Ecoacoustics : the ecological role of sounds [First edition] 9781119230724

Table of contents :
Content: List of Contributors xiii Preface xv 1 Ecoacoustics: A New Science 1Almo Farina and Stuart H. Gage 1.1 Ecoacoustics as a New Science 1 1.2 Characteristics of a Sound 1 1.3 Sound and Its Importance 2 1.4 Ecoacoustics and Digital Sensors 3 1.5 Ecoacoustics Attributes 3 1.5.1 Population Census 4 1.5.2 Biological Diversity 4 1.5.3 Habitat Health 4 1.5.4 Time of Arrival/Departure of Migratory Species 4 1.5.5 Diurnal Change 5 1.5.6 Seasonal Change 5 1.5.7 Competition for Frequency 5 1.5.8 Trophic Interactions 5 1.5.9 Disturbance 5 1.5.10 Sounds of the Landscape and People 6 1.6 Ecoacoustics and Ecosystem Management 6 1.7 Quantification of a Sound 7 1.7.1 Species Identification 7 1.7.2 Acoustic Indices 7 1.8 Archiving Ecoacoustics Recordings 8 1.9 Ecological Forecasting 9 References 9 2 The Duality of Sounds: Ambient and Communication 13Almo Farina and Stuart H. Gage 2.1 Introduction 13 2.2 Vegetation and Ecoacoustics 14 2.2.1 Vegetation Quality and Ecoacoustics 15 2.2.2 Soundscape Indices and Biodiversity 15 2.2.3 Applications of Remote Sensing of Vegetation and Ecoacoustics 16 2.3 Acoustic Resources, Umwelten, and Ecofields 17 2.4 Sounds as Biological Codes 20 2.5 Sound as a Compass for Navigation 21 2.6 Geophonies from Sacred Sites How to Incorporate Archeoacoustics into Ecoacoustics 22 2.6.1 The Characteristics of Geophonies 23 2.6.2 Geophonies and Sacred Sites 23 2.6.3 Human Versus Other Animals Perception of Sound: The Role of Archeoacoustics 24 References 24 3 The Role of Sound in Terrestrial Ecosystems: Three Case Examples from Michigan, USA 31Stuart H. Gage and Almo Farina 3.1 Introduction 31 3.2 C1 Visualization of the Soundscape at Ted Black Woods, Okemos, Michigan during May 2016 31 3.2.1 C1 Background 31 3.2.2 C1 Objectives 32 3.2.3 C1 Methods 32 3.2.3.1 C1 Soundscape Metrics 33 3.2.3.2 C1 Weather Factors Affecting Sounds 33 3.2.4 C1 Results 33 3.2.4.1 C1 Patterns of Soundscape Power for Six Frequency Intervals 33 3.2.4.2 C1 Patterns of Soundscape Indices 37 3.2.4.3 C1 Wind Patterns During May 2016 37 3.2.4.4 C1 Rain Patterns During May 2016 37 3.2.4.5 C1 Spectrogram Patterns 41 3.2.5 C1 Discussion 42 3.3 C2 Implications for Climate Change Detecting First Call of the Spring Peeper 44 3.3.1 C2 Background 44 3.3.2 C2 Methods 44 3.3.3 C2 Results 45 3.3.4 C2 Discussion 48 3.4 C3 Disturbance in Terrestrial Systems: Tree Harvest Impacts on the Soundscape 49 3.4.1 C3 Background 49 3.4.2 C3 Methods 51 3.4.3 C3 Results 52 3.4.3.1 C3 Changes in the Soundscape 52 3.4.3.2 C3 Statistical Influence of Forest Harvest 55 3.4.4 C3 Discussion 55 References 59 4 The Role of Sound in the Aquatic Environment 61Francesco Filiciotto and Giuseppa Buscaino 4.1 Overview on Underwater Sound Propagation 61 4.1.1 Sound Speed in the Sea 61 4.1.2 Transmission Loss 61 4.1.3 Deep and Shallow Sound Channel and Animal Communication 62 4.2 Sound Emissions and Their Ecological Role in Marine Vertebrates and Invertebrates 63 4.2.1 Marine Mammals 63 4.2.2 Fish 64 4.2.3 Crustaceans 65 4.3 Impacts of Anthropogenic Noise in Aquatic Environments 67 4.3.1 Main Anthropogenic Sources of Noise in the Sea 67 4.3.2 The Effects of Anthropogenic Noise on Marine Organisms 68 4.3.2.1 Acoustic Masking and Damage to Hearing System of Marine Organisms 68 4.3.2.2 Biochemical Impacts and Stress Responses 69 4.3.2.3 Behavior Alterations 70 References 71 5 The Acoustic Chorus and its Ecological Significance 81Almo Farina and Maria Ceraulo 5.1 Introduction 81 5.2 Time of Chorus 82 5.3 The Chorus Hypothesis 86 5.4 Choruses in Birds 87 5.5 Choruses in Amphibians 87 5.6 Choruses in the Marine Environment 88 5.7 Conclusions and Discussion 89 References 89 6 The Ecological Effects of Noise on Species and Communities 95Almo Farina 6.1 Introduction 95 6.2 The Nature of Noise 96 6.3 Natural Sources of Noise 96 6.4 Anthropogenic Sources of Noise 97 6.5 Effects of Noise on the Animal World 97 6.6 How Animals Neutralize the Effect of Noise 100 6.6.1 Changing Amplitude 100 6.6.2 Changing Frequency 100 6.6.3 Changing Signal Redundancy 101 6.6.4 Changing Behavior 101 6.7 Noise in Marine and Freshwater Systems 101 6.8 Conclusions 102 References 103 7 Biodiversity Assessment in Temperate Biomes using Ecoacoustics 109Almo Farina and Nadia Pieretti 7.1 Introduction 109 7.2 Sound as Proxy for Biodiversity 110 7.3 Methods and Application of Ecoacoustics 111 7.4 Acoustic Communities as a Proxy for Biodiversity 113 7.5 Problems and Open Questions 114 7.6 Ecoacoustic Events: Concepts and Procedures 116 7.7 Conclusion 122 References 122 8 Biodiversity Assessment in Tropical Biomes using Ecoacoustics: Linking Soundscape to Forest Structure in a Human-dominated Tropical Dry Forest in Southern Madagascar 129Lyndsay Rankin and Anne C. Axel 8.1 Introduction 129 8.2 Methods 131 8.2.1 Study Area 131 8.2.2 Forest Sampling 132 8.2.3 Soundscape Survey 133 8.2.4 Acoustic Indices 133 8.2.5 Mixed Model Analysis 134 8.3 Results 135 8.3.1 Seasonal Acoustic Indices 135 8.3.2 Mixed Model Analyses 137 8.4 Discussion 137 Acknowledgments 141 References 142 9 Biodiversity Assessment and Environmental Monitoring in Freshwater and Marine Biomes using Ecoacoustics 145Denise Risch and Susan E. Parks 9.1 Introduction 145 9.2 Freshwater Habitats 147 9.2.1 Rivers 147 9.2.1.1 Remote Monitoring of Biotic Signals in the Environment 147 9.2.1.2 Remote Monitoring of the Environment Using Sound in River Habitats 148 9.2.1.3 Anthropogenic Sources of Noise in River Systems 148 9.2.2 Lakes and Ponds 148 9.2.2.1 Remote Monitoring of Biotic Signals in the Environment 149 9.2.2.2 Remote Monitoring of the Environment Using Sound in Lakes and Ponds 149 9.2.2.3 Anthropogenic Sources of Noise in Lakes and Ponds 149 9.3 Marine Neritic Habitats 150 9.3.1 Estuaries and Coastal Habitats 150 9.3.1.1 Remote Monitoring of Biotic Signals in the Environment 150 9.3.1.2 Remote Monitoring of the Environment Using Sound in Estuarine and Coastal Habitats 150 9.3.1.3 Anthropogenic Sources of Noise in Estuarine and Coastal Habitats 152 9.3.2 Coral Reefs 152 9.3.2.1 Remote Monitoring of Biotic Signals in the Environment 152 9.3.2.2 Remote Monitoring of the Environment Using Sound in Coral Reef Environments 153 9.3.2.3 Anthropogenic Sources of Noise in Coral Reef Environments 153 9.4 Marine Oceanic Habitats 153 9.4.1 Open Ocean and Deep Sea Habitats 153 9.4.1.1 Remote Monitoring of Biotic Signals in the Environment 154 9.4.1.2 Remote Monitoring of the Environment Using Sound in the Open Ocean 154 9.4.1.3 Anthropogenic Sources of Noise in the Open Ocean 154 9.4.2 Polar Oceans 155 9.4.2.1 Remote Monitoring of Biotic Signals in the Environment 155 9.4.2.2 Remote Monitoring of the Environment with Sound in Polar Regions 155 9.4.2.3 Anthropogenic Sources of Noise in the Polar Regions 156 9.5 Summary and Future Directions 156 References 158 10 Integrating Biophony into Biodiversity Measurement and Assessment 169Brian Michael Napoletano 10.1 Introduction 169 10.1.1 Biodiversity and Its Parameterization 170 10.2 Biological Information in the Soundscape 171 10.2.1 Physiology: Sound Production and Detection 173 10.2.2 Communication: Medium and Context 176 10.2.3 Coordination: Evolution of the Biophony 178 10.2.4 Adaptation: Mechanization of the Soundscape 180 10.3 Ecoacoustics in Biodiversity Assessment 182 10.3.1 Developing a Soundscape Monitoring Network 182 10.3.2 Acoustic Data Processing and Management 183 10.4 Conclusion 184 References 184 11 Landscape Patterns and Soundscape Processes 193Almo Farina and Susan Fuller 11.1 An Introduction to Landscape Ecology (Theories and Applications) 193 11.1.1 Patch Size, Shape, and Isolation 193 11.1.2 Patch ]Matrix Context 194 11.2 Relationship Between Landscape Ecology and Soundscape Ecology: A Semantic Approach 195 11.2.1 The Contribution of Landscape Ecology to the Development of Ecoacoustics 196 11.2.2 Acoustic Heterogeneity in a Landscape Across Space and Time 197 11.3 Acoustic Community and Landscape Mosaics 199 11.4 Ecoacoustics in a Changing Landscape 202 11.5 Conclusion 203 References 204 12 Connecting Soundscapes to Landscapes: Modeling the Spatial Distribution of Sound 211Timothy C. Mullet 12.1 Introduction 211 12.2 Conceptualizing Soundscapes in Space and Time 211 12.3 Capturing Soundscapes in Time and Space 212 12.4 Sound Metrics and Interpreting Nature 213 12.5 A Soundscape Metric for Modeling 215 12.6 Discriminating the Components of a Soundscape 216 12.7 Generating a Predictive Soundscape Model 217 12.8 Conclusion 219 Disclaimer 221 References 221 13 Soil Acoustics 225Marisol A. Quintanilla ]Tornel 13.1 Introduction 225 13.2 Soil Insect Acoustics 226 13.3 Compost Activating Agent Acoustics 226 13.4 Soil Aggregate Slaking Acoustics 227 13.5 Conclusion 230 References 231 14 Fundamentals of Soundscape Conservation 235Gianni Pavan 14.1 Introduction 235 14.2 Nature Sounds in Science and Education 238 14.3 The Role of Sound Libraries 242 14.4 Noise Pollution, the Acoustic Habitat, and the Biology of Disturbance 243 14.5 Soundscapes, Nature Conservation, and Public Awareness 244 14.6 Marine Soundscapes 245 14.6.1 Ship Noise 246 14.7 Conclusion 251 14.7.1 Terrestrial Soundscapes 252 14.7.2 Marine and Aquatic Soundscapes 252 Acknowledgment 252 References 252 15 Urban Acoustics: Heartbeat of Lansing, Michigan, USA 259Stuart H. Gage and Wooyeong Joo 15.1 Introduction 259 15.2 Objectives 260 15.3 Methods 261 15.3.1 Sampling Design 261 15.3.2 Recording at Sample Sites 262 15.3.3 Data Conversion 262 15.3.4 Data Processing 262 15.4 Results 264 15.4.1 The NDSI 264 15.4.2 The H, ADI, AEI, ACI, and BIO Indices 267 15.5 Discussion and conclusions 267 References 271 16 Analytical Methods in Ecoacoustics 273Stuart H. Gage, Michael Towsey and Eric P. Kasten 16.1 Introduction 273 16.2 Components of an Acoustic Recording 275 16.3 Visualization of an Acoustic Recording 276 16.3.1 Frequency Analysis 276 16.3.2 Three ]Dimensional Spectrogram 277 16.4 Processing Multiple Recordings 277 16.5 Analyzing Acoustic Time Series 279 16.6 Time Series of Acoustic Indices 281 16.7 Searching and Symbolic Methods 282 16.7.1 Searching a Recording for Anomalies 284 16.7.2 Symbolic Representations and Unsupervised Learning 285 16.8 Visualization and Navigation of Long ]Duration Recordings 286 16.9 Spectrogram Pyramids 289 16.9.1 Diel Plots 289 16.10 New Approaches to Analysis 291 16.11 Web Platforms for the Visualization of Environmental Audio 291 References 293 17 Soundscape Ecology and its Expression through the Voice of the Arts: An Essay 297David Monacchi and Bernie Krause 17.1 Introduction 297 17.2 Immersive Art as a Science Dissemination Tool 299 17.3 Examples of Ecoacoustic Works by Bernie Krause 302 17.4 Examples of Ecoacoustics Works by David Monacchi 305 17.4.1 Designing Temples for the Ear: The Ecoacoustic Theater 308 17.4.2 Soundscape Projection Ambisonics Control Engine (S.P.A.C.E.) 309 17.5 Conclusion 310 References 311 18 Ecoacoustics Challenges 313Stuart H. Gage and Almo Farina 18.1 Introduction 313 18.2 Philosophical Issues 313 18.3 Ecological Issues 314 18.4 Sensor Technology 315 18.5 Acoustic Computations and Modeling 316 18.6 Public Information 316 18.7 Monetary Issues 317 References 317

Citation preview

Ecoacoustics

Ecoacoustics The Ecological Role of Sounds

Edited by

Almo Farina

Urbino University, Italy

Stuart H. Gage

Michigan State University, East Lansing, Michigan, USA

This edition first published 2017 © 2017 John Wiley and Sons 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, except as permitted by law. Advice on how to obtain permission to reuse material from this title is available at http://www.wiley.com/go/permissions. The right of Almo Farina and Stuart H. Gage to be identified as the authors of this work/the editorial material in this work has been asserted in accordance with law. Registered Offices John Wiley & Sons Inc., 111 River Street, Hoboken, NJ 07030, USA John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK Editorial Office 9600 Garsington Road, Oxford, OX4 2DQ, UK For details of our global editorial offices, customer services, and more information about Wiley products visit us at www.wiley.com. Wiley also publishes its books in a variety of electronic formats and by print-on-demand. Some content that appears in standard print versions of this book may not be available in other formats. Limit of Liability/Disclaimer of Warranty: While the publisher and authors have used their best efforts in preparing this work, they make no representations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties, including without limitation any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives, written sales materials or promotional statements for this work. The fact that an organization, website, or product is referred to in this work as a citation and/or potential source of further information does not mean that the publisher and authors endorse the information or services the organization, website, or product may provide or recommendations it may make. This work is sold with the understanding that the publisher is not engaged in rendering professional services. The advice and strategies contained herein may not be suitable for your situation. You should consult with a specialist where appropriate. Further, readers should be aware that websites listed in this work may have changed or disappeared between when this work was written and when it is read. Neither the publisher nor authors shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages. Library of Congress Cataloging-in-Publication Data Names: Farina, Almo, editor. | Gage, S. H., editor. Title: Ecoacoustics : the ecological role of sounds / edited by Almo Farina, Urbino University, IT, Stuart H Gage, Michigan State University, East Lansing, MI, US. Description: First edition. | Hoboken, NJ : John Wiley & Sons, Inc., 2017. | Includes bibliographical references and index. Identifiers: LCCN 2017003603 (print) | LCCN 2017005379 (ebook) | ISBN 9781119230694 (cloth) | ISBN 9781119230700 (pdf ) | ISBN 9781119230717 (epub) Subjects: LCSH: Landscape ecology. | Nature sounds. | Bioacoustics. | Ecosystem health. | Biodiversity. Classification: LCC QH541.15.L35 E247 2017 (print) | LCC QH541.15.L35 (ebook) | DDC 577--dc23 LC record available at https://lccn.loc.gov/2017003603 A catalogue record for this book is available from the British Library. Set in 10/12pt Warnock by SPi Global, Chennai, India Cover Design: Wiley Cover Image: Courtesy of Stuart H. Gage 10 9 8 7 6 5 4 3 2 1

v

Contents List of Contributors  xiii Preface  xv 1

Ecoacoustics: A New Science  1 Almo Farina and Stuart H. Gage

1.1 Ecoacoustics as a New Science  1 1.2 Characteristics of a Sound  1 1.3 Sound and its Importance  2 1.4 Ecoacoustics and Digital Sensors  3 1.5 Ecoacoustics Attributes  3 1.5.1 Population Census  4 1.5.2 Biological Diversity  4 1.5.3 Habitat Health  4 1.5.4 Time of Arrival/Departure of Migratory Species  4 1.5.5 Diurnal Change  5 1.5.6 Seasonal Change  5 1.5.7 Competition for Frequency  5 1.5.8 Trophic Interactions  5 5 1.5.9 Disturbance  1.5.10 Sounds of the Landscape and People  6 1.6 Ecoacoustics and Ecosystem Management  6 1.7 Quantification of a Sound  7 1.7.1 Species Identification  7 1.7.2 Acoustic Indices  7 1.8 Archiving Ecoacoustics Recordings  8 1.9 Ecological Forecasting  9 References  9 2

The Duality of Sounds: Ambient and Communication  13 Almo Farina and Stuart H. Gage

2.1 Introduction  13 2.2 Vegetation and Ecoacoustics  14 2.2.1 Vegetation Quality and Ecoacoustics  15 2.2.2 Soundscape Indices and Biodiversity  15

vi

Contents

2.2.3

Applications of Remote Sensing of Vegetation and Ecoacoustics  16 2.3 Acoustic Resources, Umwelten, and Eco-fields  17 2.4 Sounds as Biological Codes  20 2.5 Sound as a Compass for Navigation  21 2.6 Geophonies from Sacred Sites – How to Incorporate Archeoacoustics into Ecoacoustics  22 2.6.1 The Characteristics of Geophonies  23 2.6.2 Geophonies and Sacred Sites  23 2.6.3 Human Versus Other Animals’ Perception of Sound: The Role of Archeoacoustics  24 References  24 3

The Role of Sound in Terrestrial Ecosystems: Three Case Examples from Michigan, USA  31 Stuart H. Gage and Almo Farina

3.1 Introduction  31 3.2 C1 Visualization of the Soundscape at Ted Black Woods, Okemos, Michigan during May 2016  31 3.2.1 C1 Background  31 3.2.2 C1 Objectives  32 3.2.3 C1 Methods  32 3.2.3.1 C1 Soundscape Metrics  33 3.2.3.2 C1 Weather Factors Affecting Sounds  33 3.2.4 C1 Results  33 3.2.4.1 C1 Patterns of Soundscape Power for Six Frequency Intervals  33 3.2.4.2 C1 Patterns of Soundscape Indices  37 3.2.4.3 C1 Wind Patterns During May 2016  37 3.2.4.4 C1 Rain Patterns During May 2016  37 3.2.4.5 C1 Spectrogram Patterns  41 3.2.5 C1 Discussion  42 3.3 C2 Implications for Climate Change – Detecting First Call of the Spring Peeper  44 3.3.1 C2 Background  44 3.3.2 C2 Methods  44 3.3.3 C2 Results  45 3.3.4 C2 Discussion  47 3.4 C3 Disturbance in Terrestrial Systems: Tree Harvest Impacts on the Soundscape  49 3.4.1 C3 Background  49 3.4.2 C3 Methods  51 3.4.3 C3 Results  52 3.4.3.1 C3 Changes in the Soundscape  52 3.4.3.2 C3 Statistical Influence of Forest Harvest  55 3.4.4 C3 Discussion  55 References  59

Contents

4

The Role of Sound in the Aquatic Environment  61 Francesco Filiciotto and Giuseppa Buscaino

4.1 4.1.1 4.1.2 4.1.3 4.2

Overview on Underwater Sound Propagation  61 Sound Speed in the Sea  61 Transmission Loss  61 Deep and Shallow Sound Channel and Animal Communication  62 Sound Emissions and their Ecological Role in Marine Vertebrates and Invertebrates  63 4.2.1 Marine Mammals  63 4.2.2 Fish  64 4.2.3 Crustaceans  65 4.3 Impacts of Anthropogenic Noise in Aquatic Environments  67 4.3.1 Main Anthropogenic Sources of Noise in the Sea  67 4.3.2 The Effects of Anthropogenic Noise on Marine Organisms  68 4.3.2.1 Acoustic Masking and Damage to Hearing System of Marine Organisms  68 4.3.2.2 Biochemical Impacts and Stress Responses  69 4.3.2.3 Behavior Alterations  70 References  71 5

The Acoustic Chorus and its Ecological Significance  81 Almo Farina and Maria Ceraulo

5.1 Introduction  81 5.2 Time of Chorus  82 5.3 The Chorus Hypothesis  86 5.4 Choruses in Birds  87 5.5 Choruses in Amphibians  87 5.6 Choruses in the Marine Environment  88 5.7 Conclusions and Discussion  89 References  89 6

The Ecological Effects of Noise on Species and Communities  95 Almo Farina

6.1 Introduction  95 6.2 The Nature of Noise  96 6.3 Natural Sources of Noise  96 6.4 Anthropogenic Sources of Noise  97 6.5 Effects of Noise on the Animal World  97 6.6 How Animals Neutralize the Effect of Noise  100 6.6.1 Changing Amplitude  100 6.6.2 Changing Frequency  100 6.6.3 Changing Signal Redundancy  101 6.6.4 Changing Behavior  101 6.7 Noise in Marine and Freshwater Systems  101 6.8 Conclusions  102 References  103

vii

viii

Contents

7

Biodiversity Assessment in Temperate Biomes using Ecoacoustics  109 Almo Farina and Nadia Pieretti

7.1 Introduction  109 7.2 Sound as Proxy for Biodiversity  110 7.3 Methods and Application of Ecoacoustics  111 7.4 Acoustic Communities as a Proxy for Biodiversity  113 7.5 Problems and Open Questions  114 7.6 Ecoacoustic Events: Concepts and Procedures  116 7.7 Conclusion  122 References  122 8

Biodiversity Assessment in Tropical Biomes using Ecoacoustics: Linking Soundscape to Forest Structure in a Human-dominated Tropical Dry Forest in Southern Madagascar  129 Lyndsay Rankin and Anne C. Axel

8.1 Introduction  129 8.2 Methods  131 8.2.1 Study Area  131 8.2.2 Forest Sampling  132 8.2.3 Soundscape Survey  133 8.2.4 Acoustic Index  133 8.2.5 Mixed Model Analysis  134 8.3 Results  135 8.3.1 Acoustic Index by Season  135 8.3.2 Mixed Model Analyses  137 8.4 Discussion  137 Acknowledgments 141 References 142 9

Biodiversity Assessment and Environmental Monitoring in Freshwater and Marine Biomes using Ecoacoustics  145 Denise Risch and Susan E. Parks

9.1 Introduction  145 9.2 Freshwater Habitats  147 9.2.1 Rivers  147 9.2.1.1 Remote Monitoring of Biotic Signals in the Environment  147 9.2.1.2 Remote Monitoring of the Environment Using Sound in River Habitats  148 9.2.1.3 Anthropogenic Sources of Noise in River Systems  148 9.2.2 Lakes and Ponds  148 9.2.2.1 Remote Monitoring of Biotic Signals in the Environment  149 9.2.2.2 Remote Monitoring of the Environment Using Sound in Lakes and Ponds  149 9.2.2.3 Anthropogenic Sources of Noise in Lakes and Ponds  149 9.3 Marine Neritic Habitats  150 9.3.1 Estuaries and Coastal Habitats  150 9.3.1.1 Remote Monitoring of Biotic Signals in the Environment  150

Contents

9.3.1.2 Remote Monitoring of the Environment Using Sound in Estuarine and Coastal Habitats  150 9.3.1.3 Anthropogenic Sources of Noise in Estuarine and Coastal Habitats  152 9.3.2 Coral Reefs  152 9.3.2.1 Remote Monitoring of Biotic Signals in the Environment  152 9.3.2.2 Remote Monitoring of the Environment Using Sound in Coral Reef Environments  153 9.3.2.3 Anthropogenic Sources of Noise in Coral Reef Environments  153 9.4 Marine Oceanic Habitats  153 9.4.1 Open Ocean and Deep Sea Habitats  153 9.4.1.1 Remote Monitoring of Biotic Signals in the Environment  154 9.4.1.2 Remote Monitoring of the Environment Using Sound in the Open Ocean  154 9.4.1.3 Anthropogenic Sources of Noise in the Open Ocean  154 9.4.2 Polar Oceans  155 9.4.2.1 Remote Monitoring of Biotic Signals in the Environment  155 9.4.2.2 Remote Monitoring of the Environment with Sound in Polar Regions  155 9.4.2.3 Anthropogenic Sources of Noise in the Polar Regions  156 9.5 Summary and Future Directions  156 References  158 10

Integrating Biophony into Biodiversity Measurement and Assessment  169 Brian Michael Napoletano

