Sustainability Science for Watershed Landscapes. 9789814279956, 9814279951

358 89 24MB

English Pages [369] Year 2015

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Sustainability Science for Watershed Landscapes.
 9789814279956, 9814279951

Citation preview

25mm

3mm

3mm

3mm

9mm

6in

9mm

3mm

SUSTAINABILITY SCIENCE FOR WATERSHED LANDSCAPES

roumasset burnett balisacan

SUSTAINABILITY SCIENCE FOR WATERSHED LANDSCAPES

Sustainability science addresses the central challenge facing global society of how to reduce poverty and meet demands of a large human population desiring a good life while simultaneously maintaining the environment that provides the life support system on which long-term prosperity depends. This book provides evidence and insight into key elements of what is required to achieve sustainability by framing important policy questions and illustrating the consequences of policy alternatives in systems with complex interactions. — Stephen Polasky, University of Minnesota Sustainability Science can be both fundamental and practical, both deep and interdisciplinary.  This application of Sustainability Science to Pacific watersheds illustrates its promise. — Peter Vitousek, Stanford University

Southeast Asian Regional Center for Graduate Study and Research in Agriculture (SEARCA) College, Los Baños, Laguna Philippines 4031

FOR WATERSHED LANDSCAPES

James A. Roumasset Kimberly M. Burnett Arsenio M. Balisacan

PIC199h ISBN 978-981-4279-60-4

Southeast Asian Regional Center for Graduate Study and Research in Agriculture

9 789814 279604

9mm

25mm

9mm

6in

3mm

3mm

3mm

ENVIRONMENT

SUSTAINABILITY SCIENCE Editors

3mm

ISEAS Publishing Institute of Southeast Asian Studies 30 Heng Mui Keng Terrace Pasir Panjang Singapore 119614

9in

Reproduced from Sustainability Science for Watershed Landscapes edited by James A. Roumasset, Kimberly M. Burnett and Arsenio M. Balisacan (Singapore: Institute of Southeast Asian Studies, 2010). This version was obtained electronically direct from the publisher on condition that copyright is not infringed. No part of this publication may be reproduced without the prior permission of the Institute of Southeast Asian Studies. i Individual articles are available at < http://bookshop.iseas.edu.sg >

Sustainability Science for Watershed Landscapes

ii

The Southeast Asian Regional Center for Graduate Study and Research in Agriculture (SEARCA) is one of the regional research and training centres of the Southeast Asian Ministers of Education Organisation (SEAMEO), an intergovernmental body founded in 1965 to promote cooperation among Southeast Asian nations through activities in education, science, and culture. SEARCA’s programmes are designed to strengthen institutional capacities in agricultural and rural development in Southeast Asia through graduate education, short-term training, research, and knowledge exchange. It is hosted by the Philippine Government on the campus of the University of the Philippines Los Baños, which is based in Los Baños, Laguna, Philippines. It is supported by donations from SEAMEO member and associate member states, other governments, and various international donor agencies.

The Institute of Southeast Asian Studies (ISEAS) was established as an autonomous organisation in 1968. It is a regional centre dedicated to the study of socio-political, security and economic trends and developments in Southeast Asia and its wider geostrategic and economic environment. The Institute’s research programmes are the Regional Economic Studies (RES, including ASEAN and APEC), Regional Strategic and Political Studies (RSPS), and Regional Social and Cultural Studies (RSCS). ISEAS Publishing, an established academic press, has issued more than 2,000 books and journals. It is the largest scholarly publisher of research about Southeast Asia from within the region. ISEAS Publishing works with many other academic and trade publishers and distributors to disseminate important research and analyses from and about Southeast Asia to the rest of the world.

iii

Sustainability Science for Watershed Landscapes

James A. Roumasset Kimberly M. Burnett Arsenio M. Balisacan Editors

SOUTHEAST ASIAN REGIONAL CENTER FOR GRADUATE STUDY AND RESEARCH IN AGRICULTURE Laguna, Philippines

INSTITUTE OF SOUTHEAST ASIAN STUDIES Singapore

iv First published in Singapore in 2010 by ISEAS Publishing Institute of Southeast Asian Studies 30 Heng Mui Keng Terrace Pasir Panjang, Singapore 119614 E-mail: [email protected] Website: http://bookshop.iseas.edu.sg and Southeast Asian Regional Center for Graduate Study and Research in Agriculture (SEARCA) College, Los Baños, Laguna, Philippines 4031 E-mail: [email protected] All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the Institute of Southeast Asian Studies. © 2010 Institute of Southeast Asian Studies, Singapore The responsibility for facts and opinions in this publication rests exclusively with the authors, and their interpretations do not necessarily reflect the views or the policy of the publishers or their supporters. ISEAS Library Cataloguing-in-Publication Data Sustainability science for watershed landscapes / edited by James A. Roumasset, Kimberly M. Burnett and Arsenio M. Balisacan. 1. Water resources development—Congresses. 2. Watershed management—Congresses. I. Roumasset, James A. II. Burnett, Kimberly M. III. Balisacan, A. M. IV. International Conference on Sustainability Science for Watershed Landscapes, (2007 : Honolulu, Hawai‘i). HD1690.5 S96 2010 Singapore:

ISBN 978-981-4279-96-3 (soft cover) ISBN 978-981-4279-60-4 (hard cover) ISBN 978-981-4279-95-6 (E-Book PDF)

Cover page photo by Destin Bradwell available at www.destinationsgallery.com. Printed in Singapore by Utopia Press Pte Ltd

v

CONTENTS

Tables Figures Message Foreword Preface Acknowledgments List of Contributors

vii viii xii xiv xvi xvii xix

Theme 1: Sustainability Science for Resource Management and Policy 1 2 3 4

Economic Policy for Sustainable Development vs. Greedy Growth and Preservationism Majah-Leah V. Ravago, James A. Roumasset, and Arsenio M. Balisacan Integrated Watershed Management: Trees, Aquifers, Reefs, and Mud Sittidaj Pongkijvorasin and James A. Roumasset Transdisciplinary Research in Watershed Conservation: Experiences, Lessons, and Future Directions Chieko Umetsu, Makoto Taniguchi, Tsugihiro Watanabe, and Shigeo Yachi Payments for Ecological Services: Experiences in Carbon and Water Payments in the Philippines Rodel D. Lasco and Grace B. Villamor

3 47 77 103

Theme 2: Monitoring and Modelling 5 6

Cyberinfrastructure for Sustainability in Coupled Human – Environment Systems Micheal H. Kido, Kenneth Y. Kaneshiro, and Kevin N. Montgomery Watershed Management for Sustainability in Tropical Watersheds: An Integrated Hydrologic Modelling Approach Ali Fares

127 141

vi 7

Watershed Assessment, Restoration, and Protection in Hawai‘i Aly I. El-Kadi and Monica Mira

167

Theme 3: Participatory Approaches 8 9

Landscape-Scale Conservation: Fostering Partnerships through Ecosystem Service Approaches Rebecca L. Goldman and Gretchen C. Daily A Participatory Approach to Community Resource Management: Building Upon Local Knowledge and Concerns James B. Friday and Harold J. McArthur Jr.

195 219

Theme 4: Case Studies 10 Changing Land Use in the Golden Triangle: Where the Rubber Meets the Road Maite Guardiola-Claramonte, Jefferson M. Fox, Thomas W. Giambelluca, and Peter A. Troch 11 Effects of Feral Pigs (Sus scrofa) on Watershed Health in Hawai‘i: A Literature Review and Preliminary Results on Runoff and Erosion Gregory L. Bruland, Chad A. Browning, and Carl I. Evensen 12 Sustainable Water Quality Management for Pacific Island Watersheds: Illustrations from American Samoa, Hawai‘i, and Micronesia Carl I. Evensen 13 The Transboundary Management of Groundwater Resources in Kumamoto, Japan Jun Shimada

235

251

279

311

Synthesis 14 Sustainability Science: Overview and Directions for Further Research Kimberly M. Burnett and James A. Roumasset

329

vii

TABLES

2.1 2.2 3.1 3.2 6.1 6.2

8.1 11.1 12.1 12.2 12.3 12.4 12.5

Equations Used in the Model Summary of the Main Results (study state) from All Scenarios Characteristics of Research Organisations Tools for Integration Predominant Soils of the Upper Hanalei Watershed and Their Basic Characteristics Soil Erosion and Sediment Yield for October-December 2003 Estimates with Original and Modified Versions of N-SPECT, Using Two Different LS Factor Grids, and Compared to Actual Sediment Yield Summary of Lessons Learned From Case Studies Characteristics of the Eight Sites in the Mānoa Watershed Water Issues Ranked as High Priority (extremely important + very important) by Pacific Island Respondents Water Issues Ranked as Most Important (1) to Least Important (10) by Residents in Each Pacific Island Responses of Pacific Island Respondents to Sources of Drinking Water Responses of Pacific Island Respondents on Safety of and Satisfaction with Drinking Water in the Home Primary and Secondary Sources of Known or Suspected Water Quality Contaminants per Island

57 62 96 97 149

159 204 264 290 291 292 293 294

viii

FIGURES

1.1 1.2 1.3 1.4 2.1 2.2 2.3 2.4 2.5 2.6 2.7 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 4.1 4.2

Barbier’s Venn Diagram (adapted from Barbier 1987) Positive Sustainability Rent-seeking and the Iron Triangle Unsustainable and Sustainable Development Optimal Paths of Head Level with Different Marine Algae Stock Constraints Optimal Paths of Ground Water Extraction with Different Marine Algae Stock Constraints Open Access vs. Socially Efficient Effort Transition to New Steady-State Levels Due to Adding and Subtracting One Square Kilometer of Mangroves Total Revenues from Adding and Subtracting One Square Kilometer of Mangroves Revenue, Cost, and Profit of Fishery Production Under Sole Ownership Net Revenue of Depleting/Growing Area of Mangrove Under A Regulated Fishery Stages of Integrated Project-type Research Six Research Groups of ICCAP Impact Assessment Flow and Scenarios in ICCAP Framework of Analysis for Integrating SG Outputs in ICCAP Lake Biwa-Yodo River Project – Four Research Groups Four Research Groups in Lake Biwa-Yodo River Project Human Activities on Urban Subsurface Environments Six Research Groups of Urban Subsurface Environments Project Net Carbon Sequestration Under Various Scenarios of the LLDA Project in Tanay, Rizal Estimated Net Cumulative CO2e Removals by the Proposed Kalaban Reforestation Project, Philippines (Villamor and Lasco 2006)

6 12 18 27 61 61 65 66 67 68 69 80 84 85 86 87 89 92 93 111 114

ix 5.1

5.2 5.3

6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 6.9 6.10 6.11 6.12 6.13 7.1

InteleView – Virtual Earth Image of: (A) Kaua‘i in the Hawaiian Islands; (B) Project area on North Kaua‘i, Circles Indicate Locations of Sensor Deployments Transmitting Data Wirelessly to the InteleSense Server; (C) Location of Waipa Valley/Ahupua‘a System Architecture, InteleCell Hardware, and Sensor Types Currently Supported Example of Near-Real Time Data Observed in InteleView, Used to Evaluate Flood Response in Waipa Stream and Taro Irrigation Ditch, to Torrential Rainfall Occurring in Lower Waipa Valley over a five-day Period in Nov-Dec 2006 Hanalei Watershed Located in the North of the Island of Kaua‘i, Hawai‘i Measured Precipitations for the 2003-2004 Simulation Period at USGS Rain Gauges at the Watershed Outlet and on Mt. Wai‘ale‘ale Spatial Distribution of Average Annual Precipitation in Hanalei Watershed as Determined (a) Using the PRISM Model and (b) Hawai‘i Atlas Daily Average Stream Flow in Hanalei river at the USGS Gauging Station Based on 49 Years of Record (1963-2004) Land Cover Map of the Study Area AnnAGNPS Daily Precipitation and Runoff From 1999 and 2001 at the Watershed Outlet and Mt. Wai‘ale‘ale (a) Spatial Distribution of CN Values for C-CAP Land Cover and (b) Estimated Runoff Volume (per grid cell) Spatial Distribution of Average Annual Sediment Load from Hanalei Watershed Using AnnAGNPS model Total Sedimentation Contribution Versus Percent Area of Watershed Contributing to the Sediment Loads LS Factor Grids Applied in the Simulation of Soil Erosion Sediment Delivery Ratios and Sediment Yield in Hanalei Bay Watershed Predicted with the Modified Version of the N-SPECT Feral Pig Disturbance in Hanalei Watershed The Nāwiliwili Watershed, Kaua‘i, with its Main Perennial Streams and their Respective Basins (Hawai‘i State 2009)

131 132

136 145 147 148 148 150 151 152 154 155 156 158 158 160 171

x 8.1

8.2 9.1 10.1 10.2

10.3 10.4 11.1 11.2 11.3 11.4 11.5 11.6 11.7 11.8 12.1 12.2.

Three Possible Incentives or Partnerships that Could Encourage the Provision of Ecosystem Services from Working Landscapes, Depending on the Size Parcels in the Landscape and the Familiarity of the Landowners with Each Other Ecosystem Service Projects (ES project) Engage Significantly More Private, Corporate Stakeholders in Conservation Efforts than Traditional Approaches (BD project) Map of Timor Leste MMSEA Study Region (a) Location of the Experimental Catchment Nam Ken, in South Xishuangbanna, (b) Land Cover and Location of the Micrometeorological (triangles) and Soil Moisture Stations (squares) in Nam Ken Studied Land Covers: (a) Rubber, (b) Tea, (c) Secondary Forest, (d) Grassland Data on Daily Albedo Values and Evaporation Fraction ( λ) in the Rubber Plantations, 2005 Map of the Variations in Slope Across the Watershed with the Locations of All Eight Sampling Sites The Basic Layout of Each Site in Study. Each Site was Oriented to Achieve Similar Slope and Vegetative Characteristics for Both Plots Runoff Collection Apparatus with Lid Removed Mean Bulk Density Across the Eight Study Sites (error bars represent ±1 standard error) Mean Volumetric Soil Moisture across the Eight Study Sites (error bars represent ±1 standard error) Mean Monthly Total Suspended Solids Concentrations for All Eight Study Sites from November 2007 to February 2008 (error bars represent ±1 standard error) Mean Total Suspended Solid Concentrations for the Fenced and Unfenced Plots in Each Site (error bars represent ±1 standard error) Differences in Total Suspended Solid Concentrations Between Fenced and Unfenced Plots Over Time. The Negative Trend Represents Higher TSS Amounts in the Unfenced Plots. Location of the American-affiliated Pacific Islands (adapted from Carruth, 2003) Pacific Island Streams are Typically Short with Steep Gradients and Small Drainage Areas and Have Highly Variable Flow (Waipa stream, Kaua‘i, Hawai‘i; Photo by Carl Evensen)

201 202 223 236

240 243 245 265 266 267 269 270 271 271 273 281 282

xi 12.3 12.4 12.5 13.1 13.2 13.3 13.4 13.5 13.6 13.7 13.8 13.9 13.10 13.11 13.12 13.13 13.14

The Effects of Groundwater Withdrawal on a Freshwater Lens (adapted from Carruth 2003) Rapidly Developing Rural Landscapes Abound in the Pacific Islands (Amanave Village, American Samoa; Photo by Carl Evensen) Island Life is Physically, Economically, and Culturally Linked to the Surrounding Oceans (Tumon Bay, Guam; Photo by Glen Fukumoto) Location of the Study Area Surface Geologic Map of the Study Area Groundwater Potential Distribution of the No.2 Aquifer in the Kumamoto Area (October, 1993) Stable Oxygen Isotope Content (δ18O ‰) in the No.2 Aquifer Groundwater and Shira River Water (Kosaka, H. et al. 2002) Annual Changes in Groundwater Levels at the Suizenji Observation Well Annual Changes in Groundwater Levels at Ozu in the Midstream Area of the Shira River Annual Changes in Spring Discharge at Lake Ezu “Balance” of Groundwater in the Kumamoto Region Annual Changes in Groundwater Withdrawal Uses in the Kumamoto City Changes in Urban Land Use from 1965 to 1997 in the Kumamoto Area Changes in Non-agricultural Land Use in the Shira River Midstream Area from 1930 to Present (Ichikawa, 2004) Effect of Paddy Rice Irrigation on Annual Discharge Rate of the Shira River Schematic View of the Creation of Groundwater Resources through Rice Paddies Effects of Artificial Recharge through Abandoned Paddy Fields

283 302 304 312 313 314 315 316 317 317 318 319 319 320 321 323 324

xii

Message

Cynicism surrounds the concept of “sustainable development” and its approaches, for at least two reasons. Firstly, the concept, as practiced, is nebulous – meaning different things to different people, even among serious researchers and policy analysts. Indeed, many vociferous advocates of sustainable development have one thing in common – they tend to speak a language other than what we know as science. Secondly, evident success or sustained impact of approaches to sustainable development in many parts of the world is quite sparse and wanting. A useful point of departure is thus to define sustainability science at the outset. As William Clark succinctly puts it, sustainability science is a field defined by the problems it addresses rather than by the disciplines it employs, much like health science and agricultural science. In particular, it is a field that seeks to, and I quote, “facilitate a transition toward sustainability – that is, improving society’s capacity to use the earth in ways that simultaneously meet the needs of a much larger but stabilizing human population, … sustain the life support systems of the planet, and … substantially reduce hunger and poverty” (Our Common Journey: A Transition Toward Sustainability, National Research Council Policy Division Board on Sustainable Development, 1999). Clark sees sustainability science “transcend(ing) the concerns of its foundational disciplines and focus(ing) instead on understanding the complex dynamics that arise from interactions between human and environmental systems.” Bringing together the various sciences along with other dimensions of human thought to address sustainability’s concerns, including the social goals of sustainable development and humanity’s well-being, is, to say the least, exciting. This is indeed synergy of a 21st century kind! On the other hand, the challenges facing this new field are huge. For the most part, sustainability science requires detailed information synthesized into new knowledge, enriching and even revolutionizing present paradigms and methods of research and policy design. Interestingly, the transdisciplinary nature and requirements of this new science, which is a key to its promise of generating new knowledge, is in itself a challenge. Particularly for us scientists and researchers, it may mean having to change our paradigms and ways of looking at our world, modifying our way of viewing and doing science, stepping out of the neat little boxes we have fashioned for our respective disciplines, and developing new ways of talking and working with each other as well as nonscientists and other stakeholders who similarly aspire for society’s well-being. Perhaps the ultimate promise, and at the same time challenge, of this new field is its systems perspective of problem-

xiii solving – indeed, to be able to address human-and-environment concerns in ways that do not beget bigger problems nor bypass more significant issues. Just thinking of its possibility is overwhelming and exciting. Key to making this transdisciplinary initiative work is placing the policy question or issue at the forefront of the research agenda. How do the different chapters in this volume contribute to an improved understanding of what works – and what does not – in the area of watershed management? The other element of this initiative is the recognition that one-size-fits-all approach to policy design aimed at addressing sustainability concerns across landscapes seldom works in practice. What may work well in a Hawai‘i landscape will not necessarily work well in a Philippine landscape and vice versa. Indeed, because the bio-physical, socio-economic, and institutional conditions are quite diverse across landscapes, even within a country, the choice of appropriate policy and technology levers for addressing resource management issues is extremely challenging. It is widely believed that the present development path of the world is not a sustainable one. As a development organization, SEARCA sees in sustainability science a way of carrying out its mandate of agricultural and rural development for Southeast Asia yet managing the natural base on which most of these agricultural and rural activities depend.

Arsenio M. Balisacan Former SEARCA Director

xiv

FOREWORD

The goal of Sustainability Science for Watershed Landscapes was to synthesise the third wave of sustainability science and make science directly relevant to specific pressing policy questions. Current efforts are grounded in almost four decades of international concern commencing with the Stockholm Conference on the Environment (1972), and including, to mention only a few of the key initiatives, the Brundtland Commission (1983-1987) and the resulting report, Our Common Future (1987) and the four year Millennium Ecosystem Assessment (2005). The Assessment addressed environmental, social sustainability, and economic sustainability to meet fundamental human needs while preserving the earth’s life support systems. The challenges facing us early in the 21st century are so complex that they demand trans-disciplinary approaches, coupled models, multi-scale assessment, and the involvement of stakeholders in defining our basic questions. These challenges are heightened by a changing climate. This volume presents important case studies highlighting these approaches. It was logical for the East-West Center (EWC) to partner with the Southeast Asian Regional Center for Graduate Study and Research in Agriculture (SEARCA) and the University of Hawai‘i in convening the 2007 “International Conference on Sustainability Science for Watershed Landscapes” that resulted in this volume. The EWC has had a sustained partnership with SEARCA and its parent organisation, the Southeast Asia Ministers of Education Organisation (SEAMEO). Furthermore, since at least the mid-1970’s the EWC has been committed to the policy aspects of equitable access to resources in the Asia Pacific region. A milestone was the creation of what was then the Resource Systems Institute (1977) under the leadership of Harrison Brown. In the same year the Environment and Policy Institute (EAPI) was established under the leadership of William Mathews to study human interaction in tropical ecosystems. Both RSI and EAPI concentrated on international comparative studies with an Asian focus and hosted many Asian visiting researchers often leading to long-term research partnerships. Although there has been institutional change, the focus on the policy aspects of equitable access to resources is continued today by the Center’s Environment, Vulnerability and Governance concentration, led by Jefferson Fox, a co-organiser of the meeting and a contributor to this volume. Place-based understanding makes an important contribution to sustainability science. Hawai‘i is a logical place in which to have hosted this dialogue. Indigenous Hawaiian land management practices are rooted in the ahupua‘a – the watershed comprised of the mountaintop to the reef’s edge. This volume includes valuable

xv examples from O‘ahu, Kaua‘i and the Island of Hawai‘i as well as Micronesia, American Samoa, Thailand, the Philippines, and Timor Leste, Xishuangbanna, Yunnan, Japan and Guatemala and Brazil in Latin America. The East-West Center is committed to continued partnerships with colleagues and institutions in the Asia Pacific, the US and beyond in furthering the frontiers of sustainability science.

Nancy D. Lewis Director, Research Programme East-West Center

xvi

Preface

Globally, we must confront our inability to meet the demands of unchecked population growth, rapid depletion of natural resources, increasing energy needs, and the world-wide deterioration of aquatic and terrestrial ecosystems. In short, the current behaviour of humans on this planet is not sustainable. In August of 2006, the University of Hawai‘i at Mānoa (UHM) signed a Memorandum of Understanding with the Southeast Asian Regional Center for Graduate Study and Research in Agriculture (SEARCA). The goal of the UHMSEARCA initiative was to synthesise a third wave in sustainability science that goes beyond the nature of interactions in order to facilitate policy analysis. The resulting “International Conference on Sustainability Science for Watershed Landscapes,” held on November 13-14, 2007, was a product of that collaboration and with the East-West Center. One of the major accomplishments of that conference is this reference volume on methods and applications in both Hawai‘i and Asia. In 2008, a Mānoa Ad Hoc Committee of Faculty and Administrators proposed the organisation of an Institute of Sustainability Science, Technology and Policy. The mission they articulated was to apply the knowledge and talents of the University to the practical problems that our society faces in supplying ourselves with the resources of energy, water, food that a growing population requires from a diminishing resource base.  Specifically, they suggested that the Institute would build on the vision of sustainability science, a transdisciplinary method of organising research to deliver meaningful contributions to critical issues of resource management and rigorous policy analysis. With this effort the University of Hawai‘i joins many other universities in North America and Asia in leading the efforts to take us beyond sustainability. After all, it will not be enough to just sustain the current situation. Sustainability science offers a way of organising research to deliver practical solutions based on integrated and scientifically sound research. In closing, I both congratulate and thank the authors of the ensuing chapters for their efforts to stimulate the sustainability science movement and facilitate further collaboration between East and West.

Gary K. Ostrander Vice Chancellor for Research and Graduate Education University of Hawai‘i at Mānoa

xvii

Acknowledgments

United by a common goal to lay the foundations of policy analysis in the sustainability science arena, the Southeast Asian Regional Center for Graduate Study and Research in Agriculture (SEARCA), University of Hawai‘i Mānoa (UHM), and East-West Center (EWC) agreed to support an international conference on “Sustainability Science for Watershed Landscapes” on November 13-14, 2007 in Honolulu, Hawai‘i. SEARCA and UHM began planning the conference in August 2006; when the two institutions entered into a Memorandum of Understanding on sustainable resource management, risk management, and agricultural development. EWC partnered in the conference organisation and provided logistical support as well as the conference venue. This volume emerged from the conference discussions and subsequent papers. The chapters herein help to synthesise a “third wave” of sustainability science by integrating traditional interdisciplinary environmental studies with policy science. The challenge lies in the integration of the analysis of natural and human systems with policy science. Abstraction is an essential part of science, and specific simplifications should be motivated by specific policy questions. By focusing on actual resource management and policy questions and interacting with stakeholders, research is organised to inform real-world decision-making. Sustainability science must go beyond the application of scientific knowledge to specific problems and develop new methods for dealing with dynamic, spatial, behavioral, and interactive complications of resource systems under pressure. Watershed management is a natural challenge for sustainability science. Water is the unifying resource of a watershed. Watershed conservation generates onsite values such as timber and biodiversity and offsite values such as groundwater recharge and the reduction of downstream sedimentation. This volume suggests asking the policy questions up front before designing the research methodology and data protocols. Given the paucity of prototypes, the chapters provide case studies from Asia and the Pacific, including Hawai‘i. Some of them serve as a starting point for asking particular policy questions. Others provide the full gambit of resourcesystems analysis to specific policy recommendations. We especially thank the chapter authors, including prominent economists, biologists, hydrologists, foresters, and ecologists who strove to advance sustainability science beyond the description of resource systems towards the posing and solution of particular policy questions. Without their support and enthusiasm, this venture would have failed. Our appreciation also goes to the other conference speakers and session chairs, Barry Usagawa, Bruce Wilcox, Daniel Murdiyarso, Jim Jacobi,

xviii Kensuke Fukushi, Nguyen Hai Nam, Arnulfo Garcia and Nicomedes Briones, whose focus and stimulating discussion promoted the conference objectives. Special thanks also go to the UHM Sustainability Council and its chair, Mary Tiles, for the funds granted to this project. Jefferson Fox, EWC Senior Fellow, was instrumental in arranging EWC support. Rodel Lasco, Philippines Coordinator for the World Agroforestry Centre (ICRAF), facilitated the ICRAF and Center for International Forestry Research (CIFOR) funding of two of the conference speakers. This volume could not have been possible but for the concerted effort of the people at UHM, EWC, and SEARCA. Majah-Leah Ravago, UHM Economics Ph.D. student, EWC Degree Fellow, and chapter co-author, was coordinator extraordinaire, assisting the editors and facilitating communications between all parties involved. UHM Water Resources Research Center Director James Moncur and his stellar staff, Susan Yokouchi and Barbara Guieb, provided hasslefree handling of the direct costs for the conference. June Kuramoto, assisted by Arlene Hamasaki, ably accomplished administration and other logistics handled by EWC. Kevin Nishimura, from the UHM Economics Department, facilitated the travel logistics for some of the invited speakers. Arnulfo Garcia, former SEARCA Research and Development Department Manager, and his competent staff, Nyhria Rogel and Ruby Johnson, handled project management. Lily Tallafer also helped in various stages of the project. The Knowledge Management Department, headed by Maria Celeste Cadiz, together with her staff Lorna Calumpang and Bernadette Joven, helped polish the manuscript. Jane Kirton and Chris Wada also helped with the final copyediting of several chapters. We also thank SEARCA’s co-publisher, the Institute of Southeast Asian Studies (ISEAS), especially Triena Ong, Rahilah Yusuf, and Celina Kiong for their support and interest and in seeing the volume through to its present form.

James A.Roumasset Kimberly M. Burnett Arsenio M. Balisacan

xix

List of contributors

Arsenio M. Balisacan Professor of Economics School of Economics University of the Philippines Diliman Chad Browning Environmental Scientist Office of the Environment, Fish & Wildlife Bonneville Power Administration Gregory L. Bruland Assistant Professor Department of Natural Resources & Environmental Management University of Hawai‘i at Mānoa Kimberly M. Burnett Assistant Specialist University of Hawai‘i Economic Research Organisation Gretchen C. Daily Director, Center for Conservation Biology Professor, Biology Department Stanford University Aly I. El-Kadi Professor of Hydrology  Department of Geology and Geophysics and Water Resources Research Center University of Hawai‘i at Mānoa

Carl I. Evensen Chair and Extension Specialist Department of Natural Resources & Environmental Management University of Hawai‘i at Mānoa Ali Fares Professor of Watershed Hydrology Department of Natural Resources & Environmental Management University of Hawai‘i at Mānoa Jefferson M. Fox Senior Fellow Research Programme East-West Center James B. Friday Extension Forester College of Tropical Agriculture and Human Resources University of Hawai‘i at Mānoa                            Thomas W. Giambelluca Professor, Department of Geography University of Hawai‘i at Mānoa Rebecca L. Goldman Senior Scientist for Board Relations The Nature Conservancy Worldwide Office

xx Maite Guardiola-Claramonte Graduate Research Assistant Department of Hydrology and Water Resources University of Arizona Kenneth Y. Kaneshiro Director, Center for Conservation Research and Training University of Hawai‘i at Mānoa Michael H. Kido Director Hawai‘i Stream Research Center Center for Conservation Research and Training University of Hawai‘i at Mānoa Rodel D. Lasco Senior Scientist and Philippine Programme Coordinator World Agroforestry Centre

Kevin N. Montgomery Director National Biocomputation Center Stanford University Gary K. Ostrander Vice Chancellor for Research and Graduate Education University of Hawai‘i at Mānoa Sittidaj Pongkijvorasin Faculty of Economics Chulalongkorn University Majah-Leah V. Ravago PhD Student in Economics, University of Hawai‘i at Mānoa Graduate Degree Fellow East-West Center

Harold J. McArthur, Jr. Assistant Vice Chancellor for Research Relations University of Hawai‘i at Mānoa

James A. Roumasset Professor of Economics University of Hawai‘i at Mānoa Research Fellow, University of Hawai‘i Economic Research Organisation Jun Shimada Professor of Hydrology Graduate School of Science and Technology Kumamoto University

Monica Mira Researcher Water Resources Research Center University of Hawai‘i at Mānoa

Makoto Taniguchi Professor Research Institute for Humanity and Nature

Nancy D. Lewis Director, Research Programme East-West Center

xxi Peter A. Troch Professor Department of Hydrology and Water Resources University of Arizona Chieko Umetsu Associate Professor Research Institute for Humanity and Nature Grace B. Villamor Researcher World Agroforestry Centre Tsugihiro Watanabe Professor Center for Coordination, Promotion and Communication Research Institute for Humanity and Nature Shigeo Yachi Associate Professor Center for Ecological Research Kyoto University

xxii

Economic Policy for Sustainable Development vs. Greedy Growth and Preservationism

Theme 1 Sustainability Science for Resource Management and Policy

1

2

Majah-Leah Ravago, James Roumasset and Arsenio Balisacan

Economic Policy for Sustainable Development vs. Greedy Growth and Preservationism

1

Economic Policy for Sustainable Development vs. Greedy Growth and Preservationism



Majah-Leah V. Ravago,a James A. Roumasset,b and Arsenio M. Balisacan c

3

Abstract Sustainability science emerged from the felt need to employ appropriate science and technolog y in the pursuit of sustainable development. The existing sustainability science agenda emphasises the importance of using a systems approach, stressing the many interactions between natural and human systems. Despite its inertia and avowed purpose of being practical and feasible, however, sustainability science has yet to embrace the policy sciences. In pursuit of this objective, we first trace the history of thought of sustainable development, including its definition and operationalisation. Sustainable development encompasses sustainable growth and dynamically efficient development patterns. Two promising approaches to sustainable growth are contrasted. Negative sustainability counsels policy makers to offset any decrease in natural capital with at least the same value of net investment in produced capital. This sustainability criterion cannot determine how and how much to conserve natural capital nor how much to build up human and productive capital. Indeed, there is ambiguity regarding what prices to use in summing the values of diverse capital assets. To fill the void, we offer positive sustainability, which maximises intertemporal welfare while incorporating system linkages, dynamic efficiency, and intertemporal equity. This provides a solid and operational framework for sustainable growth. In addition, sustainable development must include the lessons from development theory, including how optimal patterns of production, consumption, and trade change with standards of living.

4

Majah-Leah Ravago, James Roumasset and Arsenio Balisacan

However, like Tolstoy’s unhappy families, there are many pathways to unsustainable development. We describe two broad causes of unsustainable growth – rent-seeking and preservationism. We also illustrate patterns of unsustainable development by drawing on lessons from the Philippines. While specialisation is the engine of growth, fragmentation is the anchor. In addition to natural fragmentation from natural trade barriers in an island archipelago, rent-seeking promotes economic stagnation. Low economic growth in turn exacerbates population pressure and environmental degradation – the vicious circle of unsustainable development. We give particular attention to how a resource curse can exacerbate policy distortions and rent-seeking, and how the same phenomenon can be promulgated by foreign aid, foreign direct investments, remittances, and tourism. For sustainable development not to be at odds with policy science, positive sustainability must be combined with projects and policies that promote dynamic comparative advantage and poverty reduction. We emphasise the facilitative role of government especially in transforming the vicious circle into a virtuous circle. Introduction: Sustainability science and sustainable development Sustainability science emerged from the felt need to employ appropriate science and technology in the pursuit of sustainable development. The publication of Our Common Future1 in 1987 by the World Commission on Environment and Development (WCED: the Brundtland Commission) placed research and development as an integral component of sustainable development strategies and has been espoused by a number of international scientific organisations that were formed in the 1980s (Clark and Dickson 2003). Because of its importance to the roots of sustainability science, we first trace the history of thought of sustainable development, including its definition and operationalisation. Two promising approaches to sustainable growth are compared. Negative sustainability enjoins policy makers to conserve natural capital in accordance with dynamic efficiency and to invest in productive capital such that genuine investment is positive or at least zero. This sustainability criterion is called negative sustainability because it only provides guidance on what not to do. There are no policy principles to prescribe how and how much to conserve natural capital, nor how much to build up human and productive capital. Indeed there is ambiguity about what prices to use in summing the values of diverse capital assets.

Economic Policy for Sustainable Development vs. Greedy Growth and Preservationism

5

While most scholars have focused on what should be ruled out in the name of sustainable development (negative sustainability), we put forward the idea of positive sustainability, which maximises intertemporal welfare while incorporating interlinkages within the total environomy, dynamic efficiency, and intertemporal equity. This gets us to a solid and operational framework for sustainable growth. In addition, sustainable development must include lessons from traditional development studies, including how patterns of production, consumption, and trade change with the welfare of an economy’s citisens. While specialisation is the engine of growth, fragmentation is the anchor. In addition to natural fragmentation from natural trade barriers in an island archipelago, policy and governance, driven by rent-seeking, promote economic stagnation. Low economic growth in turn exacerbates population pressure and environmental degradation – the vicious circle of unsustainable development. We focus on how a resource curse can exacerbate policy distortions and rent-seeking and how the same phenomenon can be promulgated by foreign aid, foreign direct investments, remittances, and tourism. We further emphasise the facilitative role of the government and what it can do to transform the vicious circle into a virtuous circle. Sustainable development: history of thought As early as 1980, international bodies such as the World Conservation Strategy (WCS) endeavored to integrate economic and environmental management. However, the WCS fell short of being operational – it was unable to articulate either how poor economic policies would degrade the environment or how conservation might affect economic policies (Pearce, Markandya, and Barbier 1989). Heightened concern for the environment was manifested when the United Nations (UN) formed the Brundtland Commission 2 in 1983 to examine the interrelationship between human activity and the environment, and its implications for economic and environmental policy. Before the publication of the Bruntland report, Barbier (1987) represented sustainable development with a Venn diagram of three intersecting circles for biological, economic, and social systems (Figure 1.1). A unique set of human goals is assigned for each system. The overlapping area of the three circles corresponds to the overall objective of maximising the goals across these systems through an adaptive process of trade-offs.

6

Majah-Leah Ravago, James Roumasset and Arsenio Balisacan

Figure 1.1 Barbier’s Venn Diagram (adapted from Barbier 1987)

The Brundtland Report (1987) successfully ensconced sustainability in the development arena and became the basis for an integrative approach to economic policy. The Commission defined sustainability as, “… development that meets the needs of the present without compromising the ability of future generations to meet their own needs.” This rather vague definition has been the source of considerable contention. Indeed, a host of definitions has sprung up over the years: Pearce, Markandya, and Barbier (1989) supply a gallery of definitions while Quiggin (1997) demonstrates several inconsistencies among alternative definitions. For sustainable development tourism alone, one can find over 300 definitions (Stabler and Goodall 1996). Ahmed (2009) noted that over 500 definitions of sustainable development exist. An International Environment Forum of the United Nations (UN) stated that there are one thousand definitions of sustainable development.3 Pezzey (1989) attempted to extract one single definition that could command the widest possible academic consent. Nearly a decade after, he gave up on his quest, noting that:

Economic Policy for Sustainable Development vs. Greedy Growth and Preservationism

7

“So I see little point in expanding the collection of fifty sustainability definitions which I made in 1989, to the five thousand definitions that one could readily find today.” (Pezzey 1997, p. 488) The burgeoning number of definitions is a consequence of an explosion of literature from economics, philosophy, and other disciplines that followed immediately after the Brundtland Report, much of which was an attempt to specify what sustainable development meant for public policy. Just as the definitions and implications of sustainable development were proliferating along multiple lines, Pearce, Markandya, and Barbier (1989) abandoned the Venn approach and turned the attention to articulating a vision of sustainable development that was consistent with both economics and the Report. In his background paper for the 1992 UN Conference on Environment and Development (UNCED)4, however, Munasinghe (1992, revised 1994) resurrected the Venn approach but turned it into a triangle encompassing three major objectives – economic, social, and environmental.5 The three vertices are further connected with double-sided arrows showing the interactions among domains, thus illustrating trade-offs and synergies. The picture6 is now known as the sustainable development triangle with its key elements and links. While politically popular, this portrayal was so grand that the scope includes income distribution, gender equity, culture, and a host of other political goals.7 The Venn diagram has now evolved into 178 images8 reflecting both a diversity and inconsistency among concepts. Arguably, the Venndiagram image is still the most popular. As in the original Barbier (1987) version, the three circles continue to represent three objectives: economic, environmental, and social. This version is non-operational, however, lacking separate but comprehensive performance indices of the three objectives. Moreover, in order for a strategy to be in the central section where all three circles intersect, the authors articulate minimum satisfactory performance levels for each. Three objectives, three performance indicators, and three constraints! Given that the Venn approach and most of the other images cannot be operationalised, how did they become so popular? Perhaps the very infeasibility in implementation is exactly the concept’s appeal. When nothing is well-defined, anything goes. Any organisation with a political agenda can readily turn the rhetoric of sustainable development to its own ends and not be constrained by transparency or accountability.

8

Majah-Leah Ravago, James Roumasset and Arsenio Balisacan

In the field of economics, Pearce, Markandya, and Barbier (1989) proposed what later became known as weak and strong sustainability. Strong sustainability prohibits any level of depletion of natural capital such as trees, water, or fish. The weak sustainability rule requires the summed value of produced and natural capital to remain constant or increase over time. A subsequent debate emerged over whether weak or strong sustainability was the best single criterion of sustainable development. Arrow et al. (2004) and others (e.g. Asheim 1999; Pezzey 1997) have advocated a broadened form of the weak criterion for sustainable development – the requirement that the wealth of a society, including human-capital, knowledge-capital, and natural-capital (as well as produced-capital), does not decline over time. Others, including Barbier (2007), continue to contend that strong sustainability – non-depletion of essential stocks of renewable resources – may be appropriate, especially for forms of natural capital that are essential in the sense that produced capital or other resources are poor substitutes.9 Dasgupta and Mäler (1995, p. 2394) criticise strong sustainability on the grounds that it is a category mistake, “… the mistake being to confuse the determinants of well-being (for example, the means of production) with the constituents of well-being (for example, health, welfare and freedoms)…” That is, by proposing strong sustainability as both a criterion and a constraint, the means are confounded with the ends, as opposed to deriving the rule from more fundamental objectives. As Dasgupta and Mäler (1995) conclude: “The point is not that sustainable development, even as it is defined by these authors, is an undesirable goal. It is that, thus defined, it has negligible information content. (We are not told, for example, what stock levels we ought to aim at). This is the price that has to be paid for talking in terms of grand strategies. The hard work comes when one is forced to do the ecolog y and the economics of the matter.” Regarding another formulation, that sustainable development requires both welfare and natural capital to be non-declining, Dasgupta and Mäler (1995, p. 2394) ask: “Two constraints? … [the formulation] offers no direct ethical argument for imposing either of the side constraints.”

Economic Policy for Sustainable Development vs. Greedy Growth and Preservationism

9

At first blush, it would appear that even weak sustainability involves a category mistake in the sense that the requirement not to deplete the total capital stock is a means without a distinguishable objective or criterion. Other authors have shown, however, that the weak sustainability rule (wherein total value of capital cannot decline) can indeed be derived from the sustainability criterion – the mandate that the total welfare of all future generations not be diminished (Arrow et al. 2004). That is, the weak sustainability rule is a necessary and sufficient condition for achieving the sustainability criterion. However, this still leaves the problem that no fundamental ethical argument has been provided for the sustainability criterion in the first place. Moreover, the sustainability constraint can be extremely restrictive. Suppose that society has the choice between maintaining utility at subsistence levels forever or doubling utility and then decreasing it (due to a previous decline in total capital) by 10 percent. The dominant, but non-monotonic, path is ruled out by the sustainability criterion. Nonetheless, the sustainability criterion and the (implied) weak sustainability rule have become firmly ensconced in the environmental economics literature. Genuine investment10 is commonly applied to empirically determine whether specific countries satisfy the sustainability criterion or not (Arrow et al. 2004; Dasgupta 2007; Hamilton and Clemens 1999). Genuine investment is the increase in the stock of capital assets – manufactured, human, and natural.11 The sustainability criterion is met if genuine investment is non-negative. Augmenting the measures employed by Hamilton and Clemens (1999), Arrow et al. (2004) showed that most low-per-capita-income countries, especially in sub-Saharan Africa, are unsustainable even though net investment and increases in human capital (education expenditure) are positive. On the other hand, China’s genuine investment is high, even amidst claims of heavy pollution and natural capital depletion, because of extremely high domestic net investment and human capital accumulation. By alerting countries that the total value of their capital accumulation is negative, the statistic provides a useful signal to examine the components more closely and formulate possible strategies to combat unsustainability. The criterion by itself, however, does not provide clear guidance on how much genuine investment should be increased nor how much its components should be changed. For that reason, the sustainability criterion can be classified as negative sustainability – it only tells us what not to do.

