occurring along root channels and larger pores as shown in Figure 17.9. Laterization (ferralitization) occurs in tropical and subtropical soils where high temperatures and heavy rain result in intense weathering and leaching (Figures 17.6e and 17.10). Almost all the by-products of weathering are leached out of the soil leading to the development of horizons depleted in base cations (e.g. calcium, magnesium, potassium and sodium) and enriched in silica and oxides of aluminium and iron (McRae, 1988). The red colour of these soils is due to the presence of haematite and goethite (Figure 17.10). Conversely in areas where evapotranspiration exceeds rainfall, such as arid and semi-arid areas, water is drawn to the soil surface and as water evaporates salts are precipitated at or near the surface (Figure 17.6f; see also Box 18.4 in Chapter 18). This salinization process is almost the complete opposite of leaching. Figure 17.9 A stagnohumic gley. (Source: John Conway)

Figure 17.10 A red laterite soil (also

known as an oxisol) formed under a wet, humid and warm tropical climate, Brazil. (Source: Phil Haygarth)

436

17.4  Soil formation processes

3000 2700 2400 2100 1800 1500 1200 900 600 300 0

Polar desert

Temperature (5C)

Evaporation or precipitation (mm)

Climate is perhaps the most influential factor affecting soil-forming processes as it determines the moisture and temperature regimes under which a soil develops. In addition, climate is influential in determining vegetation distribution (see Chapters 10 and 13). Rainwater is involved in most of the physical, chemical and biological processes that occur within the soil, and particularly weathering and leaching. To be effective, however, water must pass downwards through the whole of the soil profile and into the regolith. The amount of precipitation that percolates downwards through the soil is mainly related to total annual precipitation and rate of evaporation (from vegetation and soil), although topography and permeability of the parent material are also important factors. Overall, percolating

Tundra

Taiga zone

17.4.2.2 Parent material Soils may develop on the weathered surfaces of exposed, consolidated in situ rock surfaces, or unconsolidated superficial material that has been transported and deposited by gravity, water, ice or wind. Parent material influences soil formation through the process of weathering and then through the influence of the weathered material on soil processes. Rock types influence the rate of weathering through their mineralogical composition and the surface area of the rock exposed. The larger the exposed surface area, the faster the rate of weathering. Some minerals are more susceptible to weathering than others.

Semi-desert and desert

Tropical forest zone

Savannas

17.4.2.1 Climate

Savannas

The major processes involved in soil formation described above are controlled by local and regional environmental factors. In the late 1800s, Dokuchaiev, a Russian scientist, was one of the first to recognize that soils do not occur by chance but usually form a pattern in the landscape and develop as a result of the interplay of climate, parent material, organisms and time. Building on this work in the 1930s and 1940s, Hans Jenny suggested that topography was an additional important factor (Jenny, 1941).

water stimulates weathering processes, which helps to differentiate the soil into horizons and influences soil depth. The main effect of temperature on soils is to influence the rate of soil formation via mineral weathering and organic matter decomposition. For every 10°C rise in temperature, the speed of chemical reactions increases by a factor of two or three; biological activity doubles, up to around 30–35 °C, and evaporation of water increases. As rates of chemical weathering are greatest under conditions of high temperature and humidity, soils in tropical areas are often several metres deep while those in polar regions are shallow and poorly developed (Figure 17.11). In addition, soils are influenced by microclimates that are related to altitude and aspect.

Steppes

17.4.2 Factors affecting soil formation

Temperature

25 20 15 10 5 0

Evaporation n ipitatio Prec

Fresh rock Little chemical alteration Illite— montmorillonite Kaolinite Al2O3 Fe2O3 + Al2O3

Figure 17.11 Schematic representation of the variation of soil depth with climate and biome from the equator to the north polar region. Soils are

deeper in the wet humid tropics and in the temperate zone and most shallow in dry or very cold locations. The weathering products of aluminium and iron oxides are also shown. (Source: after Strakhov, 1967, as adapted in Birkland, 1999, Fig. 10.5, p. 274) 437

Chapter 17 Soils

Goldich (1938) proposed a ‘stability series’ for the silicate minerals (see Figure 14.21 in Chapter 14). This arrangement of minerals is in the same form as Bowen’s reaction series, where the silicate minerals are placed in their order of crystallization (also shown in Figure 14.21). The minerals that crystallize first form under much higher temperatures than those that crystallize last. Consequently, the minerals that crystallize first, such as olivine and pyroxene, are not as stable at the Earth’s surface, where the temperature and pressure are very different from the environment in which they form. In contrast, quartz, which crystallizes last, is the most resistant to weathering. Knowledge of rock mineralogy allows rocks to be placed in their order of susceptibility to weathering. Hard igneous rocks and Carboniferous and Jurassic sandstones weather slowly to give shallow, stony, coarse-textured soils. In contrast, softer Permo-Jurassic sandstones weather more rapidly to give deeper, less stony, loamy or sandy soils. The soils that develop on all these parent rocks are generally acidic owing to the low base cation content of these rocks or the bases are leached from the soil faster than they are replenished by weathering. Carboniferous, Permo-Triassic and Jurrassic clays, siltstones, mudstones and shales are all fine-grained rocks which weather to give silty or clayey soils which are generally slowly permeable. The weathering products of chalk and limestone are very soluble, and therefore soil depths are often shallow, particularly on steeper slopes. At the foot of the slope, where deeper soils form, they are well drained and base rich. Further information about parent material weathering can be found in Chapter 14.