10.1 Introduction  169 10.1.1 Biodiversity and its Parameterization  170 10.2 Biological Information in the Soundscape  171 10.2.1 Physiology: Sound Production and Detection  174 10.2.2 Communication: Medium and Context  176 10.2.3 Coordination: Evolution of the Biophony  178 10.2.4 Adaptation: Mechanization of the Soundscape  180 10.3 Ecoacoustics in Biodiversity Assessment  182 10.3.1 Developing a Soundscape Monitoring Network  182 10.3.2 Acoustic Data Processing and Management  183 10.4 Conclusion  184 References 185 11

11.1 11.1.1 11.1.2 11.2

Landscape Patterns and Soundscape Processes  193 Almo Farina and Susan Fuller

An Introduction to Landscape Ecology (Theories and Applications)  193 Patch Size, Shape, and Isolation  193 Patch‐Matrix Context  194 Relationship Between Landscape Ecology and Soundscape Ecology: A Semantic Approach  195 11.2.1 The Contribution of Landscape Ecology to the Development of Ecoacoustics  196 11.2.2 Acoustic Heterogeneity in a Landscape Across Space and Time  197 Acoustic Community and Landscape Mosaics  199 11.3

ix

x

Contents

11.4 Ecoacoustics in a Changing Landscape  202 11.5 Conclusion  203 References  204 12

Connecting Soundscapes to Landscapes: Modeling the Spatial Distribution of Sound  211 Timothy C. Mullet

12.1 Introduction  211 12.2 Conceptualizing Soundscapes in Space and Time  211 12.3 Capturing Soundscapes in Space and Time  212 12.4 Sound Metrics and Interpreting Nature  213 12.5 A Soundscape Metric for Modeling  215 12.6 Discriminating the Components of a Soundscape  216 12.7 Generating a Predictive Soundscape Model  217 12.8 Conclusion  219 Disclaimer  221 References  221 13

Soil Acoustics  225 Marisol A. Quintanilla‐Tornel

13.1 Introduction  225 13.2 Soil Insect Acoustics  226 13.3 Compost Activating Agent Acoustics  226 13.4 Soil Aggregate Slaking Acoustics  227 13.5 Conclusion  230 References  231 14

Fundamentals of Soundscape Conservation  235 Gianni Pavan

14.1 Introduction  235 14.2 Nature Sounds in Science and Education  238 14.3 The Role of Sound Libraries  242 14.4 Noise Pollution, the Acoustic Habitat, and the Biology of Disturbance  243 14.5 Soundscapes, Nature Conservation, and Public Awareness  244 14.6 Marine Soundscapes  245 14.6.1 Ship Noise  246 14.7 Conclusion  251 14.7.1 Terrestrial Soundscapes  252 14.7.2 Marine and Aquatic Soundscapes  252 Acknowledgment  252 References  252 15

Urban Acoustics: Heartbeat of Lansing, Michigan, USA  259 Stuart H. Gage and Wooyeong Joo

15.1 Introduction  259 15.2 Objectives  260 261 15.3 Methods 

Contents

15.3.1 Sampling Design  261 15.3.2 Recording at Sample Sites  262 15.3.3 Data Conversion  262 15.3.4 Data Processing  262 15.4 Results  264 15.4.1 The NDSI  264 15.4.2 The H, ADI, AEI, ACI, and BIO Indices  267 15.5 Discussion and Conclusions  267 References  271 16

Analytical Methods in Ecoacoustics  273 Stuart H. Gage, Michael Towsey and Eric P. Kasten

16.1 Introduction  273 16.2 Components of an Acoustic Recording  275 16.3 Visualization of an Acoustic Recording  276 16.3.1 Frequency Analysis  276 16.3.2 Three‐Dimensional Spectrogram  277 16.4 Processing Multiple Recordings  277 16.5 Analyzing Acoustic Time Series  279 16.6 Time Series of Acoustic Indices  281 16.7 Searching and Symbolic Methods  282 16.7.1 Searching a Recording for Anomalies  284 16.7.2 Symbolic Representations and Unsupervised Learning  285 16.8 Visualization and Navigation of Long‐Duration Recordings  286 16.9 Spectrogram Pyramids  289 16.9.1 Diel Plots  289 16.10 New Approaches to Analysis  291 16.11 Web Platforms for the Visualization of Environmental Audio  291 References  293 17

Ecoacoustics and its Expression through the Voice of the Arts: An Essay  297 David Monacchi and Bernie Krause

17.1 Introduction  297 17.2 Immersive Art as a Science Dissemination Tool  299 17.3 Examples of Ecoacoustic Works by Bernie Krause  302 17.4 Examples of Ecoacoustics Works by David Monacchi  306 17.4.1 Designing Temples for the Ear: The Ecoacoustic Theater  309 17.4.2 Soundscape Projection Ambisonics Control Engine (S.P.A.C.E.)  310 17.5 Conclusion  311 References  311 18

Ecoacoustics Challenges  313 Stuart H. Gage and Almo Farina

18.1 Introduction  313 18.2 Philosophical Issues  313 Ecological Issues  314 18.3

xi

xii

Contents

18.4 Sensor Technology  315 18.5 Acoustic Computations and Modeling  316 18.6 Public Information  316 18.7 Monetary Issues  317 References  317

Index  321

xiii

List of Contributors Anne C. Axel

Stuart H. Gage

Department of Biological Sciences Marshall University Huntington USA

Department of Entomology Michigan State University East Lansing USA

Giuseppa Buscaino

Wooyeong Joo

BioAcousticsLab National Research Council (IAMC-CNR) - Detached Unit of Capo Granitola (TP) Italy

Choongnam Seocheon‐gun Maseo‐Myeon Geumgang‐ro South Korea

Maria Ceraulo

Department of Pure and Applied Sciences University of Urbino Urbino Italy

Eric P. Kasten

Almo Farina

Bernie Krause

Department of Pure and Applied Sciences University of Urbino Urbino Italy

Wild Sanctuary Glen Ellen USA

Francesco Filiciotto

Conservatorio Gioachino Rossini Pesaro Italy

BioAcousticsLab National Research Council (IAMC-CNR) - Detached Unit of Capo Granitola (TP) Italy Susan Fuller

Queensland University of Technology Brisbane Australia

Michigan State University East Lansing USA

David Monacchi

Timothy C. Mullet

Ecological Services US Fish and Wildlife Service Daphne Alabama USA

xiv

List of Contributors

Brian Michael Napoletano

Marisol A. Quintanilla-Tornel

Centro de Investigaciones en Geograf ía Ambiental Universidad Nacional Autónoma de México Morelia Michoacán México

Plant and Environmental Protection Sciences University of Hawaii Manoa USA

Susan E. Parks

107 College Place Syracuse USA

Lyndsay Rankin

Northern Illinois University DeKalb USA Denise Risch

CIBRA University of Pavia Italy

Scottish Association for Marine Science (SAMS) Oban Scotland UK

Nadia Pieretti

Michael Towsey

Department of Pure and Applied Sciences University of Urbino Urbino Italy

Queensland University of Technology Brisbane Australia

Gianni Pavan

xv

Preface Discovering the importance of sound in natural processes is an important legacy of bioacoustics and human acoustics, two disciplines that have developed in the second half of the twentieth century. At that time, Aldo Leopold and Rachel Carson used acoustics to describe relevant phenomena like animal migration or the effect of chemical pollution on reproductive success of breeding birds but acoustics technology methods were rare. Their heritage is an important baseline for a new ecological perspective in the scientific investigation of sound, known as ecoacoustics, a discipline that incorporates and integrates the study of sound in both ecological and human systems. Sound is an important phenomenon including behavioral functions that range from mate performance to territory defense and social cohesion and has recently been shown to be a key issue in ecological processes. The Earth emits geological, biological, and human sounds within the biosphere, creating a sonic context that characterizes ecosystems at different spatial and temporal scales and has consequences that can affect many ecological processes. Vocal animals have a direct relationship with habitat suitability and the vocal performance of other organisms, further confirming the energy investment required to produce acoustic signals and the trade‐off between such performances, other life traits, and the availability of resources needed for their survivorship. All young disciplines, including ecoacoustics, have difficulty in tracing historical origins so there is no precise date allocated to its foundation. The use of the term “ecoacoustics” was suggested at a meeting in June 2104 at the Museum of Natural History in Paris where “soundscape ecology” was also suggested as an alternative. The assembly decided that ecoacoustics was all‐inclusive in studies of ecologically based sound and thus included soundscape ecology. With this book, we offer examples of studies, theoretical concepts, and methodologies that have evolved over the past decades in an attempt to provide a synthesis of the new discipline of ecoacoustics, although we emphasize that these are only a subset of possible examples. This book is not a celebrative edition of a consolidated ecological discipline but a contribution to transmit the principles and ideas of ecoacoustics to a wider audience. We believe that the examples of these aspects of ecoacoustics will provide an incentive for others interested in ecological sounds, including those in the sciences and the arts, to pursue their research, applying sound to solve ecological problems and to educate the next generation about the importance of ecological sounds to the survivorship of the human race.

xvi

Preface

The 18 chapters in this book cover important topics to assist others to understand the ecological significance of sounds. This introduction to ecoacoustics is intended for all who are interested in or concerned about the ecosystems in which we live and utilize for the resources that they provide. Almo Farina and Stuart H. Gage

1

1 Ecoacoustics: A New Science Almo Farina1 and Stuart H. Gage 2 1 2

Department of Pure and Applied Sciences, Urbino University, Urbino, Italy Department of Entomology, Michigan State University, East Lansing, USA

1.1 ­Ecoacoustics as a New Science Ecoacoustics is the ecological investigation and interpretation of environmental sound (Sueur and Farina 2015). It is an emerging interdisciplinary science that investigates natural and anthropogenic sounds and their relationships with the environment over multiple scales of time and space. Ecoacoustics is inclusive of the realms of ecological investigation including populations, communities, ecosystems, landscapes, and biotic regions of the Earth system. Studies of ecoacoustics in these realms can include terrestrial, freshwater, and marine systems. Ecoacoustics thus extends the scope of acoustic investigations, including bioacoustics and soundscape ecology. Ecoacoustics studies involve the investigation of sound as a subject to understand the properties of sound, its evolution, and its function in the environment. Ecoacoustics also considers sound as an ecological attribute that can be utilized to investigate a broad array of applications including the diversity, abundance, behavior, and dynamics of animals in the environment. To facilitate this emerging new science and the investigators interested in the study of ecoacoustics, the International Society of Ecoacoustics (ISE) has recently been established and details can be found at https://sites.google.com/site/ ecoacousticssociety/. For definitions of other acoustics disciplines, see Pijanowski et al. (2011) and Farina (2014).

1.2 ­Characteristics of a Sound Sound is a flow of energy in the form of lateral vibrations through a medium capable of oscillation. Sound is additive, meaning separate waves combine to form a single signal. The ear and brain manually separate this into distinct waves. The number of vibrations a sound produces per second is called frequency with a unit measurement of hertz. A spectrogram, commonly used to “see” a sound recording, is shown in Figure 1.1 where time is on the x‐axis (seconds), frequency is on the y‐axis (kilohertz), and sound intensity (energy) is on the z‐axis. The spectrogram shown is a visual representation of a Ecoacoustics: The Ecological Role of Sounds, First Edition. Edited by Almo Farina and Stuart H. Gage. © 2017 John Wiley & Sons Ltd. Published 2017 by John Wiley & Sons Ltd.

2

Ecoacoustics

Figure 1.1  A spectrogram from a recording made at site LA00 (45.53320, –84.291960 in decimal degrees) on May 4 2009 at 0600h. (See color plate section for the color representation of this figure.)

sound. The creation of a sound image requires that the sound be processed using fast Fourier transform (FFT). Creating a spectrogram using the FFT is a digital process. Digitally sampled data, in the time domain, is divided into components, which usually overlap, and Fourier transformed to calculate the magnitude of the frequency spectrum for each component. Each component then corresponds to a vertical line in the image – a measurement of magnitude versus frequency for a specific moment in time. The spectra or time plots are then “laid side by side” to form the image. The sound shown in Figure 1.1 was recorded in monaural at 22 050 Hz at site LA00 (45.53320, –84.291960 decimal degrees) on May 4 2009 at 0600h. Most of the sound in this recording occurs between frequencies 2 and 6 kHz with some high‐frequency sounds ­occurring about 8 kHz and some low‐frequency sounds at about 0.5 kHz. For those interested in the details of a mathematical treatment of acoustic signal processing, please see Hartmann (1998).

1.3 ­Sound and its Importance Hearing is one of the five key senses (hearing, vision, touch, smell, and taste) that allow organisms in the animal kingdom to relate with the environment. Hearing is an intrinsic component of the life of many organisms, including humans. Many animals use hearing to receive signals made by the environment or by other organisms. They derive meaning from these signals, which can range from danger to courtship, and these sound signals can often mean survival or a source of food. The importance of sound to humans has diminished due to evolution, since we have built habitation and created technology that we think protects us from the outside world. As our world has become louder, due to our increasing population and technological development, we

1  Ecoacoustics: A New Science

are becoming more sensitive to the importance of sound. Sound is the heartbeat of the biosphere, the places on Earth where life exists. If we can measure this heartbeat, we can determine the condition of the biosphere. When one scales from biosphere, to eco‐region, to landscape, to ecosystem or to habitat, the sounds produced within each of these realms can determine the condition of that realm if we can determine the type of sounds being emitted.

1.4 ­Ecoacoustics and Digital Sensors Ecoacoustics has been recognized as an approach to the study of species communication and census species over long periods of time. There have been significant changes in monitoring technology. Ecoacoustics has been developed thanks to instrumentation and analytical techniques. For instance, the microphone is an important sensor because this single instrument can serve many purposes for ecological investigations when connected to a recorder. The array of ecological attributes that can be determined by a microphone, which is an analog for hearing, is broad compared to other types of a­ vailable sensors (smell, taste, vision, touch). Sensors which measure other senses are important but are not yet fully applicable to the field as is the microphone, mainly due to cost. Studies of animal attributes by listening to their sounds can be a fruitful undertaking, especially if one enjoys listening to and documenting the occurrence of animal species during the dawn or nighttime chorus. However, there are many pitfalls, including change in species composition over season and time of day and the potential for misidentification of species. Errors in species identification are introduced because an observer cannot be at multiple places at the same time. Within the past decade, analog tape recorders have been replaced by digital recorders. Clocks have been added to recorders so that recordings can be made at specific times and other environmental sensors have been incorporated in the same recording machine. The length of a recording period was previously limited due to high power consumption by processors. Just a few years ago, it was not possible to record in a remote place without being there to manage the recording unit. Today, sound recorders can be programmed to suit a project’s objective, can store many recordings on removable digital media and can remain active in the field for months without intervention. This change in technology has given rise to the use of sound as an ecological attribute. Modern acoustic sensors can be used to investigate several attributes of ecological significance. These may include practical and theoretical aspects of the environment, including acoustic identification of species in terrestrial and aquatic ecosystems; the vocal behaviors of specific organisms and their physiology; the study of noise pollution; and measuring ecological processes under a climate change scenario.

1.5 ­Ecoacoustics Attributes A microphone and an automated recorder can provide an array of attributes that can have significant implications for theoretical and applied ecology. Important processes can be remotely investigated, including the number of species present, phenology of sound, trophic interactions, biological diversity, level of disturbance, diurnal and

3

4

Ecoacoustics

seasonal change of acoustic activity, level of habitat health, acoustic interactions between species, and complexity of the soundscape. 1.5.1  Population Census

Sound as a tool to survey animals has been utilized for decades (Ralph and Scott 1981). Birds are monitored by listening to the morning chorus and identifying the species based their signals at prescribed listening posts. Gage and Miller (1978) describe a long‐ term study using this method. Similar monitoring methods have utilized sound to determine species occurrence and abundance of amphibians using nighttime signaling (Karns 1986). The Breeding Bird Survey of North America (BBS) has been ongoing since the 1960s (Robbins and van Welzen 1967); it uses sound to determine avian species occurrence and this eco‐region assessment has provided one of the longest records of bird species occurrence in North America, thus enabling the assessment of change in avian species. The surveys conducted by the BBS take place during the peak of the breeding season. The BBS routes are 24.5 miles long and there are 50 stops at every 0.5 mile along the route. Routes are randomly located in order to sample habitats that are representative of the entire region (Sauer et al. 1997). Although surveys are conducted differently in Europe, sound is used to determine the occurrence of bird species in many countries. The Pan‐European Common Bird Monitoring Scheme commenced in January 2002; its main goal is to use common birds as indicators of the general state of nature using scientific data on changes in breeding bird populations across Europe (Voříšek et al. 2008). 1.5.2  Biological Diversity

Biological diversity is a complex ecological attribute to measure because it requires documentation of all species that inhabit a place. In addition, seasonal change can change biological diversity. Therefore, vegetation is commonly used as a surrogate for biological diversity. Measurement of the sound diversity at a site can begin to add information to the determination of biological diversity (Farina et al. 2005; Fuller et al. 2015; Sueur et al. 2008; Tucker et al. 2014). 1.5.3  Habitat Health

Habitat health is a relative term, but when defined by the types of sounds emitted from the site, these signals can provide an indication of the quality of that place. In fact, sounds differ in type and character depending on the types of vegetation and food available to the organisms. Benchmarks need to be established for urban, forest, grassland, and desert systems so that sounds in arrays of these systems can be compared (Fuller et al. 2015; Qi et al. 2008). 1.5.4  Time of Arrival/Departure of Migratory Species

The changing global climate is causing shifts in the arrival and departure times of animals that inhabit terrestrial and marine ecosystems (MacMynowski et al. 2007). Shifts in the areal pathways used by migratory animals to move from wintering sites to breeding sites may also be determined by measuring sounds along these marine or terrestrial routes.

1  Ecoacoustics: A New Science

1.5.5  Diurnal Change

Daily patterns of change in animal behavior can be determined by measuring sounds emitted from a place (Farina et al. 2015). Many factors can cause diurnal change and the measurement of sound along with weather information can help to describe the magnitude of the change (Gage and Axel 2013). 1.5.6  Seasonal Change

Seasonal change caused by climate shifts or physical disturbance of the Earth system due to large‐scale natural events or by land use change due to human development can be measured by recording sounds in a place. Seasonal change is also a natural occurrence. In temperate regions, there are shifts in animal behavior as seasons change. In spring, migratory populations of marine and terrestrial animals (mammals, fish, birds) move from overwintering habitats to breeding locations that can be far distant and require a great expenditure of energy. Food and habitat resources change and during this period, the sounds emitted from these organisms differ as they enter the breeding cycle (Gage and Axel 2013). 1.5.7  Competition for Frequency

The acoustic niche hypothesis (Krause 1993), an early version of the term biophony (sounds made by organisms), describes the acoustic bandwidth partitioning process that occurs in still wild biomes by which nonhuman organisms adjust their vocalizations by frequency and time‐shifting to compensate for acoustic habitat occupied by other vocal creatures. Thus each species evolves to establish and maintain its own acoustic bandwidth so that its voice is not masked (Malavasi and Farina 2013). For instance, examples of clear partitioning and species discrimination can be found in the spectrograms derived from the biophonic recordings made in most uncompromised tropical and subtropical rainforests (Krause 1993). 1.5.8  Trophic Interactions

Many species of organisms do not emit audible sounds but those that do emit acoustic signals may depend on organisms that do not. Therefore, the presence of those that do not emit sounds may be deduced by quantifying the sounds for those that produce auditory signals. Consider birds and their food source. A wood thrush sings a beautiful song in undisturbed forests and searches and feeds on worms and other food that occurs on the forest floor. Although the food sources do not make audible sounds, the wood thrush would not occur in the habitat if it were not for the resources found there. When we hear the sound of the thrush, we can infer that there are food resources nearby and thus identify trophic interactions. 1.5.9 Disturbance

Disturbance can be caused by natural events (hurricanes, volcanoes, fires, floods) or by human‐caused events (mining, urbanization, forest harvest, spraying). Such events are characterized by acoustic emissions. The measurement of sounds (noise) caused by disturbance can indicate the type and duration of the disturbance. The term technophony,

5

6

Ecoacoustics

the sounds made by machines, is used to characterize disturbance and can occur when an overabundance of machine sounds from aircraft, automobiles, watercraft, chain saws, etc. dominates a habitat. Usually technophony occurs at lower sound frequencies than biota so it is possible to use sound to quantify disturbance. 1.5.10  Sounds of the Landscape and People

Every landscape has a specific acoustic signature that is the result of the mixture of all the physical and biological acoustic agents. The measurement of the sounds emitting from a place can provide an enjoyable experience to the listener. Listening to recordings of the howl of a coyote, the yodel of a common loon or the song of a thrush can conjure up memories of a place long forgotten. Figure 1.2 provides a summary of the value of sound ranging from population census to quality assessment of the landscape for human well‐being.

1.6 ­Ecoacoustics and Ecosystem Management There are two aspects of sound that relate to ecosystem management: ●●

as a response indicator by estimating the diversity of vocal organisms; determining the relative proportions of human and natural activity; characterizing the daily and

Ecoacoustics attributes Population census Biological diversity Habitat health Time of arrival/departure of migratory species Diurnal changes Seasonal changes Competition for frequencies Trophic interactions Disturbance regime Sound of the landscape and people

Figure 1.2  Ecoacoustics has several competencies in environmental surveys, ranging from population census to quality assessment of landscape for human well‐being.

1  Ecoacoustics: A New Science

●●

long‐term trends of human and biological activity; and measuring sound in response to changes in land use. as a stress indicator by examining the effects of human activity on organism communication during critical functions (e.g. reproduction, food tracking, migration, etc.); determining the causes of natural population declines in organisms sensitive to human disturbance or to climate change (Krause and Farina 2016).

Sound can also be used as a management tool to regulate the amount of noise that is tolerable to humans (Farina 2014, pp. 263–296). Sound maps of urban areas, airports, manufacturing zones, and parks can be useful tools to guide the development of sound abatement regulations. Measurement of sound can be used to identify and characterize the amount of technology (trucks, cars, boats, ships, jet skis, snow machines) and the length and intensity of human‐kept animals (dogs, roosters) which can be a local disturbance.

1.7 ­Quantification of a Sound 1.7.1  Species Identification

One can listen to the sounds in a recording and identify the entities recorded. Haselmayer and Quinn (2000) compared field observations using the point‐count method of species identification by listening to recordings made at the time of the point‐count and found that they are highly correlated. Joo (2009) conducted a breeding bird survey and also identified species in simultaneous recordings and found a high correlation as well. Kasten et al. (2012) provide a method to catalog species heard in a recording using a web‐based tool. Automated species identification has been found to be complex due to the variability within species of songs and calls and the overlap in frequencies caused by sound emitters. Butler et  al. (2007) used signatures extracted from spectrograms to search other spectrograms for that signature providing the probability of match to that signature. Match probabilities are closer to 1 for simple signatures (insects, amphibians) compared to more complex signatures (birds). However, new approaches to this problem have made major improvements in automation of species identification (Acevedo et al. 2009; Dong et al. 2015; Duan et al. 2013). To quantify sounds recorded in the environment, the spectrogram representation can be used to create acoustics indices by dividing the spectrogram into frequency intervals and counting the pixels in each interval (Napoletano 2004). The spectrogram can also be used to select signatures of a species from the image and search a series of spectrograms for that signature (Butler et al. 2007). Since these studies were undertaken, there has been considerable improvement in the development of acoustics indices and species recognition algorithms. 1.7.2  Acoustic Indices

Acoustic indices are derived from environmental recordings that do not depend on the species that occur in the recordings but rather on the characteristics of the recording, including the diversity of the sounds in the recording, the complexity of the sounds, the degree of evenness of the sounds, or ratios of frequencies in the sounds.

7

8

Ecoacoustics

Seewave, a package in R developed by Sueur et al. (2008), provides functions for analyzing, manipulating, displaying, editing, and synthesizing time waves. This package processes time analysis (oscillograms and envelopes), spectral content, resonance quality factor, entropy, cross‐correlation and autocorrelation, zero crossing, frequency coherence, dominant frequency, analytic signal, 2D and 3D spectrograms. Seewave enables a user to compute acoustic indices including H (Sueur et al. 2008), the Acoustic Complexity Index (ACI) (Pieretti et al. 2011), and the Normalized Difference Soundscape Index (NDSI) (Kasten et al. 2012). Soundecology, another R package focusing on acoustics, was developed by Villanueva‐ Rivera et al. (2011) and enables a user to compute values for acoustic indices where one can specify the acoustic index and its parameters. Acoustics indices in R‐Soundecology include the ACI (Pieretti et al. 2011), the Acoustic Diversity Index (Villanueva‐Rivera et al. 2011), the Acoustic Evenness Index (Villanueva‐Rivera et al. 2011), the Bioacoustic Index (Boelman et al. 2007) and the NDSI (Kasten et al. 2012). These indices and other techniques used to interpret environmental recordings are discussed in Chapter  16. A  procedure to detect and identify acoustic events, the Ecoacoustic Event Detection and Identification (EEDI) developed by Farina et al. (2016). is powered by free access software, the SoundscapeMeter 2.0 (Farina and Salutari 2016).