10

Majah-Leah Ravago, James Roumasset and Arsenio Balisacan

Hamilton and Clemens (1999, p. 351) maintain that the criterion naturally leads to a management of a country’s portfolio of natural, produced, and human capital. “More optimal natural resource extraction paths will, other things being equal, boost the value of genuine savings. The policy question for natural resource management is therefore: to what extent can stronger resource policies (royalty regimes, tenure) boost the genuine rate of savings.” As appealing as it sounds, the advice is misleading in two important ways. First, maximising savings, genuine or otherwise, is inconsistent with maximising welfare, which is given by green net national product, GNNP=GS+C, where GS is genuine savings and C is aggregate consumption. Second, even maximising GNNP is not an operational guideline without the correct shadow prices of the various forms of capital relative to consumption. The correct shadow prices can only be obtained by specifying the problem and then deriving the corresponding efficiency conditions. This is the objective of the following section. Positive sustainability and sustainable growth One can note in the history of sustainable development thought, as sketched above, that somewhere along the line, attention was shifted away from development to the question of sustainable growth. We continue that focus in the present section and extend the framework to embody development in the subsequent section. Positive sustainability posits sustainable growth as neither an objective nor a constraint. Rather, it is based on the central pillar of policy science, the maximisation of a single intertemporal welfare function,

Economic Policy for Sustainable Development vs. Greedy Growth and Preservationism

11

subject to resource constraints regarding the amounts of produced capital and natural capital available. B is the benefit from the joint consumption of material goods (M) and environmental amenities (A). Net benefit is obtained after deducting per unit extraction (c) and damage (d) cost associated with the utilisation of the resource (R). Damage cost is necessarily included since using the resource may generate pollution, e.g. burning of coal contributes to the increase of greenhouse gas in the atmosphere. The sum of c and d is the total marginal cost of resource extraction. The total amount of produced capital and the amount of the natural resource consumed are arguments in the aggregate production function (e.g. Toman et al. 1995). Intergenerational equity is incorporated into the planner’s objective function by setting the planner’s pure rate of time preference equal to zero. Thus, maximisation of intertemporal welfare incorporates interlinkages within the total economic and environmental system (environomy), dynamic efficiency, and intertemporal equity. Two necessary conditions are required for positive sustainability: 1. MB = MC + MUC + MEC =MOC 2. NMPK = ηg The first is the so-called Hotelling condition for optimal resource extraction – extract an additional unit of the resource until the marginal benefit (MB) of using the resource is equal to the marginal cost (MC) plus the marginal user cost (MUC) plus the marginal externality cost (MEC). The sum of these three right hand variables is called marginal opportunity cost (MOC).12 The second condition is the Ramsey savings equation, which states that produced capital should be accumulated in any given period until its net marginal product (NMPK ) declines to equal the growth rate (g) of consumption in that period times a measure of the planner’s aversion to intergenerational inequality (η), (Ramsey 1928). Choosing the vectors of the two control variables, M and R, which maximise present value, also determines the rest of the system. Capital formation is the residual aggregate output after subtracting depreciation, consumption, and total extraction costs (c x R) . The change in the stock of the natural resource is g(A)-R , where g(A) is the natural growth of the resource (zero for a non-renewable) as a function of its own stock. Thus, consumption is balanced with conservation

12

Majah-Leah Ravago, James Roumasset and Arsenio Balisacan

of amenities and savings. Natural capital conservation is balanced against accumulation of produced capital. Total value of consumption is balanced against total value of capital.13 The positive sustainability solution provides the dynamically efficient paths of consumption, investment, and natural capital depletion/ accumulation. The contrivance of a sustainability constraint is unnecessary. All that is needed is to recognise the interdependence of the economy and the environment and to disallow intertemporal discrimination in the social welfare function (Heal 2009). Having characterised the socially optimal path of the environomy, one can then proceed to determine conditions under which the optimal path satisfies the sustainability criterion or not (Anand and Sen 1994). Critical determinants include initial conditions and model parameters, such as the substitutability between produced and natural capital. Moreover, adherence to the Hotelling and Ramsey conditions determines optimal management of the country’s aggregate produced capital and natural capital. This implies an optimal drawdown of non-renewable resources, an optimal drawdown or accumulation of renewable resources, and an optimal accumulation of produced capital. As shown in Figure 1.2,14 these conditions move the environomy to the limits of the current frontier and move the frontier outward over time. Living standards, incorporating material consumption and environmental amenities, rise as the environomy approaches a “golden-rule” steady state. Win-win efficiency is achieved, and unsustainable growth is avoided (Ayong le Kama 2001 and Endress et al. 2005). Figure 1.2 Positive Sustainability

Economic Policy for Sustainable Development vs. Greedy Growth and Preservationism

13

Implementation of win-win environmentalism requires modeling important interlinkages and dealing with political impediments to the implied public policies. Policy models of global warming, containing both climate change and the economic system, provide well developed – although not definitive – examples, inasmuch as they solve for specific policy prescriptions. Intergenerational equity can also be incorporated into this approach, as has become common in climate change economics (Heal 2009). In ruling out discrimination against future generations and allowing for the possibility of renewable alternatives to petro-chemicals and other non-renewable resources, efficient policies are compatible with increasing human welfare, eventually reaching a steady state (Ayong le Kama 2001 and Endress et al. 2005). Sustainability does not require throwing out policy science that has evolved over the last quarter millennium. Rather, it represents an injunction to extend policy analysis to encompass both system interdependence and intergenerational equity. Patterns of sustainable development Sustainable growth literature, as discussed in the preceding section, presumes allocative efficiency and focuses on the dynamically optimal balance between produced capital (including human) and natural capital. By using aggregate capital and aggregate environmental amenities, the model abstracts from issues of composition across consumption goods and across environmental amenities. But no economy is without waste. Thus, there is a need to add principles governing efficient sectoral composition – each good is produced until its marginal benefit equals marginal cost, including any negative external effects of production. What is known about dynamically-efficient patterns of development? Initially, specialisation and capital accumulation (infrastructure, processing and mechanisation) in the agriculture sector barely outstrip the dismal Malthusian forces of population pressure and diminishing labor productivity. But population pressure also induces some innovation and specialisation in the agricultural sector (Boserup 1965, 1981). Aided by capital accumulation and fortune, these forces eventually lead to the emergence of industrialisation. The possibilities for vertical and horizontal specialisation are inherently more compact in industry (especially due to economies of assembly), and the

14

Majah-Leah Ravago, James Roumasset and Arsenio Balisacan

external economies they afford eventually outstrip the negative Malthusian forces (Roumasset 2008).15 Manufacturing and the surplus from agricultural development beget capital accumulation. Specialisation and capital formation together increase the return to human capital formation, lowering fertility and augmenting the virtuous circle of industrial revolution (Lucas 1993 and 2001). Horizontal specialisation of final products is further stimulated by vertical coordination provided by big-box retailers. Size of the market allows specialisation of component varieties such that standardisation is not needed.16 This process has no natural end point in manufacturing, inasmuch as further market growth allows further vertical specialisation across intermediate products, tailored to specific end-products. A metaphor can be found in the new supermarket economics (Reardon and Timmer 2007) wherein vertical coordination begets specialisation. Farmers are increasingly linked to specific retailers by means of complex chains that transform farm products over space, time, and form. Thus, the process replaces the cumbersome and costly method of indirect coordination via inventories. Dedicated wholesalers coordinate specific farmers with specific retailers with appropriate procurement, quality, safety, and timing standards. This is important for land policy inasmuch as these arrangements confer transaction cost advantages on large farms (Roumasset, forthcoming). Changing relative factor endowments, technology, demand, and these organisational issues determine dynamic comparative advantage, i.e. the pattern of optimal specialisation over time (Chenery 1961). Moreover, specialisation co-evolves with human capital (through learning by doing) and economic organisation. These in turn create positive spillover effects (since knowledge is not entirely contained at the firm level) and further rounds of technological and institutional innovation. In short, specialisation is the engine of growth. The composition of natural capital follows a natural resource Kuznets curve (NRKC). Resource depletion increases at the early stage of industrialisation. As the comparative advantage shifts towards labor intensive manufactured goods and domestic resource prices increase, imports of natural resources increase and domestic resource extraction slows. For renewables, the reduction of stocks decreases or even reverses as conservation increases, e.g. replanting, bench terraces, fertiliser, and transition to renewable energy sources (Krautkraemer 1994; Pender 1998). These trends are augmented

Economic Policy for Sustainable Development vs. Greedy Growth and Preservationism

15

as the service sector grows relative to manufacturing, all resulting in a decline in the value of natural capital depletion, even as resource prices are increasing worldwide. In the case of China, combining the NRKC with the environmental Kuznets curve for most air pollutants yields the result that China’s economic welfare, GNNP, is growing as fast or faster than its NNP, i.e. its net national product is uncorrected for natural capital depletion and environmental pollution (Roumasset et al. 2007). The final stage of structural transformation and specialisation is often referred to as “de-industrialisation.” Developed countries and the Asian Tigers followed this path of structural transformation.17 The flying geese18 metaphor demonstrates how the co-evolution of industrialisation is attained by macroeconomic patterns of specialisation in the course of sustainable development. As wages increase in developed economies, other countries gain a comparative advantage in labor-intensive, manufactured exports. In the East-Asian model, the lead goose was Japan, followed by the New Industrialising Economies (NIE) of South Korea, Taiwan, Singapore, and Hong Kong. The third layer consists of Malaysia, Thailand and Indonesia, with the Philippines and Vietnam trailing behind. The East-Asian miracle is characterised by capital intensification and human capital accumulation whereby network externalities led to endogenous growth. What can go wrong? In contrast to the positive pathway to sustainable growth, there are many pathways to unsustainable growth, much like Tolstoy’s (1878) unhappy families.19 The first of these is growth through excessive resource depletion and/or pollution as illustrated by the example of Indonesia (Repetto et al. 1989). From 1971 to 1984, the country’s gross domestic product (GDP) grew on average 7.1 percent a year. Because it does not account for depreciation, however, GDP is not reflective of a country’s real economic growth. Just as capital depreciation is subtracted from GDP to obtain net domestic product, so should the depletion of natural capital be depleted to obtain a more accurate measure of social welfare (Weitzman and Löfgren 1997; Weitzman 2003). Indeed, Repetto et al. (1989) found that depreciation of petroleum, timber, and soil resources was valued at 4.5 percent of GDP from 19711984.

16

Majah-Leah Ravago, James Roumasset and Arsenio Balisacan

Dynamic efficiency calls for extracting resources in accordance with the extended Hotelling rule for renewable resources (Stavins et al. 2003; Endress et al. 2005). Excess depletion of natural resources, as illustrated by the example of Indonesia, results from failing to align private incentives with social priorities, especially through inappropriate property rights, e.g. the ability of military units to exploit public forests and the nationalisation of the oil business. Such institutions are often manifestations of “greedy growth,” (northwest arrow of Figure 1.2) the proliferation of government entities, projects and regulatory policies to transfer rents from taxpayers to the “iron triangle” of politicians, bureaucrats, and special interests discussed below. Greedy growth in the arena of resource management is unsustainable in the sense that resources are wasted instead of being transformed into produced capital or conserved for future generations. Panayotou (1993) illustrated such unsustainable development projects with subsidised cattle ranching operations in Brazil.20 The Superintendency for Development of Amazonia (SUDAM) was created to improve the economic development of the region. The well-intentioned programme provided certain corporations a tax credit scheme aimed at promoting livestock ranches in the Amazon. Overgrazing and ranch expansion into previously forested areas led to a loselose situation for the economy and the environment, as shown by the “greedy growth” arrow in Figure 1.2. When the ranches failed due to overgrazing and poor management, the economy suffered from commercial failure of the ranches and enormous fiscal costs, the latter amounting to more than $US1 billion between 1975 and 1986 (Volpi 2007). But the failed ranches did not revert to forests. Rather, the overgrazed pasture grasses were overtaken by invasive cogon grass (Imperata cylindrica) furthering soil erosion. Unfounded preservationism also results in a lose-lose situation for material consumption and the environment (southwest arrow of Figure 1.2), especially in developing countries. This inefficient policy alternative prevents sufficient capital formation from increasing wages, thus failing to provide a positive check on population growth. In the absence of jobs in the modern sector, the increased labor force seeks subsistence living in environmentally fragile areas. Thus, preservationism can lead to the unintended consequence of environmental degradation as well as decreasing standards of living (Roumasset and Endress 1996).

Economic Policy for Sustainable Development vs. Greedy Growth and Preservationism

17

Fragmentation has a stagnating effect on economic development. Wages may rise in favored enclaves without conferring gains in areas of underemployment and low wages. Specialisation is limited by the extent of markets. Consumers and producers pay higher prices for goods and intermediate products. Three pervasive forms of fragmentation are geographic, economic, and political. 21 1. Geographic fragmentation - Island economies are unfortunate to have these natural barriers to internal integration. 2. Economic fragmentation - Protection by tariff and non-tariff barriers are distortions that pull resources into the protected enclave, distorting factor prices, and artificially increasing the real exchange rate. These forces discriminate against agricultural exports and forward linkages into agricultural processing and packaging (e.g. Clarete and Roumasset 1987). 3. Political fragmentation - Entrepreneurs are equal under the law but some are more equal than others. Politically influential individuals and companies can obtain special favors or circumvent the law while others are subjected to bureaucratic red tape, unnecessarily increasing the cost of doing business. Fragmentation has both static and dynamic effects. Like a system of internal tariffs, high transportation/transaction costs inhibit specialisation and exchange, thus lowering national income. Specialisation begets more specialisation, due to learning-by-doing, human capital spillovers and growth externalities in general. Fragmentation impedes this process and begets stagnation instead. In addition to the natural fragmentation from natural trade barriers in an island archipelago, policy and governance, driven by rent-seeking, promote economic stagnation. A government needs extra-Smithian powers afforded by their supposed facilitative role in coordinating investments beyond the basic institutions of property, contracts, and markets. However, centralisation is easily abused. Facilitation via cozy arrangements between the government and industrial-financial conglomerates can lead to erosion of fiduciary accountability as in the East-Asian financial crisis of 1997 and the US-led world recession of 2007-2009. On the other hand, government “prizes” easily turn into mandates, subsidies, and “picking winners” with

18

Majah-Leah Ravago, James Roumasset and Arsenio Balisacan

costly repercussions to economic growth and development (The Economist 2009). In short, even limited government powers allow for strategic coalition formation to extract unproductive rents from the economy – a phenomenon known as rent-seeking. The potential for rent-seeking is greater in real-world economies whose governments are inevitably larger than the minimal state described above. Figure 1.3 illustrates rent-seeking with the famous iron triangle (Lowi 1979; McConnell 1966), representing alliances between politicians, bureaucrats, and special interests. Politicians form alliances with bureaucrats and special interest groups, who enjoy more sharply focused objectives and ease of organisation (Olson 1982; and Olson and Zeckhauser 1966). Special interest groups benefit from lax regulations and special favors. Politicians influence the legal, policy, and fiscal environment to confer rents in return for campaign contributions, favors, and other support. Bureaucrats implement rent-capturing agreements and broaden their power base. With these dynamics, the minority coalition readily tyrannises the needs of the relatively poor majority. A primary example is protectionism, which confers rents to some industries but reduces total consumer and producer welfare (Corden 1985). Protectionism decreases the price of foreign exchange, making imports cheaper and discriminating against exports. Since nonprimary-good exports are a key engine of growth (World Bank 1993), their inherent potential for specialisation and learning-by-doing is partially lost, thereby stifling economic development.22 Figure 1.3 Rent-seeking and the Iron Triangle

Economic Policy for Sustainable Development vs. Greedy Growth and Preservationism

19

Nature’s abundance can ironically be another anti-development force – a phenomenon known as resource curse. This term is commonly attributed to countries with large endowments of natural resources, such as oil and gas, whose economic development and governance have lagged behind less endowed countries (Sachs and Warner 1995 and 2001; Auty 1993, 1990). Economies suffering from a resource curse are said to have contracted the “Dutch disease”23. The theoretical development of the Dutch disease phenomenon is formalised in the classic papers of Corden and Neary (1982) and Corden (1984). Two curses could arise from having abundant natural resources: 1. Dutch Disease Mechanics: Appreciation of the Real Exchange Rate - The first curse results from real exchange rate appreciation, which lowers the relative price received by other exportables. The core model 24 explains how an intersectoral reallocation of resources is induced by the relative price change due to a boom in an extractable resource.25 A discovery (boom) of natural resource causes a contraction of other exportables such as the manufacturing sector. Resources are pulled out from manufacturing sector into mining the newly discovered resource. At the same time, income generated from the booming resource sector is spent on consumption of non-traded goods. Both effects go in the same direction, increasing the real exchange rate and contracting manufactured exports.26 The appreciation of the real exchange rate implies an increase in the opportunity cost in the production of traded goods thereby eroding the competitiveness of the manufactured exports.27 Exchange rate appreciation per se, however, is not a “disease” but an economy’s statically efficient response to changing relative prices. Due to the resource discovery or increase in world price, the economy’s comparative advantage shifts in favor of “production” of the booming traded sector in accordance with specialisation for mutual gain. Resource exports earn foreign exchange for imports more cheaply thereby driving out manufactured exports. But static efficiency is not everything. Since future prices do not exist for most goods and services, static market equilibrium fails to achieve dynamic efficiency. If the lagging sector has more potential growth externalities than

20

Majah-Leah Ravago, James Roumasset and Arsenio Balisacan

the booming sector, the boom negatively impacts growth. Recall that the engine of growth in the East-Asian Miracle countries was manufactured exports with their abundant and never-ending possibilities for continuing vertical and horizontal specialisation and learning-by-doing. Furthermore, the manufacturing sector is a source of forward and backward linkages that generates production externalities (Sachs and Warner 1995). But the resource boom discriminates against manufactured exports, thus putting the brakes on the engine. 2. The Returns to Rent-Seeking in Resource-Rich Economies - The second curse is the more virulent strain of the Dutch disease, directly unproductive rentseeking (Bhagwati et al. 1998). The political economy in resource-rich countries exacerbates the difficulty of sustaining growth (Corden 1984 and 1982; and Humphreys, Sachs, and Stiglitz 2007). The more there is inherent resource wealth, the higher the returns to lobbying, the greater the lobbying, and the greater the resulting policy distortions. Corden (1984) gives the example of industrialists adversely impacted by the appreciation of the real exchange rate increasing their lobbying effort for tariff and non-tariff protection. Protectionism is said to be the most common form of rent-seeking because industrialists are already wellorganised to lobby, and their export ox has been gored by the falling price of foreign exchange (Olson 1982). The lagging sector (other exportables) is hit by a double whammy – both the resource discovery/ boom and protectionism lower the value of the foreign exchange that exports earn. The preceding discussion focuses on natural resource abundance/boom as the source of a curse. Subsequent studies extended the curse to other sources. Palma (2008) investigated how the expansion of tourism 28 in Greece, Cyprus, and Malta and the “export” of financial services in Switzerland, Luxembourg, and Hong Kong have also increased the exchange rates in these countries. Using cross-country comparisons, Rajan and Subramanian (2006) examined foreign aid as a similar curse. Paldam (1997) studies the adverse impact of grants from Denmark on the Greenland economy. Generally, any exogenous development that brings foreign exchange earnings into the country will have adverse affects on the “lagging sector” whose exports earn a lower price of foreign exchange. If the lagging sector is an engine of

Economic Policy for Sustainable Development vs. Greedy Growth and Preservationism

21

growth, such as manufactured exports, there may be a negative effect on economic development. If the exogenous boom increases the rate of return to rent-seeking, there will be an additional negative effect on development. Whether these negative effects outweigh the original positive effects of the boom has not been predicted by theory nor demonstrated unambiguously through empirical studies. All that curses is not gold: a Philippine illustration The Philippines has been called a historical underachiever (Briones 2009), laggard among flying geese (Intal and Basilio 1998), and stray cat amongst economic tigers (Vos and Yap 1996). In this section, we illustrate particular patterns of unsustainable development by drawing on lessons from the Philippines with the overall message that rent-seeking deepens fragmentation and economic stagnation. Lucas (1993) compares the Philippines with South Korea as an illustration of opportunities lost. In the year 1960, the Philippines was at par with Korea in terms of GDP per capita, GDP composition, population, and even exceeded the latter in literacy rates. The country’s human capital in terms of educational attainment was very high. The civil, judiciary and legal institutions are also relatively well established. However, contrary to expectations, the country’s development has been substandard and has veered away from the stylized pattern of structural transformation. Low investments in R&D, rural infrastructure, and detrimental government policies were among the problems. From the 1950s to the early 1980s, protectionism and importsubstitution led to premature and distorted patterns of industrialisation (Bautista, Power and associates 1979; and Clarete and Roumasset 1987). As opposed to the structural transformation of its neighbors, particularly the Four Asian Tigers, the Philippines largely skipped the primary engine of growth – manufacturing for export. While the Philippines’ growth has been restrained, Korea’s accelerated. Lucas (1993) refers to this continuing transformation of Korea as a miracle; similar to what transpired in Taiwan, Hong Kong, and Singapore. The Tigers pursued an export-driven model of economic development. The Philippines, meanwhile, followed the import-

22

Majah-Leah Ravago, James Roumasset and Arsenio Balisacan

substituting industrialisation strategy. After an initial spurt of finishing-stage import substitution, industrialisation stagnated (Power and Sicat 1971). There have been ups and downs in Philippine economic growth in the 1990s and the current decade. Despite some years of high growth performance, total investment (public and private) has been mostly in decline, especially after the Asian crisis in 1997. Among other reasons, Bocchi (2008) suggested that the fast-growing service sectors (electronics assembly, voice-based business process outsourcing (BPO), and information and communications technology) remained profitable based on initial investments. Neither backward nor forward integration was forthcoming, nor increasing rounds of specialisation. Since many of these services, such as call centres, are exports, they also raise the exchange rate, similar to a resource boom. Rapid population growth has exacerbated the lackluster performance of the Philippine economy (Mapa and Balisacan 2004; Herrin and Pernia 2003). In the 1960s, the Philippine population was well within Asian standards at 3 percent. However, while neighboring South Korea, Thailand, and Indonesia successfully slowed down population growth to an average of 0.51.3 percent, the Philippine population is still growing at 2.1 percent in the current decade.29 As a consequence, the Philippines missed the demographic bonus enjoyed by its neighbors (Herrin and Pernia 2003). A high level of population coupled with low investment translates to declining labor productivity throughout the economy. This implies mediocre growth of human capital, high unemployment, and the so-called brain drain problems. A national consensus on the importance of population management has remained elusive. Moreover, the country’s capacity to facilitate a high impact of economic growth on poverty reduction has been comparatively weak vis-à-vis its Southeast Asian neighbors, even after accounting for differences in the level of growth that occurred (Balisacan 2007). Poverty in the country remains a rural phenomenon with three quarters of the poor still residing in rural areas and dependent on agriculture. Poverty remains stubbornly high (at 11% in 2007), and the reduction of poverty and malnutrition is slow relative to other Asian countries. As an island economy, the Philippines has natural barriers to internal integration. These are exacerbated by inadequate transportation infrastructure and misguided transport regulations, especially shipping

Economic Policy for Sustainable Development vs. Greedy Growth and Preservationism

23

(ADB 2007; Balisacan et al. 2008). As a result of these factors, the growthenhancing effects of trade liberalisation are conferred disproportionately to port cities and their environs (e.g. Metro Manila, Cebu, and Davao). Producers in rural areas are sheltered from international competition to the detriment of consumers and wage earners. In addition to geographic fragmentation, economic and political fragmentations are also pervasive. Economic fragmentation is manifested by the protection afforded to agriculture, services, and some manufacturing (e.g. steel and some petrochemicals). As a result, the agricultural sector remains inward looking and insulated from productivity-enhancing competition. Political fragmentation is also widespread. Even as trade barriers and bureaucratic red tape unnecessarily increase the cost of light manufacturing, agricultural processing/packaging, and production for export, politically influential entrepreneurs are able obtain exemptions or easy passage through the barriers. For example, even though the poultry lobby has managed to use the WTO apparatus to protect themselves from importation of low marketvalued chicken parts, McDonalds has succeeded in exempting themselves. Similarly, Coca-Cola and other large companies have managed to secure low-cost sugar, even as potential small-scale candy makers and canned fruit manufacturers are unable to access the same benefit. While Monsanto and Cargill may be able to accelerate the Bureau of Plant and Industry (BPI) quarantine procedures and other restrictions on importing seeds, farmers who want to experiment with new varieties would face great difficulty in doing so (Roumasset 2003; David, et al. 2009). Similarly, implementation of land reform can be delayed or avoided altogether through political influence. The Philippine experience also exemplifies how rent-seeking can induce other inefficiencies as well. Large conglomerates were able to raise prices and deter entry through political connections (e.g. agricultural commodity, most notably rice and sugar, transport services, electricity, and cement in the 1990s). Some of these were critical inputs to production. The whiteelephant convention and “cultural” centres,30 the mothballed Bataan Nuclear Power Plant,31 and the mercantilistic structuring of agricultural and cement industries32 further exemplify the formidable presence of rent-seeking. Other examples are the long-standing NFA rice monopoly,33 the exemption of the sugar industry from land reform and WTO provisions, government offices being controlled by political appointees, and barriers that increase

24

Majah-Leah Ravago, James Roumasset and Arsenio Balisacan

the cost of doing business (e.g. permits). Other recent examples include the NAIA Terminal III fiasco34 and the controversy on the Philippine National Broadband Network.35 Moreover, governance indicators showed that the country fared comparatively poorly among countries with similar per capita GDP levels (ADB 2007). It garnered the lowest score on control of corruption and political stability since 1996 and on rule-of-law since 2002. Furthermore, the county lost momentum in controlling corruption compared with Vietnam, which has been able to better manage corruption, and Indonesia, which is poised to overtake the Philippines in the rankings. As reviewed in a previous section, fragmentation and rent-seeking can be exacerbated by a resource curse. In what follows, we use the Philippine case to explore how a similar curse can be promulgated by foreign aid, foreign direct investments (FDIs), remittances and other activities that increase the supply of foreign currency, increasing the exchange rate. In the 1970s, the Philippines suffered from exchange rate appreciation due to the large increase in foreign borrowings used to finance its trade deficit.36 The upsurge of foreign direct investments (FDIs) in the 1990s had a similar effect. Long-term capital inflows have been on the rise37 after the passage of the Foreign Investment Act in 1991.38 The policy of deregulation and privatisation in the service sectors (e.g. water, communications, and transport), especially during 1992-1998, attracted FDIs away from the traditional manufacturing into the services sector. More recently, efforts are made to attract FDI in BPO.39 The steady increase of remittances is likely to have the same effect of increasing exchange rate. Remittances accounted for an average of 13 percent of GDP during 2003-2007 (WDI 2008). As noted in the preceding section, any activity that brings in foreign exchange earnings into the country would contract the output of the “lagging sector.” In the case of the Philippines, manufactured exports have shrunk.40 Innovation and exploitation of backward and forward linkages has been sluggish with finishing-stage semiconductors and electronic equipment, which accounts for more than 60 percent of merchandise exports (ADB 2007). Since manufactured exports serve as an engine of growth, economic development is adversely affected. If, as suspected, the lagging sector is more labor intensive than the booming sectors, downward pressure on wages and increasing unemployment may also result. This effect has been exacerbated by the high population growth in the Philippines. Outward migration serves as a safety valve, but

Economic Policy for Sustainable Development vs. Greedy Growth and Preservationism

25

higher skilled workers and entrepreneurs tend to leave the country, ergo the moniker, “brain-drain.” Unskilled workers, who tend to remain in the country, are more likely to live in poverty. In addition to the exchange rate appreciation and the possible subsequent contraction of other manufactured “exportables,” some of these booms increase the returns to rent-seeking, leading to the second curse. What really drove trade deficit between 1970 and 1980 was the chronic budget deficit that reached 4 percent of GDP in 1981. The root cause of this massive budget deficit was “crony capitalism” owing to substantial extraction of rents made possible by an iron-triangle reinforced by a strong leader and limited accountability (Roumasset 2009). The prevalence of rent-seeking activities and corruption inflated the budget deficit through increased external borrowing and misspending. Whether these foreign exchange earnings have indeed exerted upward pressure on the exchange rate remains disputable.41 Nevertheless, the fact remains that these sectoral booms add another distortion to the Philippine economy, and they may have adverse general equilibrium effects on other sectors. Remittances may have invigorated consumption and growth, but without induced specialisation, the resulting spending may not have the desirable network and other growth externalities. In the current decade, remittances have been increasingly channeled into investments in real estate partly because of the incentives created by the government through the Pag-Ibig Funds programme.42 This allows for remittances to be an easy target for taxation or extortion. One example is when remittances are used for commercial construction wherein rents are shared between the housing developer and the minor bureaucrats. As for BPO, although it claimed to have generated employment, numerical simulation indicates that the sector has very low intersectoral linkages with the rest of the economy (Magtibay-Ramos et al. 2008). Moreover, when transparency and accountability are absent, the blessing could turn into a curse because inefficiency in the system increases. Some caveats are in order. The above discussions focused on the possible adverse effects of foreign exchange earnings examined in the framework of Dutch disease. This is not to say, however, that we should get rid of these “blessings” any more than one would ban mining a new gold discovery. FDI, BPOs, and remittances may also generate growth externalities, though perhaps not at the same degree as manufactured exports. For example, FDI in banking and telecommunication, induced by deregulation in the 1990s,

26

Majah-Leah Ravago, James Roumasset and Arsenio Balisacan

may have generated some innovations, spillover effects, and learning-bydoing. If the Philippines can follow the example of India, it may also be possible for institutions, human-capital, and knowledge provided by BPOs to create a comparative advantage in knowledge process outsourcing (KPO).43 The provision of business and technical analysis, animation services, pharmaceuticals and biotechnology, and architectural design may be similarly advantaged.44 Similarly, overseas workers may return if and when the economy becomes sufficiently dynamic. These opportunities present policy challenges to facilitate warranted investments. Some general principles are explored in the next section. From vicious to virtuous circle The struggles of the Philippines do not portend permanent doom. In the words of J.R.R. Tolkien (1954): All that is gold does not glitter; all that is long does not last; All that is old does not wither; not all that is over is past. The Philippine case illustrates some of the inefficiencies associated with economic fragmentation. If these forces overcome the positive pull of total capital accumulation and dynamic efficiency, an economy can be trapped in a vicious circle of poverty, population pressure, and resource degradation. The original Brundtland concept demonstrated that sustainable development must allow for the interlinkages among poverty, population pressure, and resource degradation. Figure 1.4 (left panel) depicts the interaction of population pressure and poverty as the notorious Malthusian vicious-circle, and environmental degradation, which exacerbates that circle. Population growth, in the face of a limited resource base, exacerbates poverty by lowering the return to unskilled labor. This in turn prevents mechanisms whereby increased incomes and the rising productivity of human capital lower the demand for children. The population-poverty cycle is exacerbated as households with limited resource-access strive to eke out a living from environmentally fragile areas resulting in further environment degradation.

Economic Policy for Sustainable Development vs. Greedy Growth and Preservationism

27

The migration of labor to hillside agriculture may result in an increased importance of agriculture and natural resource in national income and employment (Balisacan and Rola 2008). However, the end result is stagnation of per capita income, increased poverty, and deterioration of the natural resource base. Moreover, diminishing agricultural productivity growth, limited employment opportunities outside the agricultural sector, high population growth, slow poverty reduction, and dysfunctional institutions are major impediments to achieving sustained macroeconomic development. The conclusion reached by the Commission is that the problems could be addressed only if the three interacting forces in the vicious circle are taken into account collectively. Figure 1.4 Unsustainable and Sustainable Development

The right panel of Figure 1.4 shows the flipside of the vicious circle – the virtuous circle. The virtuous circle incorporates sustainable growth and patterns of sustainable development. The three pillars of sustainability, total environomy, dynamic efficiency, and intertemporal equity, support the virtuous circle of innovation, specialisation, and increased social welfare. The double arrows of the circle represent the dual interactions between the three positive forces. Social welfare increases with higher per capita income, better health conditions, and improved education. These three stimulate demand for investment in children’s human capital over sheer numbers, thereby reducing fertility and decreasing population pressure (Becker et al. 1990).

28

Majah-Leah Ravago, James Roumasset and Arsenio Balisacan

What would it take to transform the vicious circle into a virtuous circle? Orbiting into the virtuous circle is not an easy task. Environmental degradation in rural areas is both a cause and effect of poverty. Unshackling the link requires sustained growth of employment opportunities in the general economy, making it attractive for the poor to move away from low-productivity annual crops on marginal, sloping lands into less erosive perennial crops (Balisacan and Rola 2008) and into the modern sector. A country must secure sources of productivity growth and income diversification through investments in R&D, irrigation, information, and education. The first requirement for transitioning from the vicious to the virtuous circle is the establishment of appropriate institutions. The New Institutional Economics and the new political economy (e.g. Acemoglu 2005; Dixit 1996; Coate and Morris 1995; Besley 2007; Rodrik 2007) can be utilised to help understand the role of institutions in sustainable development. Good institutions are prerequisite for economic cooperation, especially for contracting and market development. The core economic principle from Adam Smith (1776) to Acemoglu (2005) has been that well-functioning competitive markets exploit comparative advantage as defined by its factor endowments, technology, market structure and transportation costs (Stolper and Samuelson 1941; Jones and Scheinkman 1977; and Helpman and Krugman 1985) and transaction costs (Yang 2003). But since factor endowments and other above features change over time, so does comparative advantage. Furthermore, it is endogenous due to the role of specialisation and learning-by-doing (Yang 2003). As comparative advantage changes, new investments become profitable; but these are interdependent, involving both supply and demand-side linkages (Stiglitz 1993). To exploit these interdependencies, mechanisms for investment coordination are needed that go beyond the basic institutions of property, contracts, and markets.45 The government can either facilitate agreements between private groups and between the private and public sectors that coordinate investments (Roumasset and Barr 1992) or give “prizes” such as preferential financing for high-performing exporters (World Bank 1993; Stiglitz 1993). Getting the right institutions is a prerequisite but not a sufficient condition for development because of the presence of externalities. An example is illustrated in Bukidnon, Philippines wherein policy and institutional

Economic Policy for Sustainable Development vs. Greedy Growth and Preservationism

29

initiatives evolved from centralised to decentralised forest ownership and management. Forest clearing has been prevented but because agricultural intensification persists, soil erosion continued, soil fertility declined, and water resources were further degraded (Rola and Coxhead 2005). To correct for these, going beyond the market is required in order to adjust the prices. Thus, the second condition is to get the prices right. Inasmuch as good institutions stimulate growth due to specialisation, getting the prices right accelerates growth. Right prices should allow not only for optimally changing relative prices between consumption, capital, and resources, but changing relative prices between different consumption goods as the comparative advantage changes over time. Furthermore, the presence of externalities calls for corrective prices to bring us closer to optimal outcomes. To facilitate this, economists are increasingly viewing the economy and the environment as a single interlinked system with a unified valuation methodology (Hamilton and Clemens 1999; Dasgupta 2007). The third condition for the endurance of the virtuous circle is for transparency and accountability to become more than buzz words.46 Transparency includes information systems that reveal what the government is doing, what the economic consequences are, what groups enjoy the benefits, and who bears the costs. For example, economic benefit-cost studies inevitably increase transparency. The justice system provides market accountability through the system of contract, tort, and property law, and a commercial code. With this legal support, producers and consumers are accountable to one another. Administrative accountability subjects government officials and programmes to similar principles. User fees, benefit taxation, and constitutional limitations on excess deficit spending are examples (Roumasset 1989). Similarly, bureaucrats can be held accountable to operational performance standards. Conclusions Sustainable development encompasses sustainable growth and dynamically efficient development patterns. Sustainable growth is given a solid and operational framework by positive sustainability, which provides specific guidelines for the accumulation/depletion of produced and natural capital. Resource management and pricing is guided by the Pearce equation. Optimal

30

Majah-Leah Ravago, James Roumasset and Arsenio Balisacan

extraction occurs where the marginal benefit from consuming the resource is equal to its marginal opportunity cost. Resource pricing, ecosystem payments, and non-market valuation of environmental resources should be guided by the sum of extraction, marginal user, and marginal externality costs. In some cases, however, it is not feasible to reach an internal optimum as described by the Pearce equation, because so doing would deplete the resource below some technical threshold or safe minimum standard (Chapter 2). In such cases, the shadow price of the resource is given by its marginal opportunity cost, but not the resource price. Pursuit of sustainable development is the road less travelled. Many countries exhibit unsustainable development (Dasgupta 2007). Others are growing the environomy, but with unnecessary departures from the ideals of dynamic efficiency and intergenerational equity (Roumasset, et al. 2007). The lesson learned from the Philippine illustration is that policy and governance, driven by rent-seeking, exacerbate natural fragmentation and retard economic development. In addition, foreign exchange earnings from foreign aid, foreign direct investments, remittances, and tourism can promulgate a resource curse – causing a further drag on manufactured exports, the engine of growth in developing Asian countries, and increasing the returns to rent-seeking. Rent-seeking further aggravates economic fragmentation. The resulting low rate of economic growth exacerbates population pressure and environmental degradation, leading to a vicious circle of unsustainable development. An economy can escape from the vicious circle and transition into a virtuous circle by incorporating the principles of sustainable growth and development. For sustainable development not to be at odds with policy science, positive sustainability must be combined with efficient sectoral development and poverty reduction. The key is cultivating an economic environment that is conducive to specialisation and innovation. These, plus productive and human capital formation and the efficient conservation of natural resources are all part of sustainable development policy. Government policies should be framed by the principles of positive sustainability – dynamic efficiency and capital formation that allow for standards-of-living to perpetually increase, eventually approaching a golden-rule state wherein per capita welfare increases at the rate of technical change. The role of public policy in sustainable development is gleaned from the oldest lesson in economics – facilitate, don’t dictate (Smith 1776).

Economic Policy for Sustainable Development vs. Greedy Growth and Preservationism

31

Rent-seeking stifles innovation through fragmentation and stagnation. Transparency and accountability are the enemies of rent-seeking. A systems approach that takes into account the environomy, combined with dynamic efficiency and intergenerational equity, will promote meaningful sustainable development. notes a PhD student at the Department of Economics, University of Hawai‘i, Mānoa and Degree Fellow of East-West Center, e-mail: majah@hawaii. edu b Professor of Economics, University of Hawai‘i, Mānoa, and Environmental Director for University of Hawai‘i Economic Research Organization, Saunders 542, 2424 Maile Way, Honolulu, HI 96822 USA, e-mail: [email protected] c Professor of Economics, University of the Philippines Diliman, e-mail: [email protected] 1 Also referred in this chapter as the Brundtland Report. 2 Named after the Head of the Commission, Norwegian Prime Minister Gro Harlem Brundtland. 3 See International Environment Forum (2009) 4 Held in Rio de Janeiro and commonly known as the Earth Summit, the conference resulted in: Agenda 21, the Rio Declaration on Environment and Development, the Statement of Forest Principles, the United Nations Framework Convention on Climate Change, and the United Nations Convention on Biological Diversity. 5 Munasinghe now advocates Sustainomics, “a transdisciplinary, integrative, comprehensive, balanced, heuristic and practical framework for making development more sustainable,” see Munasinghe (1994) and also . 6 Munasinghe’s “Sustainable Development Triangle,” Encyclopedia of the Earth at http://www.eoearth.org/article/Sustainable_development_ triangle 7 See e.g. Hardi and Zdan (1997)

32

Majah-Leah Ravago, James Roumasset and Arsenio Balisacan

8 See Mann (2009) on Visualising Sustainability at (accessed August 20, 2009). Oliva, E. July 14,2007c. “Should the national broadband network be outsourced?” Philippine Daily Inquirer. http://newsinfo.inquirer.net/ breakingnews/infotech/view/20070714-76642/Should_the_national_ broadband_network_project_be_outsourced%3F (accessed August 20, 2009). Olson, M. The Rise and Decline of Nations: Economic Growth, Stagflation, and Social Rigidities. New Haven: Yale University Press, 1982. Olson, M. and R. Zeckhauser. “An Economic Theory of Alliances.” Review of Economics and Statistics 48(1966):266-79. Outsource to India. “Knowledge Process Outsourcing.” http://www. outsource2india.com/why_india/articles/KPO.asp (accessed August 20, 2009). Paldam, M. “Dutch Disease and Rent-Seeking: The Greenland Model.” European Journal of Political Economy 13(3)(1997):591-614. Palma, J. G. “De-industrialization, ‘Premature’ De-industrialization and the Dutch Disease.” In The New Palgrave Dictionary of Economics, Second Edition, S. Durlauf and L. Blume (eds). The New Palgrave Dictionary of Economics Online. Palgrave Macmillan. http://www.dictionaryofeconomics.com/ article?id=pde2008_D000268 (accessed November 7, 2008). Panayotou, T. Foreword to Green Markets: The Economics of Sustainable Development by Oscar Arias. International Center for Economic Growth, Harvard Institute for International Development, 1993. Pearce, D., A. Markandya and E. B. Barbier. Blueprint for A Green Economy. Earthscan, London, Great Britain, 1989. Pender, J. “Population Growth, Agricultural Intensification, Induced Innovation and Natural Resource Sustainability: An Application of Neoclassical Growth Theory.” Agricultural Economics 19 (1-2)(1998): 99112.