soils on steeper slopes are drier as most precipitation runs off the surface or through the upper horizons to lower ground. This produces the pattern of soil distribution illustrated in Figure 17.12a. On slopes with very permeable parent material water tends to penetrate to the subsoil, leaving the higher ground and steep slope well drained, whereas soils on the lower slopes and valley bottoms are more likely to be affected by groundwater as shown in Figure 17.12b. Milne (1935) was the first to use the term soil catena for topographically determined soil profiles in East Africa. Where there is no change in the geology along the slope, soil differences in the catena are brought about by drainage conditions, differential transport, eroded material and the leaching, translocation and redeposition of mobile chemical constituents. Aspect affects the solar energy received at the ground surface. In the northern hemisphere, south-facing slopes receive more and are therefore warmer and generally lower in moisture than north-facing slopes. Consequently, soils on the south slopes tend to be drier, less densely vegetated,

Overland flow and near-surface (shallow) runoff

(a)

Soil class

(b)

Surface water gley soils

Stagnogleyic brown earths

Winter water table

438

Stagnogleyic brown earths

Surface water gley soils

Summer water table

17.4.2.3 Topography Topography relates to the altitude, slope and aspect of the landscape and can hasten or delay the influences of climatic factors. Slope steepness is an important factor, as steeper slopes reduce the amount of water infiltrating and percolating through the soil and allow increased erosion of the surface layers. Therefore soils formed on steeper slopes tend to be thin, coarse textured and poorly developed compared with soils on gentler slopes or more level terrain. However, weathering rates tend to be greater on steeper slopes, although the weathering products do not accumulate very deeply as they are efficiently removed by erosion. For example, 90% of the dissolved material in the rivers of the Amazon basin comes from the steep Andes Mountains which only cover 12% of the basin (Gaillardet et al., 1997). On slopes with less permeable parent material, surface waterlogging causes gleying on flat ground, whereas the

Brown earths

Peat

Soil class

Brown earths or podzols

Groundwater gley soils

Peat soils

Gleyic brown earths or gley—podzols Never waterlogged, unmottled

Seasonally waterlogged, strong mottling

Occasionally waterlogged, slight mottling

Permanently waterlogged, typically grey or bluish grey

Figure 17.12 Relationship between slope, hydrology and soil forma-

tion on (a) slowly permeable parent material and (b) permeable parent material. (Source: Reproduced from National Soil Resources Institute, Bulletin 10, Jarvis et al. (1984) ‘Soils and their use in Northern England’, Soil Survey of England and Wales, Rothamsted Experimental Stations, Harpenden, Herts. Fig. 16, p. 52 © Cranfield University 1984. No part of this publication may be reproduced without the express written permission of Cranfield University)

17.4  Soil formation processes

and thus lower in organic matter. These differences are reversed in the southern hemisphere. Altitude influences climate (Chapter 9). Temperature declines with altitude and precipitation tends to increase with altitude in the middle latitudes. This leads to an excess of rainfall over evaporation and as a result leaching rates are high and waterlogging occurs where the drainage is poor. The lower temperatures also lead to a reduction in biological activity and therefore slower decomposition of organic matter. This in turn leads to the accumulation of thick organic horizons at the surface and ultimately to the formation of peat.

17.4.2.4 Organisms Organisms include plants, animals, microorganisms and humans. Vegetation extracts water and nutrients from the soil and under natural conditions returns most of the nutrients it uses to the soil in litter. The type of vegetation influences the type and amount of litter that is returned to the soil. Different soil types support different vegetation communities. Vegetation also protects the soil from water and wind erosion by intercepting rainfall, decreasing the velocity of runoff, binding soil particles together, improving soil structure and porosity, and providing a litter cover which protects the soil surface against raindrop splash. Earthworms and other small animals such as moles mix and aerate the soil as they burrow through the soil. Earthworms have been found to increase the infiltration rate of fine-textured soils and contribute towards increasing the stability of the soil structure by intermixing organic matter with mineral particles (Curtis et al., 1976). Soil organisms, including fungi, bacteria and single-celled protozoa, play a major role in the decomposition of organic matter (see Section 17.7). The end product is humus. Humans influence soil formation through manipulation of vegetation, agricultural practices such as drainage and irrigation, the additions of fertilizers, lime and pesticides, and urban and industrial development.

17.4.2.5 Time Over time soil is continually forming from the parent material, under the influence of the climate, topography, vegetation and soil organisms. Soil genesis is a long process; the formation of a layer 30 cm thick can take from 1000 to 10 000 years. During this time, the properties of the soil continually change. This is manifest by changes in the soil profile including the number of horizons, their depth and their degree of differentiation. When the rate of change of a

soil property with time is negligible, the soil is said to be in steady state. However, in reality soil rarely reaches this state because of changes in one of the environmental factors. For example, changes in the world’s climate over geological time accompanied by changes in sea level, erosion and deposition have produced large changes in the distribution of vegetation and parent material. Therefore most soils have not developed under a single set of environmental factors but have undergone successive waves of pedogenesis. The most recent large change in climate resulted in alternating glacial and interglacial periods of the Pleistocene (Chapter 4). In high and middle latitudes, glaciation removed the majority of soils and covered large areas with drift material. Therefore, soil development in these areas began again on new surfaces after the final retreat of the ice during the Holocene which began about 11 700 years ago (Chapter 5).

17.4.2.6 Combined influences It can be seen that the five factors influencing soil formation do not operate as single independent factors. Climate influences vegetation and human activities and is itself affected by topography. Vegetation is influenced by climate and parent material. The combined influence of the five factors produces a set of soil-forming processes, which results in the world’s distinctive soil profiles. Not all soils develop the same amount or combination of horizons and therefore specific combinations of horizons are used to classify soils. Box 17.1 provides details on soil