1.8 ­Archiving Ecoacoustics Recordings The new types of automated recorders can be programmed to record sounds based on project objectives. Recording may be continuous or recorders may be programmed to sample the environment by having the recorder wake up, record for a length of time, then sleep until the internal clock tells the recorder to wake and record again. There are many recording options that were not possible just a few years ago. For instance, recorders can be set to record continuously for one hour before sunrise to one hour after sunrise. One can purchase such recorders from companies like Wildlife Acoustics (www.wildlifeacoustics.com), Lunilettronik (www.lunilettronik.it/) or Frontiers Lab (www.frontierlabs.com.au/) or one can construct automated recorders (Aide et al. 2013; Farina et al. 2014; Gage et al. 2015; Mason et al. 2008; Wimmer et al. 2013). These types of recorders can amass many recordings. For example, a project which has been in operation since 2009 has made over 500 000 recordings to date from 12 sites at 30‐­minute intervals, each one minute in length (www.real.msu.edu/projects/one_proj.php?proj=la). The start and end recording dates are different depending on the intent. This requires an infrastructure to enable computation of sound metrics, storage of the sounds and their associated metrics and then retrieval of the sounds and/or the metrics for analysis. The Remote Environmental Assessment Laboratory’s Digital Acoustic Library System has these features and is described in Kasten et al. (2012), while Villanueva‐Rivera and Pijanowski (2012) described “Pumilio,” a web‐based system to archive acoustic recordings. One may ask “Why keep all these recordings?” The answer is simple: “When the project began in 2009, automated species recognition was a dream. Now it is becoming a reality.” We can then use these historical recordings to automatically identify the species in the database (Aide et al. 2013; Dong et al. 2015). To complement the issues involved in automated species identification, methods have been developed to search for specific

1  Ecoacoustics: A New Science

frequency intervals within the digital database since vocal organisms often signal within a range of frequencies (Kasten et al. 2012).

1.9 ­Ecological Forecasting We depend on the Earth’s natural resources to sustain our economies and our life support. However, we are exploiting these resources at an unprecedented rate and thus undermining our economies and life support systems. This is a critical time in human history and we have the responsibility to measure and assess the effects of biological, chemical, physical, and human‐induced change on the Earth system and its function. Ecological forecasts predict the effects of biological, chemical, physical, and human‐ induced changes on ecosystems. The ecological science community is entering a new era in which forecasts of ecological change can become commonplace if we bring to bear new tools, monitoring and observing systems, and increased understanding available today and on the horizon. We are poised to capitalize on new opportunities as we significantly change the way we anticipate and manage ecological risk. Sound is one of the key ecological attributes that can be used to monitor the heartbeat of the biosphere and thus enable ecological forecasting. The advent of automated sensors is revolutionizing environmental monitoring and leading to new thrusts in environmental research and education, including ecological forecasting (NSF 2015).

­References Acevedo, MA, Corrada‐Bravo, CJ, Corrada‐Bravo, H, Villanueva‐Riverad, LJ and Aide, TM (2009) Automated classification of bird and amphibian calls using machine learning: a comparison of methods. Ecological Informatics, 4, 206–214. Aide, TM, Corrada‐Bravo, C, Campos‐Cerqueira, M, et al. (2013) Real‐time bioacoustics monitoring and automated species identification. PeerJ, 1, e103. Boelman, NT, Asner, GP, Hart, PJ and Martin, RE (2007) Multi‐trophic invasion resistance in Hawaii: bioacoustics, field surveys, and airborne remote sensing. Ecological Applications, 17, 2137–2144. Butler, RM, Servilla, Gage, S, et al. (2007) CyberInfrastructure for the analysis of ecological acoustic sensor data: a use case study in grid deployment. Cluster Comp, 10, 301. Dong, X, Towsey, M, Truskinger, A, Cottman‐Fields, M, Zhang, J and Roe, P (2015) Similarity‐ based birdcall retrieval from environmental audio. Ecological Informatics, 29, 66–76. Duan, S, Zhang, J, Roe, P, et al. (2013) Timed Probabilistic Automaton: A Bridge between Raven and Song Scope for Automatic Species Recognition. Paper presented at the Twenty‐Fifth IAAI Conference. Farina, A (2014) Soundscape Ecology: Principles, Patterns, Methods and Applications, Springer Science+Business Media, Dordrecht. Farina, A and Salutari, P (2016) Applying the Ecoacoustic Event Detection and Identification (EEDI) model to the analysis of acoustic complexity. Journal of Mediterranean Ecology, 14, 13–42. Farina, A, Bogaert, J and Schipani, I (2005), Cognitive landscape and information: new perspectives to investigate the ecological complexity, BioSystems,79, 235–240.

9

10

Ecoacoustics

Farina, A, Pieretti, N and Piccioli L (2011), The soundscape methodology for long‐term bird monitoring: A Mediterranean Europe case‐study. Ecological Informatics, 6, 354–363. Farina, A, James, P, Bobryk, C, Pieretti, N, Lattanzi, E and McWilliam, J (2014) Low cost (audio) recording (LCR) for advancing soundscape ecology towards the conservation of sonic complexity and biodiversity in natural and urban landscapes. Urban Ecosystems, 17, 923–944. Farina, A, Ceraulo, M, Bobryk, C, Pieretti, N and Lattanzi, E (2015) Spatial and temporal variation of bird dawn choruses in a Mediterranean landscape. Bioacoustics, 24, 269–288. Farina, A, Pieretti, N, Salutari, P, Tognari, E and Lombardi, A (2016) The application of the Acoustic Complexity Indices (ACI) to Ecoacoustic Event Detection and Identification (EEDI) modeling. Biosemiotics, 9, 227–246. Fuller, S, Axel, AC, Tucker, D and Gage, SH (2015) Connecting soundscape to landscape: which acoustic index best describes landscape configuration?, Ecological Indicators, 58, 207–215. Gage, SH and Axel, AC (2013) Visualization of temporal change in soundscape power of a Michigan lake habitat over a 4‐year period. Ecological Informatics, 21, 100–109. Gage, SH and Miller, CA (1978) A Long‐Term Bird Census in Spruce Budworm‐Prone Balsam Fir Habitats in Northwestern New Brunswick. Information Report M‐X‐84. Fisheries and Environment Canada, Canadian Forest Service, Maritimes Forest Research Centre, Fredericton. Gage, SH, Joo, W, Kasten, EP, Fox, J and Biswas, S (2015) Acoustic observations in agricultural landscapes, in The Ecology of Agricultural Ecosystems: Long‐Term Research on the Path to Sustainability (eds S.K. Hamilton, J.E. Doll and G.P. Robertson), Oxford University Press, New York, pp. 360–377. Hartmann, WM (1998) Signals, Sound and Sensation (Modern Acoustics and Signal Processing), Springer, New York. Haselmayer, J and Quinn, JS (2000) A comparison of point counts and sound recording as a bird survey method in Amazonian southeast Peru. The Condor, 102, 887–893. Joo, W (2009) Environmental Acoustics as an Ecological Variable to Understand the Dynamics of Ecosystems, PhD dissertation, Michigan State University, East Lansing. Karns, DR, (1986) Field Herpetology: Methods for the Study of Amphibians and Reptiles in Minnesota, Museum of Natural History, Occasional Paper 18, University of Minnesota, Minneapolis. Kasten, E, McKinley, P and Gage, SH (2010) Ensemble extraction for classification and detection of bird species. Ecological Informatics, 5, 153–166. Kasten, EP, Gage, SH, Fox, J and Joo, W (2012) The remote environmental assessment laboratorys acoustic library: an archive for studying soundscape ecology. Ecological Informatics, 12, 50–67. Krause, B (1987) Bioacoustics, habitat ambience in ecological balance. Whole Earth Review, 57. Krause, B (1993) The niche hypothesis. Soundscape Newsletter, 6, 6–10. Krause, B and Farina, A (2016) Using ecoacoustic methods to survey the impacts of climate change on biodiversity. Biological Conservation, 195, 245–254. MacMynowski, DP, Root, TL, Ballard, G and Geupel, G (2007) Changes in spring arrival of Neararctic‐Neotropical migrants attributed to multi‐scalar climate. Global Change Biology, 13, 1–13.

1  Ecoacoustics: A New Science

Malavasi, R and Farina, A (2013) Neighbours talk: interspecific choruses among songbirds. Bioacoustics, 22(1), 33–48. Mason, R, Roe, P,Towsey, M, Zhang, J, Gibson, J and Gage, SH (2008) Towards an Acoustic Environmental Observatory. Paper presented at the 4th IEEE International Conference on e‐Science, Indianapolis, IN. DOI: 10.1109/eScience.2008.16:135‐142. Napoletano, BM (2004) Measurement, Quantification and Interpretation of Acoustic Signals within an Ecological Context. MS thesis. Michigan State University, East Lansing. NSF (2015) America’s Future for Environmental Research and Education for a Thriving Century: A 10 Year Outlook. NSF Advisory Committee for Environmental Research and Education, Washington, DC. Pieretti, N, Farina, A and Morri, D (2011) A new methodology to infer the singing activity of an avian community: the Acoustic Complexity Index (ACI). Ecological Indicators, 11, 868–873. Pijanowski, BC, Farina, A, Gage, SH, Dumyahn, S, Krause, B (2011) What is soundscape ecology? Landscape Ecology, 2, 1213–1232. Qi, J, Gage, SH, Joo, W, Napoletano, B and Biswas, S (2008) Soundscape characteristics of an environment: a new ecological indicator of ecosystem health, in Wetland and Water Resource Modeling and Assessment (ed. W. Ji), CRC Press, New York, pp. 201–211. Ralph, JC and Scott, JM (1981) Estimating numbers of terrestrial birds, Studies in Avian Biology No. 6, Cooper Ornithological Society and Allen Press, Lawrence. Robbins, CS and van Welzen, WT (1967) The Breeding Bird Survey, 1966, US Department of the Interior, Fish and Wildlife Service, Bureau of Sport Fisheries and Wildlife, Washington, DC. Sauer, JR, Hines, JE, Gough, G, Thomas, I and Peterjohn, BG (1997) The North American Breeding Bird Survey Results and Analysis, Version 96.4, Patuxent Wildlife Research Center, Laurel. Sueur, J and Farina, A (2015) Ecoacoustics: the ecological investigation and interpretation of environmental sound. Biosemiotics, 8, 493–502. Sueur, J, Aubin, T and Simonis, C (2008) Seewave: a free modular tool for sound analysis and synthesis. Bioacoustics, 18, 213–226. Sueur, J, Pavoine, S, Hamerlynck, O and Duvail, S (2008) Rapid Acoustic Survey for Biodiversity Appraisal. PLoS One, 3, e4065. Tucker, D, Gage, SH, Williamson, I and Fuller, S (2014) Linking ecological condition and the soundscape in fragmented Australian forests. Landscape Ecology, 29, 745–758. Villanueva‐Rivera, LJ and Pijanowski, BC (2012) Pumilio: a web‐based management system for ecological recordings. Bulletin of the Ecological Society of America, January. Villanueva‐Rivera, L.J, Pijanowski, BC, Doucette, J, and Pekin, B. (2011) A primer of acoustic analysis for landscape ecologists. Landscape Ecology, 26, 1233–1246. Voříšek, P, Klvaňová, A, Wotton, S and Gregory, RD (eds) (2008) A Best Practice Guide for Wild Bird Monitoring Schemes, CSO/RSPB. Wimmer, J, Towsey, M, Planitz, B, Williamson, I and Roe, P (2013) Analyzing environmental acoustic data through collaboration and automation. Future Generation Computer Systems, 29, 560–568.

11

13

2 The Duality of Sounds: Ambient and Communication Almo Farina1 and Stuart H. Gage2 1 2

Department of Pure and Applied Sciences, Urbino University, Urbino, Italy Department of Entomology, Michigan State University, East Lansing, USA

2.1 ­Introduction In this chapter we address the ontology of sounds, their nature and function. Sounds are often considered as a means to communicate, but in this chapter sound is considered as a component of the environment and a passive source of information for animals and humans. Every organism requires precise information from the external world to make the right choice at the right time and finally to intercept the necessary resources to maintain itself and to accomplish vital functions (Farina 2012). The perception of the environment is committed to a species‐specific censorship and to a successive cognitive and “cultural” elaboration of the signals that are coming from the external world (Farina et al. 2005). If visual information is dominant in diurnal species, olfactory, tactile, and magnetic information prevails in nocturnal animals, whereas the chemical information is common between many organisms, from plants to humans, regardless of the presence or absence of light. This information is mediated by olfactory and tasting sensors. Acoustic information is a further important source of knowledge utilized by diurnal and nocturnal organisms in terrestrial and aquatic systems in order to communicate and detect the surrounding environment. Acoustic information is very common in nature, and its origin may be geophonic, biofonic or technophonic. Acoustic information to which animals are sensitive has been explored for a long time with bioacoustics and behavioral approaches that describe the “anatomy” of sounds by partitioning song and call sequences into elementary components. The role of ambient sound is a subject that is rarely considered in ecological research and remains more popular in human acoustics (Davies et al. 2013). In reality, every organism is embedded in a sonic environment, the characteristics of which represent reference points to enable the accomplishment of important functions. Animals living in groups, such as titmice, use extensively heterospecific signals that inform individuals about the presence of a threat (predators, humans) (Langham et al. 2006) and avoid areas dominated by strong or permanent noise that could mask their acoustic signals (Francis et al. 2009). Acoustic communities, such as temporary aggregations of singing species, are created according to an interspecific self‐organized Ecoacoustics: The Ecological Role of Sounds, First Edition. Edited by Almo Farina and Stuart H. Gage. © 2017 John Wiley & Sons Ltd. Published 2017 by John Wiley & Sons Ltd.

14

Ecoacoustics

communication design (Farina and James 2016). Acoustic communities are sonic broadcasting centers that create a temporary sonic environment that in part copes with environmental conditions such as aspect, humidity, wind, and human sounds and that is strongly affected by the presence of active species. Acoustic communities are distributed across a landscape in a variable geometry due to the uneven distribution of resources and of interacting individuals.

2.2 ­Vegetation and Ecoacoustics There are three major components that enable a sound‐producing species to survive: a place to live, a place to reproduce, and food resources. Different species of vegetation provide these opportunities but differ in different ecosystems. For example, there are more vocal species in the tropics because there are more species of vegetation than in temperate systems. In temperate systems, there are more species in ecosystems that have a diversity of vegetation. The same is true for desert systems. The more species of vegetation, the more vocal species there are. Many vocal organisms in northern ecosystems migrate to southern areas where there are more abundant resources for survival. Because vocal animals require food for survival, more species occur in ecosystems which produce a diversity of food. The Earth system is losing both complex and simple ecosystems which contain vegetation upon which vocal species rely for existence. These losses are due to deforestation and habitat fragmentation caused by expansion of croplands and pastures. Biodiversity loss caused by humans is identified as a major and challenging problem globally (Pimm et al. 2006), and threats to species and ecosystems will continue (Pereira et  al. 2010). Examples include areas in South America, where significant changes in cropland expansion occurred between 1960 and1990 (Ramankkutty et al. 2002). Most of the cropland expansion occurred in the Brazilian Cerrado (woodland savannah), a region that has recently lost more than 9 million km (Brasil 2009; Klink and Machado 2005). In Borneo, remote sensing shows that the forest area has declined by 30.2% since 1973 (Gaveau et al. 2014). Therefore, specific long‐term datasets are required to assess the dynamics of biodiversity in the region, which at present remain largely unknown (Molleman et al. 2006). Canada, Russia, and Brazil contain 65% of all the world’s intact forest landscapes (IFL) but these forests are becoming increasingly disturbed. In Canada, four provinces, Quebec, Alberta, Ontario, and British Columbia, account for 71% of the 216 199 km2 of human disturbances (Global Forest Watch Canada 2016). The fragmented forest remnants of south‐east Queensland, Australia, are noted for high biodiversity value and increased pressure associated with habitat fragmentation and urbanization (Tucker et al. 2014). How biodiversity responds to habitat loss and fragmentation is one of the key topics in ecology and conservation biology (Sala et  al. 2000). There is little understanding about the response of species and communities to human‐induced stress (Gardner et al. 2009). Considering the rapid deforestation rate observed, it is important to develop and apply methods that can be effective for biodiversity assessment. Ecoacoustics techniques have been used in behavioral studies, and now these are also being applied to problems in conservation biology (Farina 2014; Ritts et  al. 2016; Sueur et  al. 2008; Towsey et al. 2014).

2  The Duality of Sounds: Ambient and Communication

2.2.1  Vegetation Quality and Ecoacoustics

Tucker et al. (2014) studied 10 sites defined by a distinct open eucalypt forest community dominated by spotted gum (Corymbia citriodora ssp. variegata), which were stratified based on patch size and patch connectivity. Each site underwent a series of detailed vegetation condition and landscape assessments, together with bird surveys and acoustic analysis using relative soundscape power. Univariate and multivariate analyses indicated that the measurement of relative soundscape power reflects ecological condition and bird species richness, and is dependent on the extent of landscape fragmentation. The authors concluded that acoustic monitoring technologies provide a cost‐effective tool for measuring ecological conditions, especially in conjunction with established field observations and recordings. Fuller et al. (2015) examined how soundscape patterns vary with landscape configuration and condition. The goal of the study was to examine a suite of published acoustic indices to determine whether they provide comparable results relative to varying levels of landscape fragmentation and ecological condition in 19 forest sites in eastern Australia. The study revealed that two indices, the Acoustic Complexity Index (Pieretti et al. 2011) and the Bioacoustics Index (Boelman et al. 2007), presented a similar pattern that was linked to avian song intensity, but was not related to landscape and biodiversity attributes. Two soundscape diversity indices, acoustic entropy (Sueur et al. 2008) and acoustic diversity (Villanueva‐Rivera and Pijanowski 2011), and the Normalized Difference Soundscape Index (NDSI) (Gage and Axel 2013) revealed a high incidence of nighttime sounds, as well as a peak occurrence of sound energy at dawn and dusk ­chorus. The three indices that best connected the soundscape with landscape characteristics, ecological condition, and bird species richness were acoustic entropy (Sueur et al. 2008), acoustic evenness (Villaneuva‐Rivera and Pijanowski 2011), and the Normalized Difference Soundscape Index (Gage and Axel 2013; Kasten 2012). The study showed that remote soundscape assessment can be implemented as an ecological monitoring tool in fragmented Australian forest landscapes. 2.2.2  Soundscape Indices and Biodiversity

Gasc et  al. (2015) examined the limitations and bias in acoustic biodiversity indices. They revealed that none of the indices tested was able to represent species richness accurately under field conditions. Consequently, further work is required to validate the meaning of biodiversity indices. It is reported that habitat structure can place significant constraints on the development of acoustic signals (Aylor 1971). For example, ­species from areas of dense vegetation tend to exhibit lower frequency sounds and have narrower frequency ranges (McCracken and Sheldon 1997) because these are subject to less attenuation by vegetation than high‐frequency sounds (Boncoraglio and Saino 2007). Acoustic indices exploit this to assess diversity by calculating the occupancy of different frequency bands, which can be used to represent different species (Villanueva‐ Rivera et al. 2011). The complexity of the soundscape can, therefore, be linked to the number of species present in the landscape. Although it is argued that biodiversity indices neglect the multidimensional nature of biodiversity (Purvis and Hector 2000), the development of an index is attractive because an index can distil biodiversity into one or more values that can be easily tracked through time and presented to policy makers. Sueur et al. (2008) presented two indices,

15

16

Ecoacoustics

the H index, which measures species richness, and the D index, which calculates acoustic dissimilarity between communities. Since this benchmark paper was published, there has been a proliferation of indices that have considered specific environmental traits (Depraetere et al. 2012; Rodriguez et al. 2014) and species (Pieretti et al. 2011). A recent review of the development of these indices suggests that there are 21 different alpha acoustic indices and seven beta diversity measures (Sueur et al. 2014). As described above, the majority of these indices use measures of amplitude or frequency to determine diversity (Depraetere et al. 2012; Gasc et al. 2013; Pieretti et al. 2011; Sueur et al. 2008; Villanueva‐Rivera et al. 2011). Using a different approach, Qi et al. (2008) describe how sounds are decomposed into 1 kHz frequency intervals using power spectral density (PSD) values (Welch 1967) for each frequency interval in a sound recording. The frequency interval 0–1 kHz was not included. After normalizing the PSD values (0–1) (Kasten et al. 2012), the patterns of these frequencies are shown in Gage and Axel (2013) (see Figure 4 for six frequency intervals). Then frequency intervals were assigned to mechanical signals (technophony) (1–2 kHz), and to biological sounds (biophony) (2–11 kHz). The NDSI is computed based on these metrics. Like the Normalized Difference Vegetation Index (NDVI), the NDSI uses a simple algorithm to compress a large amount of information into an index to characterize ecological disturbance. The NDSI is calculated as follows: NDSI= (biophony − technophony ) / (biophony + technophony ) where biophony is ∑ (2–11 kHz) and technophony is 1–2 kHz (Gage and Axel 2013; Kasten et al. 2012). The NDSI can then range from –1 to +1 where low values of the index indicate dominance of lower frequencies (technophony) and higher values show dominance of higher frequencies (biophony). The NDSI can be used to determine the dominance of sound types (biophony, technophony) in ecosystems and this simple index may provide a tool for rapid ecosystem assessment since it does not require species identification. Although these components were determined in a temperate environment, they have been shown to be useful in east‐central Australia (Fuller et al. 2015) and in a Brazilian forest ecosystem where the NDSI has been a useful index to characterize disturbance and landscape‐scale characteristics. 2.2.3  Applications of Remote Sensing of Vegetation and Ecoacoustics

Pekin et al. (2012) determined the relationship between acoustic diversity and metrics of vertical forest structure derived from light detection and ranging (LIDAR) data in a neotropical rainforest in Costa Rica. They used the LIDAR‐derived metrics to predict acoustic diversity across the forest landscape. Sound recordings were obtained from 14 sites for six consecutive days during dusk chorus (1800h). Acoustic diversity was calculated for each day as the total intensity across acoustic frequency bands using the Shannon Index and then averaged over the six days at each site. Acoustic diversity was modeled for forested areas (where canopy height was >20 m) at 20 m resolution using coefficients obtained from the multiple linear regression (MLR), and a hotspot analysis was conducted on the resulting layer. Acoustic diversity was strongly correlated with the LIDAR metrics (R2 = 0.75), suggesting that LIDAR‐derived metrics can be used to determine canopy structural attributes important to vocal fauna species. The hotspot

2  The Duality of Sounds: Ambient and Communication

analysis revealed that the spatial distribution of these canopy structural attributes across the La Selva forest is not random. Pekin et al. (2012) argue that this approach can be used to identify forest patches of potentially high acoustic diversity for conservation or management purposes. Skole et al. (2013) state that by stacking annual vegetation index (VI) datasets as a single remote sensing data product where clearings (harvests) and regrowth can be observed, analyzed, and reported for area extent, individual patch sizes, harvest cycle periods of industrial forests, as well as their changes over time can be observed. These methods were also developed based on a spectral mixture analysis in combination with visual interpretation to quantify the forest fractional cover. In other words, these authors used spectral endmembers analysis that produced a forest fractional cover dataset. This, in turn, could be used to identify where in forests there has been logging and forest degradation (Skole et al. 2013). The NDVI described by He et al. (2009) was used as a proxy for the vegetation structure and the NDSI (Gage and Axel 2013; Kasten et al. 2012) to compare vegetation and sound quality in two sites in Brazil which differed in these properties. Areas covered by dense vegetation, such as gallery forest, were found to have higher NDVI values whereas areas with less biomass, such as the Cerrado and grasslands, were found have lower NDVI values . The authors also found that there was a spatial association between NDSI values and noise sources regardless of the type of environment (Cerrado or gallery forest) where points located close to residential districts had a lower NDSI value. Conversely, higher NDSI values were observed away from such noise, showing that lower NDSI values were found near noisy places in both habitat types.

2.3 ­Acoustic Resources, Umwelten, and Eco-fields “Resource”, from the Latin “spring (surgere) again (re),” is a term used in both ecology and economics, and defined by the Oxford English Dictionary as: “a stock or supply of materials or assets that can be drawn on in order to function effectively.” This term may be applied to every natural element (abiotic and biotic) as well as to any man‐made element, material or immaterial, which, after having been used/consumed, regenerates itself by way of its internal, independent mechanisms (Farina 2012). Resources can be distinguished according to different categories, including the consistency of the material, the origin of the material (abiotic, biotic), the abundance of the material (limited, unlimited), and the semiotic mechanisms involved (indexical, iconical, and symbolic). Resources that are species specific are scarce in nature, and are the fundamentals that regulate the carrying capacity of a system. These resources are heterogeneously distributed in time and space and accurate interpretation of their meaning is needed in order to reduce the energetic cost of organisms. Resources include matter, energy, information, meaning, and culture (Farina 2012). According to this, sounds can be described as energy + information + meaning, and, in some cases, culture (Lemon 1975). Sounds are resources for a great variety of organisms and sounds have different levels of importance. In humans, music is a good example of a cultural resource important to create a condition of well‐being. Organisms and resources are connected by semethic interactions, which means a “personal” reciprocal knowledge between two subjects (Hoffmayer 2008). This interpretation

17

18

Ecoacoustics

is unknown in general ecology, but is strongly supported by biosemiotic models like the Umwelt theory (von Uexküll 1982) and the eco-field theory (Farina and Belgrano 2004, 2006). The Umwelt theory considers the private or subjective surroundings at the scale of species and individual (von Uexküll 1982). When interacting with the environment, organisms reshape and create their own “umwelt.” In this way, every individual has a unique umwelt and the interaction between different umwelts produces a semiospheric network (von Uexküll 1957). The umwelt is an important theory because it connects behavioral and ecological aspects of a species by recognizing the individual subjectivity of the perceived world and assures the individual‐specific reaction to external stimuli. A large part of the complexity observed in the real world is connected with the subjectivity of individuals. This concept is relevant in ecoacoustics because it may explain the ­acoustic dimension of the world that surrounds every individual. According to the physiological and social status of individuals, the sonic environment may have a different consistency, shape, and meaning. To track resources, a sequence of biosemiotic steps is set in motion, trigged by a physiological need that in turn activates the necessary function. This last step requires a spatial configuration carrier of meaning (the eco-field) to track a resource. The theory of the eco-field originated with the umwelt concept (Farina and Belgrano 2004, 2006), utilizes the Peirce semiotic process of sign (Atkin 2013). According to eco-field theory, every organism tracks a resource utilizing a biosemiotic trick. It searches a spatial configuration that facilitates the interception of the resource necessary at a specific time to accomplish a specific need (mating, roosting, singing). Need  –  Function  –  Semiotic interface – Resource describes the hierarchical mechanism necessary to track resources (Figure 2.1). A specific physiological need is the active indicator of the level of scarcity of a resource within the body of an organism. Every need produces a specific function

Need

Function

Cognitive template

Eco-field

Resource

Figure 2.1  Need and resources are connected by a chain of semiotic processes in which need activates a function that in turn elicits a cognitive template necessary to locate an eco-field where resources are expected to be.