42

Majah-Leah Ravago, James Roumasset and Arsenio Balisacan

Pezzey, J. “Economic Analysis of Sustainable Growth and Sustainable Development.” Environmental Department Working Paper No. 15. World Bank, 1989. Pezzey, J. “Sustainability Constraints versus ‘Optimality’ versus Intertemporal Concern, and Axioms versus Data.” Land Economics 73 (4)(1997): 448– 466. PinoyBlogoSphere. “Condohotel Philippines OFW Pag ibig Funds 3M,” July 5, 2009 http://www.pinoyblogosphere.com/2009/07/05/condohotelphilippines-ofw-pag-ibig-funds-3m/ (accessed August 18, 2009). Power, J. and G. Sicat. The Philippines: Industrialization and Trade Policies. London, New York: Oxford University Press, 1971. Quiggin, J. “Discount Rates and Sustainability.” International Journal of Social Economics 24(1/2/3)( 1997): 65-90. Rajan R. and A. Subramanian. “Aid, Dutch Disease, and Manufacturing Growth.” Revision of What Undermines Aid’s Impact on Growth, NBER Working Paper, No. 11657, 2006. http://www.iie.com/publications/ papers/subramanian0606.pdf (accessed on January 30, 2009). Ramsey, F.P. “A Mathematical Theory of Saving.” Economic Journal 38(152) (1928):543-559. Reardon, T. and P. Timmer. “The Supermarket Revolution with Asian Characteristics,” In Reasserting the Rural Development Agenda: Lessons Learned and Emerging Challenges in Asia, A. Balisacan and N. Fuwa (eds). SEARCA, Los Baños, Philippines and ISEAS, Singapore, 2007. Repetto, R., W. Magrath, M. Wells, C. Beer, and F. Rossini. Wasting Assets: Natural Resources in the National Income Accounts. Washington, D.C.: World Resources Institute, 1989. Rhodes, R. “Arsenals of Folly,” The New York Times, November 25, 2007. 0. We assume that an abundant but expensive alternative freshwater source exists. For example, freshwater can be produced by seawater desalination. This is a special feature when dealing with coastal groundwater management, since seawater is abundant in the area. We denote cb as the fixed cost of producing water from desalination, and bt is the amount of water produced by desalination technology at time t. The change in the head level of the aquifer is explained by natural recharge (R), the amount of water extracted (q), and the amount of water discharged to the ocean (l). Recharge is assumed constant; or in other words, the average rainfall is fixed over time. As the head level increases, water pressure and surface area from which water can leak increase. Thus, there is more water discharging from the aquifer into the ocean when the head level is high. We model discharge as a positive, increasing, convex function of head level, i.e., l(h)≥0, l′(h)>0, and l′′(h)>0. The state of the aquifer can be explained as: h˙ t = a[R - l(h t ) - q t ] , where a is a conversion factor converting volume to height.3 Assume that, at any time t, marine algae is harvested at rate mt ; and D -1m(m) is the inverse demand function of marine algae, which depends on € the amount of marine algae consumed. As the stock level decreases, it is more difficult to find and harvest the marine algae. The marginal cost of harvesting marine algae is assumed to be positive, decreasing, and convex with the stock of marine algae (S), i.e., cm(S)≥0, cm′(S)0. Natural growth of

53

Integrated Watershed Management: Trees, Aquifers, Reefs, and Mud

marine algae depends on the size of its own stock and water quality, G(S,Q).4 With respect to its own stock, the growth is assumed to have the traditional properties; strictly concave and attaining a maximum at a finite value of S. Examples of water quality indicators, which may affect the growth of marine algae, are salinity, nitrogen, and temperature. These indicators are related to the amount of freshwater discharged, Q = Q(l(h)). From these relationships, the natural growth of marine algae can be expressed in terms of marine algae stock and aquifer head level, g(S,h). Assume further that freshwater positively affects marine algae growth, gh(S,h)>0. The change in marine algae stock over time is explained by: S˙ = g(S,h) - m . The social planner’s problem is to choose the paths of groundwater extraction, desalinated water, and marine algae extraction in order to maximise the social net benefit, which is equal to the consumer surpluses € consumption, minus the costs of derived from water and marine algae obtaining the water and marine algae. Given a discount rate r, the problem can be written as: max

q t ,b t ,m t



∫e

-rt

0

[

q t +b t

∫ 0

D-1 (x)dx - c(h t )q t - cbb t +

s.t. h˙ t = a[R - l(h t ) - q t ] €

€ €

∫D 0

m

-1

(y)dy - c m (St )mt ]dt

S˙ t = g(St ,h t ) - mt

To simplify notation, define the marginal benefits of water and algae as p ≡ D-1(q+b) and pm ≡ Dm-1(m). The corresponding condition for the optimal water use can be expressed as:

p=c(h)+



mt

p˙ - a(R - l(h))c'(h) ag h (S,h)θ + r + al'(h) r + al'(h)

… (1)

where θ ≥ 0 is the co-state variable of the marine algae stock. The optimality condition in equation (1) requires equality of the marginal benefit of water extraction and the marginal cost, which consists of the extraction cost and the MUC. The MUC is composed of the forgone benefit from the higher future price, the higher extraction cost in the future, and the external cost

54

Sittidaj Pongkijvorasin and James Roumasset

of using water on the marine algae. While the Pearce equation (e.g. Pearce et al., 1989) requires p = c + MUC + MEC, this time the efficiency condition is p = c + MUC, albeit the marginal externality cost is embedded in the MUC. This is because the externality is determined by the resource stock, not the extraction. This distinction is important for corrective taxation of private resource owners, setting royalty charges for resource concessionaires on public lands, and for resource valuation. There are two relevant steady-states for our application: one without desalination and one with desalination. In the following, we first identify the two steady-states and then discuss conditions governing which steady state dominates. Without desalination, the unrestricted steady state head level can be obtained by solving the first-order conditions given that b = 0, p = 0, h = 0 (q = R-l(h)), S = 0 and m = 0 . The optimality condition at the steady state can be written as: p(q*) = c(h*) -



a(R - l(h*))c'(h*) ag h (S*,h*)θ + r + al'(h*) r + al'(h*)

… (2)

Assuming that ghh(S,h) ≤ 0, the derivative of the right-hand side of (2), with respect to the head level, is unambiguously negative (Appendix 2.1). Since the derivative of the left-hand side is zero, the steady state head level that solves equation (2), h*, is unique. However, if desalination technology is available, it will be introduced when the cost of water extraction is high enough (i.e. when the shadow price of water in the absence of desalination exceeds the cost of desalination). From equation (1), when desalination technology is used, water price will be constant and equal to cb . Substituting cb into equation (1), the steady state condition when desalination is used can be derived as: a(R - l(h b*))c'(h b *) ag h (S b*,h b *)θ cb = c(h b*) + r + al'(h b *) r + al'(h b *) … (3)



As the left-hand side of equation (3) is constant and the derivative with respect to h of the right-hand side is negative (Appendix 2.1), a head level solving equation (3) is unique (called hb*). Thus, when desalination is being used, the optimum head will be maintained at its steady state level, hb*. The optimal groundwater extraction is equal to q b* = R - l(h b*).

Integrated Watershed Management: Trees, Aquifers, Reefs, and Mud

55

From the steady-state optimal conditions in equations (2) and (3), it can be shown that the unrestricted steady-state head level (h*) and the backstop steady-state head level (hb*) are both increasing in the value of the marine algae. This implies that a higher algae demand function results in a higher steady state head level. When there is no externality, the last terms on the right-hand side of equations (2) and (3) disappear (θ = 0). Since the last terms are positive, the values of h* and hb* that solve equations (2) and (3) are lower when there is no externality. The use of desalination technology depends on the demand for water and the extraction cost. If groundwater is available to satisfy demand without drawing the aquifer down to hb* (i.e., h* > hb*, or p* < cb ), desalination will never be applied. In that case, the head level will remain at h* in the steady state. On the other hand, if h* < hb*, then desalination will be used at the steady state. The steady state head level will be at hb*. Groundwater vs. Marine Algae In this section, we provide a numerical example of the groundwater management model, including nearshore externalities. For reasons of data availability, we have chosen a study site located along the North Kona coast of the island of Hawai‘i. Data



The Kuki‘o aquifer is a thin basal or Ghyben-Herzberg lens (Duarte 2002). It can be thought of as a lens of less-dense freshwater floating on heavier saltwater. The state equation of head level can be expressed by: h˙ t = (2000 /41γWL)(R - l t - q t ) , where γ is the porosity. W is the aquifer width; and L is the aquifer length. For the Kuki‘o aquifer, the porosity is 0.3; the aquifer unit width is 6,000 m; and the aquifer length is 6,850 m. The state equation for the Kuki‘o

aquifer can be written as h˙ t = 0.00000396(R - l t - q t ) . The water (e.g. recharge, discharge, or extraction) and head level are presented in units of “thousand cubic meters per year” (tm3/yr) and “meters” (m) respectively. The leakage function is assumed to be l(h) = kh 2 , where k is a coefficient specific to €

56

Sittidaj Pongkijvorasin and James Roumasset

a particular aquifer (Mink 1980). Based on the average net recharge for the Kuki‘o aquifer (15,114 tm3/yr), the discharge function is estimated as l(h)=4800h 2. The costs of extracting groundwater are primarily due to the cost of energy needed to lift water to the ground level. Duarte (2002) estimates the unit cost of water extraction for Kuki‘o aquifer as a function of head level: c = 0.00083(403.2 – h). The future cost of desalination is estimated at ($2/m3)5. Demand for water is assumed to have a linear form. We assume that the price elasticity of water demand is -0.7 (based on Griffin, 2006; Dalhuisen et al. 2003). With expected water use of 3,809 tm3/yr priced at $1.27 /m3, the demand for water can be estimated as pt= 3.1 – 0.00048qt accordingly. At the current level of discharge, the average coastal seawater’s salinity around the area is approximately 31 ppt (parts per thousand). We assume that if there is no discharge, the salinity level will be at the average level of seawater (36 ppt). Assuming that salinity is linearly related to discharge, the relationship can be expressed as: sal = 36-0.00033l. In this paper, we choose “limu kohu” (Asparagopsis taxiformisas) as the algae species of interest because of its prevalence and popularity along the Kona coast. Data on this species is available, and limu kohu has the highest market value among all marine algae in Hawai‘i. According to the Department of Land and Natural Resources (2001), around 1,060 kg of limu kohu were commercially picked in the State of Hawai‘i in 2001, with a total value of $19,291. The same report shows that 15.3 percent of sea landing (all species) comes from the Island of Hawai‘i. From that amount, 9.5 percent comes from the study area between Keahole to Kawaihae. Applying these ratios to the limu kohu harvest, we infer that 15.4 kg of limu kohu were harvested from this area in 2001.6 Assuming that the elasticity of demand for limu kohu is -0.4,7 the linear demand function for limu kohu in the area is, then, expressed as: pm = 63.63 – 2.95 m. The initial marine algae stock is indexed at 100 percent. The natural growth function of marine algae is assumed to be in the logistic form, where natural growth depends on the net proportion growth rate (α) and the carrying capacity level (Scc), i.e., g=α(1-S/Scc)S. We assume that the carrying capacity of marine algae depends directly on the salinity level8 and can be explained by Scc=200(31/sal). Studies by Hoyle (1976) and Wong and Chang (2000) indicate that the growth rates of marine algae approximately decrease

Integrated Watershed Management: Trees, Aquifers, Reefs, and Mud

57

by 50 percent when salinity increases from 31 ppt to 40 ppt. Assuming that the growth rate is linearly related to salinity, we can write the function explaining the annual increase in marine algae as: g = (0.8412-0.0172sal)(1 - S/Scc)S. For the cost of harvesting marine algae, we assume that if the marine algae are abundant, harvesting 1 kg of marine algae requires 0.5 hours. With the open-access assumption (S = 100, cm = 18.2) and the minimum wage rate at $7.25 per hour, the unit cost can be written as: ⎛ 201⎞ cm = ⎜ 0.5 + ⎟7.25 ⎝ S ⎠



The parameters used in the model are summarised in Appendix 2.2. Table 2.1 presents the functions used in the model. Table 2.1 Equations Used in the Model Function

Equation

Aquifer state equation

h˙ t =0.00000396(R-lt-q t )

Discharge function

l(h)=4800h 2

Unit cost of water extraction

c = 0.00083(403.2 – h)

Water demand



pt= 3.1 – 0.00048qt

Salinity level

sal = 36 – 0.00033 l

Natural growth function of marine algae

g=(0.8412-0.0172sal)(1-S/Scc)S

Marine algae demand

pm = 63.63 – 2.95 m

Marine algae harvesting cost

⎛ 201⎞ cm = ⎜ 0.5 + ⎟7.25 ⎝ S ⎠

Scenario Description €

Using the above data, we can calculate the optimal water extraction rates, marginal opportunity cost of extraction (extraction cost plus marginal user cost), head levels, and other variables of interest for different scenarios. We use Excel’s Solver to solve the optimisation problems. We start by solving

58

Sittidaj Pongkijvorasin and James Roumasset

for optimal water management without accounting for the value of marine algae (optimal, O). Then, the optimal water extraction rates, taking into account the value of marine algae, are derived (marine algae, L). However, the value of marine algae in the calculation represents only the value for consumption, ignoring the ecological and cultural value. Realising that the consumption value of marine algae may not adequately capture its importance, we alternatively impose a minimum constraint on the level of marine algae stock. Particularly, the stock of marine algae is restricted; it cannot be depleted by more than a specified percentage of the current level. The minimum stock constraint can also be thought of as a “safe minimum standard” corresponding to the precautionary principle. The precautionary principle is commonly suggested by ecologists and environmental economists for dealing with species loss that involves ecological complexity and irreversibility (Harris 2005, p.147). For example, Gollier, Jullien, and Treich (2000) show that the higher scientific uncertainty about future risk is, the higher prevention society should take.9 In this chapter, we run the simulations with three different levels of the constraint on the minimum stock of marine algae. The three scenarios include the restrictions that marine algae stock has to remain at least 50 percent (L50), 75 percent (L75), and 90 percent (L90) of the initial stock level. In a first-best world, a social planner maximises net social benefit by choosing the paths of groundwater extracted, desalinated water, and algae harvested. However, in Hawai‘i as in many parts of the world, algae are harvested on an open-access basis. Controlling the quantity of algae harvested may be politically and administratively difficult. Moreover, in this analysis, the marine alga represents an indicator of the health of the nearshore ecology. In the following application, we assume that the government does not control algae harvesting. Rather, algae are considered to be an open access resource, which implies that algae is harvested until its price equals its extraction cost pm(m)=cm(S). Results For most scenarios, we present the optimisation results for 100 years10. However, due to the increase in the number of constraints and the limitation of the optimisation software, the results of the L90 scenario will be shown for 80 years. If there is no groundwater extraction, the groundwater aquifer

Integrated Watershed Management: Trees, Aquifers, Reefs, and Mud

59

will build up and reach the steady state head level of 1.77 m in 14 years with a salinity level of 31.01 ppt. Marine algae stock increases to 101.07 percent of the initial level. With the current groundwater extraction rate, the steady state head level will be at 1.71 m, lower than its current level and it will be reached after 32 years. The salinity level increases from 31 ppt to 31.37 ppt, while the stock of marine algae decreases from 100 percent to 90.4 percent. However, with the projected extraction rate, the head level will be drawn down to its steady state, 1.54 m, after 50 years. The salinity level increases to 32.27 ppt, and the stock of marine algae decreases by almost half to 55.9 percent in 100 years. Results show that small changes in salinity have significant impacts on the stock of marine algae. Optimal water management, without valuing marine algae, involves high groundwater extraction rates. The extraction rate starts from 5,763.67 tm3/yr and slightly decreases to 5,763.06 tm3/yr in 100 years. The marginal opportunity cost of water increases vary slightly from $0.3334 to $0.3337 per m3. As the cost is always lower than the cost of desalination, desalination will never be exercised. The steady state head level in the optimal scheme is 1.40 m and will be reached in 65 years. The salinity level increases to 32.91 ppt, while the stock of marine algae decreases considerably to 46 percent in 100 years. With marine algae value. When the marine algae value is taken into consideration, optimal extraction rates are slightly lower than those of the case without marine algae. In this case, extraction rate starts from 5,763.64 tm3/yr and decreases to 5,763.04 tm3/yr in 100 years. The steady state head and salinity level, however, are the same as in the previous case. Marine algae stock is equal to 46 percent after 100 years. With lower extraction rate, the cost of water is only slightly higher than that of the case without marine algae. Additional results of interest include the following. First, the optimal extraction rate is higher than the projected or current water use. This implies that the current and projected water use is too conservative and it is possible to increase welfare by extracting more water. This may be partially explained by the fact that a private owner currently operates the wells. The monopoly type of market for water may explain the lower-than-optimal extraction rate (e.g., Perman et al. 2003). Another possible explanation is that the demand for water may be overestimated in the lower ranges of price. In reality, water demand may not be linearly related to price over its entire range.

60

Sittidaj Pongkijvorasin and James Roumasset

Overestimating the demand for water may result in a high extraction rate. By using a “one-dimensional” model of the groundwater aquifer, we abstract from “three-dimensional” effects of hydraulic cones of depression and saltwater upconing on extraction costs. Including such “three dimensional” effects would tend to lower optimal extraction rates.11 Second, desalination will never be used in any market-value scenario because the steady state efficiency price is less than the desalination cost, just as explained in the analytical sections. Extraction rates with algae value taken into consideration are only slightly lower than in the case where marine algae is ignored. This is because the value of marine algae is relatively small compared to the benefit of water consumption. However, it should be emphasised that the value of marine algae in this example presents only the consumption value and ignores the other values it may have (e.g. ecological or cultural value). Safe minimum standards for marine algae stock. Imposing hard constraints on the level of marine algae stock leads to more conservative water management. Given that the stock of marine algae cannot be depleted by more than 50 percent (L50), the steady state head level is at 1.50 m (compared to 1.40 m from the case without the constraint). The steady state extraction rate is around 4,380.48 tm3/yr. The marginal cost of water starts from $0.35 and reaches $1.00 per m3 in the steady state. If the marine algae stock is not allowed to fall below 75 percent of the current level (L75), the steady state head level becomes 1.65 m, with the steady state extraction rate around 2,000 tm3/yr. In this scenario, since the groundwater extraction is low, it is optimal to use desalinated water. Desalination will be first introduced in year 38, when the marginal cost of water is equal to desalination cost, $2/m3. In the steady state, approximately 292 tm3/yr of desalinated water will be used. If the marine algae stock has to be at least 90 percent of the current level (L90), the steady state head level is 1.71 m. The steady state extraction rate is approximately 1,064 tm3/yr, while the steady state amount of desalination is around 1,227 tm3/yr. For this scenario, desalination will begin after 17 years. In both cases, desalination is used whenever the marginal cost of groundwater reaches $2/m3. Figures 2.1-2.2 show the optimal paths of head level, extraction rate, and desalination rate for each scenario.

Integrated Watershed Management: Trees, Aquifers, Reefs, and Mud

61

Figure 2.1 Optimal Paths of Head Level with Different Marine Algae Stock Constraints

Figure 2.2 Optimal Paths of Ground Water Extraction with Different Marine Algae Stock Constraints

It is interesting that time paths of the head level, groundwater extraction, and desalination can be non-monotonic. With the marine algae constraint, it is optimal to deplete the aquifer until the head level is lower than the steady state and then conserve the water later on to fill the aquifer until it reaches its steady state level. Imposing a marine algae constraint may also induce the use of desalination technology as an alternative backstop water source (e.g. L75, L90 cases). In the L75 and L90 cases, it is optimal to stop using groundwater during some periods. The main results of steady states from all scenarios are summarised in Table 2.2.

62

Sittidaj Pongkijvorasin and James Roumasset

Table 2.2 Summary of the Main Results (steady state) From All Scenarios B Head level (m) Groundwater extraction (tm3/yr)

1.71 1704.4

P

O

L

L90

L75

1.54

1.40

1.40

1.71

3809 5763.06 5763.04

1064

L50

1.65

1.50

2000 4380.48

Desalination (tm3/yr)

-

-

-

-

1227

292

-

Salinity (ppt)

31.37

32.27

32.91

32.91

31.36

31.67

32.46

Marine algae stock (% of initial stock)

90.42

55.90

46.01

46.01

90

75

50

-

1.27

0.33

0.33

2

2

1.00

Marginal cost of water ($/m3)

Mangrove vs. Offshore Fishery In most wetland ecosystems, mangroves play an important role and provide several functions, ranging from the ecological to human uses. Of particular interest are the ecological benefits such as nutrient recycling and nursery functions, in addition to the direct use of mangroves as a timber source and for harvesting fishery products on site. The ecological benefits are often externalised to areas outside the mangroves, and are likely to be ignored when a private decision is taken to convert mangroves to other uses. In particular, mangroves in Thailand have been converted to shrimp farms, construction sites and other non-mangrove uses. However, awareness of the external impacts of loss of mangrove areas has increased in recent times. There is now an active movement to replant mangroves in many areas along the country’s coastline. Many attempts have been made in the literature to address economic valuation of mangroves. A good review is Spaninks and van Beukering (1997). The study examines various mangrove functions and methods to assess their economic values, and finds that “most studies limit valuation to use values: the availability of market prices or market prices for substitutes means that the valuation of use values is relatively easy.” Spaninks and Beukering also note that “in most cases, valuation of the impacts of a management alternative on catches in off-site

Integrated Watershed Management: Trees, Aquifers, Reefs, and Mud

63

fisheries is based on somewhat arbitrary assumptions, rather than on detailed scientific information.” Previous studies on the value of mangroves in Thailand include Sathirathai (1998), Talaue-McManus et al. (2001), and Tingsabadh and Pongkijvorasin (2004). While the first two studies estimate mangrove value using static frameworks, Tingsabadh and Pongkijvorasin (2004) develop a simple dynamic model to estimate on- and off-site values. In this section, we apply a stock-to-stock model to examine the impact of mangrove reforestation on an offshore fishery using nutrient status and biological productivity in the bay area. Model and Data In this section, we develop a model integrating the relationships between economic activities, land use change, nutrient status, biological productivity and a fishery in an estuary ecosystem. We divide the land use into three categories: mangroves, shrimp farms, and other economic activities, each of which provides different profiles of carbon discharge into the bay area. The third category affects the carbon profile via biological oxygen demand (BOD) loading from wastewater discharge. The fish population is assumed to have a traditional logistic growth function. However, the carrying capacity depends on the level of organic carbon flow into the bay. We assume that there is no fishing regulation in the area, i.e. that the fishery is an openaccess resource. As a result, fish will be harvested until price is equals the marginal cost of harvesting. The Bandon Bay area of Suratthani province lies to the west of the Gulf of Thailand. It covers approximately 1,070 km2. Freshwater flows into the area from the Tapi River. The estuary area is fringed with mangrove forests, which have been reduced by conversion to shrimp farming and urban development. The nutrient status of the bay depends on the inflow of nutrients from the river, human-induced nutrient discharge from upstream urban settlements, discharge from shrimp farms, and mangrove contribution. Using 1997 data, Talaue-McManus et al. (2001) estimate the flow of organic carbon contained in BOD from anthropogenic sources in the Bandon Bay area to be around 10,642 ton C/yr. The mangrove forest’s contribution of organic carbon is approximately 1,118 ton C/km 2/yr. Shrimp

64

Sittidaj Pongkijvorasin and James Roumasset

farm waste contains organic carbon, which contributes around 39.44 ton C/yr. The area of mangroves in the bay is around 17 km2 while the shrimp farm area is around 80.88 km2. Thus total organic carbon input from land is approximately 32,837.91 ton C/yr. According to their biological study, Talaue-McManus et al. (2001) estimate that the conversion from primary to fishery production in the bay area occurs at the rate of 0.84 percent. Marine animal carbon content is assumed to be 5 percent. This implies that 1 kg of carbon can contribute to (1/0.05)(0.0084) = 0.168 kg of marine life. In this chapter, the intrinsic growth rate of fish is assumed to be 0.5. Thus, the growth function of fish biomass can be represented by: g = 0.5S(1 - S/Scc) + 0.168(CAR-CAR0 ), where S is the stock of fish, Scc represents the carrying capacity of fish, and CAR0 is the carbon stock in the baseline case. In 1993, the price of fish was 37.81 baht/kg. The amount of fish harvest was 1,545 t/yr (Sathirathai 1998). The unit cost of fishing effort is assumed to be 400 baht per pushnet-boat day (FAO 2001). The catch function is assumed to be H=1.5E0.5S0.5. The cost of fishery (C) is equal to wE, where w is the fixed per unit cost of the effort. Since effort (E) depends on harvest (H) and fish stock (S), we can rewrite the total cost function as a function of H and S:

C=

4wH2 9S

The marginal harvesting cost is equal to:





c=

8wH 9S

The net social benefit resulting from a more productive fishery depends on the management regime. Figure 2.3 contrasts this open access equilibrium, xOA, with the socially optimal harvest, x*, which occurs where price and marginal opportunity cost, MOC, are equal. MOC is total marginal cost, comprised of unit harvesting cost, c, and marginal user cost, MUC. Marginal user cost is the loss in the social value of the asset by harvesting a marginal unit. If many competing boats are exploiting the fishery, however, they will not internalise the MUC that they impose on one another, and the equilibrium will be at xOA, where price, p, equals unit harvesting cost, c.

Integrated Watershed Management: Trees, Aquifers, Reefs, and Mud

65

Figure 2.3 Open Access vs. Socially Efficient Effort

Assuming that the existing fishery is in open-access equilibrium, the initial fish stock and fish carrying capacity are calculated to be 7,266 tons and 12,643 tons, respectively. The discount rate used in this paper is 3 percent. Results First consider the effects of changing the mangrove area under conditions of open access. The effects of adding and subtracting one square kilometer (denoted as mangrove +1 and mangrove -1), respectively, from the current mangrove area are compared. From the integrative framework in the previous section, a decrease in mangrove area decreases the inflow of organic carbon, fish carrying capacity, the growth of fish, and catch per unit of effort. As a result, the quantity of fish harvested and the total revenue both decrease. If the mangrove area is reduced by one square kilometer, there will be less carbon discharged into the bay, thereby lowering growth and stock. The smaller stock shifts up the marginal cost curve, c, in Figure 2.3, resulting in lower effort and harvest. Steady-state harvest decreases from 1,545 tons to 1,384 tons (equivalent to a -10.44% change). The stock of fish decreases

66

Sittidaj Pongkijvorasin and James Roumasset

from 7,282 tons to 6,508 tons in the steady state (around -10.62%). The revenue decreases from 58.5 million to 52.3 million baht in the steady state (a decrease of 6.2 million per year). However, because the fishery sector is assumed to be open-access, the profit is zero for each year. With a one square kilometer increase in total mangrove area, the carbon discharged in the area increases by 1,118 t/yr. This results in a 190 ton additional growth of fish annually (1,118 tonsC multiplied by 0.17 tons/ton C). As a result, the catch jumps from 1,545 tons to 1,583 tons in the first year and eventually reaches the steady state at 1,669 tons (Figure 2.4). The revenue increases from 58.5 million baht to around 63.1 million baht. Compared to the baseline case, the catch and total revenue increase by approximately 7.8 percent in the steady state, which will be reached in less than 20 years. Figure 2.4 Transition to New Steady-State Levels Due to Adding and Subtracting One Square Kilometer of Mangroves

Integrated Watershed Management: Trees, Aquifers, Reefs, and Mud

67

Figure 2.5 Total Revenues from Adding and Subtracting One Square Kilometer of Mangroves

Under open-access, however, there is no welfare gain from an increase or decrease in fishery productivity. There is no consumer surplus under the assumption that the local fish price is given exogenously, and all potential fishery rents are dissipated. Increased revenues are simply offset by increased costs. Next, we examine the contrasting case of optimal management, wherein fish are harvested in order to maximise discounted net profits over time. This is also the case where the fishery is managed by a sole owner. Simulation results show that, given existing data and parameters, the catch in the first year is 648 tons and gradually increases to around 957 tons in the steady state. The stock of fish in the area increases from 7,226 tons to 10,230 tons. Total revenue at the steady state is approximately 36.20 million baht. Although the revenue is lower than the case of open-access, the cost of the fishery decreases dramatically to 15.95 million baht. Net revenue increases from 14.17 million baht in the first year to 20.27 million baht in the longrun (Figure 2.6). Discounted net benefit over 50 years is calculated at 504.66 million baht.

68

Sittidaj Pongkijvorasin and James Roumasset

Figure 2.6 Revenue, Cost, and Profit of Fishery Production Under Sole Ownership

If the mangrove is depleted by one square kilometer, optimal catch in the first year falls to 640 tons; and profit to 14.12 million baht. In the steady state, the optimal catch is 900 tons, and profits are 19.30 million baht. The fish stock decreases to 9,770 tons. Discounted net revenue over 50 years is calculated to be 481.35 million baht. A one square kilometer depletion of mangrove area results in lower net revenue of almost 1 million baht annually, 22.65 million baht over 50 years. In the opposite case where mangrove area increases by one km2 , the steady state fish stock is 10,655 tons. The optimal catch is 1,007 tons in the steady state. This creates net revenue equal to 21.16 million baht annually. The discounted net revenue over 50 years is equal to 526.19 million baht. A comparison of net revenue under different scenarios is shown in Figure 2.7. By comparing the outcomes under different fishery controls, we can draw a valuable policy lesson. From this mangrove-fishery example, we illustrate that, on one hand, the activities directed at the wetland largely affect the other interacting resources, such as an offshore fishery. On the other hand, the outcomes of mangrove conservation or restoration also depend on how interacting resources are managed, in particular the fishery. The productivity benefits of conservation or restoration of mangroves can be dissipated if the fishery is inefficiently managed.

Integrated Watershed Management: Trees, Aquifers, Reefs, and Mud

69

Figure 2.7 Net Revenue of Depleting/Growing Area of Mangrove Under a Regulated Fishery

Conclusions Optimal management of one resource has to take into account the interactions with the surrounding ecosystem. Without well-established markets for these impacts, externality problems arise, leading to inefficient resource management. We have emphasised that traditional resource economics may not be capable of explaining the problem of resource management with interacting stocks, especially for the case where the stock of externality is determined by the stock of resource, not the use. In the first section, we develop a theoretical model, specifically a regional hydrologic-ecologic-economic model concerning the interaction between groundwater use and a nearshore ecosystem. Considering the positive external impacts of the resource on the nearby ecosystem, the resource stock at the steady state must be maintained at a higher level than otherwise. The aquifer’s steady state head level increases when limu value is taken into account. Desalinated water will only be used when the shadow price of water in the absence of desalination exceeds the cost of desalination. We numerically illustrate the model using data from the North Kona Coast of the island of Hawai‘i. Two different approaches are applied in order to incorporate the limu consideration. The first approach is to include

70

Sittidaj Pongkijvorasin and James Roumasset

the market value of limu directly in the objective function. In this case, the water extraction rate is only slightly lower when the value of limu is included. This is because the value of limu is small relative to that of water. This could change if cultural and ecological values were directly estimable. Desalination technology will never be used in the market-value case. The second approach, following the precautionary principle of “safe minimum standard,” models the limu concern by imposing a minimum stock constraint. We find an interesting result that the paths of water extraction, head level, and water price (without desalination) are non-monotonic. It is optimal to deplete the aquifer below the steady state level, followed by a conservation period. Even though the algae stock cannot be maintained indefinitely at the safe minimum standard with the aquifer head below its steady state level, it is optimal to temporarily “overdraft” the aquifer and build it up later. We also provide a numerical simulation explaining the interaction between wetland management (mangrove and shrimp farming) and the offshore fishery at Bandon Bay, Thailand. A change in land use affects the nutrient mix in the bay, which in turn, has an impact on the level of fish stock. We show that, under open-access fishery management, depleting one square kilometer of mangrove area decreases the fish stock and the fishery output in the area by around 10.62 percent and 10.44 percent, respectively. The revenue from offshore fisheries decreases by 6.2 million baht ($187,878) per year in the steady state. The net revenue is equal to zero under openaccess. If the fishery is managed by a sole owner (either private or state), the catch is lower and total revenue decreases. However, the net revenue is maximised, around 20 million baht ($614,242) annually in the steady state. We found that depleting one square kilometer of mangrove area, net revenue decreases by almost 1 million baht ($30,000) annually in the steady state or 22.65 million baht ($686,363)over 50 years. This external cost should be added to the other costs of converting the mangrove areas to other purposes for a more complete social benefit-cost analysis.

71

Integrated Watershed Management: Trees, Aquifers, Reefs, and Mud

Appendix 2.1 Given that c'(h) < 0,c''(h) > 0,R - l(h) > 0,l'(h) > 0,and l''(h) > 0 , the derivative of the left-hand side of equations (2) and (3) with respect to h is equal to ac'(h)l'(h) a(R - l(h))c''(h) a 2 (R - l(h))c'(h)l''(h) c'(h) +€ + + (r + al'(h))2 r + al'(h) r + al'(h) , which is unambiguously negative. €

Appendix 2.2 Parameters used in the limu model Parameter Explanation [units]

Value

g

Porosity

0.3

W

Aquifer width [m]

6000

L

Aquifer length [m]

6850

R0

Net Recharge [tm /yr]

15114

L0

Current discharge [tm /yr]

14700

h0

Initial head level [m]

hgrd

Ground elevation [m]

EC

Energy cost of lifting 1 m3 of water up for 1 m. [$/m/m3]

cb

Desalination cost [$/m3]

q0

Current water extraction rate [tm3/yr]

qp

Projected water use [tm /yr]

3809

p0

Current water price [$/m ]

1.27

η

Long-term elasticity of demand for water

-0.7

sal0

Current salinity level [ppt]

S0

Initial limu stock [%]

100

Scc0

Current limu carrying capacity [%]

200

m0

Current limu harvesting [kg/yr]

15.4

pm0

Current limu price [$/kg]

18.2

ηm

Elasticity of demand for limu

-0.4

r

Discount rate [%]

3

3

1.75 403.2 0.00083 2

3

3

1074.4

31

3

72

Sittidaj Pongkijvorasin and James Roumasset

Notes a Faculty of Economics, Chulalongkorn University, Thailand, e-mail: [email protected] b Department of Economics, University of Hawai‘i at Mānoa, USA, e-mail: [email protected] 1 This research was supported by the United States Geological Survey (USGS) through the University of Hawai‘i Water Resources Research Centre (WRRC) with project titles (grant award number): “Coastal Groundwater Management in the Presence of Stock Externalities” (05HQGR0146), “Efficient Water Management and Block Pricing for Integrated Aquifers: Lessons from Southern O‘ahu” (2007HI183B), and “Integrated Management of Multiple Aquifers with Subsurface Flows and Inter-district Water” (2005HI103B). S. Pongkijvorasin also received travel grant from the Southeast Asian Regional Centre for Graduate Studies (SEARCA) for presentation of this chapter. The authors thank Chris Wada for comments and editorial assistance. 2 We use a dollar exchange rate of 33 baht in this chapter. 3 a is specific to the site. It depends on the shape (e.g. length and width), and the porosity of the aquifer. 4 The population dynamics of marine algae is also influenced by competition with other algal species and by herbivores. Such complicated dynamics are not considered here. 5 Pitafi and Roumasset (2009). Values are in 2001 dollars throughout this section. 6 The actual number of marine algae harvest maybe different due to two factors. First, because marine algae are not commercially harvested extensively in the area, the actual number may be lower. This is because the proportion data used here represents an all-species harvest, not only marine algae. Second, the total harvest is higher than the commercial harvest alone. With limited data, we assume here that the total harvest is equal to 15.4 kg/year. 7 Marasco (1974) reviews the elasticity of demand for fish in the U.S. He finds that it ranges from -0.65 to 0.23. 8 The carrying capacity concept applied here can represent the competitiveness among different species. When salinity is high, invasive

Integrated Watershed Management: Trees, Aquifers, Reefs, and Mud

73

algae can grow better than indigenous algae. Thus, the carrying capacity of the indigenous species decreases as salinity increases. This effect is separated from the direct effects of salinity on the growth rate. 9 Mathematically, they show that precautionary principle (as defined by previous prevention effort) is an optimal policy if the inverse of marginal utility is concave. 10 Because the numerical example is illustrated by using a finite time simulation, there is a problem of extensive resource exploitation near the end period. Accordingly, the model was run with various lengths in order to confirm that each scenario reaches its steady state. Results show that the steady state is reached within 100 years for all scenarios. In most cases, the model is run for 200 years. The results from the first 100 years are reported. 11 See e.g. Haitjema (1995) for a discussion of one, two, and threedimensional modeling of groundwater hydrology. References Brander, J.A. and M.S. Taylor. “The Simple Economics of Easter Island: A Ricardo-Malthus Model of Renewable Resource Use.” American Economic Review 88(1)( 1998): 119-138. Clark, W. and B. Kates. Foreword in Exploring Sustainability Science, A Southern African Perspective, M. Burns and A. Weaver (eds). AFRICAN SUN MeDIA, Stellenbosch 7600, 2008. Dalhuisen, J.M., R.J.G.M.Florax, H.L.P. de Groot, and P. Nijkamp.. “Price and Income Elasticities of Residential Water Demand: A Meta-analysis.” Land Economics 79(2)( 2003): 92-308. Department of Land and Natural Resources.Commercial Marine Landings Summary Trend Report: Calendar Year, 2001. http://www.hawaii.gov/ dlnr/dar/pubs/cmlstr2001.pdf. Duarte, T.K. “Long-term Management and Discounting of Groundwater Resources With A Case Study of Kuki‘o, Hawai‘i,” PhD dissertation, Massachusetts Institute of Technology, 2002. Food and Agriculture Organisation of the United Nations (FAO). Fishcode Management: Report of a Bio-Economic Modelling Workshop and a Policy Dialogue Meeting on the Thai Demersal Fisheries in the Gulf of

74

Sittidaj Pongkijvorasin and James Roumasset

Thailand held at Hua Hin, Thailand 31 May – 9 June 2000. Field Report F-16, Rome. ftp://ftp.fao.org/docrep/fao/field/006/y1160e/Y1160E00. pdf Gollier, C., B. Jullien, and N. Treich. “Scientific Progress and Irreversibility: An Economic Interpretation of the ‘Precautionary Principle.” Journal of Public Economics 75(2000): 229-253. Griffin, R.C. Water Resource Economics: The Analysis of Scarcity, Policies, and Projects. Massachusetts: The MIT Press, 2006. Harris, J.M. Environmental and Natural Resource Economics: A Contemporary Approach. 2nd ed. New York: Houghton Mifflin Company, 2005. Haitjema, H.M.. Analytic Element Modeling of Ground Water Flow. San Diego: Academic Press, 1995. Hoyle, M.D. “Autecology of ogo (Gracilaria bursapastoris) and limu manauea (G. coronopifolia) in Hawai‘i with Special Emphasis on Gracilaria Species As Indicators of Sewage Pollution,” Ph.D. dissertation, University of Hawai‘i, 1976. IPCC. “Summary for Policymakers, Climate Change 2007: The Physical Science Basis.” In Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Solomon, S., D. Qin, M. Manning,Z. Chen, M. Marquis, K.B. Averyt, M.Tignor and H.L. Miller (eds.),. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 2007. Krulce, D. L., J.A. Roumasset,, and T. Wilson. “Optimal Management of a Renewable and Replaceable Resource: The Case of Coastal Groundwater.” American Journal of Agricultural Economics 79(1997): 1218-1228. Marasco, R. “Food from the Sea: An Economic Perspective of the Seafood Market.” American Journal of Agricultural Economics 56(5)(1974): 1030-37. Mink, J.. State of the Groundwater Resources of Southern O‘ahu. Honolulu: Board of Water Supply, 1980. Nordhaus, W.D. “To Slow or Not to Slow: The Economics of the Greenhouse Effect.” The Economic Journal 101(407)( 1991): 920-937. Ostrander, G. Preface in Sustainability Science for Watershed Landscapes, J. A. Roumasset, K. M. Burnett, and A.M. Balisacan (eds.) Singapore: Institute of Southeast Asian Studies; Los Banos, Philippines: Southeast Asian Regional Centre for Graduate Study and Research in Agriculture, 2009. Pearce, D. and A. Markandya. “Marginal Opportunity Cost As A Planning Concept,” in Environmental Management and Economic Development, G.

Integrated Watershed Management: Trees, Aquifers, Reefs, and Mud

75

Schramm and J.J. Warford (eds). Baltimore: The Johns Hopkins University Press, 1989. Perman, R., Ma, Y., J. McGilvray, and M. Common. Natural Resource and Environmental Economics, 3rd ed. Pearson Addison Wesley, New York, 2003. Pitafi, B.A. and J.A. Roumasset. “Pareto-Improving Water Management over Space and Time: The Honolulu Case.” American Journal of Agricultural Economics 91(1)( 2009): 138-153. Sathirathai, S. “Economic Valuation of Mangroves and the Roles of Local Communities in the Conservation of Natural Resources: Case Study of Suratthani, South of Thailand,” EEPSEA Research Report Series, 1998. Settle, C., T.D. Crocker, and J.F. Shogren. “On the Joint Determination of Biological and Economic Systems.” Ecological Economics 42(1-2)(2002): 301-311. Settle, C.. and J.F. Shogren. “Modeling Native-Exotic Species within Yellowstone Lake.” American Journal of Agricultural Economics 84(5)( 2002): 1323-1328. Sinclair, P. “On the Trend of Fossil Fuel Taxation.” Oxford Economic Papers 46(1994): 869-877. Spaninks, F. and P. van Beukering. “Economic Valuation of Mangrove Ecosystems: Potential and Limitations,” CREED Working Paper no. 14, 1997. Stern, N. The Economics of Climate Change: The Stern Review. Cambridge University Press, 2006. Talaue-McManus, L., H.H. Kremer, and JI. Marshall Crossland.. “SARCS/ WOTRO/LOICZ: Biogeochemical and Human Dimensions of Coastal Functioning and Change in Southeast Asia,” Final report of the SARCS/ WOTRO/LOICZ project 1996–1999. LOICZ Reports and Studies No. 17, Texel, The Netherlands: LOICZ IPO, pp. 277, 2001. Tingsabadh, C. and S. Pongkijvorasin. “Economic Valuation of Mangroves for Improved Usage and Management in Thailand.” In Wetlands Ecosystems in Asia: Function and Management, M.H. Wong (ed). Amsterdam, Oxford, Elsevier, 2004. Wong, S.L. and J. Chang. “Salinity and Light Effects on Growth, Photosynthesis, and Respiration of Grateloupia filicina (Rhodophyta).” Aquaculture 182(2000): 387-395.

76

Sittidaj Pongkijvorasin and James Roumasset

Transdisciplinary Research in Watershed Conservation: Experiences, Lessons and Future Directions

77

3 Transdisciplinary Research in Watershed Conservation: Experiences, Lessons, and Future Directions

Chieko Umetsu,a Makoto Taniguchi, Tsugihiro Watanabe, and Shigeo Yachi

Abstract The importance of consilience among different disciplines to solve global environmental issues has long been discussed among researchers and decision makers. As global environmental problems become more complex in the real world, society calls for transdisciplinary approaches that analyse and solve environmental problems. The watershed is a geographical boundary of ecosystem that supports social and economic activities at the regional scale. The problems associated with watersheds are considered to have multi-faceted aspects. The integrated project-type research creates a new arena for transdisciplinary research. It is possible to integrate research through concepts, methodolog y, and a common geographical area of study. The research organisation of three transdisciplinary watershed projects is compared, specifically, the Impact of Climate Changes on Agricultural Production System in the Arid Areas (ICCAP) Project, The Lake Biwa-Yodo River Watershed Project, and Urban Subsurface Environment Project. In the three projects, the research organisations evolved to pursue mission and transdisciplinary goals. Various tools for integration were observed, which included the comparison of spatial and temporal changes, generation of indicators, development of coupled models, and communication among researchers. More importantly, a tool to integrate scientific knowledge and transdisciplinary communication with the stakeholders of the watershed was used to disseminate scientific information for resource management.

78

Chieko Umetsu, Makoto Taniguchi, Tsugihiro Watanabe, and Shigeo Yachi

Introduction The importance of consilience among different disciplines to solve global environmental issues has long been discussed among researchers and decision makers. Society calls for transdisciplinary approaches to analyse and solve these worsening global environmental problems. The watershed is a geographical boundary of ecosystem that supports social and economic activities at the regional scale. The problems associated with watersheds are considered to have multi-faceted aspects. These include the spatial and temporal aspects of material flow and balance, the present and future climate conditions, resource use, hierarchical structure of social and cultural organisations, and ownership. Thus, the sustainability of watershed requires a “synthesis of knowledge from different specialised fields of human endeavor” (Wilson 1998). This chapter tries to review some transdisciplinary research projects that have been implemented for watershed management. The projects considered effective research organisations to achieve goals. Particular focus was placed on how the project integrated scientific knowledge/findings into policy formulations or provided recommendations for watershed resource management. The chapter is organised into five sections. The first section identifies the possibility of consilience in transdisciplinary research exploring the stages of transdisciplinary project and types of integration. The second section describes experiences from the three Research Institute for Humanity and Nature (RIHN) Watershed projects. The third section considers the evolution of research organisations for transdisciplinary research. The next section shows the possible tools for integration and the last section presents the lessons learned and future directions for watershed management. Consilience and Transdisciplinary Research In a path-breaking book, Consilience: The Unity of Knowledge, a biologist named Edward O. Wilson mentioned the possibility of uniting the sciences with the humanities and synthesising knowledge from different disciplines. In his 1840 book, The Philosophy of the Inductive Sciences, William Whelwell first introduced the concept of consilience. He explained consilience as literally a

Transdisciplinary Research in Watershed Conservation: Experiences, Lessons and Future Directions

79

“jumping together” of knowledge as it links facts and fact-based theory across disciplines to create a common groundwork for explanation” (Wilson 1998). According to Wilson, the “new synthesis” of knowledge in natural and social sciences, as an example of consilience, is the unification of Darwin’s theory of evolution with the new information on genetics. The evolution theory was empirically tested by the new knowledge of genetics as the advancement of science. In case of environmental issues, Wilson asserts that political decision makers cannot solve most of the problems “without integrating knowledge from the natural sciences with that of the social sciences and humanities.” There is a need to link knowledge with environmental policy, ethics, biology and social science since most real world problems arise from the intersection of the above domains (Wilson 1998). Consilience is closely related with transdisciplinarity. Max-Neef (2005) concludes that “transdisciplinarity is more than a new discipline or superdiscipline because it is actually a different manner of seeing the world – more systemic and more holistic.” He defined transdisciplinarity in the context of disciplinarity, multidisciplinarity, pluridisciplinarity, and interdisciplinarity. Disciplinarity means the existence of a discipline with specialisation in isolation. Multidisciplinarity is the existence of multiple disciplines without cooperation. Pluridisciplinarity is the state where there is cooperation among disciplines in the same level but coordination is not present between levels. In interdisciplinarity, there is coordination between two hierarchical levels. Transdisciplinarity, therefore, according to Max-Neef (2005), means the outcome of coordination among all hierarchical levels. The hierarchical level starts with the empirical level, then goes on to the pragmatic level, then the normative level, and finally the value level. The empirical level consists of chemistry, biology, and sociology. The empirical levels are coordinated into the pragmatic levels, such as architecture, engineering, and agriculture. The pragmatic levels are grouped into normative levels consisting of planning, design, politics, and law. The top of hierarchy is the value level consisting of values, ethics, and philosophy (Max-Neef 2005). How then can consilience and transdisciplinarity be realised? The rest of the section describes the stages of project-type research and directions to realise consilience. Three stages of area studies are described in Tachimoto (2003, 2004). These include multidisciplinary, comprehensive, and integrated area studies. Although Tachimoto used these criteria to describe the stages of

80

Chieko Umetsu, Makoto Taniguchi, Tsugihiro Watanabe, and Shigeo Yachi

area studies, this chapter used the criteria to analyse the stages of integrated project-type research. Figure 3.1 depicts the three stages of integrated project-type research. The first stage is the multidisciplinary project-type research. In this stage, the members from different disciplines work together in the same project. However, the members consider the project as an arena for their own individual studies. In other words, the members are like vegetables in the salad bowl. There is almost no influence among each other for the outcome of the research. This must have been true for the early stage of the so-called “integrated project.” Figure 3.1 Stages of Integrated Project-type Research

Stage 1: Multidisciplinary project-type research

Many disciplines exist in the project but no influence among each other.

Stage 2: Comprehensive (inter-disciplinary) project-type research Project aims at comprehensive outputs and some influence by other disciplines.

Transdisciplinary Research in Watershed Conservation: Experiences, Lessons and Future Directions

81

Stage 3: Integrated (transdisciplinary) project-type research

Disciplines becomes smaller and bonds between disciplines.

The second stage is the comprehensive (pluridisciplinary) project-type research. In this stage, the project focuses more on the comprehensive and interdisciplinary outputs. The members are somewhat influenced by other members in the same project. The third stage is the integrated project-type research. This stage creates a new arena for interdisciplinary and/or transdisciplinary research. In this stage, the project serves as the melting pot for the integrated study across disciplines. A holistic approach is used to tackle issues and realise consilience. In this stage, the presence of disciplines becomes smaller and instead the bonds among disciplines become thicker. In addition, there are three typical directions of integrated research. One direction is to integrate research using common concepts and missions for environmental issues. This common ground is shared among researchers from various disciplines. Thematic concepts, such as sustainability and resilience, may be examples for this direction of integrated research. Another direction is to integrate research through a method of analysis. An effort to integrate various field of science with a unified method or a model may be one example. The last direction is to integrate research by studying the same geographical area. The area of study that has been developed, particularly after the Second World War, has this type of integration method. Experiences from RIHN Watershed projects In 1995, the Japan Science Council proposed the necessity to “examine the founding of a central research organisation that will promote integrated

82

Chieko Umetsu, Makoto Taniguchi, Tsugihiro Watanabe, and Shigeo Yachi

cooperative research toward the solution of global environmental problems.” In March 2000, the Preparation Committee of the Institute proposed the foundation of the Research Institute for the Global Environmental Sciences to promote integrated research projects. This is done by amalgamating various broad disciplines, from humanity and social sciences to the natural sciences and build networks among university researchers within and outside the country. The Research Institute for Humanity and Nature (RIHN)1 was founded in 2001 as one of the inter-university research institutes of the Ministry of Education, Culture, Sports, Science and Technology, the Government of Japan. The research topics of the initial five RIHN projects were on water issues, i.e. Seyhan River Basin, Oasis Region in Inner Mongolia, the Lake BiwaYodo River Watershed, and Global Virtual Water. This chapter discusses the experiences from the three watershed projects, titled “Impact of Climate Changes on Agricultural Production System in the Arid Areas (ICCAP) (2002-2006)”, “Multi-Disciplinary Research for Understanding Interactions between Humans and Nature in the Lake Biwa-Yodo River Watershed (2002-2006),” and “Human Impacts on Urban Subsurface Environments (2006-2010).” 1. Impact of Climate Changes on Agricultural Production System in the Arid Areas (ICCAP) (Study Period: 2002-2006) a. Outline The ICCAP project was one of five research projects that started out as the first phase of RIHN’s research activities in 2002. Their study site was the Seyhan River Basin located in the central south of Turkey. The Seyhan River extends to Adana City, and flows into the Mediterranean Sea. The total watershed area of the Seyhan River Basin is 21,734 square kilometers and rainfed wheat and barley (22.2%), pasture areas (31.8%), and forests (19.4%) dominated the land cover class in the upper and middle basin. Along the downstream of the Seyhan River, the Lower Seyhan Irrigation Project (LSIP) in Adana was initiated by the Turkish government as one of the most important irrigation projects located in southern Turkey. The Government constructed the Seyhan Dam in 1956 for the purposes of

Transdisciplinary Research in Watershed Conservation: Experiences, Lessons and Future Directions

83

irrigation, power generation, and flood protection. The reservoir can store 1.2 billion cubic metres of irrigation water to LSIP. There is a concern among local authorities, as well as farmers, that climate change might affect the local agricultural production. The project intends to address this concern by focusing on the local impacts of climate change in the 2070s. b. Objectives ICCAP aimed at assessing the impacts of climate change during the 2070s, including the possible adaptations of the agricultural production system in the Seyhan River Basin to climate change. For this purpose, the project developed a methodology or model to assess climate change impacts, and to improve the Regional Climate Model (RCM) to help generate future regional climate scenarios. ICCAP also evaluated the vulnerability of agricultural production systems from natural changes, and suggested measures to enhance the sustainability of agriculture through the integrated impact and adaptive assessment of climate change (ICCAP 2007). c. Research Organisation The ICCAP team consisted of more than a hundred researchers from Japan and Turkey, with Dr. Tsugihiro Watanabe, an irrigation engineer, as the Project Leader. The Scientific and Technological Research Council of Turkey (TÜBITAK), and Cukurova University in Adana were the major collaborating institutions. Figure 3.2 shows the research organisation of ICCAP. The project researchers are grouped into six sub-groups based on conventional disciplines – climate, hydrology and water resources, irrigation and drainage, crop production, vegetation, and socio-economics. The climate group provided the regional climate conditions during the 2070s using the RCM. Other groups assessed the impacts of climate change on river flow, crop growth, vegetation changes, irrigation water, agricultural production of farm households, and other aspects.

84

Chieko Umetsu, Makoto Taniguchi, Tsugihiro Watanabe, and Shigeo Yachi

Figure 3.2 Six Research Groups of ICCAP

ICCAP (2007)

d. Integration The member of each sub-group was quite stable from the start to the end of the project. However, there were some integral transformations at the later stages. At the same time, the RIHN researchers and the post-doctoral fellows provided integral drives across disciplines of the six sub-groups. The integration activity and transdisciplinary communication were mostly realised in the process of scenario formation for climate change (Figure 3.3).

Transdisciplinary Research in Watershed Conservation: Experiences, Lessons and Future Directions

85

Figure 3.3 Impact Assessment Flow and Scenarios in ICCAP

ICCAP (2007)

Many researchers discussed the possible climate change scenarios in the future (Nagano et al., 2007). One of the integration efforts was made through the implementation of various modeling exercises. The RCM, developed by ICCAP, utilised the output from the Global Circulation Model (GCM) and provided future climate conditions, such as precipitation, temperature, radiation, wind speed, and humidity. The SiBUC (Simple Biosphere including Urban Canopy) land surface model was one of the integrated models of ICCAP which utilised climate information from RCM to calculate river water flow. The output of SiBUC model, i.e. the water availability for LSIP, was used to analyse cropping patterns so that farmers may determine the water constraints to their cropping patterns according to the different scenarios provided (Figure 3.4).

86

Chieko Umetsu, Makoto Taniguchi, Tsugihiro Watanabe, and Shigeo Yachi

Figure 3.4 Framework of Analysis for Integrating SG Outputs in ICCAP

Umetsu et al. (2007)

ICCAP provided the final project outputs to the local water authority. The General Directorate of State Hydraulic Works and the members of water users associations in LSIP discussed and recommended the strategies against possible future climate change impacts. 2. Multidisciplinary Research for Understanding the Interactions between Humans and Nature in the Lake Biwa-Yodo River Watershed (Study Period: 2002-2006) a. Outline Another first phase project of RIHN was the Lake Biwa-Yodo River Project. The Lake Biwa-Yodo River watershed is one of the most intensively humandominated watersheds in Japan. Since the River’s water quality began to deteriorate in the 1970s, the Shiga Prefectural Government worked at reducing the human load inflow by establishing laws and promoting the development of sewage plants. At present, the control of non-point sources,

Transdisciplinary Research in Watershed Conservation: Experiences, Lessons and Future Directions

87

including agricultural drainage, has become an urgent issue. The problem of agricultural drainage also includes the turbid water flowing out from the paddy fields by means of medium and small rivers until it finally drains into Lake Biwa (Yachi 2006). In general, a watershed has a “nested” structure that encompasses the human activities that are occurring along its boundaries (Figure 3.5). Lake Biwa watershed was thus chosen as an example of this kind of nested structure. Its hierarchy consisted of three spatial levels – macroscale (the entire watershed), mesoscale (the regional community), and microscale (the local community) (Yachi 2006). Figure 3.5 Lake Biwa-Yodo River Project – Four Reasearch Groups

Lake Biwa-Yodo River Project Final Report (2007)

The project particularly focused on ways of viewing the problems that are occurring in the watershed at each level. Different views can cause conflicts of interest between levels and thus obstruct watershed management. The project believed that effective watershed management strategies were difficult to implement because of the barriers to active communication that aimed at balancing competing interests. Because of these barriers, various stakeholders were dispersed throughout the different levels.

88

Chieko Umetsu, Makoto Taniguchi, Tsugihiro Watanabe, and Shigeo Yachi

b. Objectives The project’s four objectives were achieved through interdisciplinary partnership and local execution of activities in each hierarchy. The project aimed at: (1) clarifying the entire problem; (2) developing environmental diagnostic indices for each hierarchy, and at the same time researching on a method to support adaptive management; (3) establishing a methodology to promote inter-hierarchical communication; and (4) identifying important problems related to water environment, in the case of the Yodo River watershed. The Yodo River watershed constitutes the downriver reaches of Lake Biwa. These water environment problems were identified based on the findings of the research on the Lake Biwa watershed. Based on these case studies, the project intended to provide appropriate recommendations for the management of the Lake Biwa-Yodo River watershed. It also aimed at contributing to the establishment of global environmentolog y, i.e. a new synthesis of global environmental studies. c. Research Organisation Dr. Shigeo Yachi, a theoretical ecologist, led the Lake Biwa-Yodo River watershed project. The project consisted of 48 members, including 8 RIHN researchers. Figure 3.6 shows the research groups comprising the Lake Biwa-Yodo River Project. The project consisted of four working groups (WGs), Material Cycling WG, Ecosystem WG, Social and Cultural WG, and Watershed Information System WG. The number of project members increased from 19 in 2002 to 48 in 2007.

Transdisciplinary Research in Watershed Conservation: Experiences, Lessons and Future Directions

89

Figure 3.6 Four Research Groups in Lake Biwa-Yodo River Project

RIHN Annual Report 2006 (2007)

During project implementation, various working groups were organised to support the research activities (Tanaka 2007a). The Integrated WG, seven core project members, and other members who liked to participate, discussed detailed research plans by sharing the concept of the “watershed management model.” Also, there are other working groups that are not shown in Figure 3.6. These working groups were formed to meet the project’s research needs – Extended Material Cycling WG, Social Psychology WG, Paddy Field WG, Environmental Economics WG, Yodo – River Downstream WG. The meetings of these working groups accelerated transdisciplinary communication among researchers. d. Integration Practical methods in watershed diagnosis for hierarchical watershed management included two important aspects. One was the use of indices, models, and geographic information systems (GIS). The other aspect dealt on stakeholder communication. These two aspects also represented naturally integrating methods for the Lake Biwa-Yodo River Project. Project findings revealed that small- to medium-sized rivers, in the farming areas located east of Lake Biwa, were responsible for discharging acid and minerals (i.e. sulfuric acid, nitric acids, bicarbonates, calcium, and

90

Chieko Umetsu, Makoto Taniguchi, Tsugihiro Watanabe, and Shigeo Yachi

magnesium). These acids were produced from farming activities, and have eluted minerals from the soil. These chemicals caused changes in the lake’s water quality. The project utilised GIS so that scientific information, such as turbid water discharge and water quality (stable isotope compositions), was mapped for clear visualisation of the situation. The information were then used and presented to the stakeholders. Another important aspect of integration was placed on stakeholder communication. Interviews, questionnaire-based surveys, and various workshops/meetings were implemented with the stakeholders. The researchers interviewed residents from 35 communities in the region regarding water management and use. They then integrated the data, along with the detailed data on the region and that of Lake Biwa watershed, into the GIS database. Based on the results, and the workshops that were held in the three communities, the residents themselves discussed the location of the areas, which they considered as beautiful and comfortable using maps. The workshops attempted to encourage the residents to notice the actual regional waterside environment and form desired visions for the area. All these integrated efforts were compiled in a project proposal on hierarchical watershed management system. The proposal was based on adaptive management that is appropriate for each level, and communication between levels. The hierarchical watershed management in Figure 3.5 means that communities and stakeholders at each level undertake specific functions and perform diverse activities related to the watershed at their respective level. The actions in each unit involved drawing up plans (P), monitoring the process (D and C), analysing and assessing the results of monitoring (A), and making necessary revisions to the plans. 3. Human Impacts on Urban Subsurface Environments (Study period: 2006-2010) a. Outline Securing water resources and preventing water contamination in the urban areas, caused by human activities, are global environmental issues in the 21st century. The “heat island” phenomenon, which is caused by human activities, is also a big environmental problem in addition to global warming.

Transdisciplinary Research in Watershed Conservation: Experiences, Lessons and Future Directions

91

These global environmental issues, which are caused by urbanisation, should be addressed strongly and prevented as population and density increases occur rapidly in urban areas. Most global environmental studies have focused on the environmental issues above ground, such as air pollution, global warming, seawater pollution, and biodiversity loss. Subsurface environmental issues are also important as they maintain present and future human survival. However, such issues are largely ignored because of the “invisibility” of the phenomena and the difficulty of their evaluation. Subsurface environmental problems, such as subsidence due to excessive pumping and groundwater contamination, have occurred repeatedly in major Asian cities with a time lag that depends on the development stage of urbanisation. Therefore, the project assessed future scenarios based on the evaluation of relationships between subsurface environmental problems and the development stage of the city (RIHN 2007). b. Objectives The project addressed the sustainable use of groundwater and subsurface environments to provide for better future development and human wellbeing. The primary goal of the project was to evaluate the relationships between the development stage of cities and the various subsurface environmental problems. These problems included extreme subsidence, groundwater contamination, and subsurface thermal anomalies (Figure 3.7).

92

Chieko Umetsu, Makoto Taniguchi, Tsugihiro Watanabe, and Shigeo Yachi

Figure 3.7 Human Activities on Urban Subsurface Environments

The project targeted four research topics: • •





relationship between the development stages of the cities and subsurface environmental problems, as assessed by socio-economic analysis and the reconstruction of urban areas using historical records; serious problems in subsurface environments and changes in reliable water resources, which would be studied after evaluating groundwater flow systems and changes in groundwater storage using hydrogeochemical data and in-situ/satellite-GRACE (Gravity, Recovery and Climate Experiments) gravity data; evaluation of materials (contaminants) accumulation in the subsurface environment, and their transport from land to the ocean, including groundwater pathways, using the chemical analyses of subsurface water, sediments, and tracers; and subsurface thermal contamination due to the “heat island” effect in the urban area based on the reconstruction of surface temperature history and urban meteorological analysis (RIHN 2007; Taniguchi 2007).

Transdisciplinary Research in Watershed Conservation: Experiences, Lessons and Future Directions

93

Tokyo, Osaka, Bangkok, and Jakarta were targeted as the main study cities, while Taipei, Manila, and Seoul were selected as secondary study cities, depending on the four sub-themes. The project focused on the urban subsurface environments. It treated the problems on a basin scale because subsurface water, heat, and material transports were interconnected on this scale. The project assessed the relationships between subsurface environmental changes and human activities during the past 100 years. c. Research Organisation Dr. Makoto Taniguchi, a hydrologist, headed the Urban Subsurface Environments project. It consisted of six research groups as shown in Figure 3.8. The Social Economic Group and the Urban Geography Group were organised to study the first research topic. The Water Group and the Gravity Group studied the second research topic. The Material Group and Heat Group studied the third and fourth research topics, respectively. Figure 3.8 Six Research Groups of Urban Subsurface Environments Project

RIHN Annual Report 2006 (2007)

94

Chieko Umetsu, Makoto Taniguchi, Tsugihiro Watanabe, and Shigeo Yachi

Based on the field surveys in the targeted cities and the assessments of natural and social data in each city, the six research groups built the project database/platform on GIS for further analysis of data. d. Integration One of the project’s integration methods focused on the historical changes of four research targets, from 1900 to 2000. It also used GIS to map historical data. These database/platforms were then utilised for the visualisation and analysis of urban subsurface environments in the target cities. Preliminary models, such as GRACE, groundwater flow, and DPSIR (Driving force, Pressure, State, Impact, and Response) have been established in each subtheme to provide scientific information on urban environments. To evaluate the origin and process of material loads to the subsurface, isotopes and chemical analyses of water samples, with new tracer techniques that used CFC, Kr, were introduced. Aside from GIS, integrated models and indicators of natural and social science information were developed to provide recommendations for sustainable groundwater management and management of subsurface environment of cities. Evolution of Research Organisation for Transdisciplinary Research Different types of research organisations have evolved from the RIHN projects since its establishment. The Project Leader was able to choose the type of research organisation for his/her project depending on the mission and the methodologies to address important project issues. The research organisation most commonly used by project-type research can be categorised into three types ­– discipline-, target-, and issue/ theme-oriented research organisation. Although the mission of each project differs, there are various reasons to adopt a research organisation specific to a project. The type of research organisation is critical in achieving the project’s mission and goals.

Transdisciplinary Research in Watershed Conservation: Experiences, Lessons and Future Directions

95

1. Discipline-Oriented Research Organisation (First Phase) Most of the initial RIHN projects had this type of research organisation. In this research organisation, the project is grouped into sub-groups based on the conventional academic disciplines found in universities, such as climatology, hydrology, agronomy, socio-economics, and the like. This type of research organisation is easy to establish. Also, communication/ work among members is easier as the members of the sub-group are mostly from the same discipline, academic society, or even department in the same university. This type of research organisation, however, is difficult to integrate with other groups that have different disciplines. ICCAP is a typical example of a project that has discipline-oriented type of research organisation. 2. Target-Oriented Research Organisation (Second Phase) The target-oriented research organisation appeared as the second phase of RIHN projects. This type of research organisation has particular research targets. This way, researchers from different disciplines can possibly work together as they focus on a research target and not a particular discipline. For example, the Lake Biwa-Yodo River watershed project had a group focused at achieving the research targets on material cycling, ecosystem, and social and cultural issues. The Urban Subsurface project had a research organisation that targeted research on water, heat, contamination, and the city. This type of research organisation can easily focus on a research topic. However, being target-oriented, this type of research organisation has difficulty getting a holistic view of the project. 3. Issue/Theme-Oriented Research Organisation (Third Phase) The pursuit for increased transdisciplinarity created another type of research organisation – the issue/theme-oriented research organisation. Based from experiences, discipline-oriented and target-oriented research organisations seemed not enough to achieve transdisciplinarity. With issue/theme-oriented research organisation, a transdisciplinary group tackled a particular working hypotheses/issues at the onset.

96

Chieko Umetsu, Makoto Taniguchi, Tsugihiro Watanabe, and Shigeo Yachi

The urban subsurface project may be considered a combination of target-oriented and issue-oriented research organisation as each sub-group had a particular target and focused on specific issues. The project thus made it easier to focus on issues to produce transdisciplinary outputs even at the earlier stages of the project and with a relatively small number of researchers required to pursue each goal. For this type of research organisation to be effective, transdisciplinary leadership is essential. Table 3.1 shows the characteristics of the various research organisations. Easy communication among the sub-group members is present in disciplineoriented research organisation. Research focus is better made through target-oriented research organisation. Transdisciplinary integration is easier to achieve when issue/theme-oriented research organisation is used. Table 3.1 Characteristics of Research Organisations RIHN Projects Communication TransResearch among disciplinary Focus Integration Sub-group Discipline Oriented RO Target Oriented RO Issue/Theme Oriented RO

Lake Biwa- Urban Yodo SubICCAP River surface

xxx

xx

x

xx

xx

xxx

xx

x

xx

xx

xxx

x

x

x

x

RO=Research Organisation

For example, ICCAP started by being a discipline-oriented research organisation. Through the course of project implementation, the group members transformed themselves and became a more target- and issueoriented research organisation. This occurred at the latter stages of project implementation. The Lake Biwa-Yodo River Project and the Urban Subsurface Project, meanwhile, were geared towards being target/issueoriented research organisations as compared to ICCAP. The success or failure of the integrated project outputs not only depends on how the research groups are organised. It also depends on the

Transdisciplinary Research in Watershed Conservation: Experiences, Lessons and Future Directions

97

transdisciplinary communication occurring among project members that are considered essential in pursuing the research goals. 4. Tools for integration Integrating the project outputs makes use of many practical ways and tools. Table 3.2 cites examples of tools to integrate the outputs from different disciplines. The “x” shows the relative focus of each tool within the project. They are not comparable across projects. Table 3.2 Tools for Integration Spatial changes

Historical changes

xx xx

x xx

x x

xx

xxx

xx

ICCAP Lake BiwaYodo River Urban Subsurface

Indicators Coupled models

Communication among researchers

Communication among stakeholders

x n.a.

xx xxx

xx xxx

xx

xx

x

n.a=not available

One way to integrate the outputs is to map spatial changes and try to overlay the various parameters and variables for comparison. The historical changes can also be mapped, while tracing various parameters and variables. Mapping provides easy exposition and visualisation of target indices across space and time. These mapping and overlaying methods are often done using GIS. The mapping and tracing method produces some indicators to gain new insights. The indicators can be used for both quantity and quality indices, or totally new composite indices for different perspectives. Modeling is also an important tool for integrating project outputs. The fourth method, which is called the “coupled model,” has both the social and natural sciences as components of the same model. Coupled models have been especially developing in the field of ecological economics (Resilience Alliance 2007). Mathematical biologists/ecologists are considered the pioneers in explaining human-nature behaviours using ecological modeling. This is the natural science approach of integrating social science. Also,

98

Chieko Umetsu, Makoto Taniguchi, Tsugihiro Watanabe, and Shigeo Yachi

social science modeling can embrace information from the natural sciences to integrate analyses. The fifth method of integrating project outputs is communication among researchers. The project can be better integrated if there is smooth communication among researchers from different disciplines. This is not always easy because each discipline has its own technical languages, definitions, models, and a way of approaching issues, i.e., induction or deduction. The sixth method is communication among the stakeholders. By communicating the project outputs to the stakeholders, the output becomes strongly integrated in the communities or society in general. The six integration methods are sometimes combined in project implementation. For example, all three RIHN watershed projects utilised GIS methods to indicate spatial and temporal changes in communication. In ICCAP, GIS was particularly useful in visualising the impacts of future climate change scenarios on the availability of water resources and the state of agricultural production systems in the region. The results were shared with local water authorities and water users associations in addition to university researchers (Nagano et al. 2007). In the Lake Biwa-Yodo River Project, transdisciplinary communication was very intense as the project members held various integrated working group meetings, and published and disseminated working paper series (Tanaka 2007a, 2007b). The project also focused on communication among stakeholders to showcase their research outputs using the GIS database. The Urban Subsurface Project, meanwhile, focused on generating various indicators through intensive field observations and historical overlay for long periods of time. Conclusion This chapter tried to consider the type of research organisation to implement transdisciplinary research projects. In the early stages of transdisciplinary projects, interactions among researchers may not be significant. However, as the problems of the real world become more complex, there is a demand for comprehensive and integrated research outputs from researchers and decision makers. With comprehensive and integrated research outputs,

Transdisciplinary Research in Watershed Conservation: Experiences, Lessons and Future Directions

99

the realities of global environmental problems are better understood and solutions to such problems are clearly provided. The experiences from the three watershed projects revealed that different types of research organisations were formed as a result of each project’s efforts to pursue and implement their respective research agenda. By pursuing consilience, research organisations have simultaneously evolved to focus more on transdisciplinary outputs and solutions to the environmental problems at hand. To synthesise knowledge from the different specialised fields of science, the tools that the projects adopted varied. The tools used depended on the projects’ focus – be it on mapping spatial and temporal indicators using GIS or stakeholder communications. There seemed to be no golden rule to implement transdisciplinary research, since the type of research organisation is selected based on the research goals and needs of the project. Visualising scientific knowledge, using GIS, is useful to achieve communication among researchers as well as stakeholders. Recent developments in unified models for natural and social sciences, for example, may possibly provide a common ground to explain consilience in the context of social-ecological systems. Meanwhile, management frameworks, such as the Integrated Water Resource Management (IWRM) or Adaptive Management (AM), provide general testable hypotheses on managing water resources to achieve specific goals (Medema et al. 2008). Various tools of integration that have been mentioned earlier in this chapter are useful not only in generating scientific information about specific watershed management issues but also in providing management frameworks. Once research efforts are integrated, there is now the move towards normative and value levels from the empirical and pragmatic levels as asserted by Max-Neef (2005). In analysing the impacts of future climate change scenarios in the Lower Seyhan Irrigation Project in Turkey, the transdisciplinary approach was particularly useful in making future scenarios and simulating agricultural production outcomes in the 2070s. To solve the water turbidity in Lake Biwa, a stakeholder approach was used to share scientific knowledge with the local farmers through intensive discussions. The discussions were crucial in understanding the turbidity problem and designing the vision of the communities regarding the local watershed. In the case of the historical changes occurring in the subsurface environment of Asian cities because of human activities, multidisciplinarity played an

100

Chieko Umetsu, Makoto Taniguchi, Tsugihiro Watanabe, and Shigeo Yachi

important role in developing an integrated model and identifying indicators that helped interpret the scientific information of the past hundred years. Future direction may involve the application of transdisciplinarity or integrated research approaches to help local government units and communities in designing the vision of their watershed resources, and implementing the framework for watershed management. This way, the gap between the scientific community, the one who develops the framework, and the one who actually implements it, is minimised. This improved relationship will certainly pave the way for coming up with better policies on local watershed management. notes a Research Institute for Humanity and Nature, Kamigamo Motoyama 457-4, Kita-ku, Kyoto 603-8047 Japan, Phone: +81 (0)75 707 2203; fax: 707 2506; e-mail: [email protected] 1 RIHN’s mission is to construct an integral wisdom such as consilience, and to solve global environmental issues, which RIHN sees as deeply rooted in human culture (RIHN 2007). Environmental issues, such as global warming, loss of biodiversity, and depletion of water resources, are said to be the consequences of human-nature interactions. These consequences are now manifesting themselves in various parts of the world. It is fundamentally a problem of human lifestyle, or “human culture” in the broadest sense of the word (RIHN 2007). To achieve RIHN’s mission, RIHN carries out transdisciplinary and integrated project-based research. As of 2007, five programmes/domains were organised at RIHN to lead transdisciplinary interactions and generate consilience among the domains. Five research programmes included Circulation, Diversity, Resources, Environmental history, and Ecosophy.

Transdisciplinary Research in Watershed Conservation: Experiences, Lessons and Future Directions

101

References ICCAP. The Final Report of ICCAP-The Research Project on the Impact of Climate Changes on Agricultural Production System in Arid Areas (ICCAP), 2007. The Lake Biwa-Yodo River Project. “Multi-Disciplinary Research for Understanding Interactions between Humans and Nature in the Lake Biwa-Yodo River Watershed,” Final Report. RIHN, Kyoto, 2007. Max-Neef, Manfred A. “Foundations of Transdisciplinarity.” Ecological Economics 53(2005): 5-16. Medema, W., B.S. McIntosh, and P.J. Jeffery. “From Premise to Practice: a Critical Assessment of Integrated Water Resources Management and Adaptive Management Approaches in the Water Sector.” Ecolog y and Society 13(2)( 2008):29. www.ecologyandsociety.org/vol13/iss2/art29/ ES-2008-2611.pdf. Nagano, T., Y. Fujihara, K. Tanaka, C. Umetsu, K. Hoshikawa, T. Kume, F. Kimura, and T. Watanabe.. “Generated Social Scenario and Basin Condition for the Final Integration,” the Final Report of ICCAP-The Research Project on the Impact of Climate Changes on Agricultural Production System in Arid Areas (ICCAP). RIHN, 15-18, 2007. Resilience Alliance. Assessing and Managing Resilience in Social-ecological Systems: A Practitioners Workbook. Version 1.0, 2007. RIHN. RIHN Annual Report 2005. RIHN. RIHN Annual Report 2006. Tachimoto, Narifumi. “Methodological Issues for the Area Studies in the 21st Century.” Kokusai Kenkyu 19(2003):148-156. Chubu University. (in Japanese) Tachimoto, Narifumi Maeda. “Global Area Studies and Fieldwork,” Discussion Paper No. 29. Graduate School of International Development, Nagoya University, Nagoya, 2004. Tanaka, Takuya.. “Collaboration Among Researchers in the Lake BiwaYodo River Project,” Multi-Disciplinary Research for Understanding Interactions between Humans and Nature in the Lake Biwa-Yodo River Watershed, Final Report. RIHN, Kyoto (2007a), 581-587. (in Japanese) Tanaka, Takuya. “Collaboration with the Local Community in the Project: The Case of the Lake Biwa-Yodo River Project and Inae Region,” MultiDisciplinary Research for Understanding Interactions between Humans

102

Chieko Umetsu, Makoto Taniguchi, Tsugihiro Watanabe, and Shigeo Yachi

and Nature in the Lake Biwa-Yodo River Watershed, Final Report. RIHN, Kyoto, 2007b, 588-594. (in Japanese) Taniguchi, Makoto. Progress Report 2007: Human Impacts on Urban Subsurface Environments No.4. Research Institute for Humanity and Nature, Kyoto, Japan, 2007. Umetsu, Chieko, K. Palanisami, Ziya Coşkun, Sevgi Donma, Takanori Nagano, Yoichi Fujihara, Kenji Tanaka.“Climate Change and Alternative Cropping Patterns in Lower Seyhan Irrigation Project: A Simulation Analysis with MRI-GCM and CCSR-GCM,” The Final Report of the Research Project on the Impact of Climate Change on Agricultural Production System in Arid Areas (ICCAP), 227-239. RIHN, Kyoto, Japan, 2007. Wilson, Edward O. Consilience: The Unity of Knowledge. New York: Alfred A. Knopf, Inc, 1998. Yachi, Shigeo. “A Hierarchy-based Approach to the Problem of Agricultural Water Turbidity in the Lake Biwa Watershed,” RIHN 1st International Symposium Proceedings – Water and Better Human Life in the Future. November 6-8, 2006, Kyoto International Conference Hall, Kyoto. 8187pp, 2006.

Payments for Ecological Services: Experiences in Carbon and Water Payments in the Philippines

103

4

Payments for Ecological Services: Experiences in Carbon and Water Payments in the Philippines



Rodel D. Lasco a and Grace B. Villamor

Abstract There is a rising interest globally in payments for ecological services (PES) as a way of promoting sustainable development. In this paper, we share the experience of the Philippines in payments for carbon storage and watershed protection. There are several projects on carbon sequestration in varying stages of design and implementation, most of which are in the planning stage. The initial lessons regard the country’s potential for carbon sequestration projects, the insufficiency of carbon payments to cover the full cost of plantation development, and the high transaction cost that have limited the participation of local residents. Payments for watershed protection were more advanced. Several projects, meanwhile, are ongoing. The value of water was more easily recognised at the local and national levels. However, most of the payments did not satisfy the requirements for full-fledged PES. Introduction There is a lot of interest in PES schemes around the world (Landell-Mills and Porras 2002). An environmental service payment or reward refers to compensation for service, merit or effort, and/or incentives for maintaining or enhancing environmental service functions received by the sellers or paid for by the buyers of the environmental service(s) (van Noordwijk 2005). It is a voluntary transaction in which a well-defined environmental service

104

Rodel D. Lasco and Grace B. Villamor

(or a land use likely to secure that service) is “bought” by a (minimum of one) buyer from a (minimum of one) provider if and only if the provider continuously secures the provision of the service (conditionality) (Wunder 2005). The compensation and incentives can be financial, social and moral. Compensation may be made in terms of direct payments, financial incentives, or payments in kind. The rewards and payments in kind may include the provision of infrastructure, market preference, planting materials, health and educational services, skills training, technical assistance, or other material benefits. Aside from indirect and direct monetary payments, rewards can take the form of land tenure security, which may also be considered an economic incentive. Social and moral incentives and rewards may address the non-material aspects of poverty, including the recognition and respect of the community, and personal satisfaction. These types of incentives are considered beneficial to society, likewise with the recognition of service providers in maintaining or enhancing ecosystem services. The Philippines has a severely degraded natural-resources capital base. The situation has adversely affected the environmental services they provide. In the early 1900s, it was estimated that 70 percent of the country was covered with 21 million ha of forests (Garrity et al. 1993). However, 6 M ha of forests remain as of 2004 (FMB 2004). Thus, in the last century alone, the Philippines lost almost 15 M ha of tropical forests. Since the early 1970s, when extensive reforestation efforts began in the Philippines, various incentive schemes have been devised and implemented to encourage people to plant trees on private and public lands. However, after more than three decades of support, reforestation efforts in the Philippines have largely been ineffective and inefficient (Chokkalingam et al. 2006; Garrity et al. 1993) partly because the incentives provided were either inappropriate or did not consider the long-term nature of reforestation. For instance, on public forestlands, the 25-year renewable land tenure instrument was still considered an insufficient incentive for the tenants to invest in longterm forestry and environmental protection (Garrity et al. 1993). Moreover, resource-use rights were only partially transferred. Short-term contracts and direct payments to farmers were not able to draw a genuine interest in tree planting either. Partly in response to the limited success of government-initiated programmes, a number of local governments, research organisations and

Payments for Ecological Services: Experiences in Carbon and Water Payments in the Philippines

105

non-government organisations (NGOs) in the Philippines were testing various PES schemes as a way of reversing environmental degradation. The environmental services being compensated in existing projects included water resources (i.e. RUPES Bakun), carbon sequestration (Lasco et al. 2005), seascape and landscape beauty, and biodiversity (Padilla et al. 2005). This paper discusses the experiences of the Philippines regarding payments for carbon storage and watershed protection. Payments for Carbon Sequestration in the Philippines Climate change is one of the most challenging issues facing humanity today. The Intergovernmental Panel on Climate Change (IPCC) Report (2007) concluded that there is unequivocal evidence that man’s activities are affecting the world’s climate. There is considerable interest on the role of terrestrial ecosystems in climate change, more specifically on the global carbon cycle. The world’s tropical forests, covering 17.6 M square kilometers, contain 428 Gt C in vegetation and soils. It is estimated that about 60 Gt C is being exchanged between terrestrial ecosystems and the atmosphere every year, with a net terrestrial uptake of 0.7 ±1.0 Gt C. However, land use and land-use change and forestry (LULUCF) activities, mainly tropical deforestation, are also significant net sources of CO2 (carbon dioxide), accounting for 1.6 Gt C per year of anthropogenic emissions (IPCC 2007; Watson et al. 2000). The IPCC Report (2007) concluded that the tropical region had the largest mitigation potential in the forestry sector (Nabuurs et al. 2007). For the tropics, the mitigation estimates for lower price ranges (< 20 US$/tCO2) are around 1100 Mt CO2 per year in 2040 through afforestation, reforestation, and forest conservation. In the high range of price scenarios (< 100 US$/t CO2), the mitigation estimates are in the range of 3000 to 4000 Mt CO2 per year in 2040. The Clean Development Mechanism (CDM) and Forestry Projects The Kyoto Protocol sets emission limits for six greenhouse gasses (GHGs) for the developed nations, mostly industrialised countries and economies in

106

Rodel D. Lasco and Grace B. Villamor

transition, known as “Annex 1” or “Annex B” Countries or Parties. These countries have committed to collectively reduce GHG emissions by at least 5 percent relative to their 1990 emissions. To enter into force, 55 countries must ratify the Protocol and must include 55 percent of emissions of Annex 1 Parties for 1990. On the 90th day, after the ratification by Russia, the Kyoto Protocol entered into force on 16 February 2005. The Philippines ratified the protocol in November 2003. The Clean Development Mechanism (CDM) is one of three flexibility mechanisms established to meet the goals of the Kyoto Protocol. CDM aims to assist the parties not included in Annex I in achieving sustainable development, and the parties included in Annex I in achieving compliance with their quantified emission limitation and reduction commitments by implementing projects in developing countries. The CDM essentially offers many opportunities for financing sustainable development projects in developing countries that can generate Certificates of Emission Reduction (CERs)1. CDM presents opportunities for a developing country to host projects that rehabilitate degraded lands, among others. Participants who are eligible to engage in CDM projects are individuals, groups of individuals, private companies, and non-government organisations (NGOs) that belong to a country that is a Party (signed and ratified) to the Kyoto Protocol. Under CDM, only afforestation and reforestation (A/R) projects are allowed for the first commitment period (2008-2012). However, payments for conserving existing forests (carbon storage), under the so-called Reducing Emissions from Deforestation and Degradation (REDD) are among the key issues being discussed at the United Nations Framework Convention on Climate Change (UNFCCC) conference in Bali, Indonesia in 2007. These payments may be implemented after 2012. During the first commitment period of the Kyoto Protocol, the use of carbon credits from CDM forestry projects is limited to at most 5 percent of the respective countries’ 1990 emissions. This estimated emission amount to about 231 M carbon credits (Neef and Henders, 2007). However, some Annex I Countries have refrained from using credits from forestry CDM so that actual demand has become lower. The World Bank is the largest buyer of forestry carbon credits. The BioCarbon Fund compiled a portfolio of candidate projects that were estimated to deliver 22 M carbon credits (Neef and Henders 2007). The BioCarbon Fund bought carbon credits from forestry projects at US$ 3.75-

Payments for Ecological Services: Experiences in Carbon and Water Payments in the Philippines

107

4.35 per t CO2e (or ton of CO2 equivalent) for carbon removals in a forestry project until the end of the Kyoto Protocol’s second commitment period in 2017. As of August 2007, there were more than 750 CDM projects that were registered in the UNFCCC. Of these projects, only one was an A/R project. There were several factors that accounted for the slow uptake of forestry projects under the CDM: • • •



The high transaction cost, estimated at US$200,0002 for a regular project (Neef and Henders 2007), was beyond the reach of potential proponents from developing countries. Developing countries lacked the base financing for tree planting. In the Philippines, for example, the official cost of reforestation by government projects was US$1,000 per hectare for three years. Proceeds from carbon credits were insufficient to cover the full cost of tree plantation establishment. Subsidies were rarely meant to pay the full cost of establishment. Rather, subsidies move production up the supply curve (marginal cost), thereby increasing the quantity produced. Hence, A/R projects were only profitable when combined with other forest products such as wood and fruits. In many developing countries, technical capacity was weak in designing and implementing A/R CDM projects. This was partly due to the countries’ limited experience in carbon sequestration projects.

Initial estimates showed that the life cycle cost of potential forestry projects in the Philippines, even those that were not Kyoto Protocol compliant), ranged from US$0.12 per t C to US$7.60 per t C (Lasco and Pulhin 2001). On the other hand, the cost of protecting a Philippine National Oil Company – Energy Development Corporation (PNOC-EDC) geothermal forest reservation in the island of Leyte was US$2.94 per t C (Lasco et al. 2002). In contrast, a systematic comparison of sequestration supply estimates from the national studies in the USA produced a range of US$25-US$75 per t for a programme size of 300 Mt of annual carbon sequestration (Stavins and Kenneth Richards 2005). Areas suitable for CDM projects in the Philippines include those that need to be permanently forested for legal, ecological, or social reasons. These areas are also the most likely candidates for climate mitigation projects.