2  The Duality of Sounds: Ambient and Communication

and this function changes by using a cognitive template, altering the sensorial relationship between the individual and the external world. Resources are intercepted using a semiotic interface that generally has spatial properties and recognizable configurations. A function, activated by a need (scarcity of a ­specific resource), can be considered as a string of instructions to locate the specific resource. Every function produces the emergence of a cognitive template that is compared with the spatial configuration of the environment. When the eco-field model is applied to sound, we assume the presence of acoustic eco-fields (Farina 2014, p. 22). Sounds can be used to defend a territory, to confound predators, to locate food sources, to find partners, etc. For every function activated using acoustic mechanisms, spatial acoustic configurations are activated. For instance, the distribution of a group of singing males around an individual is assessed using a specific cognitive template that fixes the distance and abundance of singing males that allows the bird to have free space to determine its territory (Figure 2.2). In the same way, the calls in a bird flock (e.g. a tit assemblage) is an eco-field that indicates the presence of required resources. For instance, in Mediterranean forests during winter, some species are acoustically active in areas rich with food (e.g. abundant

No

Cognitive template

Yes

Territory resource

Figure 2.2  According to the eco-field model, a cognitive template is activated as reference for the location of an adapt (acoustic) eco-field. In this simplified model, a “territory resource” is found when the cognitive template is coincident with the acoustic (territory) eco-field. In this model, a free space (in gray) is the territory resource surrounded by singing territorial individuals (black dots).

19

20

Ecoacoustics

fruits on ivy) and these acoustic communities attract visiting migrants and mobile species. The calls used by individuals to search for food create an acoustic community that represents an eco-field, a spatial configuration carrier of meaning, used by eavesdropping to track food resources (ivy fruit in this case) and that represent the “object” of Peirce semiotics. The theory of the eco-field applied to ecoacoustics depicts an acoustic environment that is variable according to individual needs. Every individual may find different acoustic eco-fields in the same location according to the different needs that it has for survival. Finally, the ensemble of different acoustic eco-fields becomes the acoustic habitat (Merchant et al. 2015) of that individual. The level of resource interception in a location depends on the presence of a resource and often not all resources needed for survival are available. Therefore, every location may have resource sources or sinks according to the Pulliam model (Pulliam 1988, 1996).

2.4 ­Sounds as Biological Codes According to Marcello Barbieri (2003, 2013), “information is a nominable entity, a fundamental but not‐computable observable” which is activated by the use of organic codes. Ecological codes have been defined by Farina and Pieretti (2014) as: … mechanisms that establish an arbitrary set of connections between two or more components (organisms and/or their aggregations) of a complex system. Ecological codes are tools that organisms use in everyday life to relate to the environment. Ecological codes are visual, acoustic, tactile, chemical, and cultural and exist at every scale of the living organization. Ecological codes classify external objects and process analog gradients as discrete meaningful digital units. In birds, for example, distance from a safe place is transformed into discrete distance units, each associated with a specific habit (e.g. stay confident, quick escape, alerting, etc.). Such transformations are performed by cognitive code makers and produce codes that are incorporated into the genetic or cultural reservoir of every organism (www.codebiology.org). Acoustic codes are a type of ecological code and are used by vocal animals to maintain intra‐ and interspecific communication. They are characterized by a sequence of nominal entities (syllables, words, and sentences) and by magnitude modulation, recently defined by Farina (2014) as: units of informational acoustics carried out by special organs (e.g. syrinx in birds, vocal cords in humans, tymbals in some insects) with specific sequences (song, contact calls, alarm calls, etc.) to produce meaning and have a bivalent nature (behavioral and ecological). Acoustic codes result in frequency partitioning to reduce interspecific competition. Acoustic codes are plastic, can be adapted to environmental change and are subordinated to the energetic investment of broadcasting individuals. The phenological plasticity of acoustic codes has been demonstrated in areas dominated by noise of human

2  The Duality of Sounds: Ambient and Communication

origin (see Chapter  6). The variation of codes is evident across the landscape where different types of sonic context are connected to a thermodynamic gradient (Odum 1983), by a negentropic regime (Schrödinger 1944), and by a dramatic increase in the flux of information (Reza 1961). In areas of high biodiversity, such as occur in the tropics, acoustic codes occupy narrow frequency bands to reduce the potential overlap and masking between species. Acoustic codes can be adapted to new conditions, showing great plasticity.

2.5 ­Sound as a Compass for Navigation Frog choruses have been shown to help nocturnal migratory birds orient during the trip toward polar regions along their migration routes. Observations on Rana pipiens choruses in eastern New York State have been reported by Griffin and Hopkins (1974) and Griffin (1976). On nights with light winds, frog choruses were audible at a distance of 965 m with an amplitude of 20 dB sound pressure level (SPL) in a frequency band of 1.5–2.5 kHz. These sounds offer important information to migrating birds regarding the typology of land and the velocity of wind. Probably birds can estimate their altitude based on the assumption that altitude functions as a low‐pass filter and aids in their migration. The use of sound as a probe to explore the surroundings in migrating bowhead whales (Balaena mysticetus) has been described by George et al. (1989). These authors argued that acoustic signaling by whales was used to avoid a large multi‐year ice, and that the whales were using their calls to evaluate the thickness of ice. In fact, this species breaks the ice to breathe during migration, offering open water for use by the group. In the bottlenose dolphin (Tursiops truncates), Xitco and Roitblat (1996) demonstrated the capacity for this species to eavesdrop on signals emitted by other individuals in order to recognize an object. The information gained by a dolphin using echolocation can be transferred to the sensory system of another companion. Other animals probably have this capacity (bats and some birds). The presence of biosonar has never been observed in pinnipeds, probably due to physiological constraints as argued by Schusterman et al. (2000). Physiological differences have been observed between the California sea lion (Zalophus californianus), harbor seal (Phoca vitulina), and elephant seal (Miorounga angustirostris); the sea lion hears better in air, harbor seals show no difference in hearing whether in air and water, and the elephant seal has more auditory sensitivity in water (Kastak and Schusterman 1998). In this last species, Southall et al. (2000) found that these animals have the capacity to detect signals at relatively low signal‐to‐noise ratio. These differences are explained in terms of ecological adaptation. In Arctic ringed seals (Phoca hispida) and Antarctic Weddell seals (Leptonychotes weddelli), Wartzok et al. (1992) observed a directional movement toward external acoustic cues in blindfolded animals that were able to navigate using a spatial memory to find ice holes that are essential for breathing. In birds that live in groups, the exchange of acoustic information is common. The behavior of tits and their use of alarm calls to inform each other about the presence of a diurnal or nocturnal predators and its size have been investigated (Courter and Ritchison 2010). Alarm call of tits are often eavesdropped upon by other species that

21

22

Ecoacoustics

recognize the approach of a threat (Langham et al. 2006). Eavesdropping is common for several species of animals that receive a signal indirectly from another individual or group. At the social level, the eavesdropping represents a social signal that may have several meanings (Peake 2005). Personal and public information are two faces of the same coin. A species lives in a world in which information is common currency that makes a difference in the quality of life and survival of that species. The sonic network created by individuals or acoustic communities is the ambient environment in which individuals can gather important information (Dall et al. 2005). Sonic information may be intentional, like the song emitted by a male to convince a female to become a mate, or unintentional, such as when an acoustic community provides information to an eavesdropper about the status of temporary aggregations of acoustic animals (Maynard Smith 2000). Acoustic information, including noise, is present in the environment and this information may be translated into an acoustic “illumination” (Buckingham 1992). On this theoretical basis, the soundscape in which every organism is embedded provides important information and creates an acoustic spatial dimension or acoustic habitat that is an acoustic eco-field. Often distant indistinct sound sources are considered as ambient noise (Cato 2008). Ambient noise may have a level of 20 dB but this can be higher in the presence of human technophonies. The marine system is rich in sounds that can be propagated for great distances without the degradation observed in terrestrial environments. At the same time, the anthropogenic intrusion in marine systems can create more issues with animal communication than in terrestrial systems because marine animals occur in a more restricted environment from which light is often excluded.

2.6 ­Geophonies from Sacred Sites – How to Incorporate Archeoacoustics into Ecoacoustics A recent field of acoustic study, archeoacoustics, has been dedicated to the investigation of sounds emitted by different types of artifacts, buildings, and sacred sites operating into the human emotional sphere (Garza et al. 2008; Jahn et al. 1996). This new field of research has many cultural, social, and ecological implications. The importance of sounds in ancestral societies and the value of sounds in early social organization have been emphasized for a long time but only recently received scientific interest (Devereux 2002; Eneix 2014; Reijs 2002; Watson and Keating 1999). During prehistoric times, the relationship between humanity and the environment was more intense than it is now. The hunting and gathering lifestyle shaped the biological and functional adaptation of our ancestors, thus forcing more interaction between humans and natural phenomena. The development of an acute sensorial capacity was necessary for survival in a world dominated by interspecific competition for resources. Our ancestors had a more developed sensorial capacity because they depended upon environmental resources for survival. When our ancestors heard a specific sound, it meant that there was something to eat, or that they might be eaten. Today we have lost our sensitivity to environmental sounds since we are not hunter‐ gatherers any more as we dwell in places protected from the environment and purchase food from a store.

2  The Duality of Sounds: Ambient and Communication

It seems somewhat bizarre to link archeoacoustics with ecoacoustics, but description of the past sonic environment provides the potential to better understand the current sonic environment and its meaning to humans and to other organisms. There are three main aspects of the study of archeoacoustics: ●● ●● ●●

sounds produced by historic instruments sounds of resonance in buildings geophonies in the soil at sacred sites.

In particular, the study of geophonies in the soil at sacred sites is important from an ecological perspective. The presence of different sounds of geophonic origin in sacred sites provides the opportunity to explore landscapes to find a mosaic of geophonic sources that may have a significant impact on the distribution and abundance of many species. The heterogeneity of geophonic sources to be discovered in landscapes has been neglected by focusing on the effect of vegetation structure on vocal performances (Aylor 1971; Briefer et al. 2010; Martens and Michelsen 1981). 2.6.1  The Characteristics of Geophonies

Geophonies can originate from water movement (sea waves, stream currents, rainfall), wind, volcanic activity (eruption, gas explosion, etc.), thunder, earthquakes, treefall or animal activity (digging, trampling, etc.). When geophonies are intense, they occur in the audible frequency range, but when they are less intense or occur at infra‐ and ultrasonic frequencies, they are heard only by animal auditory systems or by instruments. In particular, infrasound, which is below 20 Hz, and low‐frequency sounds, 20–100 Hz, create geophonies that originate inside the Earth by tectonic movements, volcanic activity and by underground water, and may have a major impact on the distribution of animals. For instance, it has been shown that a species of earthworm (Diplocardia ­mississippiensis) emerges from the soil after a grunting sound with a frequency of 97 Hz made on the surface (Catania 2008). Elephants have been shown to emit rumbles and foot stomps that signal at a frequency of 20 Hz and can be perceived by other groups of elephants at a distance of 16 km (O’Connell‐Rodwell 2007; O’Connell‐Rodwell et  al. 2000). Animals like wood turtles (Clemmys insculpata) and gulls have been observed to stomp in order to capture worms (Tinbergen 1960). Several species of animals intentionally produce soil vibrations to communicate or to capture prey. For these species, the importance of infrasound emission cannot be overestimated and may have an impact on the distribution of other animals as well. 2.6.2  Geophonies and Sacred Sites

Ancient people had advanced knowledge of the acoustic properties of materials, objects, and spaces which has been demonstrated by archeological evidence. There are well‐ documented sites selected for sacred ceremonies that have a peculiar and distinct source of infra‐ or ultrasound emissions. For instance, in the well of San Salvatore Abbey (Mount Amiata, Siena province), Debertolis and Bisconti (2013) found intense ultrasounds at a frequency of 26–30 kHz. This source is unique to that location because in the water bodies around the abbey, there were no ultrasound emissions. Before the erection of this abbey in medieval timea, the Etruscans probably had a temple in the same location. The body of water below the abbey was functioning as a

23

24

Ecoacoustics

concentrator of ultrasonic emissions probably produced by residual volcanic activity caused by Mount Amiata. In another case analyzed in Visocica Hill (Bosnia), the same authors found a source of infrasound between the frequencies of 10 Hz and 70 Hz with a peak at 48 Hz occurring at the top of a hill in the ruins of a medieval castle. People particularly sensitive to low frequencies can easily perceive this emission when they visit these ruins. Many holy sites have peculiar active or passive sounds. For instance, Debertolis and Gullà (2016) discovered a source of infrasound that seems to have benefits for people using the thermal water in an old Roman bath located in Lasko, Slovenia. This source of sound is probably due to the upwelling of thermal water. Some hypogeums have resonance capacity. For instance, it was shown that the hypogeum of Cividale del Friuli (Veneto, Italy) resonated at frequencies of 94 Hz and 103 Hz when female and male singers vocalized (Debertolis and Bisconti 2014). Hypogeum chambers have often been intentionally excavated into shapes that provide resonance. This effect was likely a system developed to convince people of the presence of a direct contact with a deity (Debertolis et al. 2015a,b). 2.6.3  Human Versus Other Animals’ Perception of Sound: The Role of Archeoacoustics

Today, it is believed that the sound of nature may only belong to vocal animals and that humanity is not part of this natural orchestra. However, humans are animals, too. Due to cultural processes, humanity has lost most of the capacity to listen to the breath of nature that represents an important source of information and, in many cases, of psychological well‐being. It is not by chance that holidays are spent in quiet areas far from industrial sounds. It is important to preserve areas in which human intrusion is minimal. The preservation of quiet places is growing around the world and archeoacoustics may play an important role that could help restore the value of sound in modern societies. To find that hypogeums were excavated to increase the sonic effect of instruments or the human voice, that megalithic monuments like Stonehenge had the role of resonators during sacred ceremonies, or that some sacred locations were selected for their unique soundscape as a result of subterranean energy (water, wind, gases) is an indicator of the importance of sounds in the culture of our ancestors. A rediscovery of the influence of sound on historic places could help to increase the well‐being of modern societies and protect the environment from noise pollution of human origin.

­References Atkin, A (2013) Peirce’s Theory of Signs. Stanford Encyclopedia of Philosophy. Available at: http://plato.stanford.edu/archives/sum2013/entries/peirce‐semiotics/ (accessed 12 December 2016). Aylor, D (1971) Sound transmission through vegetation in relation to leaf area density, leaf width, and breadth of canopy. Journal of the Acoustical Society of America, 51, 411–414. Barbieri, M (2003) The Organic Codes. An Introduction to Semantic Biology, Cambridge University Press, Cambridge. Barbieri, M (2013) The paradigms of biology. Biosemiotics, 6(1), 33–59.

2  The Duality of Sounds: Ambient and Communication

Barbieri, M (2015) What is information? Philosophical Transactions of the Royal Society A, 374, 20150060. Boelman, NT, Asner, GP, Hart, GP and Martin, RE (2007) Multi‐trophic invasion resistance in Hawaii: bioacoustics, field surveys, and airborne remote sensing. Ecological Applications, 17, 2137–2144. Boncoraglio, G and Saino, N (2007) Habitat structure and the evolution of bird song: a meta‐analysis of the evidence for the acoustic adaptation hypothesis. Functional Ecology, 21(1), 134–142. Brasil (2009) Relatório Técnico de Monitoramento do Desmatamento No Bioma Cerrado, 2002 a 2008: Dados Revisados, Centro de Sensoriamento Remoto Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis, Brasília‐DF. Briefer E, Oiejuk TS, Rybak F and Aubin T (2010) Are bird song complexity and song sharing shaped by habitat structure? An information theory and statistical approach. Journal of Theoretical Biology, 262, 151–164. Buckingham, MJ (1992) Theory of acoustic imaging in the ocean with ambient noise. Journal of Computational Acoustics, 181, 117–140. Catania, KC (2008) Worm grunting, fiddling, and charming: humans unknowingly mimic a predator to harvest bait. PLoS One, 3(10), e3472. Cato, DH (2008) Ocean ambient noise: its measurement and its significance to marine animals. Proceedings of the Institute of Acoustics, 30, Pt5. Courter, RJ and Ritchison, G (2010) Alarm calls of tufted titmice convey information about predator size and threat. Behavioral Ecology, 21(5), 936–942. Dall, RXS, Giraldeau, LA, Olsson, O, McNamara, MJ and Stephens, WD (2005) Information and its use by animals in evolutionary ecology. Trends in Ecology and Evolution, 20, 187–192. Davies, WJ, Adams, MD, Bruce, NS, et al. (2013) Perception of soundscapes: an interdisciplinary approach. Applied Acoustics, 74, 224–231. Debertolis, P and Bisconti, N (2013) Archeoacoustics in ancient sites. Proceedings of the First International Scientific Conference, Zilina, Slovakia, June 10–14, pp. 306–310. Debertolis, P and Bisconti, N (2014) Archeoacoustic analysis of an ancient hypogeum in Italy, in Archeoacoustics (ed. LC Eneix), OTS, Myakka City. Debertolis, P and Gullà, D (2016) Healing aspects identified by archeoacoustic techniques in Slovenia. Available at: www.sbresearchgroup.eu/index.php/en/notizie‐in‐inglese/286‐ healing‐aspects‐identified‐by‐archaeoacoustic‐techniques‐in‐slovenia (accessed 12 December 2016). Debertolis, P, Coimbra, F and Eneix, L (2015a) Archeoacoustic analysis of the Hal Saflieni hypogeum in Malta. Journal of Anthropology and Archeology, 3, 59–79. Debertolis, P, Nikolic, D, Savolainen, H, Marjanovic, G, Earl, N and Ristevski, N (2015b) Archeoacoustic Analysis of Kanda Hill in Macedonia. Available at: http://www. sbresearchgroup.eu/index.php/en/research-papers/279-archaeoacoustic-analysis-ofkanda-hill-in-macedonia (accessed 12 December 2016). Depraetere, M, Pavoine, S, Jiguet, F, Gasc, A, Duvail, S and Sueur, J (2012) Monitoring animal diversity using acoustic indices: implementation in a temperate woodland. Ecological Indicators, 13, 46–54. Devereux, P (2002) Stone Age Soundtracks: The Acoustic Archaeology of Ancient Sites, Vega, London.

25

26

Ecoacoustics

Diaz I, Armesto, J, Reid, S, Sieving, K and Willson, M (2005) Linking forest structure and composition: avian diversity in successional forests of Chiloe Island, Chile. Biological Conservation, 123, 91–101. Drake, JB, Dubayah, RO, Clark, DB, et al. (2002) Estimation of tropical forest structural characteristics using large‐footprint lidar. Remote Sensing of Environment, 79, 305–319. Eneix, LC (ed) (2014) Archeoacoustics, OTS, Myakka City. Farina, A (2012) A biosemitoic perspective of the resource criterion: toward a general theory of resources. Biosemiotics, 5, 17–32. Farina, A. (2014) Soundscape Ecology: Principles, Patterns, Methods and Applications, Springer, Dordrecht. Farina, A and Belgrano, A (2004) The eco‐field: a new paradigm for landscape ecology. Ecological Research, 19, 107–110. Farina, A and Belgrano, A (2006) The eco‐field hypothesis: toward a cognitive landscape. Landscape Ecology, 21, 5–17. Farina, A and James, P (2016) The acoustic communities: definition, description and ecological role. Biosystems, 147, 11–20. Farina, A and Pieretti, N (2014) Acoustic codes in action in a soundscape context. Biosemiotics, 7, 321–328. Farina, A, Bogaert, J and Schipani, I (2005) Cognitive landscape and information: new perspectives to investigate the ecological complexity. Biosystems, 79, 235–240. Francis, CD, Ortega, CP, Cruz, A (2009) Noise pollution changes avian communities and species interactions. Current Biology, 19, 1415–1419. Fuller, S, Axel, AC, Tucker, D and Gage, SH (2015) Connecting soundscape to landscape: which acoustic index best describes landscape configuration? Ecological Indicators, 58, 207–215. Gage, SH and Axel, AC (2013) Visualization of temporal change in soundscape power of a Michigan lake habitat over a 4‐year period. Ecological Informatics, 21, 100–109. Gardner, TA, Barlow, J, Chazdon, R, et al. (2009) Prospects for tropical forest biodiversity in a human‐modified world. Ecology Letters, 12, 561–582. Garza, C, Medina, A, Padilla, P, Ramos, A and Zalaquett F (2008) Arqueoacústica maya. La necesidad del estudio sistemático de efectos acústicos en sitios arqueológicos. Estudios de Cultura Maya, 32. Gasc, A, Sueur, J, Jiguet, F, et al. (2013) Assessing biodiversity with sound: do acoustic diversity indices reflect phylogenetic and functional diversities of bird communities? Ecological Indicators, 25, 279–287. Gasc, A, Pavoine, S, Lellouch, L, Grandcolas, P and Sueur, J (2015) Acoustic indices for biodiversity assessments: analyses of bias based on simulated bird assemblages and recommendations for field surveys. Biological Conservation, 191, 306–312. Gaveau, DL, Sloan, S, Molidena, E, et al. (2014) Four decades of forest persistence, clearance and logging on Borneo. PLOS One, 9(7), e101654. George, JC, Clark, C, Carroll, GM and Ellison, WT (1989) Observations on the ice‐ breaking and ice‐navigation behavior of migrating bowhead whales (Balaena mysticetus) near Point Barrow, Alaska, Spring 1985. Arctic, 42(1), 24–30. Global Forest Watch Canada (2016) www.globalforestwatch.ca/ Goetz, S, Steinberg, D, Dubayah, R and Blair, B (2007) Laser remote sensing of canopy habitat heterogeneity as a predictor of bird species richness in an eastern temperate forest, USA. Remote Sensing of Environment, 108, 254–263.

2  The Duality of Sounds: Ambient and Communication

Griffin, DR (1976) The audibility of frog choruses to migrating birds, Animal Behaviour, 24, 421–427. Griffin, D and Hopkins, CD (1974) Sounds audible to migrating birds. Animal Behaviour, 22(3), 672–678. Halaj, J, Ross, DW and Moldenke, AR (2000) Importance of habitat structure to the arthropod food‐web in Douglas‐Fir canopies. Oikos, 90, 139–152. He, KS, Zhang, J and Zhang, Q, (2009) Linking variability in species composition and MODIS NDVI based on beta diversity measurements. Acta Oecologica, 35(1), 14–21. Hoffmayer, J (2008) Biosemiotics. An Examination into the Signs of Life and the Life of Signs, Scranton University Press, Scranton. Jahn, RG, Devereux, P and Ibison, M (1996) Acoustical resonances of assorted ancient structures. Journal of the Acoustical Society of America, 99(2), 649–658. Karr, JR and Roth, RR (1971) Vegetation structure and avian diversity in several new world areas. American Naturalist, 105, 423–435. Kastak, D and Schusterman, JR (1998) Low‐frequency amphibious hearing in pinnipeds: methods, measurements, noise, and ecology. Journal of the Acoustical Society of America, 103(4), 2216–2228. Kasten, EP, Gage, SH, Fox, J and Joo, W (2012) The remote environmental assessment laboratory’s acoustic library: an archive for studying soundscape ecology. Ecological Informatics, 12, 50–67. Klink, CA & Machado, RB (2005) Conservation of the Brazilian Cerrado. Conservation Biology, 19(3), 707–713. Langham, GM, Contreras, TA and Sieving, KE (2006) Why phising works: titmouse (Paridae) scolds elicit a generalized response in bird communities. Ecoscience, 13(4), 485–496. Lemon, RE (1975) How birds develop song dialects. The Condor, 77, 385–406. Martens, MJM and Michelsen, A (1981) Absorption of acoustic energy by plants leaves. Journal of the Acoustical Society of America, 69(1), 303–306. Maynard Smith, J (2000) The concept of information in biology. Philosophy of Science, 67(2), 177–194. McCracken, KG and Sheldon, FH (1997) Avian vocalizations and phylogenetic signals. Proceedings of the National Academy of Sciences USA, 94(8), 3833–3836. Merchant, ND, Fristrup, KM, Johnson, MP, et al. (2015) Measuring acoustic habitats. Methods in Ecology and Evolution, 6, 257–265. Molleman, F, Kop, A, Brakefield, PM, et al. (2006) Vertical and temporal patterns of biodiversity of fruit‐feeding butterflies in a tropical forest in Uganda. Biodiversity and Conservation, 15(1), 107–121. O’Connell‐Rodwell, CE (2007) Keeping an “ear” to the ground: seismic communication in elephants. Physiology, 22, 287–294. O’Connell‐Rodwell, CE, Arnason, BT and Hart, LA (2000) Seismic properties of Asian elephant (Elaphas maximus) vocalization and locomotion. Journal of the Acoustical Society of America, 108, 3066–3072. Odum, HT (1983) Systems Ecology: An Introduction, Wiley, New York. Peake, TM (2005) Eavesdropping in communication networks, in Animal Communication Networks (ed. PK McGregor), Cambridge University Press, Cambridge, pp.13–37. Pearman, PB (2002) The scale of community structure: habitat variation and avian guilds in tropical forest understory. Ecological Monographs, 72, 19–39.

27

28

Ecoacoustics

Pekin, BK, Jung, J, Villanueva‐Rivera, LJ, et al. (2012) Modeling acoustic diversity using soundscape recordings and LIDAR‐derived metrics of vertical forest structure in a neotropical rainforest. Landscape Ecology, 27, 1513. Pereira, HM, Leadley, PW, Proença, V, et al. (2010) Scenarios for global biodiversity in the 21st century. Science, 330(6010), 1496–1501. Pieretti, N, Farina, A and Morri, D (2011) A new methodology to infer the singing activity of an avian community: the Acoustic Complexity Index (ACI). Ecological Indicators, 11, 868–873. Pimm, S, Raven, P, Peterson, A, Sekercioglu, CH and Ehrlich, PR (2006) Human impacts on the rates of recent, present, and future bird extinctions. Proceedings of the National Academy of Sciences USA, 103(29), 10941–10946. Pulliam, HR (1988) Source‐sinks, and population regulation. American Naturalist, 132, 652–661. Pulliam, HR (1996) Sources and sinks: empirical evidence and population conservation, in Population Dynamics in Ecological Space and Time (eds OE Rhodes, RK Chesser and MH Smith), University of Chicago Press, Chicago. Purvis, A and Hector, A (2000) Getting the measure of biodiversity. Nature, 405, 212–219. Qi, J, Gage, SH, Joo, W, Napoletano, B and Biswas, S (2008) Soundscape characteristics of an environment: a new ecological indicator of ecosystem health, in Wetland and Water Resource Modeling and Assessment (ed. W. Ji), CRC Press, New York, pp. 201–211. Ramankkutty, N, Foley, JA and Olejniczak, NJ. (2002) People on the land: changes in global population and croplands during the 20th century. Ambio, 31(3), 251–257. Reijs, VM (2002) MegaSound: Sound in Irish Megalithic Buildings. Presented at the First Pan‐American/Iberian Meeting on Acoustics, Session 3aAA2, p. 2284. Reza, FM (1961) An Introduction to Information Theory, Dover, New York. Ritts, M, Gage, SH, Picard, CR, Dundas, E and Dundas, S (2016) Collaborative research praxis to establish baseline acoustic conditions in Gitgaat Territory. Global Ecology and Conservation, 7, 25–38. Rodriguez, A, Gasc, A, Pavoine, S, Gradcolas, P, Gaucher, P and Sueur, J (2014) Temporal and spatial variability of animal sound within a neotropical forest. Ecological Informatics, 21, 133–143. Sala, OE, Chapin F, Armesto J, et al. (2000) Global biodiversity scenarios for the year 2100. Nature, 287, 1770–1774. Schrödinger, E (1944) What is Life? Cambridge University Press, Cambridge. Schusterman, RJ, Kastak, D, Levenson, DH, Reichmuth, CJ and Southall BL (2000) Why pinnipeds don’t echolocate. Journal of the Acoustical Society of America, 107(4), 2256–2265. Skole, DL, Samek, J, Chomentowski, W and Smalligan, M (2013) Forests, carbon, and the global environment: new directions in research, in Land Use and the Carbon Cycle: Advances in Integrated Science, Management, and Policy (eds D. Brown, D. Robinson, N. French and B. Reed), Cambridge University Press, New York. Southall, LB, Schusterman, JR and Kastak, D (2000) Masking in three pinnipeds. Underwater, low frequency critical ratios. Journal of the Acoustical Society of America, 108, 1322–1326. Sueur, J, Pavoine, S, Hamerlynck, O, Duvail, S and Reby, D (2008) Rapid acoustic survey for biodiversity appraisal. PLoS One, 3(12), e4065.

2  The Duality of Sounds: Ambient and Communication

Sueur, J, Farina, A, Gasc, A, Pieretti, N and Pavoine, S (2014) Acoustic indices for biodiversity assessment and landscape investigation. Acta Acoustica, 100, 772–781. Tinbergen, N (1960) The Herring Gull’s World. A Study of the Social Behaviour of Birds, Basic Books, New York. Towsey, M, Wimmer, J, Williamson, I and Roe, P (2014) The use of acoustic indices to determine avian species richness in audio‐recordings of the environment. Ecological Informatics, 21, 110–119. Tucker, D, Gage, SH, Williamson, I and Fuller, S (2014) Linking ecological condition and the soundscape in fragmented Australian forests. Landscape Ecology, 29(4), 745–758. Villanueva‐Rivera, LJ, Pijanowski, BC, Doucette, J and Pekin, B (2011) A primer of acoustic analysis for landscape ecologists. Landscape Ecology, 26, 1233–1246. von Uexküll, J (1957) A stroll through the worlds of animals and men: a picture book of invisible worlds, in Instinctive Behavior: The Development of a Modern Concept (ed. CH Schiller), International Universities Press, New York. Von Uexküll, J (1982) The theory of meaning. Semiotica, 42(1), 25–82. Wartzok, D and Davis, RW (1998) Under‐ice movements and the sensory basis of hole finding by ringed and Weddell seals. Canadian Journal of Zoology, 70, 1712–1722. Wartzok, D, Elsner, R, Stone, H, Kelly, BP and Davis RW (1992) Under‐ice movements and the sensory basis of hole finding by ringed and Weddell seals. Canadian Journal of Zoology, 70, 1712–1722. Watson, A and Keating, D (1999) Architecture and sound: an acoustic analysis of megalithic monuments in prehistoric Britain. Antiquity, 73, 325–336. Welch, PD (1967) The use of the fast Fourier transform for the estimation of power spectra: a method based on time‐averaging over short, modified periodograms. IEEE Transactions on Audio and Electroacoustics, AU‐15, 70–73. Wood, JD, O’Connell‐Rodwell, CE and Klemperer, SL (2005) Using seismic sensors to detect elephants and other large mammals: a potential census technique. Journal of Applied Ecology, 42, 587–594. Xitco, M Jr and Roitblat, LH (1996) Object recognition through eavesdropping: passive echolocation in bottlenose dolphins. Animal Learning and Behavior, 24(4), 355–365.

29

31

3 The Role of Sound in Terrestrial Ecosystems: Three Case Examples from Michigan, USA Stuart H. Gage1 and Almo Farina2 1 2

Department of Entomology, Michigan State University, East Lansing, USA Department of Pure and Applied Sciences, Urbino University, Urbino, Italy

3.1­  Introduction Chapter 1 describes the use of sound as an ecological attribute and discusses several types of attributes associated with sound. These include population census of animals; time of arrival/departure of migratory species; trophic interactions; biological diversity (soundscape diversity); disturbance regimes; diurnal change; seasonal change; competition for frequency; habitat health; and sounds of the biosphere. Pijanowski (2015) summarized research questions and describes several ongoing projects on sound in terrestrial systems. In this chapter we describe three case examples (C1, C2, and C3) to illustrate the use of sound in terrestrial systems. First, C1 is a visualization of the soundscape in a woodlot called Ted Black Woods, Okemos, Michigan. The second case (C2) is from Twin Lakes, Cheboygan, MI where a combination of automated metrics and manual listening was used to detect an amphibian species, the spring peeper (Pseudacris crucifer), whose vocalizations are related to weather and thus provide implications for climate change. C3, also from Twin Lakes, describes the use of sound to determine disturbance in a forest by examining before and after harvest.

3.2­  C1 Visualization of the Soundscape at Ted Black Woods, Okemos, Michigan during May 2016 3.2.1  C1 Background

Ted Black Woods is a 75‐acre deciduous woodland with associated wetlands located on Van Atta Road, Meridian Township, Okemos, MI (Figure 3.1). The forest consists of white and red oak, black cherry, maple, beech, and additional tree species. Some of the trees are more than 200 years old and reach 20 m. Three large wetlands are on the property. Access to these woods was granted by Meridian Parks. Ted Black passed away on 30 November 2007 at age 93 and donated his property to the township. Black did postdoctoral research at the University of Wisconsin Ecoacoustics: The Ecological Role of Sounds, First Edition. Edited by Almo Farina and Stuart H. Gage. © 2017 John Wiley & Sons Ltd. Published 2017 by John Wiley & Sons Ltd.

32

Ecoacoustics

Figure 3.1  C1. Location of four automated Song Meters in Ted Black Woods, Okemos, Michigan.

with Aldo Leopold, considered the founding father of wildlife ecology. Among Black’s many accomplishments was the publication of Birds of Michigan (Black and Kennedy, 2003). 3.2.2  C1 Objectives

The objective of the study was to examine and interpret patterns of the soundscape during May 2016 by time of day and day of year using soundscape power metrics and six soundscape indices. 3.2.3  C1 Methods

Song Meters (SM2) (Wildlife Acoustics 2014) were deployed near four wetlands. The SM2 acoustic recorders were used to record ecoacoustics at four sites (TB01–TB04). Song Meters were programmed to record at a 30‐minute interval for 60 seconds.

3  The Role of Sound in Terrestrial Ecosystems

Recordings written to SD media were retrieved and transferred to disk. The 5952 recordings made during May 2016 from the four sites were selected for analysis. Soundscape power and six soundscape indices were computed from the recordings using an R script which integrates output from several R packages (Gage et al. 2014; R Core Team 2012). 3.2.3.1  C1 Soundscape Metrics

Two types of soundscape metrics were used to investigate patterns in the soundscape. First, soundscape power was calculated at 1 kHz frequency intervals for each of 11 frequency intervals (F1–11 kHz) (Gage and Axel 2013). Second, six soundscape indices were calculated: the Acoustic Complexity Index (ACI) (Pieretti et al. 2011); the Acoustic Diversity Index (ADI) (Villanova‐Rivera and Pijanowski, 2013); the Acoustic Evenness Index (AEI) (Villanova‐Rivera and Pijanowski, 2013); the Acoustic Entropy Index (H) (Sueur et  al. 2008); the Bioacoustic Index (BOI) (Boelman et  al. (2007); and the Normalized Difference Soundscape Index (NDSI) (Gage and Axel 2013; Kasten et al. 2012). Statistical tests of site differences were not significant so the information for the four sites was combined. 3.2.3.2  C1 Weather Factors Affecting Sounds

Weather data at 30‐minute intervals were obtained from a Davis weather station located approximately 1 km from Ted Black Woods. Key variables included temperature, rainfall (in), and wind velocity (m/s). Weather records for May 2016 were exported to a csv file. Weather data collection began at 0300h on May 2 so subsequent analysis of weather began from that time. Soundscape power metrics, soundscape indices, and weather variables were imported into Minitab (Minitab v 16, 2010) and merged to create a single file used to visualize patterns in the soundscape. 3.2.4  C1 Results 3.2.4.1  C1 Patterns of Soundscape Power for Six Frequency Intervals

Soundscape power patterns of six frequency intervals for each of the 48 recordings made at 30‐minute intervals during May 2016 are shown in Figure 3.2. The normalized soundscape power values of frequency interval F1–2 were steady at about 0.5 at night, then dropped at 0600h to its lowest value then rose again and steadily increased until 1500h then dropped until 1800h and remained steady until 0600h the next day. Values of the frequency interval F2–3 rose rapidly at 0530h, peaked at 0600h and declined by 0700h. The values remained low until 1500h when they began to steadily rise until about 2100h and then declined. The frequency interval values of F3–4 were similar to those of F2–3, with a steeper increase at 2100h. The pattern of values of frequency intervals F4–5 was quite different. Values were low during the night then increased to 0600h and then gradually declined. Soundscape power in F5–6 dipped at 0600h, rose until 0900h then declined steadily during the day until 2100h. Soundscape power at frequency F6–7 rose at night until 0600h when values dropped and then rose steadily until 0900h and remained steady until 1800h, then dropped at 2200h and rose again. Soundscape power metrics are plotted over day of the month, based on the day count from 1 January (day of year) to examine the soundscape patterns over the months (Figure  3.3). Normalized soundscape power values for the frequency interval F1–2

33

F1-2kHz

F2-3kHz

0.6

0.6

0.5

Normalized Soundscape Power (w/kHz)

0.5

0.4

F2-3kHz

0.6

0.4

0.4

0.3 0.3

0.2

F5-6kHz

F4-5kHz

F6-7kHz

0.4 0.15

0.20 0.3 0.15

0.10

0.2

0.10

0.1

0.05

0.05

0.00

0.0 0

300

600 900 1200 1500 1800 2100

0

300

600 900 1200 1500 1800 2100

0

300 600

Time of Day

Figure 3.2 C1. Mean normalized soundscape power patterns for each of the 48 recordings made per day during May 2016.

900 1200 1500 1800 2100

F1-2kHz < 0.1 0.1 – 0.3 0.3 – 0.5 0.5 – 0.7 > 0.7

1500

1000

500

0

125

130

135

140

Day of Year

145

1500

1000

0

150

125

130

135

140

145

150

Day of Year F3-4kHz < 0.1

1500

1000

500

F4-5kHz < 0.1

2000

0.1 – 0.2 0.2 – 0.3 0.3 – 0.4 > 0.4

Time of Day

Time of Day

0.1 – 0.2 0.2 – 0.3 0.3 – 0.4 > 0.4

500

2000

0

F2-3kHz < 0.1

2000

Time of Day

Time of Day

2000

0.1 – 0.2 0.2 – 0.3 0.3 – 0.4 > 0.4

1500

1000

500

125

130

135

140

Day of Year

145

150

0 125

130

135

140

145

150

Day of Year

Figure 3.3 C1. Contour plots of normalized soundscape power values for six frequency intervals (z), day of year (x) (based on the day count from 1 January) and time of day (y) during May 2016.

F5-6kHz < 0.05

0.05 – 0.10 0.10 – 0.15 0.15 – 0.20 > 0.20

1500

1000

500

0

F6-7kHz < 0.05

2000

Time of Day

Time of Day

2000

0.05 – 0.10 0.10 – 0.15 0.15 – 0.20 > 0.20

1500

1000

500

125

130

135

140

Day of Year

Figure 3.3 (Continued)

145

150

0

125

130

135

140

Day of Year

145

150

3  The Role of Sound in Terrestrial Ecosystems

began low at the beginning of May then remained variable with minor trends until the end of the month. The values of the frequency interval F2–3 declined until day 128 then increased until then end of May. The soundscape power values for frequency interval F3–4 were high at then beginning of May, then declined, rose, declined and were then steady through the end of the month. Soundscape power values for the frequency interval F4–5 were low at the beginning of May then rose steadily during most of the month. Values for the frequency interval F5–6 were low at the beginning of May, spiked at days 129–131, declined and then rose steadily until the end of May. Values for the frequency interval F6–7 followed a similar pattern to F5–6 but were lower in intensity. 3.2.4.2  C1 Patterns of Soundscape Indices

The patterns of six soundscape indices are shown over time of day (Figure 3.4). All indices, with the exception of AEI, detect the dawn chorus at 0600h with some indices showing a strong response (NDSI, H, ACI, BOI). All indices except AEI begin to increase about 1600h. Two indices (ACI, BOI) peaked at 2100h. The soundscape indices patterns during the month of May are shown in Figure 3.5. NDSI was high early in May, declined, then increased steadily over the month. H was variable and steady then increased later in May. ACI was high at the beginning of the month then declined during the latter half of May. ADI was variable at the beginning of the month then rose and remained steady until the last few days of May when the index reached a maximum. AEI was high and variable, then became steady, then dropped and remained steady until the last three days then dropped to its lowest value. BOI was high during early May then dropped dramatically, rose to near maximum then declined again and remained steady for the rest of the month. 3.2.4.3  C1 Wind Patterns During May 2016

Wind was generally light until 1000h and variable between 1000–2000h. There were higher wind velocities (1.1–1.7) during days 129–130 and again during days 135–138. A stronger wind velocity event occurred at days 150–151. The distribution of wind during May 2016 is shown in Figure 3.6. 3.2.4.4  C1 Rain Patterns During May 2016

Time of day/rain events occurred in the early morning and in the afternoon with the largest amount of rain falling at 0330h and 1700h respectively, with the greatest rate of rainfall occurring at the same times. Rain occurred throughout May on days 123, 125, 131, 133, 134, 135, 147, and 149, with the greatest amount falling on day 135 (14 May). The rate of rainfall had the same distribution. Scattered rainfall events occurred in the morning on day 122 with a heavier event at 0330h and at about the same time on day 135. Longer rain events occurred in the afternoon between days 132–135. Light rain events, scattered throughout the day, occurred during the last day of May. The distribution of rainfall during May is shown in Figure 3.7. The distribution of rain on day 135 (14 May) is shown in Figure 3.8. Rainfall occurred on 14 May (day 135), beginning at midnight, peaking at 0330h and diminishing before the dawn chorus (0600h), and a lesser amount of rainfall occurred at 1500h (see Figure 3.8).

37

NDSI

H

0.8

ACI

168

0.80 164 0.6 0.75

160

Soundscape Index

0.4 156

0.70

0.2

ADI

2.00

AEI 0.7

1.75

152

BOI

30

0.6

1.50

20

0.5

1.25

0.4

1.00

10 0.3 0

300

600

900 1200 1500 1800 2100

0

300

600

900 1200 1500 1800 2100

0

300

Time of Day

Figure 3.4 C1. The patterns of six soundscape indices for each of the 48 recordings made per day during May 2016.

600

900 1200 1500 1800 2100

NDSI

Time of Day

1500

1000

H

1500

1000

500

500

0

< 0.40 0.40 – 0.52 0.52 – 0.64 0.64 – 0.76 0.76 – 0.88 > 0.88

2000

Time of Day

< −0.8 −0.8 – −0.4 −0.4 – 0.0 0.0 – 0.4 0.4 – 0.8 > 0.8

2000

125

130

135

140

Day of Year

145

0

150

125

130

135

140

Day of Year

150

145

ACI

Time of Day

1500

1000

1500

1000

500

500

0

ADI < 0.0 0.0 – 0.5 0.5 – 1.0 1.0 – 1.5 1.5 – 2.0 > 2.0

2000

Time of Day

< 145 145 – 150 150 – 155 155 – 160 160 – 165 > 165

2000

125

130

135

140

Day of Year

145

150

0

125

130

135

140

Day of Year

145

150

Figure 3.5 C1. Contour plots of six soundscape index values (z), day of year (x) (based on the day count from 1 January) and time of day (y) during May 2016.

AEI

Time of Day

1500

1000

500

0

BOI

< 10 10 – 22 22 – 34 34 – 46 46 – 58 > 58

2000

Time of Day

< 0.1 0.1 – 0.3 0.3 – 0.5 0.5 – 0.7 0.7 – 0.9 > 0.9

2000

1500

1000

500

125

130

Figure 3.5 (Continued)

135

140

Day of Year

145

150

0

125

130

135

140

Day of Year

145

150

3  The Role of Sound in Terrestrial Ecosystems W-Speed (mph)

< 0.5 0.5 – 1.0 1.0 – 1.5 1.5 – 2.0 2.0 – 2.5 > 2.5

20:00

Time

16:00

12:00

08:00

04:00

00:00

125

130

135

140

145

150

Day of Year

Figure 3.6  C1. The distribution of wind speed during May 2016.

3.2.4.5  C1 Spectrogram Patterns

Spectrograms from site TB01 at 0330h on day 135 (14 May) (a) and from 0330h on the next day (day 136 or 15 May) (b) are shown in Figure 3.9. Note that a high rate of rainfall masks the biological sounds in the spectrogram. Rain (in) < 0.000 0.000 – 0.001 0.001 – 0.003 0.003 – 0.005 0.005 – 0.007 > 0.007

20:00

Time

16:00

12:00

08:00

04:00

00:00

125

130

135 140 Day of Year

Figure 3.7  C1. Distribution of rainfall amount during May 2016.

145

150

41

Ecoacoustics

0.20

14 May, 2016 (Day of Year 135)

0.15 Rain (in)

42

0.10

0.05

0.00 0

300

600

900

1200

1500

1800

2100

Time of Day

Figure 3.8  C1. Distribution of rain on day 135 (14 May 2016)

A spectrogram from a recording from site TB01 at 0600h on 14 May is shown in Figure 3.10 to illustrate the characteristics of the soundscape during the dawn chorus. 3.2.5  C1 Discussion

Soundscape power in different frequency intervals can provide insight into changes in daily patterns in the soundscape. These can be used to identify the dawn and dusk choruses as well as determining different elements of the soundscape. All six frequency intervals show abrupt changes at the occurrence of the peak dawn chorus (0600h). Some soundscape power values dropped (F1–2, F5–6, F6–7) whereas others increased (F2–3, F3–4, F4–5) at the time of the dawn chorus (see Figure 3.2), illustrating that the soundscape is composed of organisms that signal at different frequencies (Gage and Axel 2013). Soundscape indices are also useful metrics for characterizing the soundscape. Some of the indices identify the dawn chorus (ACI, BOI) while others do not (see Figures 3.4 and 3.5) (Farina 2014; Fuller et al. 2015). Rain and wind can have a profound effect on the analysis of ecoacoustics patterns and must be taken into consideration. In May 2016, wind was light in the morning but occurred more often during the afternoon (see Figure 3.6). Rain occurred primarily during the first half of May (see Figure 3.7). A significant rain event occurred in the early morning on 14 May (day 135) which had a significant effect on biophony (see Figures 3.8 and 3.9(top)). The rain event at 0330h shown in Figure 3.8 was strong and blocked any evidence of biophony (see Figure 3.9(top)). Therefore, algorithms must be developed to identify wind and rain events that will affect interpretation of the soundscape. Appendix 1 documents rain, nonnormalized soundscape power, species in the soundscape, and

3  The Role of Sound in Terrestrial Ecosystems

(a)

(b)

Figure 3.9  C1. Spectrograms from 0330h on day 135 (14 May) (top) and from 0330 on day 136 (15 May) (bottom).

Figure 3.10  C1. A spectrogram from a recording made at TB01 at 0600h on 14 May 2016 (day 135).

43

44

Ecoacoustics

other sounds in the soundscape at 30‐minute intervals on 14 May (day 135). Computation of nonnormalized soundscape power is useful to detect rain but analysis of this metric showed that it was not a universal measurement to detect rainfall events.

3.3 ­C2 Implications for Climate Change – Detecting First Call of the Spring Peeper 3.3.1  C2 Background

In this case we use the spring peeper (Pseudacris crucifer) as an illustration of combining human and computational technology to find the first call of this amphibian. This anuran is one of the earliest callers among the dozen frog species found in Michigan. The spring peeper can begin to call in late March or early April when the weather begins to warm and their calls may last through May. The spring peeper is considered a harbinger of spring and can also be important as an indication of a changing climate. Therefore, the determination of the first call of the spring peeper can be a valuable attribute, not only because it is a harbinger of spring but also because it calls as temperatures rise, thus indicating the earliness of spring. Automated identification of species has been difficult because of frequency overlap by signaling species. Although there have been significant advances in automated species detection (Acevedo et al. 2009), the species detected by automated methods tend to be most accurate when those species emit monotonic calls. The best solution to date, as suggested by Wimmer et al. (2013), is to combine automated technologies and human knowledge to find acoustic events. Even when automated identifications are possible, especially when examining large numbers of recordings, it is advantageous to examine the recordings that fall within specific parameters so that only a subset of the recordings need examination to detect an event or species of interest. If, for example, the event of interest is known to occur within a specific frequency interval, then a search for the event within that frequency interval can reduce the amount of human effort and computations needed to find events that match that frequency in a database. The intensity of calling increases during humid evenings or just after rain, when many males congregate. Male frogs establish territories near the edge of permanent or vernal wetlands and call to attract mates. The faster and louder males sing, the more likely they are likely to attract a mate. Werner et al. (2009) found that pond occupancy rates for the spring peeper were relatively constant, but important source ponds in the landscape changed with drought. Ponds with extensive forest canopy cover were sink habitats for both species, and high pond connectivity appeared to be necessary to maintain species presence in these habitats. The frequency at which the spring peeper signals is a very simple, nearly tonal signal, with a single spectral peak (mean 2895 Hz) and there is little or no harmonic or internal temporal structure (Wilczynski et al. 1984). 3.3.2  C2 Methods

Recordings were made near Twin Lakes which is located in Grant Township, Cheboygan County, Michigan. Twin Lakes comprises seven small interconnected basins separated

3  The Role of Sound in Terrestrial Ecosystems

by narrow channels. Twin Lakes is located in forested land comprising mixed coniferous and deciduous trees with predominantly sandy soil. Amphibians such as the green frog (Rana clamitans), the spring peeper (Pseudacris crucifer), the northern leopard frog (Rana pipiens), and the American toad (Bufo americanus) inhabit the aquatic habitats. Many species of birds also inhabit the ecosystem, including bald eagle, osprey, Caspian tern, belted kingfisher, blue heron, common loon, merganser, pheasant, ruffed grouse, wild turkey, and other woodland birds. Mammals include black bear, white‐tailed deer, coyote, fox, beaver, river otter, marten, raccoon, rabbit, squirrel, and chipmunk. Song Meters (Wildlife Acoustics 2014) were used to record the soundscape. Each monaural recording was made at 22 050 Hz, was 1 minute in length with a recording interval of 30 minutes. The recordings made during 1 March–31 May from 2009 to 2012 at a site (LA00) located at the edge of Twin Lakes know to harbor a spring peeper population. Ecoacoustics metrics derived from each recording consist of normalized soundscape power values (watts/kHz) that range from 0–1 for each of 11 frequency intervals from 1‐11 kHz. Kasten et al. (2012) describe the technique to compute power spectral density (PSD) values from recorded sounds. A web‐based digital acoustic library for acoustic recordings has been developed to assist in computing metrics, archiving acoustic information and accessing recordings collected by sound recorders (Kasten et  al. 2012). A search tool, designed to search for specific frequency intervals at selected soundscape power values, was used to search the digital archive. Since the spring peeper signals at 2895 Hz (Wilczynski et al. 1984), the search was confined to 2–3 kHz using a soundscape power parameter of 0.5 (50% of the full range). The search was further confined to between 2000h and 0700h the following day. Since it is relatively easy to see the spring peeper signature in a spectrogram, the date and time were noted and the spectrograms for the entire day (0000–2350h) were examined to pinpoint the beginning of spring peeper calls. If spring peeper calls were detected at midnight, then the previous day was examined for spring peeper signals until the first spring peeper call was heard. All queries and results obtained to examine spring peeper first call event were conducted online using the Remote Environmental Assessment digital library system (www.real.msu.edu) (Kasten et al. 2012). 3.3.3  C2 Results

There were 9773 recordings in the library recorded by the Song Meter located at LA00 (45.53320°, -84.291960°) during March and April, 2009–2012. When a soundscape power value of 0.5 at a frequency of 2–3 kHz was used as a filter, 2703 recordings resulted. When the search was refined to 2000h, 64 recordings were found (Table 3.1). The breakdown by year was 2009 (9), 2010 (22), 2011 (10), and 2012 (22). In 2009, of the nine records that matched the criteria, the earliest record that was a spring peeper signal, based on examination of the spectrogram image and subsequently listening to the sound, occurred on 18 April. To discover the earliest spring peeper call (first event), all records for 18 April were selected, retrieved and examined for spring peeper signals. The earliest spring peeper signal occurred at 0400h on 18 April 2009 (Table 3.2). Although the spring peeper call was weak, the call in the recording one hour later (0500h) was strong. The same process used to detect the first spring peeper call in 2009 was applied to recordings made during 2010–2012. Over the four years examined, there was a 30‐day range in spring peeper first call at site LA00 (19 March–18 April).