108

Rodel D. Lasco and Grace B. Villamor

These areas include: • •



critical watersheds; forest reserves (including those being managed by other government agencies and government-controlled corporations, such as the PNOC and National Power Corporation, academic institutions and the military); forest lands under the National Integrated Protected Area System (NIPAS), including those with 50 percent slope and with altitudes of 1,000 m asl.

As of 2001, the area of the forestlands mentioned above was 5 M ha (FMB 2001) – a large portion of which needs to be either protected or rehabilitated. Another way of estimating potential areas for climate projects is to look at the extent of degraded areas that need rehabilitation. Grasslands and brushlands in the uplands cover 3.5 M ha (Lasco and Pulhin 1998). In addition, many of the supposed agroforestry lands (5.7 M ha) were shifting cultivation areas or simply degraded farmlands that need stabilisation through some form of agroforestry and soil conservation practices. Once new financing schemes are available, property rights may become important (Lasco and Pulhin 2003). Competition on the control forestlands may intensify. In the Philippines, indigenous peoples are claiming many upland areas. Such claims may be ignored in favor of establishing climatechange forests. Thus, the guidelines of the new financing schemes should have adequate provisions that respect the rights of local users. This, however, is more easily said than done in many developing countries. The issues on property rights could be adequately addressed through public consultations and community participation in project planning and implementation. The environmental impact assessment system is viewed as the main mechanism, the resolution of these issues. Existing policies and procedures, embodied in the Indigenous People’s Rights Act (IPRA), can also be considered to ensure that the rights of the indigenous peoples are fully safeguarded.

Payments for Ecological Services: Experiences in Carbon and Water Payments in the Philippines

109

Carbon Sequestration Projects in the Philippines In the last five years, there has been increasing interest in climate change mitigation projects in the Philippines. Much of this interest is probably due to the hype associated with climate change in general (as a result of recent media coverage) and CDM in particular. Three things influence whether this interest advances further: the strict requirements for a CDM project, the level of transaction costs (up to US$100,000 per project 3), and the current price of carbon from forestry projects (about US$15 per t C) vis-àvis the development cost. Three carbon sequestration projects are currently developed to qualify for UNFCCC registration and are presented below. The LLDA-Tanay Streambank Rehabilitation Project The main proponents/sellers of this project were the Municipality of Tanay, Rizal and the Laguna Lake Development Authority (LLDA). The implementers were the farmers in the Tanay watershed. The project aimed at reducing GHGs (e.g. CO2) in the atmosphere while helping to rehabilitate the Tanay watershed and provide socioeconomic benefits to the local people. Specifically, the project aimed to (Lasco and Pulhin 2006): • • •

reforest 70 ha of private lands; establish 25 ha of agroforestry farms in public lands; and sequester 10,000-20,000 t of CO2 from the atmosphere in 20 years.

The project area covered 1,000 ha. The following were the strategies used to implement the project components: Streambank rehabilitation: This component aimed to increase the riparian forest cover of Tanay River to reduce erosion. Owners of private lands were encouraged to plant trees along riverbanks within their property. Seedlings were given for free after the recipients underwent an information and education campaign and gave their pledge of commitment to the project. Provision of seedlings and support services would be contracted through

110

Rodel D. Lasco and Grace B. Villamor

the Katutubo village, an upland village comprised of indigenous Dumagat and Remontado groups. A total of 20 ha would be reforested with 33,333 trees. Ecological Enhancement in Upland Areas: This subcomponent aimed to reforest upland areas near the headwaters of Tanay River to reduce erosion. A total of 50 ha of denuded and grassland areas would be reforested with 83,333 trees using 2 x 3 m spacing. The Katutubo village would implement provision of seedlings, planting and maintenance. The tree species would be chosen by the community to provide them timber, fruit and medicinal resources. Agroforestry orchard: This subcomponent aimed to provide income for the Katutubo village through agroforestry while reducing erosion in the upland areas. This component would cover 25 ha of communal land belonging to the IP community. It will integrate mango trees at 10 x 10 m spacing with cash crops using an alley cropping design. A total of 2,500 trees would be planted. The expected GHG benefits were calculated using high and low scenarios. For the project period (2004-2014), the project will have total net carbon benefits of 3,204 t C (11,759 t CO2e) and 1,424 t C (5,230 t CO2e) under the high and low scenarios, respectively (Santos-Borja et al. 2005). The anticipated total Emission Reduction Purchase Agreement (ERPA) Value is US$31,380 for the low scenario and US$70,554 for the high scenario. The project is seeking accreditation to the UNFCCC CDM Executive Board, which will grant the credits once it is approved. The World Bank Carbon Fund is expected to buy the credits generated by the project. A third party auditor, as mandated by the CDM rules, will assess the actual amount of carbon sequestered. For the 20-year project duration, total carbon sequestration is shown in Figure 4.1 under various scenarios. The buyer of the carbon credits is the World Bank Carbon Fund, which is also providing technical assistance to LLDA and its partners.

Payments for Ecological Services: Experiences in Carbon and Water Payments in the Philippines

111

Figure 4.1 Net Carbon Sequestration Under Various Scenarios of the LLDA Project in Tanay, Rizal

CI-Philippines Sierra Madre Project (Lasco and Pulhin 2006) The proposed carbon sequestration project is part of Conservation International (CI)-Philippines’ concerted efforts to build alliances with local communities, private sector, government agencies and NGOs to facilitate the management of the Sierra Madre Biodiversity Corridor and strengthen enforcement of environmental laws. It uses a multifaceted approach to alleviate threats and restore and protect 12,500 ha of land within the Corridor. The project aims to demonstrate that a properly designed and implemented carbon offset project not only offers a financially attractive, risk-managed portfolio option, but also generates multiple benefits such as biodiversity protection, watershed restoration, soil conservation, and local income-generation. It will also demonstrate that tradeoffs, such as soil erosion, water table decrease, and loss of livelihoods can be avoided. Specifically, the project aims to: •

Protect 5,000 ha of natural forests (old growth and second growth) slated for cutting;

112





Rodel D. Lasco and Grace B. Villamor

Reduce pressure to the natural forest and provide incentives for local communities by working to establish an agroforestry project on 2,000 ha of brushland areas to supply a more stable income to the population and lessen the reliance on forest projects; and Restore 5,500 ha of grassland areas to original hardwood forests using a mix of fast-growing species and native species to aid the sequestration of carbon dioxide from the atmosphere and increase the connectivity of sensitive habitats for the world’s most threatened species.

Community-based forest management (CBFM) is the main strategy of the project with the local community/people’s organisation (PO), local NGOs, Local Government Unit (LGU), the Department of Environment and Natural Resources (DENR), the project monitoring team, and the funding organisation as the main stakeholders. After 30 years, it is expected that a total of 512,000 t C would be sequestered by the project, most of which would come from the reforestation component (453,000 t C). Kalahan Forestry Carbon Projects (Villamor and Lasco 2007) The Ikalahan Ancestral Domain covers 58,000 ha of mountainous forest and farmlands found in the provinces of Pangasinan, Nueva Ecija and Nueva Vizcaya, northern Luzon. In 2003, the area was selected as the first Rewarding Upland Poor for Environmental Services (RUPES-ICRAF) project pilot site in the country to develop a carbon sequestration payment mechanism. The Kalahan Educational Foundation (KEF), a peoples’ organisation, targeted two types of carbon markets – the regulated market, through Kyoto’s CDM, and the voluntary carbon market. To date, KEF had done preliminary activities in preparation for these markets – preparation of project idea notes (PINs) and awareness building among the members of the indigenous group. Through the PINs, KEF was actively seeking potential partners and buyers of carbon. For the Kyoto market, KEF aimed to convert the 900 ha of marginal and abandoned agriculture land to more productive tree-based system through reforestation, the only “sinks” project allowed under CDM. More specifically, the project aimed to convert the 900 ha of marginal and abandoned agriculture land to more productive tree-based system; enhance the livelihood of the communities in the proposed area through agroforestry;

Payments for Ecological Services: Experiences in Carbon and Water Payments in the Philippines

113

and protect the watershed, enhance the biodiversity, and improve land beauty of the tourism area. The project would use community-based forest management as its main strategy with the Ikalahan-Kalanguya indigenous communities, local NGOs, the DENR, project monitoring team, and the funding organisation as the key stakeholders. All project activities would be developed with the participation of indigenous communities in the project area. KEF would catalyse the community organising and development process; and manage and implement the project. The project monitoring team would quantify the carbon sequestered and assess the impacts of the project. The funding organisation would provide the financial resources for the project. The project would employ two rehabilitation technologies: agroforestry and reforestation. The agroforestry component would introduce fruit trees to existing upland farms (typically using annual crops such as corn and rice). Aside from its environmental benefits, fruit trees would provide livelihood for poor upland farmers. On the other hand, reforestation would target degraded areas that have been covered with grasses for many decades. Only native species and those that have been introduced in the Philippines for the last 10 years would be used, with priority for species already growing in and around the project area. For reforestation, the following plant species have initially been identified: mostly indigenous Dipterocarp species, with Bischofia javanica, and Alnus nepalensis observed to be favorable to wildlife. The fast growing species A. nepalensis was intended to rapidly establish vegetative cover to the area especially in highly degraded areas. Indigenous species would be planted in more favorable areas and underneath fast growing nurse trees. It was estimated that the 900 ha area would sequester 89,776 t CO2e for 20 years under the medium tree growth scenario (Figure 4.2). This estimate was based on Philippines tree growth rates (Lasco et al. 2004) and was consistent with IPCC values. More site-specific based estimates can be done in the future as local growth rates are being analysed.

114

Rodel D. Lasco and Grace B. Villamor

Figure 4.2 Estimated Net Cumulative CO2e Removals By The Proposed Kalahan Reforestation Project, Philippines (Villamor and Lasco 2006)

The voluntary carbon-offset4 markets aimed to maintain 10,000 ha of secondary forests for production and carbon sequestration. This type of project is not allowed under CDM. Hence, project implementers plan to tap into the voluntary carbon market. Between 2006 and 2007, the number of voluntary markets accelerated dramatically with more than 70 organisations in five continents involved in all stages of the supply chain – from developers, aggregators, developers and retailers. Under the voluntary carbon offset market, prices varied from US$1.50 to almost US$5 (Hamilton et al. 2007). The KEF prepared a PIN with focus on enrichment planting and rigid implementation of Forest Improvement Technology (FIT) developed by the Ikalahans to enhance carbon sequestration. Initial estimates showed that the forest area could sequester 1.7 Mt of CO2 for a period of 20 years. Growthrate studies of the indigenous trees of Kalahan forests are currently being completed which can be used to calculate site-specific carbon sequestration rates. Watershed Services Payments Watershed functions are considered to be the first environmental service functions that have been recognised for payments due to their immediate

Payments for Ecological Services: Experiences in Carbon and Water Payments in the Philippines

115

relevance to the people (van Noordwijk 2005). Communities from different parts of the world are benefiting from the commodities derived from watersheds, such as water flow regulation, water quality maintenance, erosion and sediment control, land and salinisation reduction/water table regulation and maintenance of aquatic habitats (Landell-Mills and Porras 2002). Countries, such as Columbia, Ecuador and Costa Rica are among the countries with established payment schemes for such kind of functions. The major issue in watershed services payments is a lack of clarity about the impact of different land uses on water (Kaimowitz 2004). Until recently, suppliers of watershed services have generally lacked leverage for demanding payment. This situation is changing with new government regulations for improved water quality in more developed countries, as well as improved buyer understanding of the benefits provided by watersheds and the growing threats they are facing (Grieg-Gran and Bann 2003). There are also scientific doubts about forest-water linkages. To be workable, it has been suggested that simple approaches are needed to make payments, such as a flat rate per unit area, that are not necessarily linked to the actual service provided (Kaimowitz 2004). In some societies, access to water is seen as a fundamental right (FAO 2006). The development of markets for water requires defining property rights, which in many countries is not a trivial issue. There is also concern that payment systems may exclude access of poor people to water. For large watersheds with many users, transaction costs for PES can be very high (FAO 2006). Intermediary organisations are usually needed to link producers and users. This is true especially in watersheds where institutions are weak and development of PES for water can be costly. Finally, there is a question of whether the market is a better mechanism for delivering watershed services than the tried and tested regulatory systems (LandellMills and Porras 2002). The literature provides little insight on this issue. In spite of its limitations, many cases of PES for water have been tried in developing countries. For example, in Nepal people residing in the Kulekhani watershed are conserving forests and undertaking other conservation activities to ensure consistent water flow (Huang 2007). These environmental services provide more water to the reservoir, which in turn increases electricity revenue and reduces maintenance costs of the hydropower developer. In early 2006, Makwanpur DDC and the local government body of the district housing Kulekhani I and II hydropower plants established

116

Rodel D. Lasco and Grace B. Villamor

an Environmental Management Special Fund (EMSF). EMSF receives 20 percent of the hydropower royalty received by Makwanpur DDC from the Kulekhani hydropower plants, amounting to about US$55,000 per year. The fund would be used to support conservation and development programmes proposed by the upland communities of the Kulekhani watershed. In the Philippines, under the RUPES research programme, the conditions for developing payments for watershed services mechanism have been studied. Results of the three case studies from the RUPES experience are presented below. 1. Bakun Watershed, Benguet (Villamor G. et al. 2006) Bakun is home to the Kankaney-Bago indigenous tribe. It is the first indigenous area in the country that was awarded the Certificate of Ancestral Domain Title (CADT). CADT is a product of decades of struggle by the Indigenous Peoples in the Philippines for recognition of their unique culture and property rights (Espaldon 2005). The domain’s rights cover 29,000 ha of land area in the Cordillera ranges of northern Philippines. The Bakun indigenous people (composed mainly of close-knit members of the BagoKankanaey tribe) are predominantly poor, and 90 percent of the local people are engaged in rice and vegetable farming. Despite being underprivileged, Bakun has a rich sociocultural heritage. Their indigenous way of life governs how they relate with the land, the forests and among themselves. They manage and utilise natural resources using indigenous knowledge systems and practices. The Bakun watershed is the main source of domestic water supply of the local community. More importantly, it is the source of irrigation water for the rice fields and the expanding vegetable farms in the area. Four major rivers and several tributaries drain the watershed area, two of the rivers are currently supporting the operation of the Luzon Hydropower Corporation (LHC) and the Northern Mini Hydro Corporation (NMHC). Bakun receives payments for water functions from two hydroelectric companies, the LHC and the NMHC. These payments are mandatory taxes, which are predetermined under the Philippine Energy Law of 2001 (Republic Act No. 7638) as imposed by the government. Based on the said policy, the host communities benefit from the electrification fund, livelihood development and watershed rehabilitation funds, and possible employment. Moreover, several scholarships and

Payments for Ecological Services: Experiences in Carbon and Water Payments in the Philippines

117

internships are offered to students from the host communities in Bakun. These payments or compensations are paid to the Municipal Government of Bakun. The municipal government identifies where and how to distribute the payment for the welfare of the community. Furthermore, a Memorandum of Agreement (MOA) between the Province of Benguet, LHC and National Power Corporation (NPC) was prepared to implement the Local Government Code and its implementing regulations on the payment of the realty tax and national wealth tax which are equivalent to one percent of the gross revenue of the power plant. The taxes are used for the utilisation of water which are paid to the host local government unit based on the following sharing scheme: 20 percent to the provincial government, 45 to the municipality government, and 35 percent to the barangay or local government unit (Espaldon 2005). 2. Water Levy for the Rehabilitation of Baticulan Watershed, Negros Occidental (Villamor and Lasco 2007) The Baticulan Watershed is located within the boundary of San Carlos City, Negro Occidental. It lies between the towns of Palampas and Rizal, covering a total of 428 ha. Due to shifting cultivation in the watershed, soil erosion and flooding became the serious problems in the nearby lowlands. The widespread degradation on the uplands of Baticulan Watershed has motivated the City Government of San Carlos, Negros Island to formulate a city ordinance in 2004 regulating the operation of the City Waterworks of the City and creating the Watershed Development and Protection Fund for other related purposes. The ordinance provides for a special levy for environmental fee of PhP0.75 for every cubic meter of water billed. The proceeds would go to a special account known as the “Watershed Development and Environmental Protection Fund” which supports the implementation of the City’s Master Development Plan (MDP). This fund considers the negative externalities incurred in the production and consumption of water. The price of water should include the cost of externalities to address the negative impacts of water utilisation on the environment. It was also estimated that budget allocation per year for the project was set at approximately Php1.2 M. The ordinance took effect in 2005. Local water consumers composed of households, local industrial firms, and small-scale farmers, are now paying the said amount. The City Waterworks Department automatically deducts the environmental fee and places it into

118

Rodel D. Lasco and Grace B. Villamor

the special account. Section 3 of the ordinance specifies that the Watershed Development and Environmental Fund can only be disbursed according to the implementing rules and regulations which have been submitted by the City Waterworks Department and approved by the Sanggunian or local government council. A Memorandum of Understanding (MOU) between the City Government of San Carlos and the San Carlos Development Board, Inc (SCBDI) was prepared and signed to leverage the said funds for watershed rehabilitation. 3. Sibuyan Island (Villamor and Lasco 2009) The Sibuyan Island is the second largest island of Romblon Province with a total land area of approximately 45,600 ha – 75 percent of which is forested (van de Veen 2004). In the southern half of the island, watershed arrangements and payment mechanisms for two watersheds, namely the Cantingas and Panangcalan watersheds, are being developed. Both these watersheds provide environmental services, which benefit both the upland and lowland dwellers. Below are the ecosystem services brought by these watersheds (Sibuyan Report 2006): • • • • • •

Forest cover regulates the flow of both surface and groundwater, thereby slowing the rate of runoff in the watershed and creating difficulty in determining the recharge of the water table; Forests protect the soil, reduce the incidence of soil erosion and landslides, and control the sedimentation of waterways; Forest soils filter contaminants, control nutrient and chemical load, and maintain water quality; Forests contribute to both farm and fishery production; Forests provide habitat for endangered species; and Forests have potentials for carbon storage and sequestration.

Among these environmental services, mitigating the occurrences of floods and sustaining the production of good quality water for drinking and recreation purposes are considered the most important. The World Wildlife Fund (WWF)-Philippines developed payment schemes in collaboration with the LGU of San Fernando and Sibuyan

Payments for Ecological Services: Experiences in Carbon and Water Payments in the Philippines

119

Mangyan Tagabukid (SMT), an indigenous tribal organisation. Since 2000, the WWF has been working towards providing livelihood strategies for the Mangyan Tagabukid, the main indigenous group in the area. Starting 2005, both the LGU of San Fernando and WWF have set aside Php350,000 each to establish the Cantingas Water Fund to finance watershed management in Panangcalan and Catingas watersheds, respectively. The fund provided seed money for the protection of the watersheds. It set up the scheme based on the basic ecological principle that good watershed could help maintain good water quality and continuous water supply in the area. Following the example of the Baticulan Watershed, the LGU has started drafting an ordinance for the water levy. Environmental fees for the use of the said services are being practiced in the area. For example, in Panangcalan watershed, each household reportedly pays Php15 per month to the office of the municipal engineer to avail of the water system. In the Cantingas watershed, a total of 560 ha of rice lands were reportedly irrigated at Php360 per hectare per year to be added to the water fund. Lessons Learned After 10 years of research and limited project development in carbon sequestration by trees in the Philippines, several lessons have emerged. First, the Philippines has a great potential for climate change mitigation projects in forestry. The research showed that planted trees in the Philippines could sequester significant amounts of carbon. The country has a long experience in reforestation and tree farm development albeit with mixed success. Second, initial economic studies have shown that income from carbon credits was not sufficient to recover the cost of tree planting (using standard DENR costs). This implies that carbon credits are best used as a supplemental source of income for farmers and project developers. Third, the initial or base costs (including upfront costs, establishment and administrative costs) of engaging in forestry CDM projects are enormous (estimated at US$100,000 per project) and could prove to be the most significant barrier to project fruition. One way of overcoming this barrier is to partner with a potential buyer who may be able to shoulder the

120

Rodel D. Lasco and Grace B. Villamor

upfront costs as in the case of LLDA and the World Bank Carbon Fund. Also, government institutions, particularly the DENR Forest Management Bureau, must find ways to encourage project developers by simplifying rules and regulations for forestry carbon projects. As it is, forestry projects have few takers because of their complexity and high transaction costs. The research community must also ensure that relevant information is made available to project developers. Among the knowledge gaps that need to be filled concern: carbon sequestration rates of Philippine trees, especially in various agro-ecological zones of the country; economic analysis of forestry carbon projects; and production system models (e.g. agroforestry) that would optimise carbon and financial benefits. Several lessons have also emerged from the initiatives on payments for watershed services. First, the value of payments for water services is more easily recognised at various levels from local to national and by different stakeholders. Second, various forms of payments exist but most of them do not satisfy the two main criteria as set by Wunder (2005) – voluntary and conditional. Third, the involvement of the government, especially local government units, is important for PES schemes to be successful. Fourth, PES schemes work when threats (e.g. water scarcity), values (e.g. strategic point for commerce), opportunities (e.g. people see the benefits from ES in watershed rehabilitation) and trust are met (e.g. local trust between government, local people and buyers). NOTES a World Agroforestry Centre (ICRAF), Khush Hall, IRRI Campus, College, 4031 Laguna, Philippines, e-mail: [email protected] 1 The Kyoto protocol gives financial value to GHG emission reductions. One CER is equivalent to 1 ton of CO2 emission reduced. CERs are often referred to as “carbon credits.” 2 The costs of conducting the CDM process covers the development of the project design document or PDD; validation, registration, monitoring, verification/certification, and issuance of CER; and registration. 3 The upfront costs to engage CDM projects.

Payments for Ecological Services: Experiences in Carbon and Water Payments in the Philippines

4

121

Carbon offset is defined as a mechanism by which the impact of emitting a ton of CO2 can be negated or diminished by avoiding the release of a ton elsewhere, or absorbing a ton of CO2 from the air that otherwise would have remained in the atmosphere. For voluntary carbon offset, this market refers to entities (companies, government, NGOs, individuals) that purchase carbon credits for purposes other than meeting regulatory targets.

References Chokkalingam, U., A.P. Carandang, J.M. Pulhin, R.D. Lasco, R.J.J. Peras, and T. Toma. One Century of Forest Rehabilitation in the Philippines: Approaches, Outcomes and Lessons. Bogor, Indonesia: Center for International Forestry Research (CIFOR), 2006. FAO. The New Generation of Watershed Management Programmes and Projects. “A Resource Book for Practitioners and Local Decision-makers Based on the Findings and Recommendations of a FAO Feview,” FAO Forestry Paper 150(2006), Rome. http://www.fao.org/docrep/009/ a0644e/a0644e00.htm Garrity D.P., D.M. Kummer, and E.S. Guiang. The Upland Ecosystem in the Philippines: Alternatives for Sustainable Farming and Forestry. National Academy Press, Washington DC, 1993. Grieg-Gran, M. and C. Bann. “A Closer Look at Payments and Markets for Environmental Services.” In From Goodwill to Payments for Environmental Services: A Survey of Financing Options for Sustainable Natural Resource Management in Developing Countries, Pablo Gutman (ed). WWF Macroeconomics for Sustainable Development Programme Office, 2003. Espaldon, M.V. Looking Through the Eyes of the Future. The RUPES Bakun, Benguet, Philippines, 2005 (Unpublished report). FMB. Philippine Forestry Statistics. Philippines: Forest Management Bureau, Department of Environmental and Natural Resources, 2001. FMB. Forestry Statistics 2003. Philippines: Forest Management Bureau, Department of Environmental and Natural Resources, 2004. FAO. “The New Generation of Watershed Management Programmes and Projects. A Resource Book for Practitioners and Local Decision-makers

122

Rodel D. Lasco and Grace B. Villamor

Based on the Findings and Recommendations of a FAO Review,” FAO Forestry Paper 150(2006). Rome. http://www.fao.org/docrep/009/ a0644e/a0644e00.htm Hamilton, K., R. Bayon, G. Turner, and D. Higgins. “State of the voluntary carbon market 2007: picking up steam. New Carbon Finance and Ecosystem Marketplace,” 2007. http://ecosystemmarketplace.com/ documents/acrobat/StateoftheVoluntaryCarbonMarket17July.pdf Huang, M. and S.K. Upadhyaya. “Watershed-based Payment for Environmental Services in Asia.” Working Paper No. 06-07(2007). Sustainable Agriculture and Natural Resource Management Collaborative Research Support Programme (SANREM CRSP) Office of International Research, Education, and Development (OIRED), Virginia Tech. 27. IPCC Working Group I. Summary for Policy Makers. Intergovernmental Panel on Climate Change, 18, 2007. Kaimowitz, D. “Forests in the Pressure of Global Policy Making.” In Forest Research Crossing Borders, V. Joensuu (ed). Finland, European Forest Institute. EFI Proceedings 50(2004). 67-69p. http://www.efi.fi/ publications/Proceedings/. Landell-Mills, N and Porras, T.I. “Silver Bullet or Fools’ Gold? A Global Review of Markets for Forest Environmental Services and Their Impact on the Poor. Instruments for Sustainable Private Sector Forestry Series.” International Institute for Environment and Development, London, 2002. Lasco, R.D., F.B. Pulhin, and R.N. Banaticla. “Potential Carbon Sequestration Projects in the Philippines.” In PES: Sustainable Financing for Conservation and Development. Proc from the National Conference-Workshop on Payments for Environmental Services: Direct Incentives for Biodiversity Conservation and Poverty Alleviation, J.E. Padilla, E.E. Tongson, and R.D. Lasco (eds)1-2 March 2005. WWF, ICRAF, REECS, UP-CIDS, UPLB-ENFOR, CARE. 126-132pp, 2005. Lasco, R.D. and F.B. Pulhin. “LUCF in the Philippines: CC Impacts and Mitigation Potential.” In Disturbing Climate, J.T. Villarin (ed). Philippines: Manila Observatory, 2001. Lasco, R.D. and F.B. Pulhin. Philippine forest ecosystems and climate change: Carbon stocks, rate of sequestration and the Kyoto Protocol. Annals of Tropical Research 25(2)(2003):37-51pp.

Payments for Ecological Services: Experiences in Carbon and Water Payments in the Philippines

123

Lasco, R.D. and F.B. Pulhin. Philippine Forestry and CO2 Sequestration: Opportunities for Mitigating Climate Change. Philippines: Environmental Forestry Programme (ENFOR), CFNR, UPLB, 24pp, 1998. Lasco, R.D., I.Q. Guillermo, R.V.O. Cruz, N.C. Bantayan, and F.B. Pulhin. Carbon Stocks Assessment of A Tropical Secondary Forest in Mt. Makiling, Philippines. Journal of Tropical Forest Sci 16(2004):35-45. Lasco, R.D., J.S. Lales, M.T. Arnuevo, I.Q. Guillermo, A.C. de Jesus, R. Medrano, O.F. Bajar, and C.V. Mendoza. “Carbon Dioxide (CO2) Storage and Sequstration of Land Cover in the Leyte Geothermal Reservation.” Renewable Energ y 25(2002): 307-315. Nabuurs, G.J., O. Masera, K. Andrasko, P. Benitez-Ponce, R. Boer, M. Dutschke, E. Elsiddig, J. Ford-Robertson, P. Frumhoff, T. Karjalainen, O. Krankina, W.A. Kurz, M. Matsumoto, W. Oyhantcabal, N.H. Ravindranath, M.J. Sanz Sanchez, and X. Zhang. “Forestry In Climate Change 2007: Mitigation.” In Contribution of Working Group III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, B. Metz, O.R. Davidson, P.R. Bosch, R. Dave, L.A. Meyer (eds). Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 2007. Neef, T. and S. Henders. Guidebook to Markets and Commercialization of Forestry CDM Projects. Turrialba, Costa Rica, Centro Agronomico Tropical de Investigacion (CATIE), 2007. Padilla, J.E., E.E. Tongson, and R.D. Lasco. “PES: Sustainable Financing for Conservation and Development.” Proceedings from the National Conference-Workshop on Payments for Environmental Services: Direct Incentives for Biodiversity Conservation and Poverty Alleviation. 1-2 March 2005. WWF, ICRAF, REECS, UP-CIDS, UPLB-ENFOR, CARE. 279pp. Santos, A.C. and R.D. Lasco. “Microwatershed Enhancement Through Community Participation: A Pilot Approach to Carbon Finance.” In Carbon Forestry: Who Will Benefit, D. Murdiyarso and H. Herawati (eds). CIFOR. Bogor, Indonesia. 149-155pp, 2005. Sibuyan Report. Unpublished report. Sibuyan Island, Philippines, 2006. Stavins, R. and K. Richards. The Cost of U.S. Forest-based Carbon Sequestration. Pew Center Report, 2005. Van de Veen, H. How to Care for the Casualties of Conservation? Laboring To Improve Livelihoods on Sibuyan Island, Philippines. DGIS-WWF. Gland, 2004.

124

Rodel D. Lasco and Grace B. Villamor

van Noordwijk. RUPES Typolog y of Environmental Service Worthy of Reward. Bogor: World Agroforestry Centre, 2005. Villamor, G. B., R.V. Cruz, R.D. Lasco, and P.A. Sanchez. “RUPES: Payments for Watershed Functions-Application of Rapid Hydrological Appraisal (RHA) in Bakun Watershed, Philippines.” Paper presented during the 1st eSURED meeting on International Water Resource Management, Quezon City, 2006. Villamor, G. and R.D. Lasco. “The Ikalahan Ancestral Domain in the Philippines.” In Community Forest Management As A Carbon Mitigation Option: Case Studies, D. Murdiyarso and M. Skutsch (eds). Bogor, Indonesia: CIFOR. 43-50 pp, 2006. Villamor, G. and R. D. Lasco. “Water Levy As Financial Scheme for Watershed Protection-A City Government Initiative to Rehabilitate the Baticulan Watershed, Philippines.” Paper presented at the International Forum on Water Environmental Governance in Asia. Bangkok, Thailand. 125-129pp, 2007. Villamor, G. and R.D. Lasco (In press). “Rewarding Upland People for Forest Conservation: Lessons Learned and Experiences From The Case Studies in the Philippines.” Journal of Sustainable Forestry. Watson R., I. Noble, B. Bolin, and 32 co-authors. “Summary for Policymakers: Land-use, Land Use Change and Forestry.” A Special Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, 2000. Wunder, S. “Payments for Environmental Services: Some Nuts and Bolts.” CIFOR Occasional Paper No. 42(2005). Center for International Forestry Research, Jakarta, Indonesia.

Cyberinfrastructure for Sustainability in Coupled Human-Environment Systems

Theme 2 Monitoring and Modelling

125

126

Michael H. Kido, Kenneth Y. Kaneshiro and Kevin N. Montgomery

Cyberinfrastructure for Sustainability in Coupled Human-Environment Systems

127

5 Cyberinfrastructure for Sustainability in Coupled Human-Environment Systems

Michael H. Kido,a Kenneth Y. Kaneshiro,b and Kevin N. Montgomery c

Abstract A critical challenge facing scientists, policy makers, and citizens of communities is to affect a transition into sustainability in coupled human-environment systems. This transition must be science-driven and effective enough to meet projected human development needs and at the same time protect the earth’s life support systems. Knowledge systems that are generated by science, however, are often de-coupled from decision-making and the translation of information into action and policy. Emerging cyberinfrastructure, viewed as the coordinated aggregate of computer hardware and software, and other technologies, which changes data into knowledge, has great potential to connect science with policy action. This is accomplished through the effective integration and delivery of information to end users about the condition of human-environment systems at the global scale. This paper provides a brief overview of a new cyberinfrastructure, which was originally developed for monitoring human-environment systems at the regional scale in the Hawaiian Islands. This emerging technolog y successfully synthesised the use of wireless sensor technologies, grid computing with 3D geospatial data visualisation and exploration, and a secured Internet portal user interface. This platform has significant power to crossconnect science and decision makers beyond boundaries to promote sustainable development at the global scale. A use case example is provided in which native Hawaiian residents of Waipa Valley (Kaua‘i) utilised the technolog y in monitoring the effects of regional weather

128

Michael H. Kido, Kenneth Y. Kaneshiro and Kevin N. Montgomery

variations on surface water quality and quantity response to better understand their local hydrologic cycle and promote the sustainable use of biophysical resources in their “mountainto-sea” environment. Introduction Land conversion and resource use is a trait that is characteristic of human society. Collective human action has resulted in the accelerated pace of ecosystem degradation at the global level and the loss of the services they provide. To reverse this alarming trend, scientists, policy makers, and citizens of communities must facilitate the transition of society into a sustainable state to meet human development needs, protect the earth’s life support systems, and maintain ecosystem health and integrity (e.g. Cash 2000; Pfirman 2003). To achieve these goals, there is a need to understand the boundaries, dynamics, and interactions of human-environment systems (e.g. Walker et al. 2002) and acquire the technical ability to track long-term trends in environment and development (e.g. Kates and Parris 2003). While science and technology have contributed in the creation of “knowledge systems” that begin to address sustainability needs, for example with El Niño forecasting, management of aquifer depletion, and the regulation of ocean fisheries, there has been a general disconnection between science and policy. This rift has impeded the translation of data into action for sustainability (Cash et al. 2003). Because of these so-called “boundaries” (e.g. Guston 1999) between scientists and decision makers, knowledge and action are de-coupled because of failures to integrate, communicate, and translate information effectively into action (Cash et al. 2003). As a result, knowledge systems are used ineffectively. This has caused crop yields, fishery stocks, and water levels in aquifers to decline. Because of such failures, scientific information is unjustly viewed by the public as lacking in “salience, credibility, and legitimacy” (Cash et al. 2003). Cyberinfrastructure can help bridge these boundaries by integrating the use of computing systems, data collection and storage capacities, data analyses and visualisation, communications and networking, and interoperable suites of software applications (e.g. Bajcsy 2001; Kaur et al. 2005). This emerging technology has a great potential for connecting science

Cyberinfrastructure for Sustainability in Coupled Human-Environment Systems

129

with policy by making possible the effective integration of information about human-environment systems at the global scale in near real-time. The technologies required to develop this capability are quickly evolving in both academic and commercial laboratories. For example, significant research efforts have focused on the use of wireless sensor networks to monitor environmental variables coupled with ecological habitats (e.g. Szewczyk et al. 2004). The standardised template that has been developed involved the use of untethered, battery-operated sensor nodes that are distributed into a mesh network, which collects an integrated view of the area being monitored (Santi 2005). Data is transmitted through neighboring nodes to gateways that interface with networking infrastructures (e.g. Ethernet or 802.11). However, to date practical applications of this standard template for environmental monitoring have been applied at relatively small spatial scales. Recent improvements in the use of digital computing, information technologies, and communications (e.g. Kaur et al. 2005) promise to deliver the end-to-end digital capability required to process the massive quantities of sensor data generated by automated systems. New networking technologies are better able to integrate environmental data sources across scientific disciplines, and interface clientele with geographically distributed user communities that require rapid access to data and analytical capability (e.g. Karasti et al. 2006). Similarly, advances in grid infrastructure technologies enable the processing and management of large datasets, distributed resource sharing and collaboration, and data intensive computation in a wide range of scientific domains where data resources are widely distributed (e.g. Chervenak et al. 2000). Progress in the use of portal technologies to coordinate Internet and grid infrastructures that link a diverse range of information resources can now afford users secure access to data that have been integrated across physical and intellectual domains (e.g. Jaeger-Frank et al. 2006). Finally, extended 3D tools used for global-scale exploration, visualisation and animation of environmental data have also become readily available in the public domain with the releases of NASA’s World Wind and Google Earth (Verbree and Zlatanova 2007). However despite significant technological advancements in the individual cyberinfrastructure components necessary in creating a platform to support knowledge systems for sustainable development in human-

130

Michael H. Kido, Kenneth Y. Kaneshiro and Kevin N. Montgomery

environment systems, synthesis of these components into a working model has been slow. This paper intends to provide readers with an overview of a new cyberinfrastructure that has been initially developed for integrated environmental observing applications. More recently this evolving technology has been expanded into a robust platform that effectively harnesses knowledge systems for user communities to translate information into action. A use case example is provided to demonstrate how this platform is being applied by a native Hawaiian community on Kaua‘i to more effectively manage their mountain-to-sea environment for sustainability. InteleSense Technologies-based Cyberinfrastructure Overview The Hawaiian Islands represent a unique opportunity of both benefits and challenges for the development of an advanced cyberinfrastructure for sustainability in coupled human-environment systems. Its geographical topology (mountains, steep valleys, and oceans), ecology (native biodiversity in terrestrial, freshwater, and marine ecosystems), and human development systems (natural, managed, and urban) confined within limited islandbound landscapes represent a unique microcosm of continental scale systems. Because of the compact size of watersheds, it is thus feasible to study ecosystems on mountain-to-sea scales not possible on continents (Kaneshiro et al. 2005). Recognising this potential, an initial focus area of the EPSCoR (Experimental Projects to Stimulate Competitive Research) Research Infrastructure Improvement (RII) Grant, awarded to the University of Hawai‘i in FY 2003 by the National Science Foundation, was to develop an advanced capability to collect-analyse-visualise environmental data in near real-time in coupled terrestrial-freshwater-marine ecosystems on mountain-to-sea scales (Figure 5.1). A unique interdisciplinary collaboration, that brought together ecology and engineering expertise, ultimately produced a working environmental cyberinfrastructure that has been rigorously tested in environmental monitoring applications throughout the Hawaiian Islands (Kido et al. 2008).

Cyberinfrastructure for Sustainability in Coupled Human-Environment Systems

131

Figure 5.1 InteleView – Virtual Earth Image of: (A) Kaua‘i in the Hawaiian Islands; (B) Project Area on North Kaua‘i, Circles Indicate Locations of Sensor Deployments Transmitting Data Wirelessly to the InteleSense Server; (C) Location of Waipa Valley / Ahupua‘a

To foster industry-based research and development outside of the university setting, InteleSense Technologies, Inc., a Hawai‘i-based corporation, was founded in 2005 with production and state-of-the-art server facilities located in Silicon Valley, California. The environmental cyberinfrastructure that has been developed successfully synthesised the use of wireless sensor technologies (e.g. Hill et al. 2004) and internet portal-based grid computing (e.g. Jaegar-Frank et al. 2006) with 3D geospatial visualisation and exploration capability (e.g. Vebree and Zlatanova 2007). Through several iterations in test bed sites, InteleSense Technologies designed and manufactured a new type of wireless data acquisition device called an InteleCell (Figure 5.2). These custom devices resolved various technological issues (e.g. Estrin et al. 1999; Hill et al. 2004; Ravi et al. 2004; Golab and Ozsu 2003) that previously constrained the application of wireless sensors to relatively

132

Michael H. Kido, Kenneth Y. Kaneshiro and Kevin N. Montgomery

small areas. The generic communications interface of InteleCells enabled the creation of heterogeneous networks (Hill et al. 2004) that could monitor environmental parameters at macro and micro-scales, as well as leverage upon new innovations that may arise in existing sensing platforms and mainstream commercial sensor products. The wireless networking capabilities of meshed InteleCells used an innovative frequency-hopping, spread spectrum technique to move data (including images) intelligently between nodes, at rates as high as 115 kbps, to an internet gateway without interference over distances of up to potentially 40 miles (with directional antennas). Therefore, sensor network arrays may conceivably span hundreds of miles as data hops from node-to-node to a distant internet gateway (Figure 5.2). Figure 5.2 System Architecture, InteleCell Hardware, and Sensor Types Currently Supported

Cyberinfrastructure for Sustainability in Coupled Human-Environment Systems

133

Stable, long-term power to the backbone network is provided by efficient, solar-powered InteleCells that are able to store up to 4 gigabytes of data per unit. The solar-powered backbone reduces power consumption in linked embedded sensor arrays which no longer need to store or process data. InteleCells also handle other low-level communication tasks such as localisation (with built-in GPS), on-off triggering of sensors, system monitoring, clock synchronisation, and data routing/caching (using secure 256-bit AES encryption) minimising communication costs to the overall network to further conserve power, bandwidth, and resources. Firmware, in field-deployed InteleCells, may also be upgraded remotely to provide flexible support for the addition of new sensors or adapt to topological changes in the sensor network. This advanced capability allows previously deployed monitoring systems to be essentially “future proof” to evolving research needs. To handle the large amounts of environmental data generated by distributed sensor arrays and provide a mechanism to integrate data from distant sources, the InteleSense Grid was developed which resides on a mirrored 14-terabyte server that is currently managed by InteleSense Technologies. The InteleSense Grid integrates grid-computing capability organically into the cyberinfrastructure platform so that data from deployed wireless sensor arrays are seamlessly linked into a federated set of information resources and middleware. This integration facilitates information resource sharing, coordination, management, and processing using the latest in emerging grid-based technologies (e.g. Chervenak et al. 2000). The Grid infrastructure handles the metadata, system catalogs, and data stores, providing authenticated user access to information and tools through powerful internet-based portals. These portals are built upon a free, versatile, open source, modular framework and content management system called Drupal (Mercer 2008). Portals may be readily customised to satisfy specific user requirements and interface with a “data-sharing engine” that outputs directly to a 3D visualiser (InteleView) via a set of linked computer servers. InteleView, constructed upon a NASA World Wind (Leslie 2005) platform, serves as the 3D data visualisation/exploration engine (Figures 5.1 and 5.2), which is seamlessly integrated into the cyberinfrastructure platform. A stand-alone PC version of InteleView may be downloaded with registration from the InteleSense Technologies website (http://www.intelesense. net/). An internet-based version (InteleView Java) running on the widely used

134

Michael H. Kido, Kenneth Y. Kaneshiro and Kevin N. Montgomery

Java platform (Standard Edition 6) was recently released and will eventually replace the PC version. Within portals a sophisticated user registration engine, managed by the Grid authenticates users by their names and passwords to provide portal administrators with a means of securing access to sensitive information. Although cyberinfrastructure development initially focused on the needs specific to regional-scale environmental monitoring, it was initially designed as an open-source technology readily expandable by user communities. Therefore, its base set of features were left highly configurable for exploring advanced concepts, potential applications to a variety of usecase scenarios, and flexibility in its abilities to address the next generation of challenges in environmental research and education. Leveraging upon its power to utilise and share information, and stimulate collaboration and innovation, this emerging cyberinfrastructure has significant potential to cross-connect scientists and decision makers beyond boundaries to promote sustainable development at the global scale. Waipa, KauA‘I: Use Case Example of Hydrological Cycle Monitoring Waipa is a relatively small Hawaiian “mountain-to-sea system” of about 1,500 acres (607 ha). It is located on north Kaua‘i in a valley that has been drastically altered by agricultural development (Figure 5.1). Except for remnant patches near the headwaters of Waipa Stream, human activities have decimated nearly all of the native forests in the valley replacing it with alien pasture grasses. About 30 years ago, the non-profit Waipa Foundation was established by native Hawaiian residents to promote sustainable development and the perpetuation of Hawaiian culture in their ahupua‘a. This goal is tightly linked to the restoration of the biophysical resources of Waipa and the revitalisation of the traditional cultivation of wetland (i.e. flooded pond field) taro (Colocasia esculenta) as a food staple. Today, about 20 acres (8.1 ha) of taro are cultivated in Waipa with an estimated 75% of the base flow of Waipa Stream used for its irrigation. Water, intrinsically linked to the localised hydrologic cycle, was the single most important issue in the traditional understanding and management of

Cyberinfrastructure for Sustainability in Coupled Human-Environment Systems

135

Hawaiian ahupua‘a. Traditional ahupua‘a typically had their surface water resources fully developed from sequential irrigation systems along the stream courses to constructed fish ponds that enclose shoreline springs and reef flats. However, these systems were managed within strict societal rules that enforced sustainable use of the resource. Today, there is a limited supply of surface water to support the expansion of wetland taro agriculture in the region. To complicate matters, historic changes in forest structure and composition in Waipa’s watershed have changed the natural dynamics of surface water capture, retention and release which has made the stream system more vulnerable to flooding. Therefore, Waipa residents are uncertain as to the supply and condition of freshwater resources, the alteration of hydrological cycles by historic land conversion, and their ability to achieve sustainable water resource use over the long-term in their ahupua‘a. To assist the Waipa community in addressing sustainability issues, a network of InteleSense-based weather stations, as well as water quality and stream flow sensors, were deployed in a prototype system throughout the ahupua‘a. The stations were designed to relate climate variation to surface water movement along watershed gradients to surface water response in Waipa Stream. Through a customised internet portal, residents are provided with authenticated access to near real-time data on weather parameters and stream / ditch conditions in Waipa which will help them to better understand and monitor their valley’s hydrologic system (Figure 5.3). As the sensor networks and associated cyberinfrastructure are further developed in Waipa, predictive models that can forecast hydrologic trends and trajectories will be possible. Future applications, such as an early flood warning system, can also be developed to help Hawaiian communities mitigate the devastating effects of flood disasters.

136

Michael H. Kido, Kenneth Y. Kaneshiro and Kevin N. Montgomery

Figure 5.3 Example of Near-Real Time Data Observed in InteleView, Used to Evaluate Flood Response in Waipa Stream and Taro Irrigation Ditch, to Torrential Rainfall Occurring in Lower Waipa Valley over a five-day Period in Nov-Dec 2006

In a broader context, the Waipa use case demonstrates a technological ability to monitor human impacts on “real-world” environments at extended spatial scales to address complex questions about the nature of coupled human-environment systems. A direct connection between science and technology and decision-making and action is made possible by a readily accessible cyberinfrastructure that delivers information to users effectively. With this capability further enhanced, the Waipa community can readily assess their environmental impacts and gauge their progress towards achieving their goal of sustainable development in their ahupua‘a.

Cyberinfrastructure for Sustainability in Coupled Human-Environment Systems

137

Cyberinfrastructure Implications for Sustainability The emerging InteleSense-based cyberinfrastructure described in this paper has significant potential for linking science with decision-making, and translating information into action. Its underlying strength lies in its ability to cross-connect among technologies, scientific disciplines, and organisational structures to initiate new forms of transdisciplinary collaboration (MaxNeef 2005) among the academe, industries, government agencies, private organisations, and communities. For example with appropriate investment in cyberinfrastructure, water resource managers are provided with the capability to compare the quantity and quality of all drinking water sources in a region and relate these data to recharge rates from local surface and groundwater sources monitored in real-time by distributed sensor arrays. Water resource trends may then be compared to regional weather variation occurring at local-to-regional-toglobal scales to monitor and evaluate, for example, the effects from severe storms or extended droughts predicted in climate change scenarios. With sufficient network and bandwidth capacity, raw data feeds may be analysed through models run in distributed supercomputers and rapidly displayed via the internet in 3D simulations running on desktop computers located anywhere in the world. Cyberinfrastructure may thus be globally applied to understand and monitor all facets of human-environment systems from biogeochemical cycles to pollution flows to the spread of infectious diseases. If developed appropriately, cyberinfrastructure has the potential to tear down boundaries between science and policy. It can establish a framework for knowledge systems, which rapidly delivers pertinent information to decision makers that is easily understood and interpreted. Fed by grid-based information technologies linked together over the internet, knowledge systems can quickly query, integrate, and interpret data from globally distributed information repositories through robust computational models that describe human-environment systems and their interactions. As environmental sensors become even more sophisticated and globally distributed, they will provide rapid and comprehensive feedback to knowledge systems about the condition of the earth’s life support systems. The use of cyberinfrastructure can thus revolutionise the manner in

138

Michael H. Kido, Kenneth Y. Kaneshiro and Kevin N. Montgomery

which information about human-environment systems may be obtained, interpreted, and applied to sustainable development. There is an urgent need for the creation of more effective knowledge systems for sustainability that are merged with vulnerability frameworks focused on interactions in human-environment systems (Cash et al. 2003). However, this initiative has been stymied by the slowness of institutions to accept new ideas, increase technological capacity, and implement new policies. In our view, emerging cyberinfrastructure will eventually overcome these challenges as it transforms all data into useable knowledge and delivers it rapidly through the internet to decision makers. Translation of science and technology products into policy has never been easy for institutions. However, the effective use of cyberinfrastructure can help bridge this boundary through its wide-reaching, transdisciplinary knowledge framework. After all, our ultimate survival depends on our ability to sustainably deal with the next generation of challenges arising from emerging infectious diseases (e.g. Wilcox and Colwell 2005), the effects of global warming (e.g. Malcolm et al. 2002), loss of biocomplexity (e.g. Colwell 1998), and increasing demand for natural resources, such as water (e.g. Vreke 2004). NOTES a Centre for Conservation Research and Training, University of Hawai‘i, 3050 Maile Way, Gilmore 406, Honolulu, HI 96822, USA, e-mail: [email protected] b Centre for Conservation Research and Training, University of Hawai‘i, 3050 Maile Way, Gilmore 406, Honolulu, HI 96822, USA, e-mail: [email protected] c National Biocomputation Centre, Stanford University, 701A Welch Road, Suite 1128, Palo Alto, CA 94304, e-mail: [email protected] References Bajcsy, R. NSF Blue Ribbon Panel on Cyberinfrastructure. Advisory Committee on Cyberinfrastructure, 2001.

Cyberinfrastructure for Sustainability in Coupled Human-Environment Systems

139

Cash, D. W. “Distributed Assessment Systems: An Emerging Paradigm of Research Assessment and Decision-making for Environmental Change.” Global Environmental Change 10(2000):241-244. Cash, D.W., W.C. Clark, F. Alcock, N.M. Diskson, D. H. Guston, J. Jager, and R. B. Miller. “Knowledge Systems for Sustainable Development.” Proceeding of the National Academy of Science 100(2003):8086-8091. Chervenak, A., I. Foster, C. Kesselman, C. Salisbury, and S. Tuecke. “The Data Grid: Towards An Architecture for the Distributed Management and Analysis of Large Scientific Datasets.” Journal of Network and Computer Applications 23(2000):187-200. Colwell, R.R. “Balancing the Complexity of Planet’s Living Systems: A Twenty-first Century Task for Science.” BioScience 48(1998):786-787. Estrin, D., R. Govindan, J. Heidemann, and S. Kumar. “Next Century Challenges: Scalable Coordination in Sensor Networks.” Proceedings of the 5th Annual Association for Computing Machinery Conference on Mobile Computing and Networking. ACM Press, New York, p. 263270, 1999. Golab, L. and M. T. Ozsu. Issues in Data Stream Management. SIGMOD Record 32(2003):5-14. Guston, D.H. “Stabilizing the Boundary Between US Politics and Science: The Role of the Office of Technology Transfer As A Boundary Organization.” Social Studies of Science 29(1999):87-112. Hill, J., M. Horton, R. King, and L. Krishnamurthy. “The Platforms Enabling Wireless Sensor Networks.” Communications of the ACM 47(2004): 41-46. Jaeger-Frank, W., C. J. Crosby, A. Memon, V. Nandigam, J. R. Arrowsmith, J. Conner, I. Altintas, and C. Baru. “A Three Tier Architecture for LIDAR Interpolation and Analysis.” In ICCS 2006, Part III, LNCS 3993, V.N. Alexandrov (ed),, 920-927pp. Springer-Verlag Berlin / Heidelberg. Kaneshiro, K.Y., P. Chinn, K.N. Duin, A.P. Hood, K. Maly, and B.A. Wilcox. “Hawai‘i’s Mountain-to-Sea Ecosystems: Social-ecological Microcosms for Sustainability Science and Practice.” EcoHealth 2(2005):349-360. Karasti, H., K. S. Baker, and E. Halkola. “Enriching the Notion of Data Curation in E-Science: Data Managing and Information Infrastructuring in the Long Term Ecological Research (LTER) Network.” Computer Supported Cooperative Work 15(2006):321-358. Kates, R.W. and T.M. Parris. “Long-term Trends and A Sustainability Transition.” Proceedings of the National Academy of Science 100(2003):8062-8067.

140

Michael H. Kido, Kenneth Y. Kaneshiro and Kevin N. Montgomery

Kaur, T., J. Singh, W.M. Goodale, D. Kramar, and P. Nelson. “Developing A Cyberinfrastructure for Integrated Assessments of Environmental Contaminants.” Ecotoxicolog y 14(2005):275-281. Kido, M.H., C.W. Mundt, K. N. Montgomery, A. Asquith, D.W. Goodale, and K. Y. Kaneshiro. “Integration of Wireless Sensor Networks Into Cyberinfrastructure for Monitoring Hawaiian “Mountain-to-Sea” Environments.” Environmental Management 42(2008):658-666. Leslie, M. “The Earth in Your Computer.” Science Magazine 308(2005):933. Malcolm, J.R., A. Markham, R.P. Neilson and M. Garaci. “Estimated Migration Rates Under Scenarios of Global Climate Change.” Journal of Biogeography 29(2002):835-849. Max-Neef, M.A. “Foundations of Transdisciplinarity.” Ecological Economics 53(2005): 5-16. Mercer, D. “Building Powerful and Robust Websites with Drupal 6.” Packt Publishing, Birmingham, U.K, 2008. Pfirman, S. and the AC-ERE. “Complex Environmental Systems: Synthesis for Earth, Life, and Society in the 21st Century.” NSF Foundation, 2003. Ravi, S., A. Raghunathan, P. Kocher, and S. Hattangady. “Security in Embedded Systems: Design Challenges.” ACM Transactions on Embedded Computing Systems 3(2004):461-491. Santi, P. “Topology Control in Wireless Ad Hoc and Sensor Networks.” ACM Computing Surveys 37(2005):164-194. Szewczyk, R., E. Osterweil, J. Polastre, M. Hamilton, A. Mainwaring, and D. Estrin. “Habitat Monitoring with Sensor Networks.” Communications of the Association for Computing Machinery 47(2004):34-40. Verbree E. and S. Zlatanova. “Positioning LBS to the Third Dimension.” In Location Based Services and Telecartography, G. Gartner, W. Cartwright, M. P. Peterson (eds). Springer Publishing, Berlin, 107-118pp. Vreke, J. “Optimal Allocation of Surface Water in Regional Water Management.” Water Resources Management 8(1994):137-153. Walker, B., S. Carpenter, J. Anderies, N. Abel, G. Cumming, M. Janssen, L. Lebel, J. Norberg, G. D. Peterson and R. Pritchard. “Resilience Management in Social-ecological Systems: A Working Hypothesis for A Participatory Approach.” Conservation Ecolog y 6(1)(2002):14. http://www. consecol.org/vol6/iss1/art14 Wilcox, B.A. and R.R. Colwell. “Emerging and Reemerging Infectious Diseases: Biocomplexity As An Interdisciplinary Paradigm.” Ecohealth 2(2005): 244-257.

Watershed Management to Sustain Tropical Watersheds: An Integrated Hydrologic Modeling Approach

141

6 Watershed Management for Sustainability in Tropical Watersheds: An Integrated Hydrologic Modeling Approach

Ali Fares a

ABSTRACT Competition for natural resources between various stakeholders within the same geographical area is becoming a reality. An integrated, multidisciplinary modeling approach is needed to efficiently manage these resources. A watershed management approach, using hydrologic modeling, enjoys ample acceptance in coastal areas and small islands because of the smallsized watersheds in these areas. Water quality and quantity are essential parts of the livelihood of these areas. A deep understanding of watershed hydrolog y and water quality is necessary for sustainable planning, management, and protection of natural resources. Hydrologic models offer practical tools for watershed management as they optimise two precious assets – time and money. These models lessen the number of field experiments required to implement and underscore major parameters and variables that most influence this system. This chapter summarises the results of evaluating the performance of AnnAGNPS (Annualised Agricultural Nonpoint Source Pollution) and N-SPECT (Nonpoint Source Pollution and Erosion Comparison Tool) under tropical and coastal watershed settings. The models identified “hotspots” to be targeted to enhance watershed sustainability. The models were found to have strong potential in simulating the runoff and sediment load from the Hanalei River Basin in Kaua‘i, Hawai‘i. However, these models need to be improved to make their predictions and capabilities in watersheds of steep slope and spatially variable rainfall more precise.

142

Ali Fares

Introduction Rapid growth of global population and changes in economic environments trigger land-use change that can be linked to changes in climate, biodiversity, and water quantity and quality. The impacts of these changes have more pronounced effects on coastal watersheds, especially those of small islands, i.e. Caribbean, Pacific, and the Hawaiian Islands. The concept of watershed sustainability originated from attempts to manage renewable resources by measuring annual increases in resources without reducing physical stocks (Dixon and Fallon 1989). Collaborative watershed planning and management, with the help of local groups and institutions, is a solution to developing sustainable programmes in a watershed. The Hawaiian Islands are characterised by a large number of small and steep watersheds with highly permeable volcanic rocks and soils. Rainfall is spatially and temporally variable. For example, on the island of Kaua‘i, the annual rainfall increases from 500 mm near Kekaha to over 11,000 mm in Mt. Wai‘ale‘ale, an average gradient of 0.42 mm/m (Giambelluca, et al. 1986). Because of the high degree of diversity in weather and land characteristics, and the compressed spatial and temporal scales in which on-site and off-site impacts of land use and management are manifested, an integrated watershed approach is vital to successfully managing these ecosystems. Sound scientific tools are required to predict and mitigate the detrimental impacts. Field experiments at different scales have been implemented to evaluate alternative land use management practices. However, such studies are typically site-specific, weather-dependent, and costly in time and resources (Fares et al. 1996). Numerical models offer practical tools to optimise time and money (Salama et al. 1993). Modeling decreases the number of field experiments required to implement and underscore major parameters and variables that most influence this system (Fares et al. 1997). Land-based pollutants are identified as one of the primary threats to coral reef ecosystems. In the Hanalei watershed on Kaua‘i Island, soil erosion yield was estimated at 1.4 Mg ha-1 y-1 (Calhoun and Fletcher 1999). The primary sources of pollutants are feral ungulates and alien plants that increase erosion in the upland Hanalei watershed, urban cesspools, and septic systems, agricultural operations (taro ponds), water bird impoundments, and cattle grazing. Traces of herbicides used to control invasive species and weeds in the area were found in river water only below method detection limits (Orazio et al. 2001).

Watershed Management to Sustain Tropical Watersheds: An Integrated Hydrologic Modeling Approach

143

The US Department of Agriculture (USDA) Natural Resources Conservation Service (NRCS) is providing technical assistance to the community of Hanalei, Hawai‘i to assess and analyse stream sediment-related issues through the Cooperative River Basin Study programme. The longterm objective of these activities is to develop watershed Best Management Practices (BMPs) to reduce sediment discharge from the watershed and minimise sediment impacts to the estuary, beaches, and near shore reef areas of Hanalei Bay. The watershed-scale modeling supports many elements of the Watershed Action Plan, which was prepared and updated by the Hanalei Watershed Hui. The Watershed Action Plan was initially developed in 1999 when the Hanalei River was designated by President Clinton as one of 14 American Heritage Rivers (AHR). Though the five-year programme life of the AHR initiatives has eclipsed, the community organisation, spurred by the focus of the EPA Targeted Watershed Initiatives and other federal programmes in the area, has transformed itself into the Hanalei Watershed Hui. Hanalei watershed management aims to improve water quality and reduce stress on riverine, estuarine, and coral reef ecosystems. A number of watershed-scale hydrologic models have been developed (Borah and Bera 2004). The Annualised Agricultural Nonpoint Source Pollution model, AnnAGNPS, (Binger and Theurer 2005) and its single event predecessor AGNPS have been widely used for runoff, sediment, and nutrient load predictions (Suttles et al. 2003) and as decision support systems (Mitchell et al. 1993). Another model specific for coastal research, the Nonpoint Source Pollution and Erosion Comparison Tool, N-SPECT, has the capability to examine land cover, soil characteristics, topography, and precipitation data to estimate surface water runoff, pollutant, and sediment load locally and on a watershed scale (National Oceanic and Atmospheric Administration; NOAA 2004). AnnAGNPS has been used and calibrated for a variety of conditions (e.g., Baginska et al. 2003; Grunwald and Norton 2000; Yuan et al. 2001). The model was tested in Europe (Pekarova et al. 1999), North America (Perrone and Madramootoo 1997; Yuan et al. 2001), Australia (Baginska et al. 2003), and in Africa (Leon et al. 2003). However, studies have not been conducted in tropical, volcanic, and island environments, which are characterised by high rainfall variability, complex terrain, and highly weathered soils. Likewise, N-SPECT has not been tested on the watershed scale to estimate runoff, pollutant, and sediment load.

144

Ali Fares

This study aimed at testing an integrated modeling approach for sustainable water management in tropical watersheds. The study specifically i) calibrated and validated AnnAGNPS and N-SPECT models for the Hanalei watershed, and ii) identified critical areas or “hot spots” of the watershed and their sediment loading to the river. Materials and Methods Study Area Hanalei River Basin is a 61-km2 area located on the north shore of Kaua‘i that extends from the top of Mount Wai‘ale‘ale, at 1,569 m, to the coral reefs of Hanalei Bay (Figure 6.1). The watershed is represented by heavily vegetated steep mountain slopes that plunge into deep fluvial valleys, through which flows the upper 20 km of Hanalei River with its tributaries. Approximately 7560 ± 2910 t of sediments per year are removed from the Mount Wai‘ale‘ale plateau by the Hanalei River. This situation translates into a denudation rate of approximately 0.07-0.16 mm y-1 (Calhoun and Fletcher 1999). The watershed supports an array of land uses: residential, recreational, agriculture, biodiversity, and preservation of the native agricultural practices. The Hanalei Bay coastal area is listed in section 303(d) of the Clean Water Act as an impaired water body where majority of the corals are suffering from sedimentation, and nutrient and other chemical contaminants. Nonpoint sources of pollution derive from the lack of a centralised wastewater treatment system, stream bank erosion from invasive species, and nutrient deposition from taro farms. Watershed Models The study used AnnAGNPS and N-SPECT simulation models. AnnAGNPS is a distributed-parameter, continuous simulation watershed scale model developed jointly by USDA Agricultural Research Service (ARS) and NRCS as a management decision tool for agricultural watersheds. The model includes a suite of modules that simulate water, sediment, and pollutant transport as results of precipitation, irrigation, and snowmelt. The model operates on a daily time step with a minimum simulation period of one year.

Watershed Management to Sustain Tropical Watersheds: An Integrated Hydrologic Modeling Approach

145

Figure 6.1 Hanalei Watershed Located on the Island of Kaua‘i, Hawai‘i

N-SPECT is a Geographic Information System (GIS)-based watershed scale model developed by the National Oceanic and Atmospheric Administration (NOAA) Coastal Services Centre (CSC). It examines land cover, soil characteristics, topography, and precipitation data to estimate surface water runoff, and pollutant and sediment load locally and on a watershed scale. The model facilitates the comparison of management scenarios by providing a rapid assessment of the relative impacts of current and future land-use changes. It also allows users to easily identify critical areas, or “hot spots,” in the watershed from the GIS output layers. The model operates on an annual and event basis. AnnAGNPS and N-SPECTS use the NRCS Curve Number (CN) method to determine surface runoff, and Technical Release-55 method to calculate peak flow. Sheet and rill soil erosion were determined by the Revised

146

Ali Fares

Universal Soil Loss Equation (RUSLE), and sediment yield was obtained using a sediment delivery ratio (SDR) prediction equation. N-SPECT incorporates the SDR equation, obtained from Williams (1977), which is dependent on relief-length ratio, CN, and drainage area. AnnAGNPS uses the Hydro-geomorphic Universal Soil Loss Equation (HUSLE) to determine the delivery ratio for total sediments. Weather and Climate Climate in Hawai‘i is characterised by large spatial and temporal variations of rainfall. The long-term average rainfall is 9,508 mm y-1 and 2,089 mm y-1 on Mount Wai‘ale‘ale and in the Hanalei watershed outlet, respectively. The two-year period of January 2003 through December 2004 was selected as the simulation period for the study. The two gauges located within this watershed (USGS Station 16103000 on the Hanalei River [-159.5010 W; 22.0740 N], and USGS Station 220427159300201 at Mount Wai‘ale‘ale [-159.5010 W; 22.0740 N]) were used to derive daily precipitation input for the models. Rainstorms that occurred in the watershed during this period were characterised by high frequency and intensity (Figure 6.2). The average annual precipitation over the two-year period was 3,403 mm y-1 at the watershed outlet and 9,130 mm y-1 on Mount Wai‘ale‘ale. Rain occurred for 88 percent of the days over the two-year period, and 17 percent of these events had precipitation rates exceeding 50 mm d-1.

Watershed Management to Sustain Tropical Watersheds: An Integrated Hydrologic Modeling Approach

147

Figure 6.2 Measured Precipitation for the 2003-2004 Simulation Period at USGS Rain Gauges at the Watershed Outlet and on Mt. Wai‘ale‘ale Watershed Outlet

Mt. Wai‘ale‘ale

To empirically specify a format which provides spatial rainfall variation across the watershed, two alternative maps of average annual precipitation for the Island of Kaua‘i were used (Figure 6.3). The first map was obtained using the PRISM model, which is heavily influenced by elevation (Daly et al. 1994). The second map was obtained using historic island-wide data from the Hawai‘i Rainfall Atlas (Giambelluca et al., 1986). Precipitation was interpolated over space using information from existing weather stations.

148

Ali Fares

The water flow in Hanalei River was extremely varied with an average daily discharge of 5.3×105 m3 (Figure 6.4). Daily minimum and maximum discharges were reported at 8×104 m3 and 2.0×107 m3. Figure 6.3 Spatial Distribution of Average Annual Precipitation in Hanalei Watershed as Determined Using the (a) PRISM model and (b) Hawai‘i Rainfall Atlas

a)

b)

Figure 6.4 Daily Average Stream Flow in Hanalei River at the USGS Gauging Station Based on 92 Years of Record (1913-2004)

Watershed Management to Sustain Tropical Watersheds: An Integrated Hydrologic Modeling Approach

149

Soil and Land Cover The cell map was overlaid with a digital soil map obtained from the Soil Survey Geographic (SSURGO) database (NRCS 2004). The soil series that occupied the largest area within a given cell was assigned to that cell. Major physical and chemical properties used for each soil layer included texture, hydrological soil group, saturated conductivity, organic matter, pH, and K factor (Table 6.1). Table 6.1 Predominant Soils of the Upper Hanalei Watershed and Their Basic Characteristics Soil

Area HSD (%)

Clay (%)

K sat (mm h-1)

OM (%)

pH K factor

Rough mountainous land

56.8

D

30-35

100

20-30

3.5-5 0.026

Rock outcrop

17.1

D

-

-

-

- 0.003

Hulua gravelly silty clay loam, 3-25% slopes

10.6

D

35-40

100

10-20 5.1-5.5 0.007

Hulua gravelly silty clay loam, 25-70% slopes

7.6

D

35-40

100

10-20 5.1-5.5 0.007

Hanamaulu stony silty clay, 10-35% slopes

2.7

B

50-60

100

2-6

3.6-5

0.013

Pooku silty clay, 2540% slopes

2.1

B

35-60

100

4-9 3.5-5.5

0.013

Other

3.1

Legend: HSD - hydrological soil group; Ksat - saturated hydraulic conductivity or permeability; OM - organic matter

Most of the soil types in the watershed were classified as silty clay loam. The dominant type of soil in the study area was Rough Mountainous Land, occupying 56 percent of the simulated area. Land cover was represented by a 30-m resolution baseline map of Kaua‘i developed from Landsat ETM

150

Ali Fares

satellite imagery (NOAA. 2000). Five land cover types including grassland, evergreen forest, scrub/shrub, palustrine shrub, and water were identified (Figure 6.5). Figure 6.5 Land Cover Map of the Study Area

After the overlay with the cell map, a dominant land cover type, either evergreen forest or shrub, was assigned to each cell. The areas under grassland, palustrine shrub, and water were too small and spatially dispersed to dominate in any individual cell. Additional land cover characteristics, necessary for model input, included canopy height and density, percentage of ground cover, and root mass.

Watershed Management to Sustain Tropical Watersheds: An Integrated Hydrologic Modeling Approach

151

Results and Discussion Runoff Simulation Annual runoff predicted by AnnAGNPS for the 2003 calendar year was similar to the measured runoff (1094 mm y-1 and 1149 mm y-1, respectively) (Figure 6.6). Although the year’s total simulated runoff was 95 percent of the measured value, the model performed less accurately on a monthly (R 2 = 0.76) and daily (R 2 = 0.53) basis, where R 2 is the coefficient of correlation between measured and predicted values. The relatively poor runoff prediction on an event basis was attributed to uncertainties in spatial variation of rainfall. Monthly totals were within 40 percent of observed values except for March and August, when precipitation amounts were relatively low. A similar trend was observed in a watershed in Mississippi when during relatively dry periods the model results deviated greatly from the measured values (Yuan et al. 2001). Figure 6.6 AnnAGNPS

(a)

(b)

152

Ali Fares

The surface runoff component of N-SPECT was calibrated and validated with 1999 and 2001 rainfall-runoff data, respectively (Figure 6.7). Calibration was conducted by adjusting two input parameters: NRCS curve numbers (CN) and the number of rain days. Changes in CN’s reflect the land cover condition and antecedent soil moisture. The definition of a rain day differed between the two methods, specifically in the threshold values used and the amount of canopy interception (CI) accounted for. Canopy interception is the difference between gross and net precipitation (throughfall and stemflow). It occurs with every single rainfall event and can significantly affect the amount of overland flow. Giambelluca et al., (2004) found CI to be 35 percent of incidental rainfall for Maui, Hawai‘i. Runoff for the N-SPECT model was calibrated with two methods as described below. Figure 6.7 Daily Precipitation and Runoff from 1999 and 2001 at the Watershed Outlet and Mt. Wai‘ale‘ale

Watershed Management to Sustain Tropical Watersheds: An Integrated Hydrologic Modeling Approach

153

Method I: The first calibration was conducted using the old soil classification. Initially, SCS CN’s were adjusted within a reasonable range of values as specified in the NRCS TR-55 Manual (USDA 1986). Then the number of rain days was modified to improve the accuracy of the runoff prediction. A rain day is defined as a day during which total precipitation depth is greater than specified threshold values: 0.25 mm, 1.27 mm, 2.54 mm, 7.62 mm, and 12.70 mm. Simulations were conducted at three sub-basin levels and a rain day threshold value of 0.25 mm of rain. Initial simulations of the model, using default CNs, greatly underestimated runoff for all simulation periods. The default set of CNs assumed average runoff condition with land cover in good condition and an antecedent moisture content II (AMC II). Since the Hanalei watershed is characterised by high intensity and frequent rainfalls, it was logical to use an AMC III condition. In addition, to account for the model’s inability to allow for reduced initial abstraction, Ia, (such as 0.1S instead of 0.2S), the CNs were further increased by assuming fair to poor condition for all land cover classes. Daily initial abstraction is the amount of rainfall that does not contribute to runoff. It is defined in the SCS curve number equation as equivalent to 20 percent of the potential maximum

154

Ali Fares

retention (S). With the final calibrated set of CNs, the model underestimated runoff by 1 percent for calibration period (1999) and 5 percent for validation period (2001). Method II: The second method was developed and preferred over the first method because the rain day definition and threshold values have physical meanings during this calibration. In addition, the simulations under this method used the new soil classification. The number of rain days was the sum of days with precipitation exceeding Ia. Rainfall loss was attributed to evapotranspiration, CI, and/or infiltration. The range of Ia used were 7.62 to 15.24 mm in 1.27 mm increments. Daily precipitation data were preprocessed to exclude days with rainfall less than Ia during which runoff would not occur. Some of the remaining precipitation reflected total annual effective rainfall or rainfall that would contribute to surface runoff. SCS CNs were then adjusted to achieve better runoff estimation. Similar to the results from Method I, the default set of CNs, at an Ia threshold of 7.62 mm, underestimated runoff. Curve numbers were increased to reflect land cover conditions of fair and AMC III (Figure 6.8a), while the Ia threshold was increased to 13.97 mm. The results (Figure 6.8b) showed an improved measured and estimated runoff during calibration (0.8% overestimation) and validation (4.8% underestimate). Figure 6.8 (a) Spatial Distribution of CN Values for C-CAP Land Cover and (b) Estimated Runoff Volume (per grid cell)

a) Curve numbers

b) Runoff depth

Watershed Management to Sustain Tropical Watersheds: An Integrated Hydrologic Modeling Approach

155

Sediment Load Prediction Sediment load generated by AnnAGNPS from individual cells was highly variable (Figure 6.9). Approximately a third of the watershed contributed very little (0-1 t ha-1 y-1) to the total sediment load, and thus was not a major sediment-generating area. However, 5 percent of the watershed had sediment load rates in excess of 5 t ha-1 y-1. Figure 6.9 Spatial Distribution of Average Annual Sediment Load from Hanalei Watershed Using AnnAGNPS

Most of the watershed area was covered by shrub vegetation. There was no significant spatial correlation between the Landsat land cover type and sediment yield. Both forest and shrub provided dense canopy and ground residue protection for the soil. Slope gradient was also not significantly

156

Ali Fares

correlated with sediment yield. Soil type, however, was the single most important factor affecting sediment generation. Soils classified as rough mountainous land, found on 57 percent of the watershed, generated 84 percent of the sediment delivered through the watershed outlet. The cells, dominated by rough mountainous land soil, characterised by relatively high erodibility (K = 0.026), generated three to 10 times more sediment than the cells with other soil types but with a similar LS (slope length and steepness) factor. Total monthly sediment yield varied within 50 percent of the observed values, except for May 2004. Fifty one percent of the total area of the watershed contributed less than 10 percent of the total sediment load. However, 49 percent of watershed generated over 90 percent of the total sediments (Figure 6.10). These results were based on the use of original NRCS soil classifications and USGS land cover map with baseline curve numbers. Figure 6.10 Total Sediment Contribution Versus Percent Area of Watershed Contributing to the Sediment Load

In July 2006, NRCS soil classification was preliminarily updated and the hydrologic soil group (HSG) of rough mountainous land was changed from D to B and that of the rock outcrop, from D to C. Approximately 89 percent of the original soil classification for the study area was from HSG D, where

Watershed Management to Sustain Tropical Watersheds: An Integrated Hydrologic Modeling Approach

157

the soils have low infiltration rates and therefore high runoff potential. In the updated soil classification, over half of the soil types belong to HSG B, which has moderate infiltration rates and lower runoff potential. Direct surface runoff simulated for 2003 with the updated soil classification was 842 mm – a 23 percent decrease from the value estimated with the original soil classification. Therefore, the updated data i.e., soil type, its hydrological groups, and land cover or management type, should always be used during evaluation of a model. The sediment component of N-SPECT was calibrated and validated with the October-December 2003 and January-September 2004 data, respectively. Actual average suspended sediment (SS) concentration in the watershed outlet for 2003 was 8.8 mg/L which translated to accumulated sediments of 1,629.5 tons. For 2004, the average SS concentration was 11.5 mg/L (8,184.6 tons of sediment). Accumulated sediment predicted by the original N-SPECT model for 2003 was 2.15x106 tons, almost 1.2x105 percent larger than the actual amount. Modification of the R and C factors did not improve prediction accuracy. It is strongly believed that the SDR prediction equation was the source of this overestimation. To solve this problem, N-SPECT was modified. RUSLE was used to calculate soil erosion and two LS factor grids were applied. The first grid was created during watershed delineation within N-SPECT (Figure 6.11a). The second grid was derived directly from the AML script by Bob Hickey at Central Washington University (Figure 6.11b). Delivery ratios, computed for the Hanalei basin, ranged from 0 to 10 (Figure 6.12a). Estimated sediment yield (Figure 6.12b) was 100 times higher than the actual amount (Table 6.2). A sediment delivery ratio above 1 means that more sediment was delivered from the watershed than it was actually produced by soil erosion.

158

Ali Fares

Figure 6.11 LS Factor Grids Applied in the Simulation of Soil Erosion

a) LS factor grid from N-SPECT

b) LS factor grid from AML script

Figure 6.12 Sediment Delivery Ratios and Sediment Yield in Hanalei Bay Watershed Predicted with the Modified Version of N-SPECT

a) Distribution of sediment delivery ratios b) Sediment yield distribution

Watershed Management to Sustain Tropical Watersheds: An Integrated Hydrologic Modeling Approach

159

Table 6.2 Soil Erosion and Sediment Yield for October-December 2003 Estimates with Original and Modified Versions of N-SPECT, Using Two Different LS Factor Grids, and Compared to Actual Sediment Yield N-SPECT Version LS Factor Grid Source Original Modified

N-SPECT AML script

Soil Erosion (ton)

Sediment Yield (ton)

N/A

1,630

N/A

2,146,434

81,627

161,670

78,921

156,923

In the Hanalei Watershed, there was increased sedimentation as a result of feral pig activities. Berg and Rosener (2006) developed a feral pig damage matrix by distributing different levels of pig disturbance on ground surface at varying elevations and slopes (Figure 6.13). The matrix showed that feral pigs were common in areas with 0 to 50 percent slopes and elevations of < 762 m. With nearly 90 percent of the study area affected by feral pig activities, the simulated sedimentation was 4,170 t or 2.5 times larger than the value obtained without pig damage. This substantial increase in sedimentation might be due to the high sensitivity of the RUSLE model to the surface residue cover parameter. Nearly half of the surface residue coefficients were decreased below 60 percent of their original values when pig damage was not accounted for.

160

Ali Fares

Figure 6.13 Feral Pig Disturbance in Hanalei Watershed

Sensitivity Analysis The sensitivity analysis of AnnAGNPS revealed that sediment yield from the watershed was closely related to vegetation root mass, average canopy fall height, soil erodibility, percentage of ground residue cover, and canopy cover ratio. Percentage of ground residue cover and the canopy cover ratio had a significant influence on sediment yield with absolute values of senility index, SI (Nearing et al. 1990) -2.9 and -2.4, respectively. The negative sign of SI indicates an inverse relationship – a decrease of ground residue cover from 95 percent to 60 percent resulted in a five-fold increase in sediment yield. A policy implication of this observation relates to the problem of invasive species in Hawai‘i. When invasive species such as Miconia (Miconia calvescens) replace native vegetation, surface litter, an important factor in preventing severe erosion, tends to decrease. Therefore, to minimise sediment yield,

Watershed Management to Sustain Tropical Watersheds: An Integrated Hydrologic Modeling Approach

161

best management practices (BMP) in a watershed should provide and sustain ground residue and canopy cover. Sensitivity analysis also showed that areas with low clay content and steep slopes produced high sediment. These “hotspot” areas should be managed under BMP with great care and priority. Vegetation, root mass, canopy fall height, and soil K-factor had approximately similar influences on the model output, yielding SI values of 0.9, 0.9, and 1.0, respectively. Sensitivity analysis of N-SPECT was carried out with runoff calibration methods I and II. With method I, simulated runoff was examined using three sub-basin levels and rain day threshold values of 0.25 mm, 1.27 mm, 2.54 mm, 7.62 mm, and 12.70 mm. The number of rain days were converted to relative ratios and plotted against simulated runoff percent error. Relative ratio is the ratio of the number of rain days based on one of five threshold values to the total number of days with rain. A 13 percent increase in the proportion of simulated rain days resulted in a nearly 25 percent decrease in runoff, which was consistent for all three sub-basin levels. Runoff was also simulated at a rain day threshold of 0.25 mm and without sub-basin separation for five canopy intercept (CI) levels: 15 percent, 20 percent, 25 percent, 30 percent, and 35 percent. Simulated runoff decreased by 12 percent at the first CI interval and dropped consistently to 1 percent for subsequent CI levels for all sub-basin levels and rain day thresholds. Canopy interception had a slightly lower impact on runoff than the number of rain days. When rain days were represented using method II, weighted average overestimated runoff by 10 percent, whereas the arithmetic average method underestimated runoff by 16 percent. Since weighted averages of rain days were consistently lower than those of arithmetic averages, runoff was overestimated. All three rain day distribution methods showed negative correlation in which runoff decreased with increasing rain days or decreasing Ia thresholds. Among the rain day distribution methods, the model was most sensitive to the spatial distribution of rain days, since a change in rain days affects the unique values representing the single grid cell.

162

Ali Fares

Summary and Conclusions Performance of AnnAGNPS In the initial validation phase, existing generalised soil and land cover information was used to measure and simulate annual surface runoff. Results were similar as soil runoff measured 2,213 mm, land cover runoff measured at 2,405 mm. Monthly totals were predicted yielding an R2 = 0.90. However, there was a 50-percent difference between the actual and simulated data during the dry season (May and July). Prediction of daily runoff yielded a low coefficient of correlation (0.55) and higher variation during smaller precipitation events. Comparison between predicted and observed sediment yield on a daily basis showed low correlation (R 2 = 0.5). The model generally overestimated sediment load for the small magnitude events, while the opposite was true for larger events. There was a close agreement between simulated and predicted monthly total sediment yields (R 2 = 0.85). Sediment yield from individual cells to the watershed outlet was highly variable as approximately one third of the watershed contributed 0-1 t ha-1 y-1 to the total sediment yield. However, 5 percent of the area had sediment yield rates in excess of 5 t ha-1 y-1. Sediment yield from the watershed was closely related to vegetation root mass, average canopy fall height, soil erodibility, percentage of ground residue cover, and canopy cover ratio. The latter two had the greatest influence on sediment yield. The entire watershed area was covered by dense vegetation, which protects the soil from direct rainfall impact. Under these conditions, high sediment yields were observed in areas with low clay content and along steep slopes. RUSLE erosion factor K, which is directly related to soil properties, influenced spatial variability of sediment losses. Incorporation of feral pig damage in the estimates highlights the need for an accurate estimation of the impact of pig damage on residue cover throughout the watershed. Performance of N-SPECT The number of rain days was spatially distributed throughout the watershed to accurately predict surface runoff. Results of the sensitivity analysis showed

Watershed Management to Sustain Tropical Watersheds: An Integrated Hydrologic Modeling Approach

163

at least a 15-percent difference between the simulated runoff using a single number of rain days and that which uses spatial distribution of rain days throughout the watershed. Calibrated CN assumed fair condition for all types of land cover with wet antecedent soil moisture. The model performed reasonably well during calibration and validation, 0.8 percent overestimate and 4.8 percent underestimate in runoff, respectively. However, the model failed to produce realistic sediment output. Accumulated sediment predicted by the model was almost 1,200 percent larger than the actual amount. A close examination revealed that the SDR prediction equation was the cause for such unrealistic values. The SDR model used by N-SPECT seemed inapplicable to steep tropical watersheds such as those in Hawai‘i. Based on the results derived, the component of the model performed well when the Hanalei watershed was simulated with spatial distribution of rain days and curve numbers that reflected wet antecedent soil moisture. Simulations showed that a policy implication of this observation was related to the problem of invasive species in Hawai‘i. When an invasive species i.e., Miconia (Miconia calvescens) replaces native vegetation, there is lack of surface litter, which is important in preventing severe erosion. Therefore, to minimise sediment yield, best management practices (BMP) in a watershed should provide and sustain ground residue and canopy cover. Since the models were used to primarily apply the results to conservation planning and risk analysis, the performance of AnnAGNPS and N-SPECT was adequate. Overall, the models seemed to be suitable for the comparative assessments of tropical watersheds, including approximate gross estimation of runoff and sediment loads, and identification of landscape “hot spots” (i.e., the areas of the watershed that are responsible for significant portions of the sediment delivered). AcknowledgmentS The author acknowledges the funding support of the United States Department of Agriculture Natural Resources Conservation Service, United States Department of Agriculture Tropical and Sub-tropical Agricultural Research, and National Oceanic and Atmospheric Administration. The author also acknowledges the expertise shared by Farhat Abbas, Samira

164

Ali Fares

Fares, Viktor Polyakov, Chui Cheng, and Sanjit K. Deb from University of Hawai‘i-Mānoa; Dudley Kubo, Chris Smith, Katina Hanson, Dan Moore, Fred Theurer, and Mike Robotham from NRCS; Jim Jacobi from the United States Geological Survey; and Carl Berg from the Hanalei Watershed Hui. NOTES a Department of Natural Resources and Environmental Management 1910 East West Rd, University of Hawai‘i at Mānoa, Honolulu, HI 96822 Ph: 808 956-6361, e-mail: [email protected] References Baginska, B., W. Milne-Home, and P.S. Cornish. “Modelling Nutrient Transport in Currency Creek, NSW with AnnAGNPS and PEST.” Environmental Modelling & Software, 18(8-9)(2003):801-808. Berg, C. and M. Rosener. 2006. http://www.hawp.org/downloads/Hanalei_ compressed.pdf Binger, R.L. and F.D. Theurer. “AnnAGNPS Technical Processes Documentation.” http://www.ars.usda.gov/Research (accessed 3 April 2006). Borah, D.K. and M. Bera. “Watershed-scale Hydrologic and Nonpointsource Pollution Models: Review of Applications.” Transactions of the ASAE, 47(3)(2004): 789-803. Calhoun, R.S. and C.H. Fletcher. “Measured and Predicted Sediment Yield From A Subtropical, Heavy Rainfall, Steep-sided River Basin: Hanalei, Kaua‘i, Hawaiian Islands.” Geomorpholog y, 30(3)(1999): 213-226. Daly, C., R.P. Neilson, and D.L. Phillips. “A Statistical Topographic Model for Mapping Climatological Precipitation Over Mountainous Terrain.” Journal of Applied Meteorolog y, 33(2)(1994): 140-158. Dixon, J.A. and L.A. Fallon. “The Concept of Sustainability: Origin, Extensions and Usefulness for Policy.” Society and Natural Resources 2(1989): 73-84. Fares, A., A.K., Alva, S. Paramasivam, and P. Nkedi-Kizza. “Soil Moisture Monitoring Techniques for Optimizing Citrus Irrigation.” Trends in Soil Science. 2(1997): 153-180.