45

Ecoacoustics

Table 3.1  Initial search criteria and number of recordings of the spring peeper determined by filtering the frequency in which spring peeper calls occur Search criteria

Recordings

All recordings

9773

F2–3 kHz >=0.5

2703

F2–3 kHz >=0.5 and time = 2000h

64

Table 3.2  Date and time of first call of the spring peeper (FSPC) at Twin Lakes, Cheboygan, MI

Year

Site

Recorder start

First spring peeper (FSPC)

Time FSPC (h)

Day of year

Mean March temperature (°F)

2009

LA00

25 March

18 April

0400

108

29.1

2010

LA00

28 February

1 April

0030

91

36.3

2011

LA00

18 February

12 April

2030

102

27.5

2012

LA00

1 January

19 March

2030

78

38.7

Due to the 30‐day difference in the time of first spring peeper call, temperature records were obtained from the Cheboygan County Airport to examine the mean monthly temperature from March to May when spring peepers were calling. Figure 3.11 shows the mean monthly temperature for March–May 2009–2012, indicating that March was cooler in 2009 and 2011 compared to 2010 and 2012. Cheboygan County Airport Mean Monthly Temperature (F)

60

50 Temperature (F)

46

40

30

20 Month Year

3

4 2009

5

3

4 5 2010

3

4 5 2011

3

4 5 2012

Figure 3.11  C2. Mean monthly temperature for March, April, and May for 2009, 2010, 2011, and 2012.

3  The Role of Sound in Terrestrial Ecosystems

Day of First Spring Peeper Call

110

2009

105

2011

100 95

2010

90 85 80

2012

26

28

30

32

34

36

38

40

Mean March Temperature (F)

Figure 3.12  C2. First spring peeper call (day of year) versus mean March temperature (F) at site LA00 in 2009–2012.

Based on the above criteria used to identify the first call of the spring peeper, the day of the year when the first call was detected at site LA00 was plotted against mean March temperature for each of the four years when recordings were made at 30‐minute intervals. Calls were detected earlier in 2010 and 2012 than in 2009 and 2011 (Figure 3.12), indicating that there was a relationship between March temperature and first call of the spring peeper. Audio recordings were made at eight nearby sites at Twin Lakes during this period (2009–2012). The date and year when each recorder was established depended on whether the location was accessible by boat (after ice‐out on the lake) since six sites (LA01–LA06) were located on the island in Twin Lakes. Sites LA07 and LA08 were established in 2011 on the mainland to characterize the soundscape in a nearby state forest. These sites provided limited but additional information about spring peeper first event calls. Table 3.3 shows the first event data from each site where records were available and shows the set‐up time, first spring peeper date (using the search‐filtering process described in the methods), time of day for the first spring peeper call, and the mean monthly temperature for March from the Cheboygan County Airport. Figure 3.13 examines the apparent trend between mean monthly March temperature and the day of first spring peeper call. 3.3.4  C2 Discussion

The relationship between mean March temperature and first spring peeper call determined by the search‐filter process based on this study was strong. As noted by Roloff et  al. (2011), the best detection models for American toad (Anaxyrus americanus), wood frog (Lithobates sylvaticus), and spring peeper (Pseudacris crucifer) consistently included a positive relationship with temperature and precipitation. However, male spring peepers prefer to produce advertisement calls on nights with ideal weather conditions. Gregory et  al. (1997) set out to determine these conditions by taping peeper choruses from a single wooded marsh just south of the Maple River near Pellston, MI, and recording the weather conditions over a three‐week period starting

47

Ecoacoustics

Table 3.3  Expanding the filtering process to eight additional sites to determine first spring peeper call. An * indicates that a first spring peeper call (FSPC) was detected the same day the recorder was set up. Sites LA01–LA06 could not be set up earlier due to logistics (unsafe ice conditions on the lake)

Year

Site

Recorder start

First spring peeper (FSPC)

Time FSPC (h)

Day of year

Mean March temperature (°F)

2010

LA01

27 March

1‐ April

2030

91

36.3

2011

LA01

12 April

12 April*

2030

102

27.5

2012

LA01

24 March

24 March*

2000

83

38.7

2011

LA02

12 April

12 April*

1330

102

27.5

2010

LA03

30 March

31 March

2330

90

36.3

2011

LA03

12 April

12 April*

2030

102

27.5

2010

LA04

30 March

31 March

2200

90

36.3

2011

LA04

12 April

12 April*

2100

102

27.5

2010

LA05

30 March

31 March

2230

90

36.3

2011

LA05

12 April

12 April*

2030

102

27.5

2010

LA06

27 March

31 March

2030

90

36.3

2011

LA06

12 April

12 April*

2030

102

27.5

2011

LA07

31 March

11 April

2100

101

27.5

2011

LA08

31 March

11 April

2200

101

27.5

110 Day of First Spring Peeper Call

48

105 100 95 90 85 80 26

28

30

32

34

36

38

40

Mean March Temperature (F)

Figure 3.13  C2. Day of first spring peeper call versus mean March temperature (F) (all sites).

in mid‐May. Air temperature was found to have a negative correlation with the volume of chorusing males, while humidity showed a positive correlation. We suspect examining the relationship between temperature and spring peeper calls in May is likely too late. Ospina et  al. (2013) found that calling activity of Eleutherodactylus coqui and E.  cochranae was positively correlated with temperature, while E. brittoni and

3  The Role of Sound in Terrestrial Ecosystems

E.  ­juanariveroi responded negatively to temperature and precipitation. They concluded that difference in response to temperature and precipitation could be related to differences in body size and the location of calling sites among the four species. Wimmer et al. (2013) found that the most appropriate method for accurate species identification is a combination of machine and human event detection. We found that the time to identify the first call of the spring peeper using filtering via soundscape power in the target’s frequency interval and subsequent time selection based on biology of the target, along with human visualization of spectrograms, was very accurate and very rapid as long as the data were in a searchable database. The number of recordings was reduced from 9773 to 64 using simple search methods, making it feasible to identify the target and time rapidly (see Table 3.1). There have been attempts to automate calls by amphibians because of their importance as indicators of habitat health and climate and also because their signals are relatively simple. Brandes (1970) showed that probability maps create a dramatic increase in the bioacoustic signal‐to‐noise ratio within the spectrogram. Probability maps along with threshold filtering provide a means for image segmentation of the spectrogram, creating blocks of pixels that represent bioacoustic signals, facilitating feature and signal extraction. This methodology is applied to natural sound recordings of three quality types in a wide range of signal‐to‐noise ratios. In each instance, the probability mapping greatly increases the signal‐to‐noise ratio and, when applied as a threshold filter, is far more selective with pixel inclusion than threshold filtering applied based on sound level. Suggested applications include automated call recognition of birds, frogs, and insects from field recordings within a wide range of ambient noise. Acevedo et al. (2009) found that support vector machine (SMV) was an efficient data reduction technique in conjunction with high classification accuracy and thus is a promising combination for automated species identification by sound. Recently, Duan et al. (2013) developed new methods for birdcall identification which can be applied to amphibians.

3.4 ­C3 Disturbance in Terrestrial Systems: Tree Harvest Impacts on the Soundscape 3.4.1  C3 Background

Disturbance can be caused by natural events (hurricanes, volcanoes, fires) or by human‐ caused events (mining, urbanization, forest harvest) (Hobson and Schieck 1999). The measurement of sounds caused by disturbance can indicate the type and duration of the disturbance. Part of the land surrounding Twin Lakes is public land and is thus subject to forest management by the Department of Natural Resources. A forest harvest plan was established in the mixed forest (predominantly red pine (Pinus resinosa)) adjacent to Smith Road and Twin Trails, resulting in trees being marked for preservation. A contract was let for a selective harvest in the forests in the designated harvest area based on a forest plan (Forest Resources Division 2013). The harvest began in late winter 2014 and lasted throughout spring of 2014. Prior to harvest, four recorders were

49

50

Ecoacoustics

(a)

(b)

Figure 3.14  C3. Forest stand before (a) and after harvest (b). Ring on pine denotes acoustic monitoring tree to be left for seed production.

3  The Role of Sound in Terrestrial Ecosystems

placed on marked trees (trees not for harvest). The harvest operation was completed by June 2014. Recorders were placed at the same locations in order to record before and after the forest was harvested. Monitoring continued in 2015 to determine if the forest soundscape changed due to the selective harvest. The objective of case 3 is to show how a partial forest harvest has an effect on the forest soundscape. Figure 3.14 shows the forest before and after tree harvest. 3.4.2  C3 Methods

Four SM2 Song Meters (Wildlife Acoustics 2014) were placed in the upland mixed forest designated for selective harvest one year prior to harvest and the same four Song Meters were placed in the same location for two years after the forest was harvested. The soundscape was monitored at half‐hour intervals for one minute, resulting in 48 recordings per day. Recordings were made at 22 050 Hz 16 bit in monaural. Recordings were collected monthly and transferred via ftp to the REAL server where they were integrated into the digital database (Kasten et al. 2012). The NDSI was used to determine change in the soundscape because it has been shown to indicate ecosystem conditions (Fuller et al. 2015) and thus was used to determine if the soundscape changed as a result of the harvest (Fuller et al. 2015; Gage and Axel 2014; Kasten et al. 2012). A Matlab script was used to compute the NDSI and other soundscape metrics from 87 071 recordings made prior to and after trees were harvested (Gage et al. 2013). The NDSI is used as an indicator of soundscape quality where higher values of NDSI indicated higher habitat quality and low values indicated lower quality. NDSI values can range from –1 to +1: values below 0 indicate the dominance of low‐frequency sounds (primarily human made) and values greater than 0 indicate the dominance of high‐frequency sounds

Table 3.4  C3 – Site, location, year, start day, end day, and number of recordings Site

Location

Year

Start day of year

End day of year

Audio recordings

LA09

Red Pine Trail

2013

125

295

8092

LA09

Red Pine Trail

2014

158

279

6772

LA09

Red Pine Trail

2015

123

274

7228

LA10

Red Pine Landing

2013

125

295

6608

LA10

Red Pine Landing

2014

158

300

6818

LA10

Red Pine Landing

2015

123

274

7217

LA11

Twin Trail

2013

125

295

8096

LA11

Twin Trail

2014

158

300

6820

LA11

Twin Trail

2015

123

274

7227

LA12

Twin End

2013

125

295

8096

LA12

Twin End

2014

158

300

6822

LA12

Twin End

2015

123

274

7226

Note: the forest harvest began in late winter, 2014.

51

Ecoacoustics

(primarily made by birds). Minitab (Minitab v 16, 2010) was used to visualize the soundscape and to compute statistical comparisons to determine significance. 3.4.3  C3 Results

Table 3.4 shows the number of recordings at each location before (2013) and after (2014–2015) partial tree harvest with the start day and end day of year. In 2013, prior to tree harvest, 30 891 recordings were made. Thereafter 27 232 (2014) and 28 948 (2015) recordings comprised the dataset. 3.4.3.1  C3 Changes in the Soundscape

Figure 3.15 shows changes in the soundscape after forest harvest using the NDSI. It shows that the overall NDSI was greatest in 2013 (prior to harvest) and declined in both 2014 and 2015 (after the harvest was completed). There were also differences All Months

0.7 0.6 NDSI

0.5

12

11

LA

10 20

15

LA

09

LA

LA

12

11

LA

10

14

LA

09

LA

20

20

Year

LA

12

11

LA

10

LA 13

LA

LA

0.3 Site

09

0.4

(a) All Sites

0.7 0.6 NDSI

52

0.5 0.4 0.3

2013

2014 Year (b)

2015

Figure 3.15  C3. Normalized Difference Soundscape Index (NDSI) values for each site (a) and combined NDSI for all four sites (b) for the preharvest year (2013) and postharvest years (2014, 2015).

3  The Role of Sound in Terrestrial Ecosystems

in the sites that were recorded (see Figure 3.15a). Prior to harvest (2013), sites LA09–LA11 had about the same NDSI whereas the NDSI was higher at site LA12. Although the NDSI values were lower in 2014 and 2015, the same pattern of NDSI occurred in 2014 whereas in 2015 the NDSI in LA09 was lowest, followed by LA10 and LA11. Site LA12 maintained the highest NDSI value prior to harvest and after harvest years. The NDSI values for June, July, August, and September are shown in Figure 3.16. In each of these four months, the NDSI was greater before harvest. June is the peak breeding period for many species of birds and July is when many species produce young. The soundscape deteriorated after harvest and in June, continued to decline one year after harvest (2015). In July, the soundscape quality declined after harvest to a greater extent than in June and remained low in 2015. The soundscape patterns were similar but a less dramatic decline occurred in both August and September after harvest.

June 0.7

NDSI

0.6 0.5 0.4 0.3

2013

2014 Year (a)

2015

July

0.7

NDSI

0.6 0.5 0.4 0.3

2013

2014 Year (b)

2015

Figure 3.16  C3. NDSI values before (2013) and after harvest (2014–2015) based on recordings made at 30‐minute intervals for 60 seconds at four sites.

53

Ecoacoustics August

0.7

NDSI

0.6 0.5 0.4 0.3

2013

2014 Year (c)

2015

September

0.7 0.6 NDSI

54

0.5 0.4 0.3

2013

2014 Year (d)

2015

Figure 3.16  (Continued) Table 3.5  C3 – Analysis of variance of NDSI versus year (all data) Source

DF

Adj. SS

Adj. MS

F-value

P-value

Year

2

590.2

295.124

3362.50

0.000

Error

87 019

7637.6

0.088

Total

87 021

8227.8

DF, degrees of freedom; MS, mean squares; SS, sums of squares.

Table 3.6  C3 – Tukey pairwise comparisons NDSI versus year (all data) where grouping information is based on 95% confidence. Means that do not share a letter are significantly different Year

N

Mean

Grouping

2013

30 892

0.58622

A

2014

27 232

0.43376

B

2015

28 898

0.40011

C

3  The Role of Sound in Terrestrial Ecosystems

Table 3.7  C3 – Analysis of variance of NDSI versus year (June only) Source

DF

Adj. SS

Adj. MS

F-value

P-value

776.68

0.000

Year

2

162.7

81.3367

Error

15 827

1657.5

0.1047

Total

15 829

1820.1

DF, degrees of freedom; MS, mean squares; SS, sums of squares.

Table 3.8  C3 – Tukey pairwise comparisons of NDSI versus year (June only) where grouping information is based on 95% confidence. Means that do not share a letter are significantly different Year

N

Mean

Grouping

2013

5757

0.64624

A

2014

4313

0.50320

B

2015

5760

0.40997

C

3.4.3.2  C3 Statistical Influence of Forest Harvest

An ANOVA shows that there was a difference in the NDSI in all years (Table 3.5). A Tukey means comparison shows that each year was statistically different (Table 3.6). The same tests were computed for the month of June as this is the primary breeding season for birds in this area of Michigan and there was a difference between years for June (Table 3.7) and a Tukey comparison showed that the mean NDSI was significantly different in each year during June (Table 3.8). 3.4.4  C3 Discussion

The impact of harvest on the NDSI at site LA12 was less than at the other three sites (see Figure 3.15a). This site was closer to the edge of the area designated for harvest area and, in addition, fewer trees were harvested at this site, causing less disturbance than at the other sites. The harvest of timber from this area was selective as there were trees marked for nonharvest. In addition, the harvest did not include younger trees although much of the vegetation, including trees, was damaged due to the harvest process. It is not clear what the effect of a clear‐cut would be on biological sounds. In addition, this study only examined one index, the NDSI, and did not include the effect on species of organisms occurring in the habitat. However, all recordings are in a digital archive and a species inventory can be made in the future to determine which species were most affected. Since the soundscape was examined over a three‐year period (the harvest year and two subsequent years), the long‐term implications are unknown. The NDSI values were highest in the preharvest year and were significantly lower two years after harvest. The June NDSI values indicated a further decline in the second year after harvest (0.50 in 2014 versus 0.41 in 2015; see Table 3.8). The NDSI appears to be a good index for evaluating the effect of forest disturbance on the soundscape. There are other

55

56

Ecoacoustics

Appendix 3.1  C1 – Time, rainfall, nonnormalized soundscape power, species and other sounds that occurred at site TB01 on 14 May (day 135) at 30‐minute intervals

Time of day

Rain (listening)

Rain (in)

Mean soundscape power (nonnormalized)

0000

Light

0.03

5.56

Spring peeper, northern tree toad

0030

Medium

0.02

4.97

Spring peeper, northern tree toad, American toad

Drip

Spring peeper

Train (loud)

0100

Light

0.01

4.66

0130

Light

0.01

2.39

Biophony

Other

Roar

0200

Light

0.03

1.13

Spring peeper

Roar

0230

Medium

0.05

10.50

Cricket frog

Wind, train

0300

Medium

0.11

28.41

Spring peeper

Wind

0330

Heavy

0.21

260.68

Spring peeper

Drip

0400

Medium

0.04

22.91

Cricket frog

Wind (heavy), drip

0430

Medium

0.02

9.46

Cricket frog, American Roar, wind toad, spring peeper

0500

Light

0.01

7.43

Cricket frog

0530

None

0.00

2.87

Cricket frog, spring peeper

0600

None

0.00

5.92

Red cardinal, red‐winged blackbird, song sparrow

0630

Light

0.01

6.34

Red‐winged blackbird, yellowthroat, red cardinal, UB

0700

None

0.00

4.04

Red cardinal, yellowthroat, red‐winged blackbird

0730

None

0.00

6.21

Canada goose, red‐winged blackbird, red cardinal

0800

Light

0.00

5.05

Wind (heavy) Red cardinal, yellowthroat, red‐bellied woodpecker, red‐ winged blackbird, UB

0830

None

0.00

5.57

Red‐winged blackbird, red cardinal, yellow throat, yellow shafted flicker, American robin

Drip

Automobile

3  The Role of Sound in Terrestrial Ecosystems

Appendix 3.1  (Continued)

Time of day

Rain (listening)

Rain (in)

Mean soundscape power (nonnormalized)

0900

None

0.00

6.59

American robin, red‐winged blackbird, yellowthroat, tufted titmouse, red‐bellied woodpecker

0930

None

0.00

4.95

Red‐winged blackbird, black‐ throated green warbler

1000

None

0.00

13.03

Red cardinal, American crow, red‐winged blackbird

Automobile

1030

None

0.00

3.57

Red‐winged blackbird, red cardinal, American robin, tufted titmouse, song sparrow

Automobile

1100

None

0.00

11.00

American crow, red cardinal, red‐winged blackbird, yellowthroat

Automobile

1130

None

0.00

7.83

Red‐winged blackbird, red cardinal, American robin, yellowthroat, UB

1200

None

0.00

9.11

Red‐winged blackbird, song sparrow,

Bell, automobile

1230

None

0.00

11.60

Red‐winged blackbird

Automobile, wind (heavy)

1300

None

0.00

2.64

Red‐winged blackbird, red cardinal

Wind, automobile

1330

None

0.00

7.75

Song sparrow, red‐winged blackbird, Canada goose

Wind, automobile

1400

Light

0.00

3.51

Red cardinal, red‐winged blackbird, red‐bellied woodpecker

Automobile, wind

1430

None

0.00

5.13

Red‐winged blackbird

Wind, automobile

Biophony

Other

(Continued)

57

58

Ecoacoustics

Appendix 3.1  (Continued)

Time of day

Rain (listening)

Rain (in)

Mean soundscape power (nonnormalized)

1500

Light

0.02

1530

None

1600

Biophony

Other

2.75

Red cardinal, red‐bellied woodpecker

Wind

0.00

2.40

American crow, Canada goose, red‐bellied woodpecker

None

0.00

3.26

Black‐capped chickadee, red‐bellied woodpecker

Automobile

1630

None

0.00

3.00

Red‐bellied woodpecker, red cardinal

Automobile

1700

None

0.00

8.36

1730

None

0.00

1.50

Red‐bellied woodpecker, American crow

Automobile

1800

None

0.00

2.57

American crow, red‐bellied woodpecker

Automobile, wind

1830

None

0.00

21.57

Canada goose, red‐shafted flicker, red‐winged blackbird

Automobile

1900

None

0.00

1.09

Red‐winged blackbird, red cardinal, blue jay

Automobile

1930

None

0.00

2.82

UWP, song sparrow, Canada goose

Automobile

2000

None

0.00

12.54

Canada goose, red‐winged blackbird, song sparrow

Automobile

2030

None

0.00

2.68

Red‐winged blackbird

Automobile

2100

None

0.00

2.08

Spring peeper, red‐bellied woodpecker

Train

2130

None

0.00

0.49

Spring peeper

Automobile

2200

None

0.00

1.70

Spring peeper

Automobile

2300

None

0.00

0.68

Spring peeper

Automobile, wind

2330

None

0.00

0.61

Spring peeper

UB, unidentified bird; UWP, unidentified woodpecker.

Bell

3  The Role of Sound in Terrestrial Ecosystems

soundscape metrics and indices that could be utilized but as described in Fuller et al. (2015), the NDSI appears to be helpful for examining landscape disturbance, landscape quality, and biodiversity.

­References Acevedo, MC, Corrada‐Bravo, CJ, Corrada‐Bravo, H, Villanueva‐Rivera, LJ and Aide TM (2009) Automated classification of bird and amphibian calls using machine learning: a comparison of methods. Ecological Informatics, 4, 206–214. Black, T and Kennedy, G (2003) Birds of Michigan, Lone Pine Publishing, Edmonton. Boelman, NT, Asner, GP, Hart, PJ and Martin, RE (2007) Multi‐trophic invasion resistance in Hawaii: bioacoustics, field surveys, and airborne remote sensing. Ecological Applications, 17, 2137–2144. Brandes, TS (1970) Probabilistic bioacoustic signal extraction within spectrograms. Journal of the Acoustical Society of America, 127(3), 1970. Dong X, Towsey, M, Truskinger, Cottman‐Fields, AM, Zhang, J and Roe, P (2015) Similarity‐ based birdcall retrieval from environmental audio. Ecological Informatics, 29, 66–76. Duan, S, Zhang, J, Roe, P, et al. (2013) Timed Probabilistic Automaton: A Bridge Between Raven and Song Scope for Automatic Species Recognition. Proceedings of the 25th Innovative Applications of Artificial Intelligence Conference, pp.1519–1524. Farina, A (2014) Soundscape Ecology: Principles, Patterns, Methods and Applications, Springer, Dordrecht. Forest Resources Division (2013) Seeing the Forest, the Trees and Beyond. Strategic Plan 2014–2018, Michigan Department of Natural Resources, East Lansing. Fuller, S, Axel, AC, Tucker, D and Gage, SH (2015) Connecting soundscape to landscape: which acoustic index best describes landscape configuration? Ecological Indicators, 58, 207–215. Gage, SH and Axel, AC (2013) Visualization of temporal change in soundscape power of a Michigan lake habitat over a 4‐year period. Ecological Informatics, 2, 100–109. Gage, SH, Joo, W and Kasten, EP (2013) Biophony: A Matlab program to produce values of soundscape power and optional visualizations from multiple acoustic recordings. Remote Environmental Assessment Laboratory R‐16. Michigan State University, East Lansing. Gage, SH, Axel, AC (2014) The Biophony‐R program to compute soundscape power and soundscape indices based on R‐Seewave and R‐Soundecology packages. Remote Environmental Assessment Laboratory R‐17. Michigan State University, East Lansing. Gregory, BA, Johnson, KL, Penland, JR and Warn, EA (1997) The effect of weather on the relative volume of chorusing Spring Peepers (Hyla crucifer). Available at: https://deepblue. lib.umich.edu/handle/2027.42/54732 (accessed 12 December 2016). Hobson, KA and Schieck, J (1999) Changes in bird communities in boreal mixed wood forests: Harvest and wildfire effects over 30 years. Ecological Applications, 9, 849–863. Kasten, EP, Gage, SH, Fox, J and Joo, W (2012) The remote environmental assessment laboratorys acoustic library: an archive for studying soundscape ecology. Ecological Informatics, 12, 50–67. Minitab v. 16 (2010) Statistical Software, Minitab, Inc., State College, PA.