Watershed Management to Sustain Tropical Watersheds: An Integrated Hydrologic Modeling Approach

165

Fares, A., R.S. Mansell, and S.A. Bloom. “Hydrological/environmental Impacts of Tree Harvesting within Flatwood Pine Forest Upon Local Wetlands.” In Subsurface Hydrological Responses to Land Cover and Land Use Changes, M. Taniguchi (ed). Kluwer Academic Publishers Co, 1996. Giambelluca, T. W., M.A. Nullet, and T.A. Schroeder. “Rainfall Atlas of Hawai‘i. Report R-76.” Honolulu: Hawai‘i State Department of Land and Natural Resources, 1986. Giambelluca, T.W., J.K. Delay, M.A. Nullet, M.A. Scholl, and S.B. Gingerich. “Cloud Water Interception At Two Tropical Mountain Cloud Forest Sites in Hawai‘i.” American Geophysical Union Fall Meeting, San-Francisco, 2004. Grunwald, S. and L.D. Norton. “Calibration and Validation of A Non-point Source Pollution Model.” Agricultural Water Management 45(1)(2000):1739. Leon, L.F., D.C. Lam, C.McCrimmon, and D.A. Swayne. “Watershed Management Modelling in Malawi: Application and Technology Transfer.” Environmental Modelling & Software 18(6)(2003): 531-539. Mitchell, J.K., B.A. Engel, R. Srinivasan, and S.S.Y. Wang. “Validation of AGNPS for Small Watersheds Using An Integrated AGNPS/GIS System.” Water Resources Bulletin, 29(5)(1993):833-842. Nearing, M. A., L. Deer-Ascough, and L. M. Laflen. 1990. “Sensitivity analysis of the WEPP hillslope profile erosion model.” Transaction of the ASAE, 33(3): 839-849. NOAA. Main Eight Hawaiian Islands Land Cover. National Oceanic and Atmospheric Administration, Coastal Services Center, 2000. NOAA. Nonpoint Source Pollution and Erosion Comparison Tool (N-SPECT) Technical Guide. National Oceanic and Atmospheric Administration Coastal Services Center, 2004a. NRCS.Soil Survey Geographic Database. Natural Resources Conservation Service, United States Department of Agriculture, 2004. Orazio, C. et al. “Survey of Organic Chemical and Contaminants in Water, Sediment and Biota Sampled from the Hanalei River, Kaua‘i,” Report CERC-8335. United States Geological Survey, Honolulu, HI, 2001. Pekarova, P., A. Konicek, and P. Miklanek. “Testing of AGNPS Model Application in Slovak Microbasins.” Physics and Chemistry of the Earth Part B-Hydrolog y Oceans and Atmosphere 24(4)(1999):303-305. Perrone, J. and C.A. Madramootoo. “Use of AGNPS for Watershed Modeling in Quebec.” Transactions of the ASAE 40(5)(1997):1349-1354.

166

Ali Fares

Salama, R.B., D. Laslett, and P. Farrigton. “Predicting Modeling of Management Options for the Control of Dryland Salinity in the Wheatbelt of Western Australia.” Journal of Hydrol 143(1993):19-40. Suttles, J.B., G. Vellidis, D.D. Bosch, R. Lowrance, J.M. Sheridan, and E.L. Usery. “Watershed-scale Simulation of Sediment and Nutrient Loads in Georgia Coastal Plain Streams Using the Annualized AGNPS Model.” Transactions of the ASAE 46(5)(2003):1325-1335. USDA. “Urban Hydrology for Small Watersheds.” Technical Release 55, USDA-NRCS, Washington, DC, 1986. Williams, J.R. “Sediment Delivery Ratio Determined with Sediment and Runoff Models.” In Erosion and Solid Matter Transport in Inland Waters, pp168-179, IAHS-AISH pub No. 122, 1977. Yuan, Y.P., R.L. Bingner, and R.A. Rebich. “Evaluation of AnnaGNPS on Mississippi Delta MSEA Watersheds.” Transactions of the ASAE 44(5) (2001): 1183-1190.

Watershed Assessment, Restoration, and Protection in Hawai‘i

167

7 Watershed Assessment, Restoration, and Protection in Hawai‘i

Aly I. El-Kadi a and Monica Mira

Abstract This study evaluated efforts on watershed assessment, restoration, and preparation of protection plans as cornerstones of resource sustainability. The assessment presented a specific study in the Nāwiliwili Watershed, Kaua‘i, Hawai‘i, USA. The proposed restoration and protection plan covered the nine elements required by the U.S. Environmental Protection Agency for watershed-based plans for impaired waters. The elements can be grouped into five categories: identification of sources and level of contaminants, design of restoration and protection strategies including estimation of expected reduction of contaminant levels, information and educational programme development, assessment of the financial implications of the plan, and setting up of guidelines to manage the plan. The discussion also includes limitations of the plan and its suitability to address sustainability issues. Introduction According to the US Environmental Protection Agency (EPA), over 40 percent of assessed waters in the United States do not meet water quality standards set by the local jurisdictions, including states, territories, and authorised tribes. The number of water bodies affected exceeds 20,000 individual river segments, lakes, and estuaries. These impaired waters include approximately 300,000 miles of rivers and shorelines and approximately 5 M

168

Aly I. El-Kadi and Monica Mira

acres of lakes. An overwhelming majority of the population - 218 M - lives within 10 miles of the impaired waters. Sediments, excess nutrients, and harmful microorganisms result in quality impairment. Local governments are required to develop lists of these impaired waters and establish priority rankings for waters on the lists and develop total maximum daily loads (TMDLs) for these waters. A TMDL specifies the maximum amount of a pollutant that a water body can receive and still meet water quality standards, and allocates pollutant loadings among point and nonpoint pollutant sources. By law, EPA must approve or disapprove the lists and the TMDLs established by the local governments. If the submission is inadequate, EPA must establish the list or the TMDL. The objective of this chapter is to assess the suitability of the EPA watershed plan as a means of achieving watershed sustainability. The assessment uses the plan developed for the Nāwiliwili Watershed on Kaua‘i, Hawai‘i, USA. It includes: (1) validation and documentation of existing environmental data (Furness, et al. 2002), (2) identification of existing sources of pollution and contamination (El-Kadi et al. 2003), and (3) development of the restoration and protection plan (El-Kadi et al. 2004). The Gallows Run Watershed in Pennsylvania (Genesee/Finger Lakes Regional Planning Council 2007) and the Breton Bay Watershed in Maryland (The Centre for Watershed Protection 2007) have similar protection plans. The Nāwiliwili study is different as it deals with a tropical watershed that is also a tourist destination. Although typical water quality problems exist due to nutrients and bacteria, the nature of the watershed influences restoration and protection approaches. For example, the bacterial water-quality indicators developed for the US mainland are not suitable for tropical environments (Fujioka and Shizumura 1985). Watershed Sustainability In this chapter, the suitability of the EPA’s watershed restoration and protection plan developed for the Nāwiliwili Watershed by El-Kadi et al. (2004) is assessed as a vehicle for watershed sustainability. The main concerns include severity of problems in the watershed and the challenges involved. According to Bruntland (1987), sustainable development should meet the needs of the present without compromising the ability of future generations

Watershed Assessment, Restoration, and Protection in Hawai‘i

169

to meet their own needs. Although the Bruntland study is addressing global issues facing the world as a whole, many of the concepts and ideas are applicable for a much smaller unit – the watershed. The criteria developed by the Laird Norton Foundation (2007) served as basis in evaluating the watershed plan. By meeting these criteria, the plan will likewise satisfy the measures set by the Bruntland study. The criteria include the following conditions: 1. Community Engagement – involves local or regional engagement of different stakeholders in the planning, implementation, and evaluation of watershed management practices; 2. Ecosystem Function – puts emphasis on practices of ecological protection and restoration dependent upon complex and integrated biological, chemical and physical systems within a watershed; 3. Best Science – incorporates credible scientific assessment as a basis for determining and prioritising management practices; 4. Adaptive Management – uses long-term monitoring and evaluation to reset and adjust measurable restoration practices according to results; 5. Transferable Concepts - includes practices, which encourage the replication of successful strategies in other watershed locations. Finally, sustainability science is an essential element, as watershed research is cross-disciplinary and includes natural and social sciences, practices, and policies (Wilson 2005). EPA’s watershed restoration and protection plan As required by the U.S. Environmental Protection Agency (USEPA), watershed-based plans for impaired waters must include the nine elements listed below: 1. An identification of the causes and sources that need to be controlled to achieve contaminant load reductions; 2. A description of the nonpoint-source (NPS) management measures that need to be implemented to achieve load reductions and an identification of the critical areas where the measures need to be implemented;

170

Aly I. El-Kadi and Monica Mira

3. An estimate of the load reductions expected for the suggested management measures; 4. An estimate of the amounts of technical and financial assistance needed, associated costs, and the sources and authorities that will be relied upon to implement the plan; 5. An information/education component to enhance public understanding of the project, and encourage the public’s early and continued participation in selecting and designing the NPS management measures to be implemented; 6. A schedule that is reasonably expeditious to implement NPS management measures identified in the plan; 7. A description of interim measurable milestones to determine whether NPS management measures or other control actions are being implemented; 8. A set of criteria that can be used to determine whether loading reductions are being achieved over time and whether substantial progress is being made toward attaining water quality standards; and 9. A monitoring component to evaluate the effectiveness of the implementation efforts over time, as measured against the criteria established as described above. NĀwiliwili watershed assessment Details about the Nāwiliwili Watershed are provided in El-Kadi (2004). Three major streams form the watershed: Hule‘ia, Nāwiliwili and Pu‘ali, with Hule‘ia having the largest flow (Figure 7.1). In accordance with Section 303(d) of the Federal Clean Water Act, the Hawai‘i Department of Health (DOH) currently lists Nāwiliwili Bay, Nāwiliwili Stream, and Hule‘ia Stream as water bodies where water quality is impaired by excessive turbidity. Nāwiliwili Bay is also listed as a water body impaired by excessive nutrients, enterococci, and chlorophyll a. Many of the streams in the Nāwiliwili Watershed drain into Nāwiliwili Bay and can cause pollution. Residents and tourists use streams, harbor sites, and beaches for kayaking, swimming, and other recreational activities. In this regard, Kalapaki Beach is a primary swimming area in Nāwiliwili Bay.

Watershed Assessment, Restoration, and Protection in Hawai‘i

171

Figure 7.1 The Nāwiliwili Watershed, Kaua‘i, with its Main Perennial Streams and their Respective Basins (Hawai‘i State 2009)

To assess quality problems of the watershed, Furness et al. (2002) measured turbidity, salinity, temperature, nitrate and phosphate, and fecal indicator bacteria. Data collected for the year determined point- and nonpointsource contributions of nutrients and bacteria. The data determined whether the water within the watershed met water quality standards for the above parameters or if it represented a health hazard. The study also listed potential sources of pollution in the watershed. Results identified sediment, nutrient, and bacterial-contamination problems in the watershed and bay. Sediment sources included agricultural practices, construction sites, channel alteration, erosion, quarrying, and urban runoff. Nutrients originated from agricultural practices, golf courses, cesspools, forested areas, urban runoff, and wastewater treatment spills. Bacterial contamination originated from cesspools, forested areas, urban runoff, and wastewater treatment spills. Hydrological models assessed contributions of point and nonpoint sources of contamination in the watershed.

172

Aly I. El-Kadi and Monica Mira

The study concluded that Nāwiliwili Bay is subject to accidental sewage, chemical, fertiliser, and oil/gasoline spills. There is a chance, though, that other chemicals from various sources are also present. The study focused on nutrient and sediment contamination. El-Kadi et al. (2003) describe the details of the severity of problems related to nutrients, sediments, and bacteria. Most water samples obtained from stream sites greatly exceeded the current USEPA recreational water quality standards for fecal coliform (200 CFU/100 ml) and enterococci (33 CFU/100 ml). However, USEPA standards for bacterial indicators are not suitable for assessing Hawai‘i’s water quality. Strategies and actions for improving water quality in the NĀwiliwili Watershed Managing Water Flows and Quality Unabated urban storm water runoff discharged at rapid velocities and high volumes was identified as a priority concern in the Nāwiliwili Watershed. Best management practices (BMPs) are used to integrate storm water management practices. Provisions were made for habitat protection, water quality, and water quantity [e.g., American Society of Civil Engineers (2001)]. Sample BMPs include natural drainage swales, detention basins, ponds, and constructed wetlands. El-Kadi et al. (2004) suggested specific management measures for the Nāwiliwili Watershed. A future study can also be conducted to determine the feasibility, cost, and effectiveness of each BMP through site evaluation and series of demonstrations or test projects. Other management practices include better site-design practices (BSDPs), policy changes, and education and outreach programmes. BSDPs differ from BMPs in terms of their potential to prevent pollutants and runoff volumes from reaching waterways. BMPs only help treat runoff and reduce pollutant loads. BSDPs can protect a watershed by conserving its natural features and resources, using low-impact site design techniques, reducing impervious land cover, or utilising natural features for storm water management (Atlanta Regional Commission 2001). In Hawai‘i, a combination of a growing population and limited water resources is reducing the availability and quality of the drinking water supply.

Watershed Assessment, Restoration, and Protection in Hawai‘i

173

There are also environmental problems and financial costs that result from the disposal of wastewater. Increasing the safe use of recycled water can address all these problems. DOH has long been an advocate of water reuse as long as it does not compromise public health and valuable water resources (DOH 2002). An example of water reuse in the Nāwiliwili Watershed is Kaua‘i Lagoons, which uses up to 1.2 mgd of R-2 water from the Līhu‘e Waste Water Treatment Plant for golf course irrigation. Average reuse is about 1 mgd. There is a need to increase the use of recycled water. Such uses will have to conform to guidelines for the treatment and use of recycled water (DOH 2002). Restrictions include suitability of the type of recycled water for specific uses. Specific actions to encourage water recycling include the (1) promotion and support of current recycling efforts, (2) offering of incentives such as awards and tax breaks or other types of credits, (3) producing and distributing a pamphlet that offers water recycling projects and ideas, and (4) launching of an educational campaign to use gray water for landscape irrigation. Some of these ideas can be incorporated into educational opportunities for architects, plumbers, contractors, homebuilders, and do-it-yourself activities. Storm water management manuals were designed to reflect policies. However, effective policies require effective enforcement. Many of Kaua‘i’s policies, laws, and manuals are outdated. The Storm Water Runoff System Manual for the County of Kaua‘i is now available. There have been some positive revisions to the drainage manual, such as the requirement of detention basins for projects of a certain size. Unfortunately, many policies ignore new technologies that are currently available. In addition, amendments have not been made to address old problems, such as the contribution of pollutants in storm water runoff that is currently being discharged by existing infrastructure. The Storm Water Runoff System Manual is designed to maintain the current water quality of a watershed. However, the manual does not address situations when a watershed does not meet water quality standards, and is listed as one of the impaired waters. Kaua‘i’s current policies may not reflect the changes that are necessary to improve water quality, or make the waterways comply with the Clean Water Act. Volunteer work by the county and landowners proved to be inadequate in keeping water quality in the Nāwiliwili Watershed from exceeding the standards set by the state government. Residents of the Nāwiliwili Watershed

174

Aly I. El-Kadi and Monica Mira

recommended that efforts must be enhanced to find ways to help enforce existing policies. They also recommended that policies must be expanded or changed to include the implementation of more and better BMPs to maintain and improve water quality. It is prudent for the County of Kaua‘i to recognise and authorise watershed councils or neighborhood boards to participate in developing natural resource conservation plans. Finally, there is a need for more coordinated efforts between state and county agencies. In the past, the division of responsibility has led to serious problems (KRP Information Services 1993). Managing watersheds requires the development of a water budget, which accounts for all inflows, outflows, and storage changes within a system. Inflows and outflows may include water from tributaries, ditches, irrigation diversions, and other inputs. Water, which may be flowing out of a source watershed and into an entirely different basin, also needs to be estimated with reasonable accuracy. Irrigation systems and groundwater withdrawals change natural flow patterns. Developing water use plans and assessing water quality are difficult tasks, if not impossible ones, in the absence of a water budget. Like many other Hawai‘i watersheds, the Nāwiliwili Watershed has become a complex network of interconnected ditches, irrigations systems, diversions, flumes, and reservoirs. When the sugar industry was emerging, massive quantities of water were needed to irrigate the crop. In some regions, local water supply was limited, so counties brought water in from elsewhere. This required the building of a complex network of irrigation systems that still exists today. Although few sugarcane crops are being grown today, water is still flowing through these systems. There is a lack of knowledge regarding the condition of these systems and the volumes of water diverted into other watersheds. A water budget may be prepared by first consulting the Commission on Water Resource Management’s (CWRM) database. A field survey is needed to verify the information contained in the database. In addition, large landowners need to be consulted to determine if all diversion works have been accounted for. Preventing Soil Erosion and Sedimentation from Agricultural Lands All-terrain-vehicle (ATV) riding and eco-tours exacerbate erosion. These four-wheel vehicles dislodge sediment from roads. Intensive agriculture,

Watershed Assessment, Restoration, and Protection in Hawai‘i

175

such as sugarcane, causes periodic disturbances in the soil during harvesting and tilling. Meanwhile, ranching and permanent crop agriculture may reduce these soil disturbances and improve water quality. Soil from agricultural roads is subject to erosion and can end up in streams and waterways if not properly managed. Restoration efforts can include the promotion of videos produced by soil and water conservation districts, expansion of the use of conveyor belt water bars to prevent erosion, location of water troughs for cattle away from streams, development of a “working farm” to demonstrate BMP implementation, and implementation of solutions for ATV riding and eco-tourism erosion. Controlling Invasive and Non-Native Species Invasive plant and animal species pose a threat to Hawai‘i’s watersheds, water resources, tourism-based economy, agriculture, health, and quality of life. Habitat destruction and the introduction of alien species have been the predominant causes of biodiversity loss in Hawai‘i for over a century. More native species have been eliminated in Hawai‘i than anywhere else in the United States (Kaua‘i Invasive Species Committee 2003). Native species comprise only a small portion of the total species composition in the Nāwiliwili Watershed. Examples include the invasive red mangrove (Rhizophora mangle) that is actively spreading in the Nāwiliwili Watershed. Hule‘ia estuary and ‘Alekoko Fishpond have been inundated with this mangrove. The rock walls of the fishpond are being torn apart by mangrove roots, and the estuary itself appears to be shrinking in size as mangroves close in. The introduced mangrove appears to facilitate the establishment of opportunistic exotic species, such as the Samoan crab, while concurrently enhancing local species richness (Demopoulos 2003). Therefore, controlling the spread of mangrove species and removing some of their existing range are necessary steps to protect the native species (David Smith 2003). To control the spread of mangrove species, it may be useful to develop public interest for the use of mangrove, such as firewood, mulch, and as a source of building or crafting materials. However, care must be taken in designing such incentives, which can increase the spread of mangrove for profit. In the meantime, floating booms or other devices could be used to trap propagules and control further spreading.

176

Aly I. El-Kadi and Monica Mira

Feral ungulates, such as pigs, negatively affect water quality. In upper areas of the watershed, pigs destroy habitat and remove remaining native vegetation, thereby promoting the growth of aggressive species. Pig activities cause erosion and sedimentation when soil is routed near stream banks. Moreover, pigs create some health risks. A control plan for pigs may include more hunting in the mauka sections of the watershed. Fencing is another option for controlling pigs. A priority site for pig fencing is the Hule‘ia National Wildlife Refuge. Expected Load Reductions Due to Management Measures El-Kadi et al. (2003) used modeling in assessing the expected load reductions for nutrients and sediments in the watershed, based on suggested remediation strategies. The Generalised Watershed Loading Function (GWLF) model (Haith and Shoemaker 1987) was used to estimate load reductions when cesspools were replaced by septic systems or sewer systems . The reduction in yields amounted to 25 percent and 16 percent for dissolved and total nitrogen, respectively; and 92 percent and 62 percent for dissolved and total phosphorus, respectively. The approach used by Wenger (1999) designed a variable-width riparian buffer plan. Designed buffer widths ranged from 16 to 28 m. For buffers with a width greater than 15 m, the removal efficiency for total suspended solids, total nitrogen, and total phosphorus can be conservatively estimated at greater than 50 percent. Economic implications and plan management Cost Estimation Precise and reliable data on costs for recommended measures would require extensive engineering studies. However, extant literature reveals some general notions. Two reports are particularly useful – the Kailua Bay Advisory Council (KBAC) watershed management plan (Tetra Tech EM Inc. 2003) and the storm water pollution management practices report by the Texas Statewide Storm Water Quality Task Force (2007). El-Kadi et al. (2004) provide a cost estimate of the main items dealing with the proposed

Watershed Assessment, Restoration, and Protection in Hawai‘i

177

restoration activities. The authors emphasized that all estimates serve as guidelines and that more elaborate studies are needed to finalise such estimates. Uncertainties are related mostly to the absence of site-specific data on land price and size of installations. A good approach would be to subcontract construction activities to qualified non-profit and for-profit entities through bidding procedures. For example, the problem of cesspools contributing to pollution in the watershed can be alleviated by connecting residences to sewer systems or, where that alternative is prohibitively costly, by replacing cesspools with septic tanks. Outside urban areas, sewage systems are prohibitively expensive. They are also less necessary. Due to the low density in rural areas, efficient and effective operation of septic tanks is possible without causing residual contamination. Using available price quotes from the continental U.S., the total cost of septic tanks for the Nāwiliwili Bay Watershed would range from $235,000 to $517,000 for tanks, plus $940,000 to $2.82 M for installation. The average total cost would be about $2.3 M. Land owners/operators would be responsible for the annual maintenance cost. Considering Hawai‘i’s aboveaverage living expenses, the actual cost would most likely be higher than the averages given. Yet, the numbers provided here serve as guidelines for more accurate analysis. Sources of Funds Public funds such as the DOH’s Environmental Planning Office, Funding Sources for Communities – Watershed Focus, and many others can be tapped to support watershed plan implementation. In addition, the government can institute a wide variety of regulations, subsidies, and tax schemes. For example, North Carolina has a system of incentive and bonus payments for landowners (Wossink and Osmond 2002). With federal and state participation, the Conservation Reserve Enhancement Programme provides payments to owners of agricultural lands for up to 15 years, as well as subsidies for BMP installation. All installation costs may be covered if the contract is a permanent one. The draft Kailua Waterways Improvement Plan, Volume II (Tetra Tech EM Inc. 2003) contains a list of potential grant programmes and funding sources that are available to support the implementation programme.

178

Aly I. El-Kadi and Monica Mira

Potential funding is provided where available. Finally, cost-sharing and lowinterest loans are effective incentives to implement restoration activities. Examples include that of the Virginia agricultural incentives programme (Virginia State 2007). Restoration and Protection Plan Management The study proposed the establishment of the Nāwiliwili Watershed Restoration Office, with a structure similar to that proposed for the Kailua Protection Plan (Tetra Tech Em Inc. 2003). The office will manage and supervise all restoration activities, and integrate the efforts of watershed stakeholders concerned with the watershed’s health and welfare. The stakeholders include watershed residents; landowners; commercial, industrial, and agricultural businesses; community groups; schools and academic institutions; and county, state, and federal agencies. The efforts of all parties need to be synchronised toward implementing a holistic watershed management programme using the plan proposed here as a guiding framework. Developing and Implementing Education and Outreach Programmes The watershed education programme aims to increase awareness of watershed conditions and provide opportunities for community members to participate in resolving the problems. Simple concepts, like stewardship, can teach young children their responsibilities in protecting the watershed and the consequences of their actions within the watershed. Education programmes for children can help reach out to the adult population as well. Effective adult education and outreach programmes could be done through hands-on activities. Efforts could be made via local television programmes, eco-tours, or community-participation projects, such as beach cleanup days and storm drain stenciling. Some specific projects are outlined below: 1. The design of watershed curricula for schools or expansion and use of online curricula developed by Pat Cockett, a Kaua‘i high school teacher, could be implemented at some level in all schools in the Nāwiliwili Watershed;

Watershed Assessment, Restoration, and Protection in Hawai‘i

179

2. The implementation of workshops or production of videos to educate eco-tour guides and boat captains about general initiatives to monitor and educate visitors and locals could be supported with volunteer research tours or summer class/research expeditions at Kaua‘i Community College, and educational tour packages by eco-tour companies could be offered at reduced prices to local residents. 3. Educational plaques could be placed in strategic locations that are frequented by tourists (e.g. near the Nāwiliwili Stream estuary next to the Marriott Hotel). 4. Restoration of the ‘Alekoko fishpond can be initiated to function as an aquaculture demonstration, training, education, and research centre. Priorities and Implementation Schedule In implementing the plan, priority activities include: • • • • • •

Implementation of erosion control and BMPs, including those related to nutrient load reduction; Preparation of a water budget and setting of instream flows; Development and implementation of watershed curricula; Posting of educational plaques; Revision of the National Pollutant Discharge Elimination System files; and Collaboration between state and county agencies.

The timeline to implement the plan should integrate community inputs. Securing funds, forming an advisory committee, and hiring a coordinator are also major issues that need to be addressed. Assigning the position of coordinator to someone from a state organisation, such as DOH, should be explored to minimise cost, especially if overhead expenses are needed in setting up a new office. The tasks of the coordinator include measures to evaluate the plan’s success, participation in project reviews, and presentation of project results to the community and project advisors. Short-terms projects, of less than a year, should be initiated. These may include the establishment of buffer zones and replanting of riparian vegetation, creation of educational materials, production of videos and implementation of workshops, and initiation of educational programmes in schools.

180

Aly I. El-Kadi and Monica Mira

An ATV road stabilisation project would probably take two years to complete. Large-scale projects, such as constructed wetlands and those related to policy changes, would take about five years due to the need for multiagency cooperation. Other complicating factors include the need for extensive development plans and potential diversity of funding resources. A major, multiphase project related to the restoration of ‘Alekoko Fishpond may extend over 10 years. Eventually, this project may have to spin off and become an entity in itself. Measures for evaluating plan success Major project milestones, tasks, and deliverables should be assessed regularly. Measures of success must aim for improved surface water quality based on analytical results, improved habitat quality and eradication of invasive species, enhanced physical stabilisation of streams and other water bodies, and delisting of water bodies in the watershed under Section 303(d) of the Clean Water Act. Delisting must be done in cooperation with the state’s total maximum daily load (TMDL) programme. Other important measures of progress include the establishment of an advisory group, preparation of funding applications and securing funds, creation of the Nāwiliwili Watershed Restoration Office and hiring of staff, design and construction of demonstration projects and implementation of BMPs, creation and implementation of public outreach and education programmes, including a training programme, undertaking contracts with vendors, adherence to programme schedules and budgets, and publication of research findings. Plan evaluation Complementing the plan’s measures of success on water quality is a monitoring programme. The programme will regularly assess water quality through surface water monitoring and sampling of shallow groundwater (e.g., springs, seeps, and agricultural wells). BMPs could use influent and effluent from demonstration projects to assess efficiencies. Monitoring results should be compared to state water

Watershed Assessment, Restoration, and Protection in Hawai‘i

181

quality criteria and TMDLs, where applicable, to track the progress on meeting water quality requirements and designated uses in the watershed. A new round of sampling should be conducted prior to implementing any demonstration project or BMP, in order to provide baseline information for future assessment. Periodic sampling should be done throughout the duration of the implementation programme. The proposed implementation programme should be a living document that can be revised and refined as the programme matures. Periodic updates and revisions are anticipated in response to levels of funding, agency and public participation, future conditions and developments, and lessons learned. The measures of success discussed in the previous section should be incorporated into the plan to present an accurate picture of current conditions. In addition, changes in supporting conditions–such as infrastructure upgrades, development patterns and infrastructure capacity, land-use patterns, water quality regulations, and critical habitat areas– should be incorporated into the plan to support valid and effective recommendations for future work. Monitoring plan Monitoring assesses existing conditions and tracks the progress of various restoration and protection strategies. The design of specific remediation activities might require the installation of new monitoring equipment and collection of data from other sources, including US Geological Survey, DOH, Natural Resources Conservation Service, US Environmental Protection Agency, County of Kaua‘i, and community groups. Professionals and volunteers should carry out monitoring. Using volunteers will help stretch scarce funding while giving the public a real stake in managing the programme and the watershed. Data collection and monitoring should be conducted under a rigorous quality assurance programme to ensure quality and useable data to facilitate implementation and evaluation of the restoration programme. Data collection, sampling, and monitoring should be conducted according to a sampling and analysis plan. A quality-assurance project plan should be developed. Data should be compiled into a project water-quality

182

Aly I. El-Kadi and Monica Mira

database maintained by staff of the Nāwiliwili Watershed Restoration Office. The database can be accessed online to enable participating agencies and researchers to use the data, and at the same time inform the public about the implementation programme and the water quality in the watershed. Watershed plan and sustainability Criteria based on Bruntland (1987) and The Laird Norton Foundation (2007) were adopted to examine the Nāwiliwili restoration and protection plan. Elements of the EPA’s watershed plan appear to meet such criteria. Several limitations hinder development and application of the plan. Discussions on how to address these limitations are presented below. Watershed Modeling Limitations The results from using models are limited in terms of the equations used and assumptions made that limit accuracy, and the errors that arise from the input data should be controlled by the modeler. The GWLF model used to design the watershed plan utilises the Soil Conservation Service Curve Number (CN) technique for runoff estimation, a modified Universal Soil Loss Equation, along with a delivery ratio and transport capacity equations, to determine sediment yield, and a very simplified lumped parameter nutrient-simulation model. Accuracy is influenced by the inherent limitations of the equations. For instance, the CN technique does not consider rainfall intensity of individual storms. In addition, the model operates on a daily time-step. Thus, multiple storms, occurring on a single day, are lumped into one daily storm. Storm duration within a day is not considered. The initial abstraction or surface retention value for each storm is calculated based on the CN. The values can differ from the actual initial abstraction for each storm. The model estimates evapotranspiration based on general cover coefficients provided in the input data and based on the daily temperature value. The model does not consider the soil layer’s physical properties when calculating infiltration and soil moisture retention. A more physically based model that considers soil layer properties would provide better estimates of soil moisture accounting, but with additional input data requirements.

Watershed Assessment, Restoration, and Protection in Hawai‘i

183

The model’s sediment yield results are dependent on the accuracy of streamflow prediction. The model’s nutrient yield is, in turn, dependent on sediment yield and streamflow prediction. Thus, streamflow estimates have the highest accuracy, followed by sediment yield estimates, and then by nutrient yield estimates. The Universal Soil Loss Equation ( USLE), which is used in sediment yield assessment, requires an average invariable input for various factors involved, while variability is dominant for some. Nutrient values are based on unit area loading and an enrichment ratio estimate. The lumped characteristics of nutrient loading may incorporate errors in load estimates. To compare model results or verify the simulation results’ accuracy, observed output for streamflow, sediment yield, and nutrient loads in the stream outlet over a sufficiently long period should be observed. These data were not found in sufficient amounts for the Nāwiliwili Watershed for calibration or comparison. Input data quality affects modeling results. The primary dataset driving the GWLF simulation is the weather dataset. Accurate precipitation input is necessary to simulate values for streamflow. Rainfall is highly variable throughout the watershed. For the North Wailuā River Watershed, rainfall ranged from 1,000 cm/yr near Mt. Wai‘ale‘ale to 150 cm/yr near the outlet of the watershed. High rainfall variability is difficult to incorporate into a model that uses one average rainfall amount for the entire watershed. Moreover, there are limited number of rainfall stations that have continuous long-term weather data within a watershed. The weather station points may not accurately capture rainfall variability over the watershed on a given day thereby incorporating errors in the streamflow estimate. The weather station has missing data on some days. This creates additional uncertainties and errors in the input precipitation data. A better spatially variable estimate of rainfall input may be used to improve model accuracy. An example input could be based on the calibrated Next-Generation Radar data, also known as NEXRAD data, available at a network of 159 highresolution Doppler weather radars operated by the US National Weather Service. This study used a 10 m x 10 m horizontal resolution DEM. Given the generally steep terrain in Hawai‘i, the model should provide sufficient accuracy to calculate the input parameters for the GWLF model. Parameters of the USLE include the rainfall and runoff factor R, the soil erodibility factor K, the slope length-gradient factor LS, the crop/vegetation and management factor C, and the support practice factor P. Among these, the

184

Aly I. El-Kadi and Monica Mira

LS factors calculated by GIS, along with watershed delineations, should be one of the more accurate input parameter estimates. As the GWLF model only requires the K factor value and the hydrologic soil groups estimate, the 1:24,000-scale Soil Survey Geographic (SSURGO) Database map should provide a sufficiently detailed estimate of the two parameters. Providing a more detailed soil and DEM map will not drastically improve the results of the model. Land-use coverage used in the study was NOAA’s 1:250,000-scale landsat classified remote sensing imagery. The data has a relatively coarse resolution of 30 m x 30 m grid cell size. The land-use data affect the estimates of CNs and C factors. A better resolution land-use map can improve the estimates of the input parameters. If the effect of invasive tree species in the Nāwiliwili Watershed needs to be studied, a detailed map of the different vegetation in the forests will be required. If a detailed map of tree cover is available, then the C factors can be independently calculated using the RUSLE 2.0 model and input into the GWLF model. In addition, there was lack of information on agricultural activity within the Nāwiliwili Watershed. Improving the landuse and management datasets could result in better estimates of sediment yield. The input datasets for nutrient loads are rough estimates obtained from default values in the model’s documentation and from other GWLF studies. The dissolved loads and nutrient buildup loads in urban areas of Nāwiliwili need to be estimated to get an accurate description of the actual watershed condition. Erosion and Sedimentation Studies have shown that sediment supply from mass wasting, such as landslides and debris flows, can dominate the storage and routing of sediment in some streams (Dietrich and Dunne 1978; Caine and Swanson 1989; Benda 1990). Erosion in many watersheds in Hawai‘i is dominated by mass wasting because of the steep slopes. Slides and slope failures are common after heavy rainfall events. Usually affected are the edges of alluvial terraces. This is due to a combination of saturation, gravitational forces, and direct erosion by stream action along the toe of the stream bank. Slope failures occur after heavy rainfall and are most prevalent in convex and very steep slopes near the heads of drainages

Watershed Assessment, Restoration, and Protection in Hawai‘i

185

and along ridges. Slope failures may be caused by the development of excess pore pressure forces in soils overlying less permeable bedrock. Disturbed areas may be a continuing source of sediments, particularly fine-grained silts and clays that are easily suspended and transported by streams. Models must be developed suitable to the specific nature of Hawai‘i watersheds. One approach may include the application of a deterministic slope stability predicting tool to map the potential landslide hazard and vulnerability conditions of the watershed (Deb and El-Kadi 2009). Another approach may involve modelling the interaction of landslides, debris flows/ sediment transport, and the channel network to delineate/predict the hydrogeological risk situations in the watershed areas. Bacterial Contamination As discussed earlier, most water samples obtained from watershed stream sites greatly exceeded the current USEPA recreational water quality standards for fecal coliform and enterococci. However, it appeared that sources of fecal bacteria in Kaua‘i are environmental in nature and not necessarily from sewage sources. Thus, it might not be practical to use the USEPA water quality standards in reducing levels of bacteria. Moreover, concentrations of fecal indicator bacteria are not related to health risks from sewage contamination (Hardina and Fujioka 1991). DOH is using C. perfringens water-quality standards to determine when waters are contaminated with sewage. Until completely reliable indicators are identified, it seems sanitary surveys and bacterial source tracking may provide the only definitive answers to assess sewage contamination. Sanitary surveys, however, cannot detect underground leakage from a sewage source. Economic Issues The economic impacts of the watershed plan include potential expenditures by the agricultural, recreation and tourism, and household sectors. These impacts are also felt throughout Kaua‘i and the entire state. Identifying these impacts and estimating their links with other sectors of the economy will require considerably more resources than are available in the project. Difficulties in economic assessments include problems in estimating non-market water uses. The value inherent in non-financial considerations, such as aesthetic enjoyment, is difficult to estimate.

186

Aly I. El-Kadi and Monica Mira

Water quality must also consider the cost of health care associated with illnesses traceable to swimming and other water-based activities. Unfortunately, epidemiological studies have not been implemented for Nāwiliwili on the health effects of existing contamination or the associated costs of medical care. Programmes designed to maintain water quality are effective only if derived benefits exceed costs. People might need to pay more for water quality, otherwise, the resources used to achieve this might be better used elsewhere. Measuring benefits against cost is also difficult. Laws and regulations may or may not reflect the values of all those who underwrite the costs or those who receive the benefits of enhanced water quality. Nevertheless, laws and regulations define requirements that guide project implementers. Finally, any changes made to regulations to better achieving water quality would have economic effects. These effects may be studied through input– output models (HDBEDT 1997) or more sophisticated general equilibrium methods. This might be difficult to do in the case of Nāwiliwili study, as requisite data are not yet available. BMP Limitations The study estimated riparian buffer sizes based on the approach of Wenger (1999). However, riparian buffers might not be effective in land slopes greater than 25 percent. The majority of areas in Nāwiliwili Watershed have slopes greater than 25 percent. For such areas, it might be best to extend the scope of the buffers beyond the high-slopes zones to cover less steep areas. This design might not be suitable so alternative buffering measures are needed. Moreover, accessibility to some areas may not even be possible to implement these measures. Priorities In general, plan implementation activities deal with nutrient and sediment load reduction, water resource assessment, and education. These three groups of activities overlap. For example, the preparation of the water budget, setting of instream flows, and reduction of sediment and nutrient loads are critical in restoring stream aquatic health. The effective design of BMPs greatly depends on having accurate knowledge of the water budget of

Watershed Assessment, Restoration, and Protection in Hawai‘i

187

the watershed, including instream flows. Education is also the cornerstone of the protection plan, which greatly depends on community and visitors’ good will and participation. It was difficult to establish definitive priorities for the restoration activities due to minimal participation of people in community meetings. The preparation of a water budget and the setting of instream flows were identified as priority concerns. These activities may create the need to reevaluate prior studies as they may have overlooked the uncertainties of instream flows and the watershed budget. However, community participants agreed that all restoration activities identified were important and they may need to be prioritized according to cost and feasibility. Government agencies should lead community restoration activities. CONCLUSIONS The EPA’s watershed plan is suitable for the sustainable management of the watershed. The plan seems to meet expectations in terms of setting ecological functions, achieving flexibility and the adaptive nature of the watershed, and securing active community participation. However, data and scientific limitations hinder success towards a holistic and credible scientific approach, and the possibility of transferring approaches to other watersheds. Data were either missing or not detailed enough to support modeling. Scientific approaches that deal with sediment and erosion due to mass wasting were insufficient. Studies indicated the need for suitable bacteria indicators other than those defined by EPA for watershed assessment. Estimates of plan costs are also uncertain due to the lack of data and the absence of practical approaches to assess the values of such items as the value of aesthetic enjoyment and health care costs associated with illnesses, which could be traced to swimming and other water-based activities. There is need to develop an appropriate methodology for estimating riparian buffer sizes that can be used for the areas with slopes greater than 25 percent as such areas dominate the Nāwiliwili Watershed.

188

Aly I. El-Kadi and Monica Mira

Acknowledgments This chapter covers results of the project titled “Assessment and Protection Plan for the Nāwiliwili Watershed.” Principal investigator Aly I. El-Kadi, and co-principal investigators Roger Babcock, Roger S. Fujioka, Clark C.K. Liu, Jacquelin N. Miller, James E.T. Moncur, and Philip Moravcik are all with the Water Resources Research Centre, University of Hawai‘i. The project investigators acknowledge the support of the US Environmental Protection Agency, under Section 319(h) of the Clean Water Act, and the Hawai‘i State Department of Health, Clean Water Branch, and the community members who have shared their knowledge and expertise. NOTES a Water Resources Research Centre, University of Hawai‘i, USA. Also with the Department of Geology and Geophysics, e-mail: elkadi@ hawaii.edu References American Society of Civil Engineers (ASCE). “A Guide for Best Management Practice (BMP) Selection in Urban Developed Areas.” ASCE Publications, Reston, Virginia, 2001. Atlanta Regional Commission and Georgia Department of Natural Resources Environmental Protection Division, Georgia Stormwater Management Manual, 2001. Benda, L.E. “The Influence of Debris Flows on Channels and Valley Floors in the Oregon Coast Range, U.S.A.” Earth Surf. Processes Landforms 15(1990):457– 466. Bruntland, G. (ed). Our Common Future: The World Commission on Environment and Development. Oxford University Press, 1987. Caine, N. and F.J. Swanson. “Geomorphic Coupling of Hill Slope and Channel Systems in Two Small Mountain Basins.” Z. Geomorphol. 33(1989): 189– 203.

Watershed Assessment, Restoration, and Protection in Hawai‘i

189

Center for Environmental Research and Service, Troy State University. Considerations for Storm Water and Urban Watershed Management, Developing a Programme for Complying with Stormwater, Phase II MS4 Permit Requirements and Beyond. Department of Biological and Environmental Sciences, Troy State University, Troy, Alabama, 2001. City of Bellevue. Water Quality Protection for Bellevue Businesses. City of Bellevue Utility Department, Bellevue, Washington, 1993. Deb, S.K., and A.I. El-Kadi. “Susceptibility Assessment of Shallow Landslides on O‘ahu, Hawai‘i, Under Extreme-rainfall Events.” Journal of Geomorpholog y 108(2009): 219–233. Demopoulos, A.W.J. “Mangrove Forest Ecosystems.” Department of Oceanography, University of Hawai‘i at Mānoa, Honolulu, Hawai‘i, 2003. Dietrich, W.E., and T. Dunne. “Sediment Budget for A Small Catchment in Mountainous Terrain.” Z. Geomorphol., 29(1978):191-206. El-Kadi, A.I., R.S. Fujioka, C.C.K. Liu, K. Yoshida, G. Vithanage, Y. Pan, and J. Farmer. Assessment and Protection Plan for the Nāwiliwili Watershed: Phase 2 – Assessment of Contaminant Levels, WRRC-2003-02. Water Resources Research Center, University of Hawai‘i, Honolulu, 2003. El-Kadi, A.I., M. Mira, S. Dhal, and J.E.T. Moncur. Assessment and Protection for the Nāwiliwili Watershed: Phase 3 – Restoration and Protection Plan, WRRC2004-05. Water Resources Research Center, University of Hawai‘i, Honolulu. EnvironEDGe Technologies, Inc. http://www.environedge.com/ Fujioka, R.S. and L.K. Shizumura. “Clostridium perfringens, A Reliable Indicator of Stream Water Quality.” J. Water Pollut. Control Fed., 57(1985):986–992pp. Furness, M., A.I. El-Kadi, R.S. Fujioka, and P.S. Moravcik. Assessment and Protection Plan for the Nāwiliwili Watershed: Phase 1 – Validation and Documentation of Existing Environmental Data, WRRC-2002-02. Water Resources Research Center, University of Hawai‘i, Honolulu. Genesee/Finger Lakes Regional Planning Council. www.grwabucks.org/report. htm Haith, D.A. and L.L. Shoemaker. “Generalized Watershed Loading Functions for Stream Flow Nutrients.” Water Resources Bulletin 23(1987):471– 478pp.