59

60

Ecoacoustics

Ospina, OE, Corrada‐Bravo, CJ, Villanueva‐Rivera, LJ and Aide, TM (2013) Variable response of anuran calling activity to daily precipitation and temperature: implications for climate change. Ecosphere, 4(4), 1–12. Pieretti, N, Farina, A and Morri, D (2011) A new methodology to infer the singing activity of an avian community: the Acoustic Complexity Index (ACI). Ecological Indicators, 11, 868–873. Pijanowski, BC (2015) Terrestrial soundscapes: status of ecological research in natural and human‐dominated landscapes. Advances in Experimental Medicine and Biology, 875, 839–846. Pijanowski, BC, Farina, A, Gage, SH and Dumyahn, S (2011) What Is soundscape ecology? Landscape Ecology, 26, 1213–1232. Price, SJ, Marks, DR, Howe, RW, Hanowski, JM and Niemi, GJ (2005) The importance of spatial scale for conservation and assessment of anuran populations in coastal wetlands of the Western Great Lakes, USA. Landscape Ecology, 20(4), 441–454. R Core Team (2012) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. www.R‐project.org/ Roloff, GJ, Grazia, TE, Millenbah, KF and Kroll, AJ (2011) Factors associated with amphibian detection and occupancy in southern Michigan forests. Journal of Herpetology, 45(1), 15–22. Sueur, J, Aubin, T and Simonis, C (2008) Seewave: a free modular tool for sound analysis and synthesis. Bioacoustics, 18, 213–226. Villanueva‐Rivera, LJ and Pijanowski, BC (2013) http://ljvillanueva.github.io/ soundecology/ Werner, EE, Relyea, RA, Yurewicz, KL, Skelly, DK and Davis, CJ (2009) Comparative landscape dynamics of two anuran species: climate‐driven interaction of local and regional processes. Ecological Monographs, 79, 503–521. Wilczynski, W, Zakon, HH and Brenowitz, EA (1984) Acoustic communication in spring peepers. Journal of Comparative Physiology, 155(5), 577–584. Wildlife Acoustics (2014) www.wildlifeacoustics.com Wimmer, J, Towsey, M, Planitz, B, Williamson, I and Roe, P (2013) Analysing environmental acoustic data through collaboration and automation. Future Generation Computer Systems, 29, 560–568.

61

4 The Role of Sound in the Aquatic Environment Francesco Filiciotto and Giuseppa Buscaino BioAcousticsLab, National Research Council (IAMC-CNR) - Detached Unit of Capo Granitola (TP), Italy

4.1 ­Overview on Underwater Sound Propagation 4.1.1  Sound Speed in the Sea

Sound consists of mechanical vibrations that travel as a wave. In fluid such as sea water, sound propagation is only longitudinal so water particles vibrate along the direction of energy propagation (the fluids cannot support shear wave). Sound travels about 1500 meters per second in sea water (much more than in air where the speed is about 340 m/s) and can propagate for thousands of kilometers, more than other energy forms such as electromagnetic, chemical or thermal. Sound speed can be determined from temperature, salinity, and depth using the formula of del Grosso (1974). A simplified formula by Medwin (1975) calculates the speed of sound as: c = 1449.2 + 4.6T − 0.055T 2 + 0.00029T 3 + (1.34 − 0.01T )(S − 35) + 0.016 z where c = sound speed (m/s); T = temperature (°C); S = salinity (0/00, i.e. parts per thousand); and z = depth (m). The formula is valid for realistic combinations of T, S, and z in the ranges, 0 ≤ T≤ 35 °C, 0 ≤ S ≤ 45 ppt, 0 ≤ z ≤ 1000 m. Variation of sound as a function of depth is called the sound velocity profile (SVP) (Urick 1983). One type of SVP can propagate thousands of kilometers, and another type of SVP propagates only a few kilometers. Therefore, the study of different SVPs is important for oceanographic, ecoacoustics, and ecological aspects linked to animal communication. 4.1.2  Transmission Loss

A sound signal travelling in sea water suffers from decay of acoustic intensity, called transmission loss. This is due to spreading and absorption (or attenuation). Spreading loss consists of the spreading of acoustic energy over a larger area so that the acoustic intensity, the power per unit of area, decreases as the wave propagates from the source. Spreading loss is spherical (inverse square loss), if we ignore reflection and refraction effects (i.e. for a short distance or in deep sea). In spherical spreading, the acoustic Ecoacoustics: The Ecological Role of Sounds, First Edition. Edited by Almo Farina and Stuart H. Gage. © 2017 John Wiley & Sons Ltd. Published 2017 by John Wiley & Sons Ltd.

62

Ecoacoustics

energy decreases with the square of radius of the sphere (being the surface of a sphere proportional to the square of its radius) where the radius is the distance between the sound source (center of the sphere) and the point at which we want to calculate the transmission loss. In shallow water, spreading loss is cylindrical because the sea surface and bottom reflect energy back into the water column so that the sound can only propagate horizontally. In this case, the spreading occurs only on the surface of a cylinder. Acoustic energy is spread over an area proportional to the distance r. Therefore, the propagation loss in decibels for spherical spreading equals 10 log r2, and for cylindrical spreading 10 log r. Sound also suffers conversion of acoustic energy into heat and this is called absorption loss (a). There are three main causes of absorption loss: ●● ●●

●●

viscosity caused by rubbing between the molecules change in molecular structure with dissociation and recombination of ions after the sound travels over the molecules (in the frequency range 10–200 kHz the dominant absorption is due to the ionic relaxation of MgSO4) heat conduction, even though this process is negligible.

Absorption depends on seawater properties such as temperature, pressure, salinity, and acidity as well as the frequency of the sound (Fisher and Simmons 1977). Absorption loss below 20 m can be ignored. Considering spherical spreading, the total transmission loss is expressed as: TL = 20log R + aR 4.1.3  Deep and Shallow Sound Channel and Animal Communication

Frequencies below 100 Hz in the deep ocean can be detected at ranges of thousands of kilometers. Sound speed at mid‐latitude is at a minimum at about 1000 m depth and this creates a sound channel that lets sound travel long distances in the ocean. In this channel, sound is trapped by refraction caused by different oceanographic conditions existing at different depths and in combination there are extremely low levels of absorption leading to low levels of transmission loss. This channel is called the SOund Fixing And Ranging, or SOFAR (Urick 1983). In the polar regions, because of the presence of sea ice and a positive gradient of sound speed along the depth, sound can propagate for long ranges at a shallow depth. In shallow water (10–200 m), there is another sound channel because the bottom serves as a reflective boundary that interacts with sound waves. However, sound transmission in shallow water is strongly influenced by local conditions (depth, surface roughness due to wind and bottom substrate) (Bass and Clark 2002). There are some hypotheses concerning the use of sound channels by whales and other animals to communicate longer distances. Payne and Webb (1971) proposed that certain balaenopterid whales have evolved to produce loud and infrasonic sounds that can be heard over hundreds or even thousands of miles (in some cases across ocean basins), exploiting the SOFAR channel. Bass and Clark (2002) discussed this using a simple explanation regarding the relationship between general acoustic characteristics of different whale species that inhabit either shallow and deep water and acoustic environments. They found that coastal species use sounds that matched a spectral

4  The Role of Sound in the Aquatic Environment

signature of ambient noise, using the frequency band with lower level of noise. Moreover, the three coastal species considered, bowhead, humpback and right whales, produce sounds in the range of 100–500 Hz, a frequency band where propagation is better and ambient noise is low. Deep sea species, like fin and blue whales, produce most of their sounds in lower frequencies (15–30 Hz). The shallow water would act as a high‐pass filter with the cut‐off frequency being inversely related to water depth for these species. Bass and Clark (2002) found that “sounds from shallow water and deep water species occupy different regions of the frequency spectrum, and these spectra are well‐matched to the general ambient noise and transmission properties of the two environments.”

4.2 ­Sound Emissions and their Ecological Role in Marine Vertebrates and Invertebrates Many marine animals use acoustic signals to enact a wide range of biological activities. Sounds are used by these animals to communicate, protect themselves, locate food, navigate under water, and/or understand their environment. In communication, sounds are particularly useful because they are able to convey a great deal of information quickly and over long distances. Changes in rate and structure of sounds are suited to different messages. Many aquatic organisms, including marine mammals, fish, and invertebrates (primarily crustaceans), produce sounds with emission frequencies ranging from infrasound to ultrasound. Their signals are integrated with the natural environment by forming complex and various acoustic fields. 4.2.1  Marine Mammals

Among all aquatic species, marine mammals, especially dolphins, have developed the most sophisticated and specialized sound‐producing and ‐receiving system in nature. Dolphins communicate using a combination of visual, tactile, acoustic, and possibly chemosensory channels (Herman and Tavolga 1980). Acoustic signals are the primary means by which delphinids mediate social processes that involve group coordinated behaviors such as foraging and predator defense, navigation, and maintaining contact between widely dispersed individuals. Cetaceans, particularly dolphins, in which the most extensive studies on the interpretation of the acoustic signals have been made, produce a variety of sounds, which can be classified broadly into three categories: echolocation click trains, burst pulse click trains, and whistles. The echolocation clicks of most dolphin species are extremely short (from 50 to about 100 µs) and generally show a wide frequency band which extends from 2 to over 200 kHz, and a source level variable between 140 and 220 dB 1 μPa at 1 m, as reported by Au (1993). These signals, repeated in long sequences (trains), are used to locate prey and other objects placed within short distances. Burst pulse signals are also generally associated with echolocation, and because they are often recorded during periods of high social activity, burst pulse click trains are thought to play an important role in communication (Herzing 1988; Overstrom 1983; Popper 1980). The main difference between echolocation clicks and burst pulse signals is the number of clicks produced per unit of time and the amplitude of the signal (Buscaino et al. 2015).

63

64

Ecoacoustics

The third category of signals produced by many dolphin species is long‐duration, frequency‐modulated whistles, which are also associated with social communication (Herman and Tavolga 1980), and which can be emitted simultaneously with burst pulse or echolocation clicks (Cranford 2000). Whistles are among the most variable signals produced by dolphins, varying in frequency from 1 to 20 kHz. Some whistles produced by the species Tursiops truncates have a kind of subjective “signature” that allows individuals to identify themselves (Janik and Slater 1998; Smolker et al. 1993). The acoustic ability of dolphins (May‐Collado and Wartzok 2008) has been revealed in their ability to imitate vocalizations of conspecifics (Janik 2000), modify signals in relation to environmental cues and pollution noise sources (La Manna et al. 2013; May‐ Collado and Wartzok 2008; Morisaka et al. 2005), and modulate signals in relation to behavioral (Hawkins 2010), ecological, and genetic factors (Papale et al. 2014). 4.2.2 Fish

Fish produce sounds for several reasons, including staying in contact with the school, alerting conspecifics about the presence of treats, to attract, stimulate and communicate with conspecifics, to intimidate predators which might consume their eggs and juveniles and probably, in few species, to perform echolocation activity. Fish produce different types of sounds using different mechanisms for different reasons. Sounds (vocalizations) may be intentionally produced as signals to predators or competitors, to attract mates, or as a fright response. Sounds are also produced unintentionally including those made as a by‐product of feeding or swimming. The three main ways fishes produce sounds are using sonic muscles located on or near their swim bladder (drumming); striking or rubbing together skeletal components (stridulation); and by quickly changing speed and direction while swimming (hydrodynamics). Almost all species produce sound using their teeth or by using muscles on or near their swim bladder (also called gas bladder). In some species, there is a duct between the swim bladder and esophagus (the pneumatic duct) and these are called “physostomous” fish. Fish with this structure come to the surface and “gulp” air that is directed via this duct into the swim bladder. Other fishes, including all those that live deep in the ocean, have a special gas gland and rete mirabile within the wall of the swim bladder, which is called a “physoclistous” swim bladder. Vocal fish produce sounds that commonly comprise low‐frequency pulses (the majority of sounds are at a low frequency, typically less than 1000 Hz) that vary in duration, number, and repetition rate (Myrberg et  al. 1978; Winn 1964). Most fish show very limited amplitude and frequency modulation in their sounds (Kaatz 1999; Ladich 1997; Lugli et al. 1997) and few acoustical repertoires. Toadfish are probably the best investigated fish family in terms of behavioral and neurophysiological aspects of sound production. The majority of species in this family, both males and females, produce agonistic grunts, which are broadband sounds made up of short duration pulses, but only males emit tonal courtship sounds (boat whistles or hums) produced by the extremely fast contraction of the intrinsic sonic muscles on the swim bladder (Fine et al. 2001; Skoglund 1961). The marine group of codfish also contain a large number of vocal species that emit sounds in agonistic and reproductive contexts. Agonistic sounds have been described for the cod (Gadus morhua), the haddock (Melanogrammus aeglefinus), the pollack

4  The Role of Sound in the Aquatic Environment

(Pollachius pollachius), and the more distantly related gadiformes. Pollack and tadpole fish emit grunts, cod and haddock produce knocks and grunts, and shore rockling makes thump‐like sounds in agonistic or alarm situations (Almada et  al. 1996; Amorim 1996; Fish and Mowbray 1970; Hawkins and Rasmussen 1978; Midling et al. 2002). The diversity of sounds produced by fishes show differences between species of the same family and sometimes within species (individual differences), as well as variation associated with the context of sound production. This variability is mainly based on temporal patterning of sounds or pulses within a sound and on frequency variation (sometimes modulation). Such variability has been found to play a role in the social life of fishes. Variability among sounds produced by conspecifics or interspecifics may be used in reproductive isolation contexts, mate choice (many fish, usually males, emit sounds to attract their partner(s) to the spawning site), evaluation of opponents, or identification of competitors. 4.2.3 Crustaceans

Several species of marine crustaceans have evolved various sound production mechanisms such as stick and slip friction (Meyer‐Rochow and Penrose 1976; Patek 2001; Patek and Baio 2007), percussion or rubbing (Imafuku and Ikeda 1990), stridulation (Boon et al. 2009), carapace vibrations (Patek and Caldwell 2006), snaps (Knowlton and Moulton 1963), emission of bubbles (Crane 1966), mandible grinding (Meyer‐ Rochow and Penrose 1976) and contraction of internal muscles (Henninger and Watson 2005). Spiny lobsters are able to produce highly specialized acoustic signals called rasp. These signals are produced by a specialized stridulating organ composed of two parts: a movable plectrum attached to the last segment of the antennal peduncle and a rigid file (Meyer‐Rochow and Penrose 1974, 1976; Patek 2002; Patek and Baio 2007; Patek and Oakley 2003). The lobster draws the plectrum up the file by moving the antenna base posteriorly to produce sound. Friction between the soft underside of the plectrum and the anteriorly projecting scales of the file causes the moving plectrum to alternately stick and slip, producing a pulse of sound with each slip (Meyer‐ Rochow and Penrose 1974, 1976; Patek 2001, 2002). Although some authors have assumed that lobsters produce sounds in antipredator contexts (Bouwma and Herrnkind 2009; Meyer‐Rochow and Penrose 1974; Patek and Oakley 2003), the role of lobster stridulation and the potential receivers remain unclear. Buscaino et  al. (2011) revealed a further ultrasound signal (screech) produced by lobsters, having lower duration, number of pulses per signal, bandwidth, and peak intensity and higher pulse rate and peak frequency than the rasp. The authors hypothesized that screech signals, also emitted under nonstressful conditions, may play a role in intraspecific communication. Semi‐terrestrial crabs produce sounds that are transmitted through the air and the substrate. These crabs perceive substrate vibration through mechanoreceptors and some species perceive air‐borne sound through pressure‐sensitive membranes. Some crabs use stridulation to generate sounds. Multiple anatomical structures on the claws, walking legs, and the carapace of fiddler crabs are used for stridulation. Mangrove crabs also produce stridulation. These crabs have hard ridges, or tubercles, and rows of

65

66

Ecoacoustics

bristles, called setae, on the dorsal side of their enlarged claws. When one claw rubs against the other, a rasping sound is generated. The crabs may also place one claw on the substrate and move the other claw up and down against the stationary claw, producing sound. Semi‐terrestrial crabs also produce sound by vibrating their appendages (legs, claws, tail), dancing, drumming two body parts together, and/or drumming a body part against the substrate. While the sounds of each species differ in temporal pattern as well as spectral energy distribution, they are similar in temporal characteristics. Each acoustic signal consists of several (3–10) “pulses,” with consecutive sounds produced at intervals ranging from 2 seconds to 15 seconds (Salmon 1983). Males produce sounds at slow rates (calling) if not stimulated. When stimulated by other crabs, rates of sound production can double. These faster and typically louder emissions are known as “courtship” sounds (Salmon 1965, 1967). Most of the spectral energy in the calls is confined to frequencies between 300 Hz and 3 kHz (Popper et al. 2001). Higher spectral energies are typical of rapping or stridulation, while lower frequencies are produced by leg vibrations (e.g. fiddler crabs) (Salmon 1967). Male fiddler and ghost crabs produce rapping or honking sounds as one method to court females for reproduction. During the day, male fiddler crabs wave their enlarged claws to attract females to their burrow. Once a female crab approaches, the male switches to sound production from within or just outside his burrow. To produce sound, the male crabs will strike the substrate with the lower base of their enlarged claw, drum on the substrate with both claws, and/or tap the ground with their walking legs. Calling bouts can last several minutes. Buscaino et al. (2015) observed that Ovalipes trimaculatus produces wide frequency band multipulse signals, with significant differences between males and females. The sound emissions were not accidental events correlated with locomotor activities. In trials involving precopulatory females, the total number of sounds was significantly higher compared to trials with noncopulatory females, indicating that in O. trimaculatus, sound emissions play a role in intraspecific communication related to sexual attraction. Although marine crabs produce sound primarily for courtship, it is possible to assume that some sounds are warning signals to communicate information to other crabs about the presence of a nearby predator. Among the several species of snapping shrimp, only specimens belonging to genera Alpheus and Synalpheus are able to produce vigorous snapping sounds (Johnson et al. 1947). Their sounds are extremely broadband with energy from a few hundred Hz to frequencies of above 200 kHz (Ahn et al. 1993; Cato and Bell 1992), and amplitudes measured in excess of 190 dB 1 μPa at 1 m (Au and Banks 1998; Schmitz 2002). Large aggregations of snapping shrimp can produce a loud crackling sound that is a major source of biological noise in coastal areas (Hulbert 1943; Urick 1983). Snaps are used by shrimps as signals in conspecific territorial interactions, but also in stunning prey or deterring predators (Knowlton and Moulton 1963). Furthermore, aggression (Knowlton and Keller 1982), and intra‐ and interspecific agonistic (Hughes 1996) behaviors have been well documented for some species. There is also growing evidence that snapping shrimp sound production may be modulated by abiotic factors, such as water temperature and dissolved oxygen concentration (Jung et  al. 2012; Watanabe et al. 2002).

4  The Role of Sound in the Aquatic Environment

4.3 ­Impacts of Anthropogenic Noise in Aquatic Environments Noise exposure is considered a threat to human health, causing not only annoyance or hearing impairment but also hypertension, ischemic heart disease and diabetes. Like humans, many animals do hear and can be affected by noise (Johansson et al. 2015). Although there is sufficient scientific evidence to suggest that noise exposure can threaten biodiversity (Slotte et al. 2004; Tyack 2008), this environmental risk has only recently gained attention among resource managers and policy makers. 4.3.1  Main Anthropogenic Sources of Noise in the Sea

In recent decades, anthropogenic activities have led to increased sea noise pollution and background sea noise (Hildebrand 2009; Ross 2005), and changing the acoustic characteristics of marine ecosystems (coastal, pelagic, and deep water) globally. In view of this, anthropogenic noise is now recognized as a major pollutant, appearing in the United Nations Convention on the Law of the Sea (UNCLOS) and in European legislation such as the Marine Strategy Framework Directive 56/2008 CE. Anthropogenic sounds emitted from different human activities vary significantly in terms of both frequency and intensity of the pollution source (Buck 1995; Conservation and Development Problem Solving Team 2000; Richardson et  al. 1995). Bearing this in  mind, the sources of anthropogenic noise pollution can be categorized into two main types: high‐intensity impulsive noise and low‐frequency continuous noise. High‐­ intensity sources are usually produced by pile driving, seismic testing, and active sonar application (Codarin et al. 2009). Sounds associated with pile driving are predominantly sited in shallow water where the docks and other overwater infrastructure are being developed, as well as the sounds associated with construction of offshore wind farms, liquid natural gas ports, and harbor construction. Seismic prospecting for hydrocarbon deposits is currently performed by using airguns (Gausland 2003; Wardle et al. 2001). Their operation consists of the production of high‐intensity sounds used to generate detailed descriptions of the ocean floor and its underlying geological formations (Gausland 2003). Sonar generating noise at various intensities is widely used by navies, commercial ships, the fishing industry, and marine research organizations (Popper and Hastings 2009). Low‐frequency continuous noise is generally produced by ships and vessels, including fishing vessels, recreational boats, personnel transport ships, and oceangoing freighters that transport large cargoes. Anthropogenic noise is due to vessel traffic mainly in coastal areas (Codarin et al. 2009; Picciulin et al. 2010), particularly at low frequencies (100 kPa) from a seismic water gun (impulse peak pressure levels ranged from approximately 8 to 200 kPa or 198–226 dB re 1 μPa peak pressure). In addition, in bottlenose dolphins (Tursiops truncatus), seismic airgun noise (44–207 kPa or 213–226 dB re 1 μPa peak pressure) produced a nervous reaction, a significant increase in aldosterone, and immune system response with significant decrease in monocytes. Filiciotto et al. (2013) studied the effects of anthropogenic noise in fish and observed higher levels of serum cortisol, glucose, red blood cell counts, hematocrit values, and hemoglobin content and reduced levels of immunity such as a decrease in white blood cells in gilthead sea bream juveniles (Sparus aurata) exposed to onshore aquaculture system noise compared with noise from offshore aquaculture systems. Buscaino et al. (2010) demonstrated a perturbance effect on glucose, lactate, and hematocrit levels in gilthead sea bream (S. aurata) and European sea bass (Dicentrarchus labrax) after exposure to a lab‐generated tone (0.1–1 kHz linear sweep, 150 dB re 1 μPa rms). Wysocki et al. (2006) showed a significant cortisol stress increase in different freshwater fish such as Cyprinus carpio, Gobio gobio, and Perca fluviatilis exposed to underwater ship noise. Smith et al. (2004) observed noise‐induced (white noise, 160–170 dB re 1 μPa) alterations in physiological stress by measuring plasma cortisol and glucose levels in goldfish (Carassius auratus). Celi et al. (2016) analyzed biochemical responses in gilthead sea bream after a 100day exposure to noise derived from original recordings of motorboats (recreational boats, hydrofoil, fishing boat, and ferry boat). The exposure to noise produced significant variations in almost all the plasma parameters assessed, including adrenocorticotrophic hormone (ACTH), cortisol, glucose, lactate, hematocrit, heat shock proteins (Hsp70), cholesterol, triglycerides, and osmolarity.