190

Aly I. El-Kadi and Monica Mira

Hardina, C.M. and R.S. Fujioka. “Soil: The Environmental Source of E. coli and Enterococci in Hawai‘i‘s Streams.” Environ. Toxicol. Water Qual., 6:185–195pp. Haub, A. and L. Hoenig. “Aquatic Habitat Evaluation & Management Report.” City of Olympia, Olympia, Washington, 1999. Hawai‘i Pacific Engineers, Inc. Draft Supplemental Environmental Impact Statement: Waimanalo Wastewater Facilities Plan, Koolaupoko, O‘ahu, Hawai‘i, 1998. Hawai‘i State, Department of Business, Economic Development and Tourism (HDBEDT). “The Hawai‘i Input-Output Study: 1997 Benchmark Report.” Honolulu, Hawai‘i, 2002. Hawai‘i State. http://www.state.hi.us/dbedt/gis/. 2009. Hawai‘i State, Department of Health (DOH). Guidelines for the Treatment and Use of Recycled Water. Wastewater Branch, Honolulu, Hawai‘i, 2002. HDBEDT, www.hawaii.gov/dbedt/gis/ Kaneshiro, K.Y., P. Chinn, K.N. Duin, A.P. Hood, K. Maly, and B.A. Wilcox. “Hawai‘i Mountain-to-Sea Ecosystems: Social-ecological Microcosms for Sustainability Science and Eractice.” EcoHealth J. 2:349-360pp. Kaua‘i Invasive Species Committee (KISC), Action Plan, Kilauea, Hawai‘i, 2003. KRP Information Services. Water Quality Management Plan for the County of Kaua‘i, Honolulu, Hawai‘i, 1993. Laird Norton Foundation. http://www.lairdnorton.org/ May, C.W., E.B. Welch, R.R. Horner, J.R. Karr, and B.W. Mar. “Quality Indices for Urbanization Effects in Puget Sound Lowland Streams, Water Resources Series Technical Report No. 154, University of Washington.” Ecolog y Publication No. 98-04, 1997. Moncur, J.E.T. The Value of Recreation Areas on O‘ahu. University of Hawai‘i Center for Governmental Development, Honolulu, Hawai‘i, 1972. Moncur, J.E.T. “Estimating the Value of Alternative Outdoor Recreation Facilities Within A Small Area.” J. Leisure Res., 7(4)(1975):301–311pp. National Watershed Manual, National Soil Survey Handbook. www.nrcs.usda. gov/technical/references NRCS. http://www.nrcs.usda.gov/technical/land/meta/m5058.html Ocean Arks International. http://www.oceanarks.org/wetlands/ Russell, C.S. Applying Economics to the Environment. Oxford University Press. 332–337pp, 2001.

Watershed Assessment, Restoration, and Protection in Hawai‘i

191

Tetra Tech EM Inc. Draft Kailua Waterways Improvement Plan: A Watershed Approach for Improving Water Quality in the Kailua Waterways System, 2003. Texas Statewide Storm Water Quality Task Force. http://www.txnpsbook.org/ About.htm The Center for Watershed Protection www.dnr.state.md.us/watersheds/surf/proj/ wras.html U.S. Environmental Protection Agency (USEPA). Storm Water Technology Fact Sheet Storm Water Wetlands, EPA 832-F-99-025, Washington, D.C, 2002. USEPA. http://www.epa.gov/owow/nps/ordinance USFDA. http://vm.cfsan.fda.gov/~mow/chap11.htm Virginia. http://vmirl.vmi.edu/ev2000/PPT/bayless.ppt Water Tanks. http://www.watertanks.com/category/35/ Wenger, S. “A Review of the Scientific Literature on Riparian Buffer Width, Extent and Vegetation.” Office of Public Service and Outreach, Institute of Ecology, Georgia, 1999. Wilson, E.O. Editorial on Sustainability e-Journal.  Sustainability: Science, Practice, & Policy 1(1)(2005):1-2. http://ejournal.nbii.org/archives/vol1iss1/ editorial.wilson.html. (published online March 31, 2005). Wossink, G.A.A. and D.L. Osmond. “Cost Analysis of Mandated Agricultural Best Management Practices to Control Nitrogen Losses in the Neuse River Basin, North Carolina." J. Soil Water Conserv, 57(2002):213–220pp.

192

Aly I. El-Kadi and Monica Mira

Landscape-Scale Conservation: Fostering Partnerships through Ecosystem Service Approaches

Theme 3 Participatory Approaches

193

194

Rebecca L. Goldman and Gretchen C. Daily

Landscape-Scale Conservation: Fostering Partnerships through Ecosystem Service Approaches

195

8 Landscape-Scale Conservation: Fostering Partnerships through Ecosystem Service Approaches



Rebecca L. Goldman a and Gretchen C. Daily b

Abstract The growing size and resource demands of the human population make it more important than ever to broaden options for integrating conservation into resource decisions and practices. Fostering collaboration through partnerships can engage resource users under a common framework. Focusing on ecosystem services can diversify the types of organisations and resource users who invest in, are supportive of, and participate in conservation efforts. Exploring partnerships emerging from these ecosystem service conservation projects can help inform future efforts for successful cooperation. Using two case studies – Sierra de las Minas in Guatemala and the Piracicaba, Capivari, and Jundiai (PCJ) watershed in Brazil – the formation, approaches, and strategies used are explored to contribute to the success of future partnerships. In both cases, the partnerships were initiated through third party leadership, governed by a committee including all stakeholders, and used direct payments for ecosystem services to encourage conservation efforts.

196

Rebecca L. Goldman and Gretchen C. Daily

Introduction Sustainability science refers to research that examines the mechanisms for and challenges of achieving sustainable development – reconciling human needs with environmental limits over the long term (Clark and Dickson 2003). The global pattern of expansion is such that sustainability science is critical in maintaining some level of human and environmental prosperity. As population grows, natural resources are consumed to maintain well-being. Resource use often entails land use and land conversion. Human population is expected to reach about 10 B within the next 50 years (United Nations 2007). This expansion comes with demands for natural resources (for food, shelter, fiber, fuel, etc.) (Foley et al. 2005). The use of such resources is coupled with conversion and/or intensive land use. Deforestation for agriculture is one example – one estimate holds that the global expansion of croplands since 1850 has converted some 6 M km2 of forests/woodlands (Ramankutty and Foley 1999). Food production via agriculture is arguably the most important activity in the world today. Yet agriculture is the most important proximate cause of biodiversity loss and ecosystem service degradation worldwide, both directly via land conversion and indirectly through the use of fertilisers and agricultural technologies (Matson et al. 1997; Tilman et al. 2001; Foley et al. 2005). Forty percent of the earth’s land surface has already been converted to agriculture (Asner et al. 2004; Foley et al. 2005). If humans continue to depend on agricultural products as they have done in the past, scholars estimate that, by 2050, as much as a billion hectares of natural ecosystems may be converted to agriculture with significant increases in nitrogen and phosphorus inputs into waterways (Tilman et al. 2001). As human needs for agricultural products and other human impacts continue to grow, setting aside lands in natural reserves, i.e. a protectionist strategy that prevents local communities from accessing resources, becomes untenable (Daily 2001; Heal et al. 2001; Polasky et al. 2005). In addition, reserves are too few, too small, too isolated, and often subject to change. This makes areas large enough to uphold ecosystem function difficult to maintain (Pickett and Thompson 1978; Newmark 1987; Daily and Ellison 2002; Rosenzweig 2003a, 2003b; Sanchez-Azofeifa et al. 2003). Thus, developing and deploying mechanisms to promote economically viable conservation on working landscapes are critical to make conservation commonplace and

Landscape-Scale Conservation: Fostering Partnerships through Ecosystem Service Approaches

197

effective. Land conversion and land use can degrade the ability of natural systems to conserve biodiversity and produce ecosystem services (benefits from ecosystems) such as clean water, clean air, soil fertility, and pollination (Daily 1997; Matson et al. 1997; Pimm and Raven 2000; Bennett et al. 20001; MA 2005; Brosi et al. 2006). Thus, the conservation of both working and unbuilt landscapes is necessary and interdependent (Kremen et al. 2002). As such, there is much scope for reconciling conflicting demands by creating economic and other incentives for integrating conservation practices into human-dominated landscapes. To realise the potential of this strategy, incentives would need to target the broader landscape, rather than individual parcels since resource provision rarely follows property and/or national boundaries (Goldman et al. 2007a). This is particularly true for watershed conservation. Many watersheds face the confluence of various global concerns – diverse groups of stakeholders, human-modified land parcels for production, and declining biodiversity. Human production within a watershed can also harm an ecosystem’s ability to provide water services – quality, quantity, and timing of flow – for flood mitigation and in some cases hydropower. Further, watersheds can span vast areas, which can lead to multiple land users in the landscapes all with different interests, concerns, and land uses. Influencing land use to enhance the provision of watershed services can enhance economic and human health conditions, particularly in developing countries where over a billion people lack access to clean drinking water (WHO 2002). This human-environment interface requires conservation strategies that correlate environmental protection with the protection or enhancement of human well-being. What organisational approaches encourage strategies that can foster diverse partnerships across a variety of landscapes and land users in conservation? Watershed conservation involves landscape-scale, cooperative, cross-boundary conservation efforts that engage multiple stakeholders in partnerships to encourage land use change. Challenges facing such conservation efforts include stakeholder participation, payment for conservation measures, and financial and institutional mechanisms to encourage watershed conservation. This chapter illustrates how field research and case studies inform the use of ecosystem service approaches to create broad collaboration. Specifically, two case studies are explored in depth – the creation of a water fund in Sierra de las Minas, Guatemala and the payment for environmental services

198

Rebecca L. Goldman and Gretchen C. Daily

(PES) scheme in Piracicaba, Capivari, and the Jundiai (PCJ) watershed in the Atlantic forest of Brazil. The two case studies answer crucial questions. How did these partnerships form? How are the partnerships maintained? What institutional structure is used? What strategies have been adopted in these areas to link science and policy and promote changes in land use within a watershed? The cases demonstrate entrepreneurship and the importance of strong leadership and ingenuity for conservation success. This chapter is divided into three parts. Part I briefly introduces the ecosystem service approach, the incentives to encourage ecosystem service provision, and the benefits conferred by this approach. Part II describes the two case study sites: the geography, landscapes, and stakeholders. Part III uses these case studies to explore the lessons learned from the approach and the strategies used to promote watershed conservation. Ecosystem Services Agricultural landscapes are rapidly dominating human land uses, and are therefore critical in conservation efforts worldwide. Agricultural landscapes can and do provide a number of ecosystem services. By encouraging particular management techniques, agricultural landscapes can continue to provide these services or enhance these services into the future (Goldman et al. 2007b). Changing land use and conservation management entail costs, and developing financial incentives for landowners who are willing to engage in such practices can help offset these costs. First, broad types of incentives are introduced that could encourage ecosystem service flow from agricultural landscapes. Second, focus must be made on ecosystem service approaches which have been used on the ground through conservation projects. Ecosystem Service Provision – Theoretical Incentives There is a relatively extensive literature assessing the likelihood of cooperation and coordination in governing the use of common pool resources, including analysing the benefits of face-to-face communication and the relevance of landowner homogeneity (e.g. Ostrom 1990, Ostrom et al. 1992, Ostrom 1997; Parkhurst et al. 2002). These examples, among others, detail the complexity surrounding voluntary, cooperative behaviour. The provision

Landscape-Scale Conservation: Fostering Partnerships through Ecosystem Service Approaches

199

of ecosystem services requires landscape-scale consideration (Goldman et al. 2007a), which unless a single landowner owns a large area of land, leads to the need for cooperation. Such cooperation can be encouraged or induced through the design of appropriate incentives that, by virtue of their structure, help reduce the open-access problem (Ostrom 1990) by attaching a value to resource conservation. A variety of variables can govern the type of incentives which might work best in a particular landscape, such as farm size, social context, landowner characteristics and values, crop type, and desired conservation outcome, among others (e.g. Ostrom 1990; Olson 1971). Several theoretical incentives are introduced to encourage cooperative behaviour, and examples are given of how the aforementioned variables might determine which incentive is likely to be the most effective. These incentives might not all induce conservation behaviour but may present simple ways to illustrate a suitable framework. Three theoretical incentives that could encourage cross-boundary efforts include cooperation bonus, the competitive design/entrepreneurial model, and the ecosystem service district (Goldman et al. 2007a). The cooperation bonus is similar to the agglomeration bonus, as suggested by Parkhurst and Shogren (2003). Incentives are structured to reward contiguous conservation acreage, and cross-boundary conservation is rewarded. For example, if one landowner decides to plant trees on an acre of his land, an agglomeration or cooperative incentive would give him a bonus if he further plants trees on the acre next to the acre that already has trees. It would thus be to the landowner’s advantage to convince his neighbour to plant trees on his land. Both landowners would receive the bonus although each only has to take one acre out of production (Parkhurst and Shogren 2003; Goldman et al. 2007a). The competitive design/entrepreneur incentive creates the potential for a trusted entrepreneur or third party to encourage conservation behaviour. In this scenario, an incentive designed to reward the “best” landscape for service provision, such as water purification, might be appropriate. Landowners would have to rally as a group, which could be mediated by a third party or a trusted landowner in the area, and design a landscape to submit in a competitive reward process (Goldman et al. 2007a). Finally, the ecosystem service district incentive is a more legal framework to encourage cooperation. Here, an incentive would be rewarded to ecosystem

200

Rebecca L. Goldman and Gretchen C. Daily

service districts – a legal agreement between landowners who agree to engage in conservation behaviour to promote a particular service or services by a majority vote. Once the agreement is formed, each landowner in the district is legally bound to participate, thereby solving any hold-out problems and enforcing landscape-scale, cooperative conservation behaviour (Goldman et al. 2007a). Other group decision-making rules could also be explored, e.g. allocating the number of votes in accordance with farm size, as discussed in other common-property contexts (Olson 1971; Olson 1993). As indicated, Ostrom (1990) introduces a number of complexities in governing common pool resources and assesses a number of possibilities to encourage cooperative behaviour (Ostrom 1998). These concerns are simplified by analysing the effectiveness of these incentives with respect to landowner characteristics – in particular their social familiarity and value systems. This is only one variable that might influence the type of incentive to use and at the same time provides an example of how to think through these issues. Both social familiarity and heterogeneity of value systems and management beliefs could affect the incentive to apply (Figure 8.1). In areas where there are multiple landowners who value conservation and trust each other, the cooperation bonus incentive can be effective (Goldman et al. 2007a) since landowners can communicate and encourage one another to participate (Olson 1998; Shogren et al. 2002) with relative ease. When landowners are less familiar with each other and their management ideals have greater heterogeneity, there is a role for a trusted third party to group relatively similar landowners as would be promoted in the competitive design incentive (Figure 8.1). Finally, landowners who are unfamiliar with each other and/or mistrustful of one another and who maintain relatively heterogeneous management styles and value systems, a more legal structure and relationship might be best to ensure that each landowner can engage in elected practices.

Landscape-Scale Conservation: Fostering Partnerships through Ecosystem Service Approaches

201

Figure 8.1 Three Possible Incentives or Partnerships that Could Encourage the Provision of Ecosystem Services from Working Landscapes, Depending on the Size Parcels in the Landscape and the Familiarity of the Landowners with Each Other

Social Context

Unfamiliar

Ecosystem service district Entrepreneur/ Group design

Familiar

Cooperative Bonus Homogenous

Heterogeneous

Landowner Values

This figure shows three possible incentives or partnerships that could encourage the provision of ecosystem services from working landscapes. The type of incentive depends on the size parcels in the landscape and the familiarity of the landowners with each other. Ecosystem Service Approaches in Practice The incentives discussed above are theoretical and are the types of incentives that could be integrated into US conservation policy, such as the farm bill, but could also be used by a non-government conservation organisation. Recently, there has been a rise in conservation projects that focus on ecosystem service objectives thereby opening the potential to institute incentives, like those outlined above, to encourage water services in watersheds. In a case study of projects at The Nature Conservancy (TNC), when compared to more traditional biodiversity approaches (projects focusing on only species and habitat conservation with no service goals), ecosystem service approaches expand conservation opportunities by targeting converted agricultural landscapes, working with private land users, and expanding financing opportunities for conservation (Goldman et al 2008). This confers particular advantages for watershed conservation. Lands within watersheds

202

Rebecca L. Goldman and Gretchen C. Daily

often include a diversity of stakeholders with varying interests, water quality and quantity and are often severely threatened by agriculture or other land uses, and flood mitigation services are diluted through land-use changes (e.g. FIFMTF 1992; Postel et al. 1996; Vitousek et al. 1997; Chan et al. 2006). Thus, there has been a proliferation of ecosystem service strategies on the ground thereby engaging new stakeholders in conservation efforts. There has also been greater funding levels and new funders directed at these efforts (Goldman et al. 2008). Stakeholder engagements range from private, for-profit companies, such as Coca-Cola, Inc., to small regional conservation organisations such as Fundacion Defensores de la Naturaleza (FDN). In particular, ecosystem service projects engage significantly more private corporations in conservation efforts than do biodiversity projects (Figure 8.2). Looking more closely at these projects and assessing how these partnerships formed and what led to stakeholder engagement can provide methods for reproducing these approaches worldwide which are essential in making conservation mainstream. Figure 8.2 Ecosystem Service Projects (ES project) Engage Significantly More Private, Corporate Stakeholders in Conservation Efforts than Traditional Approaches (BD project)

*

* p 50 km2 (Pavlov 1991). Pigs are not considered nocturnal, but they are typically most active during the evening, from early dusk until twilight (McIlroy et al. 1989). Feral pigs reproduce rapidly and are capable of year-round breeding. On average, sows first breed at 12 months, but in areas with prolific food supply, such as Hawai‘i, breeding can occur as early as 6 months (Singer 1981; Pavlov et al. 1992). Sows usually produce litters of 5-7 piglets and having two litters per year is not unusual (Pavlov et al. 1992). In Hawai‘i, with little predation and a variety of food sources, survivorship of piglets is high (Diong 1982). Effects of Pig Activity on Watersheds Waterborne Pathogens Feral pigs present numerous water quality issues through the propagation of several different types of pathogens. It should be noted that feral pigs do not solely spread most pathogens. Other mammals (i.e. mongoose and rats) also play a role in pathogen distribution. Fecal coliform, E. coli, and Enterococci are all types of bacteria that are present in human and animal waste. When released into a watershed, these bacteria can negatively affect water quality. Humans exposed to water contaminated with such pathogens are prone to contracting a range of illnesses including diarrhea, meningitis, and pneumonia. Streams in Hawai‘i regularly surpass the allotted EPA limits for such bacteria (Roll and Fujioka 1997). Tropical soils are the suspected

Effects of Feral Pigs (Sus scrofa) on Watershed Health in Hawai‘i: A Literature Review and Preliminary Results on Runoff and Erosion

255

primary source of the bacteria and these soils are thought to support high levels of these bacteria (Hardina and Fujioka 1991). Whether these bacteria are found naturally in the soil or occur as a result of mammalian fecal inputs is still not well known. Either way, feral pigs play a role both as a source of pathogens and also by promoting erosion of soils high in fecal bacteria. Outbreaks of E. coli have been more directly linked to the presence of feral pigs because strains found in E. coli can be matched with strains from feral pigs. A 2006 E. coli outbreak in spinach, which caused the death of three people and illnesses in about 200 others, was believed to be caused by feral pig activity on a spinach farm in California (Associated Press 2006). The spread of pathogens can also affect livestock. For example, on the island of Moloka‘i, feral pigs passed bovine tuberculosis onto cattle, resulting in the slaughter of 6,000 cows and substantial economic loss to the livestock owners (Hills 1988). In Australia, feral pigs are considered one of the primary vectors for the transmission of foot and mouth disease and could cost Australia $3 B if an outbreak is to occur (McIlroy et al. 1989). Impacts to Native Bird Populations Bird species endemic to the Hawaiian Islands evolved in seclusion for millennia and originally included 71 taxa. With the arrival of the Polynesians, Europeans, and the associated introductions of non-native biota, the number of native bird species has drastically decreased. Atkinson et al. (1995) estimated that more than 75 percent of the original bird species are either extinct or endangered. Of the many reasons cited for this decline, habitat destruction and the introduction of avian diseases are thought to be directly linked to the activity of feral pigs. Pigs have been shown to damage native plant communities and therefore disrupt the symbiotic relationships between native birds and plants. In turn, the destruction of native plants has led to an increase in non-native plant species that are supported by ungulate grazing. Feral pigs enhance proliferation of non-native plants as seeds are dispersed through their hair and feces. Furthermore, pig wastes enhance nitrogen (N) levels in soils that were once N-poor and supported native species adapted to lownutrient conditions. The influx of N from feral pigs diminishes the adaptive advantage native plants once had over invasive species (Cuddihy and Stone 1990). Pigs are most notorious in promoting the spread of invasive banana

256

Gregory L. Bruland, Chad A. Browning, and Carl I. Evensen

polka (Passiflora mollissima) and strawberry guava (Psidium cattleianum). Feral pigs have also increased the abundance of the highly-invasive fire tree (Morella faya), which comprised 5 percent of the average feral pig stomach content in one study (Stone and Taylor 1984). Management efforts to combat the spread of invasive plants have involved removal of feral pigs. Following their removal, a state of near equilibrium has been reached between native and introduced plant species (Jacobi 1981; Stone and Loope 1987). The correlation between invasive plant presence and feral pig activity is so coupled that Aplet et al. (1991) suggested the possibility of using field observations of plant composition to estimate the relative amount of pig activity. Moreover, the introduction of mosquitoes and their capacity to spread diseases, such as avian malaria, has had adverse impacts on native bird population. The increased density of mosquitoes and resultant spread of avian malaria into higher elevations and deeper into forests has been related to the formation of mosquito breeding grounds caused by pig wallowing and browsing. Wallowing is a common activity of feral pigs. Pigs dig and roll in mud-filled puddles to cool body temperature. In this process, wallows are created that retain rainwater and provide breeding pools for mosquitoes. In Hawaiian forests, pigs also have a propensity to consume the inner cores of native ferns, leaving a hollow space inside the core. After rainfall fills these voids, they can support thousands of mosquito larvae (United States Geologic Survey 2006). Most of the remaining endangered native bird species are now only found in the higher reaches of their original habitats where cooler temperatures and the absence of breeding pools have kept mosquito populations in check. Increased Runoff and Soil Loss Perhaps the most threatening of all impacts posed by feral pigs is their ability to disrupt the basic hydrologic functions of a watershed. Feral pigs threaten the ability of the watershed to regulate water quantity. In some areas of Hawai‘i, the majority of the water comes directly from surface runoff. Mosses, ferns and other understory plants and litter that line the forest floor act as sponges that collect, store, and slowly release water. Degradation of these plants and litter increase the speed of runoff after rainfall events. Water rapidly flowing into the sea leaves less water to be utilised for human and

Effects of Feral Pigs (Sus scrofa) on Watershed Health in Hawai‘i: A Literature Review and Preliminary Results on Runoff and Erosion

257

agricultural consumption. It also increases the potential of severe flooding during times of heavy rainfall. The state and federal governments have spent considerable funds to thwart such effects. One example was a flood prevention project of the Army Corps of Engineers in 2006 which cost over $23 M (United States Army Corps Of Engineers 2007). Pigs also increase runoff through soil compaction. Compaction occurs as a result of wallowing, formation of trails, trampling of vegetation during browsing, and during nest creation. Browsing, in particular, is the most damaging. In the areas immediately surrounding native ferns, a 30-percent increase in soil bulk density was observed due to pig activity (Vtorov 1993). In a seven-year soil recovery study, Vtorov (1993) found that soil bulk density within a pig exclosure decreased more than three times by the end of the study period. That same study also illustrated the correlation between feral pig activity and the absence of native microarthropods that depended on pore spaces and abundant organic matter. These microarthropods are not only good indicators of soil quality, but also perform a vital role in nutrient cycling processes. One of the most readily-observable impacts of feral pigs on watersheds is their tendency to disturb soil cover. Pigs are considered omnivores and feed on fleshy plants and soil invertebrates. A great deal of digging and rooting is required for pigs to locate food. The term “digging” generally pertains to the pig’s search for soil invertebrates, whereas “rooting” implies the removal of plant roots. Rooting is such a regular pursuit of feral pigs that it is commonly used as an indicator of pig population density (Hone 1988). The fact that feral pig rooting activity decreases soil cover is well known and not unique to Hawai‘i. This phenomenon has been observed worldwide. One observation from Singer (1981) described sample stations in Poland that were surveyed prior to pig occupation and resurveyed after eight years of pig inhabitation. The difference between the same sites over the eight-year period was dramatic. Plant cover was reduced by 87 percent. Subsequently, bare ground dominated the forest floor where no bare ground had been noted prior to pig introduction. Also noted were the 36 percent reduction in litter horizon depth, and the intermixing of the upper two soil horizons. Another study found that digging by pigs could disturb as much as 30 percent of the forest floor (Cooray and Mueller-Dombois 1981). The combination of the factors discussed above creates an environment that is susceptible to soil erosion. This can be particularly dangerous in

258

Gregory L. Bruland, Chad A. Browning, and Carl I. Evensen

Hawaiian watersheds because areas typically inhabited by pigs have steep slopes (Cuddihy and Stone 1990). The south shore of Moloka‘i is a prime example of the amount of erosion feral pigs can cause, in addition to that caused by goats and deer. Massive amounts of sediment have washed out to sea and buried large areas of coral reefs. One estimate of erosion on Moloka‘i have reported that up to four inches of topsoil per year runs off into estuaries, fish ponds and coral reefs, at a rate more than four times greater than that of the last four centuries (Hills 1988). The resulting effects of sedimentation are well known and will not be covered in detail. It is important to note that sedimentation has multiple negative impacts on watersheds and adjacent coastal areas. These impacts include the associated nutrient influx carried by sediments, the damage caused to streams and estuaries from increased turbidity, and the cost of dredging coastal waterways. Given these impacts, it is important to quantify how feral pigs contribute to runoff and erosion in order to identify proper watershed management strategies. Our review indicates knowledge gaps on the effects of feral pigs on runoff and soil loss. While rough estimates have been theorised and visual evidence is abundant, few attempts have been made to quantify these effects. Control versus Eradication When contemplating the damage caused by feral pigs in Hawai‘i, one solution seems to be the complete eradication of feral pig populations. This, however, is quite a complex cultural, economic, and political issue. Pig hunting is a popular recreational activity in Hawai‘i, which has occurred since feral pigs first inhabited the islands. Pig hunting has been passed down through multiple generations. Many native Hawaiians claim pig hunting as a cultural right that links them to the land of their ancestors. Although a smaller minority in Hawai‘i, pig hunters also have substantial political influence (Hills 1988). This is apparent in the success that hunters have had in maintaining stable pig populations on all the main Hawaiian Islands except Lana‘i. Other socioeconomic factors come into play as well. The use of pig hunting to subsist in a state that has a high cost of living is also a valid argument. In addition, hunted pigs are used in festivities that bring communities together (Diong 1982).

Effects of Feral Pigs (Sus scrofa) on Watershed Health in Hawai‘i: A Literature Review and Preliminary Results on Runoff and Erosion

259

The link of pig hunting to Hawai‘i’s culture and politics has led entire communities, not just hunters, to resist pig control measures. Thus, total eradication of feral pigs in Hawai‘i is not likely a viable option. Thus the issue becomes one of determining the optimal or desirable population of feral pigs that can be supported. Most likely, the optimal population will vary across islands and stakeholder groups, further complicating the issue. Control Measures: Background The current mission of Hawai‘i’s Department of Fish and Wildlife (DOFAW) is to maintain a balance between preserving native biota and at the same time to provide a sustained population of game animals for hunters. This mission is not only difficult, but also controversial. On O‘ahu, anti-pig sentiment has grown as pig populations have increased and interactions of pigs and landowners have become more common. On average, the O‘ahu DOFAW Branch Office receives 12 phone calls per month regarding pig nuisance (Hawai‘i Department of Land and Natural Resources 2006). Due to the growing concern over pigs, citizens have become more vocal in their call for government action to reduce pig populations. To date, the federal government and state of Hawai‘i have implemented several different control programmes either to eradicate pigs in sensitive native ecosystems, or maintain pig populations at a level where their impact is minimised. These control efforts include hunting, trapping, poisoning, and fencing. The control method used varies based on the preferred objective. Complete eradication is the desired objective in the most vulnerable areas, such as national parks, where maintaining native habitat is a priority. In contrast, some areas are managed explicitly for game management. In these areas, sustained hunting is the sole control method. There are many areas that do not fit into either category. In these areas, pigs are tolerated as long as their populations are kept at a manageable level. Hence, control methods vary widely. Deciding on the management objective of these areas then becomes a key question, as natural resource managers are increasingly being asked to decide where to allocate limited funds to best manage growing invasive species challenges.

260

Gregory L. Bruland, Chad A. Browning, and Carl I. Evensen

Hunting There are currently over 40 game management and public hunting areas maintained by DOFAW across Hawai‘i. Feral pigs are the most abundant and popular game animal. Other game mammals (i.e. feral goats, mouflon sheep, axis deer and black-tailed deer) exist on different islands, although these other species are not found in O‘ahu. Popular techniques of pig hunting include archery, rifle, and the dogand-knife method. The most effective hunting technique is the dog-andknife method. This method allows dogs to sniff out and chase down pigs. Attempting to eradicate pigs through hunting is a more viable endeavor if fences are first installed to enclose the area targeted for protection. Once fenced, hunters can be employed to remove pigs that remain in the enclosed area. This method is effective as long as it is involves smaller scale fenced enclosures (14 pigs km-2). Snaring is recommended in areas where rugged terrain reduces the success of traditional hunting. The use of snares and kill traps are not without controversy. Pigs are usually found well after the time of death causing carcasses to be unfit for consumption. Also there exist the possibilities of snare malfunctions (i.e. snaring a pig’s limb instead of its neck). Such situations can lead to prolonged death, which is viewed negatively by the public, especially by animal rights groups (Leone 2001). Live traps also vary, but the concept is similar among all types. Pigs are baited into a confined space where, by trap design, the only route of escape is closed off. Live traps are more popular with the public because they are less likely to harm non-target species, and the trapped animals can be processed for meat. This is the preferred method in areas of high human population such as residential neighborhoods (Hawai‘i Department of Land and Natural Resources 2006). One downfall of live trapping is that traps tend to be bulky, thereby making transport and setup in rough terrain difficult. Many live-trapping programmes, including those implemented by DOFAW, allowed captured pigs to be released in other areas to make them available for hunters. As with hunting, eradication through trapping alone appears to be incapable of overcoming the rapid population growth of feral pigs. One final consideration is the use of trapping instead of hunting. This often does not sit well with hunters, most of which are willing to pay for the opportunity to hunt these areas (Hubrecht 2006). Poisoning Another approach to feral pig management involves poisoning. The ease of distribution is the biggest strength of poisoning compared to other methods. Significant reductions in the number of pigs have been observed with poisoning (Saunders 1990). There are two types of poisons that have been studied in controlling pig populations – sodium monofluoroacetate (1080) and warfarin. Sodium monofluoroacetate was first introduced to control feral pig populations in Australia. Problems with 1080 arose soon after introduction

262

Gregory L. Bruland, Chad A. Browning, and Carl I. Evensen

as researchers observed bait shyness, tolerance by pigs that had ingested large amounts, and adverse impacts on non-target species (McIlroy et al. 1989; Hone 1992). Nonetheless, 1080 is still being used in some areas to reduce pig numbers. One benefit of 1080 is its lethal effect within 3-80 hours of ingestion (McIlroy et al. 1989). Issues with the use of 1080 led to the development of warfarin. Due to improvements in efficiency (i.e. less bait shyness), warfarin is currently more popular with managers than 1080. A trial poisoning reported an extermination success rate of 94 percent (McIlroy et al. 1989) which illustrated the effectiveness of this approach. However, the study also found that fatal results were incurred much later (approximately 5-10 days) compared to 1080. Arguments against poisoning are abundant. Poison is indiscriminate and can potentially harm hunting dogs, children, and other non-target organisms. Poisoning is not popular with hunters or rural communities in Hawai‘i, where pigs and humans may coexist in close proximity (Diong 1982). Another argument against the use of poison in Hawai‘i is that most successful poisoning programmes were conducted in areas with low food supply, unlike Hawai‘i. These reasons suggest that poisoning is best implemented in inaccessible areas where interaction with humans is unlikely. Fencing One of the most effective techniques, especially when combined with one or more of the previous methods, is the use of fencing. Fencing requires complete eradication of pigs from one side of the fenced boundary, which makes it unique from the other control methods. Many suggest that fencing should keep pigs from higher elevated areas, thus enabling these regions to perform necessary watershed functions and provide undisturbed habitat for the remaining native species (Hills 1988). Under this scenario, soil erosion would also be reduced because pigs would not inhabit steeply-sloped areas. This would allow public hunting to continue under designated fence lines where pig populations could be better controlled. Fencing steeper Hawaiian watersheds is extremely costly. Most areas deemed crucial for protection of watershed services and endangered species are found in the higher reaches of Hawaiian forests where slopes can average greater than 30 percent (Hills 1988). This makes installation of fences

Effects of Feral Pigs (Sus scrofa) on Watershed Health in Hawai‘i: A Literature Review and Preliminary Results on Runoff and Erosion

263

time-consuming, labor-intensive, and expensive. A 1988 cost estimate for transporting the fencing materials into designated areas ranged from $10,000 to $100,000 per mile (Hills 1988). In 1998, the cost of actual fence installation was estimated at $40,000 per mile (Kaiser et al. 1999). Costs do not end with installation, as continuous monitoring is needed to ensure that integrity of the fences is sustained. Control Method Summary The preceding material suggests that a single method does not hold the key to effective feral pig management. Rather, a combination of methods seems most useful in achieving a balance among social, political, and environmental concerns. It is also important to mention that the execution of any management plan must have the cooperation of the entire community. Failure to incorporate all stakeholder groups could translate into continued unsuccessful implementation of control policies. Finally, due to the remoteness of most feral pig habitats, the consequences of unchecked feral pig activity on a watershed are not easily observed. Likewise, the negative side effects of increased runoff and sedimentation are not always easy to observe and it is even more difficult to identify feral pigs as a contributor to such damage. Therefore, any control method should also be accompanied by education and outreach efforts to illustrate the impacts of pigs on watershed health to the general public. EXPERIMENTAL EFFECTS OF FENCING ON EROSION AND RUNOFF MATERIALS AND METHODS Study Site: Mānoa Watershed The Mānoa watershed encompasses approximately 2,528 ha in southeast O‘ahu. It contains both flat coastal and steep mountainous lands within a relatively small area. Runoff from Mānoa watershed is carried by two main streams, Mānoa and Palolo, which join each other before emptying into the Ala Wai Canal. Although highly-developed in the lower and middle

264

Gregory L. Bruland, Chad A. Browning, and Carl I. Evensen

reaches, the upper reaches of Mānoa watershed remain mostly forested and uninhabited by people. Site Selection To quantify runoff and soil loss due to feral pig activities in the Mānoa watershed, eight sites within the upper-forested areas were selected based on their slope, accessibility and location. Slope was the primary factor behind site selection to accommodate the construction of fencing and the collection of accurate runoff data. Pig activity was not expected to be high in areas of intense slope (>30%). Also, the standard by which most runoff plots are constructed is approximately 9 percent (Mutchler et al. 1994). Using a GIS map of slope throughout the watershed, areas with slopes between 5-30 percent were identified as potential sites (Figure 11.1). Final site selection was based on accessibility from existing trails and the presence of areas with homogeneous vegetative conditions determined during onsite visits. The eight sites selected along with associated characteristics are listed in Table 11.1. Table 11.1 Characteristics of the Eight Sites in the Mānoa Watershed Site

Elevation (m)

Slope (%)

Lyon

215

15-16

Mānoa Cliffs

450

8

Mānoa Falls

171

16-18

Palolo

225

25-27

Pauoa Flats

538

6

Puu Pia

209

26

Round Top

340

25-26

Wa‘ahila Ridge

340

14

Effects of Feral Pigs (Sus scrofa) on Watershed Health in Hawai‘i: A Literature Review and Preliminary Results on Runoff and Erosion

265

Figure 11.1 Map of the Variations in Slope Across the Watershed with the Locations of All Eight Sampling Sites

Site Layout Each of the eight sites measured approximately 10 m x 10 m. Within each site, two paired plots were established which consisted of a 10 m x 5 m exclosure and an adjacent 10 m x 5 m unexclosed plot (Figure 11.2). A 14-gauge utility fencing with a height of 0.91 m was used to construct the exclosures. Plots were oriented to run down slope to effectively capture the natural processes of hill-slope erosion and runoff. Slopes were measured with a handheld clinometer and ranged from 5-27 percent. Slopes for paired fenced and unfenced plots were within 2 percent of each other for every site.

266

Gregory L. Bruland, Chad A. Browning, and Carl I. Evensen

Figure 11.2 The Basic Layout of Each Site in Study. Each Site was Oriented to Achieve Similar Slope and Vegetative Characteristics for Both Plots

The eight sites represented four different soil series. The soils of three sites (Lyon, Mānoa Falls and Pu‘u Pia) were classified as the Lolekaa series. These soils are well-drained, with slow to rapid runoff and moderately rapid permeability (NRCS 2008). The Manana series was found at the Wa‘ahila Ridge site. These soils display slow to rapid runoff. Another site, Round Top, was located in an area with soils of the Tantalus series. These soils are well-drained with medium to rapid runoff (NRCS 2008). The final three sites were located in areas classified as rRT. This is not a soil series, but rather a designation given to remote unclassified areas that are typically defined by steep land with intermittent drainage channels. Runoff Plot Design After fence installation, runoff plots were established within both unfenced and fenced plots. Runoff plots measured 4.2 m x 1.2 m, and were oriented to run parallel to the slope (Figure 11.3). Plots were sheltered from additional runoff by 15 cm tall plastic pieces which were buried approximately 7.5cm deep. These pieces formed the framework that channeled runoff to a

Effects of Feral Pigs (Sus scrofa) on Watershed Health in Hawai‘i: A Literature Review and Preliminary Results on Runoff and Erosion

267

collector. At the down slope end of the plot, a triangular metal collector funneled all runoff into an 18.9-L bucket via a feed tray bolted to the collector (Figure 11.3). Water-tight lids were placed on top of each bucket to prevent rain from falling directly into the buckets. Figure 11.3 Runoff Collection Apparatus with Lid Removed

Soil Sampling Upon runoff plot installation, three soil cores were collected from each of the 16 fenced and unfenced plots (48 total soil cores). Core locations were selected randomly and no cores were collected from within the runoff plots. Cores were collected from the upper 20 cm of the soil profile and taken back to the laboratory for analysis. Cores were dried to determine soil moisture and bulk density (BD). Moisture was quantified by drying the soil at 105oC for 24 hours. The dry mass was divided by the core volume to determine volumetric water content. Bulk density was calculated by dividing dry mass by the core volume.

268

Gregory L. Bruland, Chad A. Browning, and Carl I. Evensen

Collection of Runoff Runoff samples were collected monthly from November 2007 to February 2008. For each month, a two-day period was targeted to activate all plots for runoff collection. To activate each plot, ground and canopy cover were estimated, and collection buckets were emptied. Collection times were determined by observing an online USGS rain gauge located in Mānoa Valley. Runoff samples were collected when recorded rainfall displayed a significant increase (typically >2 cm) from the time of activation. Upon collection, sites were evaluated for total runoff, and sub-samples were collected for analysis of total suspended solids (TSS). Measurement of TSS was conducted by pouring 100 mL of solution onto a Whatman 5 filter paper inside a Buchner funnel attached to a vacuum flask. As the solution was pulled through the filter, suspended particles ≥ 2.5µm were captured. Filters were dried for 24 hours at 105oC. Dried filters were weighed and the original weight was subtracted to calculate TSS in g/L. Estimation of Antecedent Conditions To quantify the amount of throughfall received at each site, a throughfall gauge was mounted approximately 1.2 m above the ground on the centermost fencepost between the two plots (Figure 11.2). Throughfall amount was recorded as a depth in cm. Throughfall gauges were capable of holding a total of 290 cm of throughfall, an amount that was not surpassed throughout the study. Results and Discussion of Soil and Runoff Data Soil Properties As feral pigs increase soil compaction, and compaction influences infiltration rate and runoff, it was important to document the BD of each plot at the start of the study. BD data was first analysed for differences between fenced and unfenced plots. A statistical comparison of fenced versus unfenced soils using a one-way ANOVA showed that there were significant differences

Effects of Feral Pigs (Sus scrofa) on Watershed Health in Hawai‘i: A Literature Review and Preliminary Results on Runoff and Erosion

269

between the two groups. Higher values were found in the unfenced sites indicating that exclusion of feral pigs may have already started to reduce compaction in the unfenced plots. A comparison of all eight sites indicated that Mānoa Cliffs had the lowest BD (Figure 11.4). The slope at Mānoa Cliffs was only 8 percent. This was a flatter area that served as a drainage point of the immediate subwatershed. After heavy rainstorms, standing water was observed in this area on multiple occasions. This standing water impeded decomposition rates due to periods of anaerobic conditions, especially when compared to the other sites with greater slope and less runoff. It was likely that this site has accumulated greater soil organic matter and consequently has a lower BD. Figure 11.4 Mean Bulk Density Across the Eight Study Sites (error bars represent ±1 standard error)

Bulk Density (g cm -3)

1.0 0.8 0.6 0.4 0.2 0.0 Lyon

Manoa Manoa Palolo Cliffs Falls

Pauoa Puu Pia Round Waahila Flats Top Ridge

Site

Round Top, meanwhile, had the highest BD. Unlike Mānoa Cliffs, Round Top accumulated little, if any organic matter. Bare soil was the most prevalent surface cover at this site. There was only a small percentage (