69

70

Ecoacoustics

In crustaceans, the hemolymph reflects physiological status as well as environmental fluctuations through the regulation of its components. The noise from boats led to a significant increase in hemolymph glucose (probably due to the high values of locomotor activities) and protein levels of European spiny lobsters (Palinurus elephas) (Filiciotto et al. 2014). Moreover, acoustic disturbance induced Hsp70 overexpression in P. elephas hemocytes and a reduction of the total hemocyte count (THC), suggesting the possibility of immune depletion as well as an increased risk of infection. Filiciotto et al. (2016) reported significant changes in total protein concentrations in the hemolymph and brain, DNA integrity, and expression protein levels of Hsp27 and 70 in brain tissue in the common prawn (Palaemon serratus) exposed to noise resembling a marine area with high anthropogenic acoustic pollution, revealing a stressful effect of noise on this crustacean species. All the data reported here may be particularly relevant when considering the potential effect of acoustic pollution on the biological and ecological activities of marine organisms. In fact, long‐term stress exposure could consequently compromise other elements, such as egg survival and reproductive and growth rates (Banner and Hyatt 1973; Lagardère 1982). 4.3.2.3  Behavior Alterations

Anthropogenic noise can generate different types of perturbations on the individual and collective behavior of marine organisms. However, data are lacking not only on the immediate behavioral effects but also on the effect of sound on animals far from a source or close to a source (i.e. alterations such as startle responses). Moreover, nothing is known about the long‐term effects of exposure to sound on the behavior of marine organisms or about the effects of cumulative exposure to loud sounds. Bailey et al. (2010) evaluated the distances within which various marine organisms could be affected by the noise generated by pile driving. Although auditory injury of Tursiops truncatus occurred within 100 m of the pile driving location, behavioral perturbance could be expected up to 50 km. Similarly, the behavioral alteration of minke whales (Balaenoptera acutorostrata) may occur as far as 40 km from the noise source (Bailey et al. 2010). The Blainville’s beaked whale, Mesoplodon densirostris, showed disruption of foraging behavior and avoidance responses when exposed to naval mid‐frequency sonar exercises (Tyack et al. 2011). Similarly, an induced avoidance response occurred in loggerhead turtles, Caretta caretta, from noise generated with a seismic airgun (array peak source level 252 dB re 1 μPa) (DeRuiter and Larbi Doukare, 2011). Animals react to cues from predators, and avoiding them can obviously prevent harmful encounters. A reflex or startle response to new stimuli that causes a need to investigate a source or a readiness to escape is an orientating response (Barrows 2000). Smith et al. (2004) noted that goldfish exhibited an initial startle response with a rapid burst of erratic swimming followed by general increased swimming activity with the onset of an experimental noise (bandwidth ranging from 0.1 to 10 kHz at 160–170 dB re 1 μPa total sound pressure level). Kastelein et al. (2008) determined the behavioral startle response for eight marine fish species to tones of 0.1–64 kHz. Response threshold levels varied per frequency within and between species. For sea bass, the 50% reaction threshold occurred at signals of 0.1–0.7 kHz, for thicklip mullet 0.4–0.7 kHz, for pout 0.1–0.25 kHz, for horse mackerel 0.1–2 kHz, and for Atlantic herring 4 kHz. For cod,

4  The Role of Sound in the Aquatic Environment

pollack, and eel, no 50% reaction thresholds were reached. These results showed that fish species react very differently to sound, as well as to the spectrum and level of anthropogenic sounds. The reactions of fish to anthropogenic sounds depend on several variables such as location, temperature, physiological state, age, body size, and school size so that generalizations about the effects of sound on fish should be made with caution. Picciulin et al. (2010) found that playback of underwater noises produced by a tourist ferry and a fiberglass boat did not produce short‐term behavioral reaction (aversion) on Gobius cruentatus and Chromis chromis. However, a time‐budget analysis revealed a significant change in the total time spent in caring for their nests (C. chromis) or time spent inside their shelters (G. cruentatus). Sarà et al. (2007) observed behavioral changes in bluefin tuna exposed to sound generated by three different boat types in a natural context and found that the sound generated by different boat engines and propellers exceeded ambient sound levels (sound pressure of about 100 dB re 1 μPa in the bandwidth from 70 to 20 000 Hz). Startle response is considered to be an immediate reaction to a threat in crustaceans, but their responses to minimize stressful and noxious stimuli have not been the focus of many studies (Barr et al. 2008; Bullock 1984; Buscaino et al. 2011; Brown et al. 2005; Dingemanse et al. 2002). Wale et al. (2013) found that crabs exposed to ship noise playback were slower to retreat to a shelter than those experiencing ambient noise playback, suggesting that anthropogenic noise can increase the risks of starvation and predation for this species. Several studies report a clear effect of acoustic stimulation on motility patterns of crustaceans. For example, Filiciotto et  al. (2016) observed in Palaemon serratus that exposure to motorboat noise produced significant changes in locomotor patterns. The results also revealed that more time is spent outside the shelter. Resting in prawns exposed to noise showed no significant differences in startle responses. Finally, tested animals showed significantly lower values of encounters between subjects. Furthermore, Celi et al. (2013) reported that specimens of Procambarus clarkii reduced the number of their encounters if exposed to acoustic stimuli (a linear sweep with a frequency range of 0.1–25 kHz; peak amplitude 148 dB rms re 1 μPa at 12 kHz) for 30 minutes. Finally, very few studies have been conducted on the impact of noise on molluscs. Donax variabilis, a clam, showed a reaction to sound stimuli in a laboratory aquarium by jumping out of the sand, lying on top of it for several seconds, and then digging in again (Ellers 1995). Among bivalves, Roberts et al. (2015) recently analyzed the behavioral responses of Mytilus edulis to anthropogenically generated noise, even though mussels have not yet been investigated in terms of the potential effects of noise on both behavior and biochemistry.

­References Ahn, JY, Hamada, H, Saito, K (1993) Study of positioning of snapping shrimp and the source levels of their sounds. Journal of Tokyo University of Fisheries, 75–81. Almada, VC, Amorim, MCP, Pereira, E, Almada, F, Matos, R and Godinho, R (1996) Agonistic behaviour and sound production in Gaidropsarus mediterraneus (Gadidae). Journal of Fish Biology, 49, 363–366.

71

72

Ecoacoustics

Amorim, MCP (1996) Acoustic Communication in Triglids and Other Fishes. PhD thesis, University of Aberdeen, Scotland. Amoser, S and Ladich, F (2003) Diversity in noise‐induced temporary hearing loss in otophysine fishes. Journal of the Acoustical Society of America, 113, 2170–2179. André, M, Solé, M, Lenoir, M, et al. (2011) Low‐frequency sounds induce acoustic trauma in cephalopods. Frontiers in Ecology and the Environment, 9, 489–493. Au WWL (1993) The Sonar of Dolphins, Springer Verlag, New York. Au, WW and Banks, K (1998) The acoustics of the snapping shrimp Synalpheus parneomeris in Kaneohe Bay. Journal of the Acoustical Society of America, 103, 41–47. Au, WW, Floyd, RW, Penner, RH and Murchison, AE (1974) Measurement of echolocation signals of the Atlantic bottlenose dolphin, Tursiops truncatus Montagu, in open waters. Journal of the Acoustical Society of America, 56, 1280–1290 Bailey, H, Senior, B, Simmons, D, Rusin, J, Picken, G and Thompson, PM (2010) Assessing underwater noise levels during pile‐driving at an offshore windfarm and its potential effects on marine mammals. Marine Pollution Bulletin, 60, 888–897. Banner, A and Hyatt, M (1973) Effects of noise on eggs and larvae of two estuarine fishes. Transactions of the American Fisheries Society, 1, 134–136. Barr, S, Laming, P, Dick, J and Elwood, W (2008) Nociception or pain in a decapod crustacean? Animal Behavior, 75, 745–751. Barrows, EM (2000) Animal Behavior Desk Reference: A Dictionary of Animal Behavior, Ecology, and Evolution, 2nd edn, CRC Press, Boca Raton. Bass, AH and Clark, CW (2002) The physical acoustics of underwater sound communication, in Acoustic Communication (eds AM Simmons, A Popper and R Fay), Springer, New York. Boon, PY, Yeo, DCJ and Todd, PA (2009) Sound production and reception in mangrove crabs Perisesarma spp. (Brachyura: Sesarmidae). Aquatic Biology, 5, 107−116. Bouwma, PE and Herrnkind, WF (2009) Sound production in Caribbean spiny lobster Panulirus argus and its role in escape during predatory attack by Octopus briareus. New Zeland Journal of Marine and Freshwater Research, 43, 3−13. Brown, C, Jones, F and Braithwaite, V (2005) In situ examination of boldness‐shyness traits in the tropical poeciliid Brachyraphis episcopi. Animal Behavior, 70, 1003–1009. Bruintjes, R and Radford, AN (2014) Chronic playback of boat noise does not impact hatching success or post‐hatching larval growth and survival in a cichlid fish. Peer Journal, 2, e594. Brumm, H and Slabbekoorn, H (2005) Acoustic communication in noise. Advances in the Study of Behavior, 35, 151–209. Buck, EH (1995) Acoustic Thermometry of Ocean Climate: Marine Mammal Issues, Congressional Research Service, Washington, DC. Bullock, TH (1984) Comparative neuroethology of startle, rapid escape, and giant fibermediated responses, in Neural Mechanisms of Startle Behavior (ed. RC Eaton), Springer, New York, pp.1–13. Buscaino, G, Filiciotto, F, Buffa, G, et al. (2010) Impact of an acoustic stimulus on the motility and blood parameters of European sea bass (Dicentrarchus labrax L.) and gilthead sea bream (Sparus aurata L.). Marine Environment Research, 69, 136–142. Buscaino, G, Filiciotto, F, Gristina, M, et al. (2011) Defensive strategies of European spiny lobster Palinurus elephas during predator attack. Marine Ecology Progress Series, 423, 143–154.

4  The Role of Sound in the Aquatic Environment

Buscaino, G, Buffa, G, Filiciotto, F, et al. (2015) Pulsed signal properties of free‐ranging bottlenose dolphins (Tursiops truncatus) in the central Mediterranean Sea. Marine Mammal Science, 31(3), 891–901. Buscaino, G, Gavio, A, Galvan, D, et al. (2015) Acoustic signals and behaviour of Ovalipes trimaculatus in the context of reproduction. Aquatic Biology, 24, 61–73. Cato, DH and Bell, MJ (1992) Ultrasonic Ambient Noise in Australian Shallow Waters at Frequencies Up to 200kHz, Materials Research Laboratory, Australia. Celi, M, Filiciotto, F, Parrinello, D, et al. (2013) Physiological and agonistic behavioural response of Procambarus clarkii to an acoustic stimulus. Journal of Experimental Biology, 216, 709–718. Celi, M, Filiciotto, F, Vazzana, M, et al. (2015) Shipping noise affecting immune responses of European spiny lobster Palinurus elephas (Fabricius, 1787). Canadian Journal of Zoology, 93, 113–121. Celi, M, Filiciotto, F, Maricchiolo, G, et al. (2016) Vessel noise pollution as a human threat to fish: assessment of the stress response in gilthead sea bream (Sparus aurata, Linnaeus 1758). Fish Physiology and Biochemistry, 42, 631–641. Chitre, MA, Potter, JR and Ong SH (2006) Optimal and near‐optimal signal detection in snapping shrimp dominated ambient noise. IEEE Journal of Oceanic Engineering, 31, 497–503. Clark, CW, Ellison, WT, Southall, BL, et al. (2009) Acoustic masking in marine ecosystems: intuitions, analysis, and implication. Marine Ecology Progress Series, 395, 201–222. Codarin, A, Wysocki, LE, Ladich, F and Picciulin, M (2009) Effects of ambient and boat noise on hearing and communication in three fish species living in a marine protected area (Miramare, Italy). Marine Pollution Bulletin, 52, 1880–1887. Conservation and Development Problem Solving Team (2000) Anthropogenic Noise in the Marine Environment, Potential Impacts on the Marine Resources of Stellwagen Bank and Channel Islands National Marine Sanctuaries. Graduate Program in Sustainable Development and Conservation Biology University of Maryland, College Park. Crane, J (1966) Combat, display and ritualization in fiddler crabs (Ocypodidae, genus Uca). Philosophical Transactions of the Royal Society B, 251, 459−472. Cranford, TW (2000) In search of impulse sound sources in odontocetes, in Hearing by Whales and Dolphins (eds WWL Au, AN Popper and R Fay), Springer, New York, pp.109–155. Del Grosso, VA (1974) New equation for the speed of sound in natural water (with comparison to others equations). Journal of the Acoustical Society of America, 56, 1084–1091. DeRuiter, S and Larbi Doukare, K (2012) Loggerhead turtles dive in response to airgun sound exposure. Endangered Species Research, 16, 55–63. Dingemanse, NJ, Both, C, Drent, D, van Oers, K and van Noordwijk, AJ (2002) Repeatability and heritability of exploratory behaviour in great tits from the wild. Animal Behavior, 64, 929–938. Ellers, O (1995) Discrimination among wave‐generated sounds by a swash‐riding clam. Biological Bulletin, 189, 128–137. Engås, A (1994) The effects of trawl performance and fish behaviour on the catching efficiency of demersal sampling trawls, in Marine Fish Behaviour in Capture and Abundance Estimation (eds A Ferno and S Olsen), Fishing News Books, Oxford, pp.45–68.

73

74

Ecoacoustics

Engås, A, Løkkeborg, S, Ona, E and Soldal, AV (1996) Soldal effects of seismic shooting on local abundance and catch rates of cod (Gadus morhua) and haddock (Melanogrammus aeglefinus). Canadian Journal of Fisheries and Aquatic Sciences, 53, 2238–2249. Enger, PS (1981) Frequency discrimination in teleosts: central or peripheral? in Hearing and Sound Communication in Fishes (eds WN Tavolga, AN Popper and RR Fay), Springer, New York, pp.243–255. Everest, FA (1948) Acoustical characteristics of noise produced by snapping shrimp. Journal of the Acoustical Society of America, 20, 137. Fay, RR and Popper, AN (eds) (1999) Comparative Hearing: Fishes and Amphibians, Springer, New York. Filiciotto, F, Giacalone, VM, Fazio, F, et al. (2013) Effect of acoustic environment on gilthead sea bream (Sparus aurata): sea and onshore aquaculture background noise. Aquaculture, 414–415, 36–45. Filiciotto, F, Vazzana, M, Celi, M, et al. (2014) Behavioural and biochemical stress responses of Palinurus elephas after exposure to boat noise pollution in tank. Marine Pollution Bulletin, 84, 104–114. Filiciotto, F, Vazzana, M, Celi, M, et al. (2016) Underwater noise from boats: measurement of its influence on the behaviour and biochemistry of the common prawn (Palaemon serratus, Pennant 1777). Journal of Experimental Biology and Ecology, 478, 24–33. Fine, ML, Malloy, KL, King, CB, Mitchell, SL and Cameron, TM (2001) Movement and sound generation by the toadfish swimbladder. Journal of Comparative Physiology Part A, 187, 371–379. Fish, MP and Mowbray, WH (1970) Sounds of the Western North Atlantic Fishes, Johns Hopkins University Press, Baltimore. Fisher, FH and Simmons, VP (1977) Sound absorption in seawater. Journal of the Acoustical Society of America, 62, 558–564. Gausland, I (2003) Seismic Surveys: Impact on Fish and Fisheries. Report for Norwegian Oil Industry Association (OLF), Stavanger, Norway. Graham, AL, and Cooke, SJ (2008) The effects of noise disturbance from various recreational boating activities common to inland waters on the cardiac physiology of a freshwater fish, the largemouth bass (Micropterus salmoides). Aquatic Conservation: Marine and Freshwater Ecosystems, 18, 1315–1324. Hastings, MC, Popper, AN, Finneran, JJ and Lanford, PJ (1996) Effects of low‐frequency underwater sound on hair cells of the inner ear and lateral line of the teleost fish Astronotus ocellatus. Journal of the Acoustical Society of America, 99, 1759–1766. Hawkins, AD and Myrberg, AA Jr. (1983) Hearing and sound communication under water, in Bioacoustics: A Comparative Approach (ed. B Lewis), Academic Press, San Diego, pp.347‐405. Hawkins, AD and Rasmussen, KJ (1978) The sounds of gadoid fish. Journal of the Marine Biological Association of the United Kingdom, 58, 891–911. Hawkins, E (2010) Geographic variations in the whistles of bottlenose dolphins (Tursiops aduncus) along the east and west coasts of Australia. Journal of the Acoustical Society of America, 128(2), 924–935. Henderson, D, Hu, B and Bielfeld, E (2008) Patterns and mechanisms of noise‐induced cochlear pathology, in Auditory Trauma, Protections, and Repair (eds J Schacht, AN Popper and RR Fay), Springer Science+Business Media, New York, pp.195–217.

4  The Role of Sound in the Aquatic Environment

Henninger, HP and Watson, WH (2005) Mechanisms underlying the production of carapace vibrations and associated waterborne sounds in the American lobster, Homarus americanus. Journal of Experimental Biology, 208, 3421−3429. Herman, LM and Tavolga, WN (1980) The communication systems of cetaceans, in Cetacean Behavior: Mechanisms and Functions (ed. LM Herman), Wiley, New York, pp.149–209. Herzing, DL (1988) A quantitative description and behavioral association of a burst‐pulsed sound, the squawk, in captive bottlenose dolphins, Tursiops truncatus. MSs thesis, San Francisco State University, San Francisco. Hildebrand, JA (2009) Anthropogenic and natural sources of ambient noise in the ocean. Marine Ecology Progress Series, 395, 5–20. Holles, S, Simpson, SD, Radford, AN, Berten, L and Lecchini, D (2013) Boat noise disrupts orientation behaviour in a coral reef fish. Marine Ecology Progress Series, 485, 295–300. Hughes, M (1996) The function of concurrent signals: visual and chemical communication in snapping shrimp. Animal Behavior, 52, 247–257. Hulbert, EO (1943) An underwater sound of natural origin. Journal of the Acoustical Society of America, 14, 173–174. Imafuku, M and Ikeda, H (1990) Sound production in the land hermit crab Coenobita purpureus Stimpson, 1858 (Decapoda, Coenobitidae). Crustaceana, 58, 68−174. Janik, VM (2000) Whistle matching in wild bottlenose dolphins Tursiops truncatus. Science, 289, 1355–1357. Janik, VM and Slater, PJB (1998) Context specific use suggests that bottlenose dolphin signature whistles are cohesion calls. Animal Behaviour, 56, 829–838. Johansson, K (2011) Impact of Anthropogenic Noise on Fish Behaviour and Ecology, Available at: http://pub.epsilon.slu.se/8366/ (accessed 13 December 2016). Johansson, T, Hallander, J, Karlsson, R, Långström, A and Turesson, M (2015) Full Scale Measurement of Underwater Radiated Noise from a Coastal Tanker. Presented at OCEANS’15, Genova, Italy. Johnson, MW, Everest, FA and Young, RW (1947) The role of snapping shrimp (Crangon and Synalpheus) in the production of underwater noise in the sea. Biological Bulletin, 93, 122. Jung, SK, Choi, BK, Kim, BC, et al. (2012) Seawater temperature and wind speed dependences and diurnal variation of ambient noise at the snapping shrimp colony in shallow water of Southern Sea of Korea. Japan Journal of Applied Physics, 51, 07GG09. Kaatz, IM (1999) The Behavioral and Morphological Diversity of Acoustic Communication in a Clade of Tropical Catfishes (Superfamily Ariodei). PhD thesis, SUNY, Syracuse, New York. Kastelein, RA, Heul, S, Verboom, WC, Jennings, N, Veen, J and de Haan, D (2008) Startle response of captive North Sea fish species to underwater tones between 0.1 and 64 kHz. Marine Environmental Research, 65, 369–377. Knowlton, N and Keller, BD (1982) Symmetric fights as a measure of escalation potential in a symbiotic, territorial snapping shrimp. Behavioral Ecology and Sociobiology, 10, 289–292. Knowlton, RE and Moulton, JM (1963) Sound production in the snapping shrimps Alpheus (Crangon) and Synalpheus. Biological Bulletin, 125, 311–331. Ladich, F (1990) Vocalization during agonistic behavior in Cottus gobio L. (Cottidae): an acoustic threat display. Ethology, 84, 193–201.

75

76

Ecoacoustics

Lagardère, JP (1982) Effects of noise on growth and reproduction of Crangon crangon in rearing tanks. Marine Biology, 71, 177–185. La Manna, G, Manghi, M, Pavan, G, et al. (2013) Behavioural strategy of common bottlenose dolphins (Tursiops truncatus) in response to different kinds of boats in the waters of Lampedusa Island (Italy). Aquatic Conservation, 23, 745e757. Lammers, MO, Brainard, RE, Au, WWL, Mooney, TA and Wong, KB (2008) An ecological acoustic recorder (EAR) for long‐term monitoring of biological and anthropogenic sounds on coral reefs and other marine habitats. Journal of the Acoustical Society of America, 123, 1720. Legg, M, Duncan, A, Zaknich, A and Greening, M (2007) Analysis of impulsive biological noise due to snapping shrimp as a point process in time. Oceans, 7, 1–6. Løkkeborg, S and Soldal, AV (1993) The influence of seismic exploration with airguns on cod (Gadus morhua) behaviour and catch rates. ICES Marine Science Symposia, 196, 62–67. Lonsbury‐Martin, BL, Martin, GK and Bohne, BA (1987) Repeated TTS exposures in monkeys: alterations in hearing, cochlear structure, and single‐unit thresholds. Journal of the Acoustical Society of America, 54, 1750–1754. Lucke, K, Siebert, U, Lepper, PA and Blanchet, MA (2009) Temporary shift in masked hearing thresholds in a harbor porpoise (Phocoena phocoena) after exposure to seismic airgun stimuli. Journal of the Acoustical Society of America, 125, 4060–4070. Lugli, M, Torricelli, G, Pavan, G and Mainardi D (1997) Sound production during courtship and spawning among freshwater gobiids (Pisces, Gobiidae). Marine and Freshwater Behavior and Physiology, 29, 109–126. May‐Collado, LJ and Wartzok, D (2008) A comparison of bottlenose dolphin whistle in the Western Atlantic Ocean: insights on factors promoting whistle variation. Journal of Mammalogy, 89, 205–216. McCauley, RD, Fewtrell, J and Popper, AN (2003) High intensity anthropogenic sound damages fish ears. Journal of the Acoustical Society of America, 113, 638–642. McDonald, JI, Wilkens, SL, Stanley, JA and Jeffs, AG (2014) Vessel generator noise as a settlement cue for marine biofouling species. Biofouling, 30, 741–749. Medwin, H (1975) Speed of sound in water: a simple equation for realistic parameters. Journal of the Acoustical Society of America, 58(6), 1318–1319. Meyer‐Rochow, VB and Penrose, JD (1974) Sound and sound emission apparatus in puerulus and postpuerulus of the western rock lobster Panulirus longipes. Journal of Experimental Zoology, 189, 283−289. Meyer‐Rochow, VB and Penrose, JD (1976) Sound production by the western rock lobster Panulirus longipes (Milne Edwards). Journal of Experimental Marine Biology and Ecology, 23, 191−209. Midling, K, Soldad, AV, Fosseidengen, JE and Øvredal, JT (2002) Calls of the Atlantic cod: does captivity restrict their vocal repertoire? Bioacoustics, 12, 233–235. Morisaka, T, Shinohara, M, Nakahara, F and Akamatsu, T (2005) Effects of ambient noise on the whistle of Indo‐Pacific bottle nosed dolphin populations. Journal of Mammalogy, 86, 541e546. Myrberg , AA Jr (1990) The effects of man‐made noise on the behavior of marine animals. Environment International, 16, 575–586. Myrberg, AA Jr, Spanier, E and Ha, SJ (1978) Temporal patterning in acoustical communication, in Contrasts in Behavior (eds E Reese and FJ Lighter), Wiley‐ Interscience, New York, pp. 137–179.

4  The Role of Sound in the Aquatic Environment

Nachtigall, PE, Supin, AY, Pawloski, J and Au, WW (2004) Temporary threshold shifts after noise exposure in the bottlenose dolphin (Tursiops truncatus) measured using evoked auditory potentials. Marine Mammal Science, 20, 673–687. National Oceanic and Atmospheric Administration (2000) Potential Impacts on the Marine Resources and Stellwagen Bank and Channel Islands National Marine Sanctuaries, University of Maryland, College Park, pp.18–19. NRC (National Research Council) (2000) Marine Mammals and Low‐Frequency Sound: Progress Since 1994, National Academy Press, Washington, DC. Overstrom, NA (1983) Association between burst‐pulse sounds and aggressive behavior in captive Atlantic bottlenose dolphins Tursiops truncatus. Zoo Biology, 2, 93–103. Papale, E, Azzolin, M, Cascao, I, et al. (2014) Acoustic divergence between bottlenose dolphin whistles from the Central‐Eastern North Atlantic and Mediterranean Sea. Acta Ethologica, 17, 155–165. Patek, SN (2001) Spiny lobsters stick and slip to make sound. Nature, 411, 153−154. Patek, SN (2002) Squeaking with a sliding joint: mechanics and motor control of sound production in palinurid lobsters. Journal of Experimental Biology, 205, 2375−2385. Patek, SN and Baio, JE (2007) The acoustic mechanics of stickslip friction in the California spiny lobster (Panulirus interruptus). Journal of Experimental Biology, 210, 3538−3546. Patek, SN and Caldwell, RL (2006) The stomatopod rumble: low frequency sound production in Hemisquilla californiensis. Marine and Freshwater Behavior and Physiology, 39, 99−111. Patek, SN and Oakley, TH (2003) Comparative tests of evolutionary trade‐offs in a palinurid lobster acoustic system. Evolution, 57, 2082−2100. Payne, R and Webb, D (1971) Orientation by means of long range acoustic signalling in baleen whales. Annals of the New York Academy of Science, 188, 110–141. Picciulin, M, Sebastianutto, L, Codarin, A, Farina, A and Ferrero, EA (2010) In situ behavioural responses to boat noise exposure of Gobius cruentatus (Gmelin, 1789; fam Gobiidae) and Chromis chromis (Linnaeus, 1758; fam Pomacentridae) living in a marine protected area. Journal of Experimental Marine Biology and Ecology, 386, 125–132. Popper, AN (1980) Sound emission and detection by delphinids, in Cetacean Behavior: Mechanisms and Function (ed. LM Herman), Wiley‐Interscience, New York, pp.1–52. Popper, AN and Clarke, NL (1976) The auditory system of the goldfish (Carassius auratus): effects of intense acoustic stimulation. Comparative Biochemistry and Physiology A Physiology, 53, 11–18. Popper, AN and Fay, RR (1993) Sound detection and processing by fish: critical review and major research questions. Brain, Behavior and Evolution, 41, 14–38. Popper, AN and Hastings, MC (2009) The effects of human‐generated sound on fish. Integrative Zoology, 4, 43–52. Popper, AN, Salmon, M and Horch, KW (2001) Acoustic detection and communication by decapod crustaceans. Journal of Comparative Physiology Part A, 187, 83−89. Popper, AN, Fay, RR, Platt, C and Sand, O (2003) Sound detection mechanism and capabilities of teleost fishes, in Sensory Processing in Aquatic Environments (eds SP Collin and NJ Marshall), Springer, New York, pp.3–38. Popper, AN, Smith, ME, Cott, PA, et al. (2005) Effects of exposure to seismic air gun use on hearing of three fish species. Journal of the Acoustical Society of America, 117, 3958–3971.

77

78

Ecoacoustics

Richardson, WJ, Finley, KJ, Miller, GW, Davis, RA and Koski, WR (1995) Feeding, social and migration behavior of bowhead whales, Balaena mysticetus, in Baffin Bay vs. the Beaufort Sea – regions with different amount of human activity. Marine Mammal Science, 11, 1–45. Roberts, L, Cheesman, S, Breithaupt, T and Elliott, M (2015) Sensitivity of the mussel Mytilus edulis to substrate‐borne vibration in relation to anthropogenically generated noise. Marine Ecology Progress Series, 538, 185–195. Romano, TA, Keogh, MJ, Kelly, C, et al. (2004) Anthropogenic sound and marine mammal health: measures of the nervous and immune systems before and after intense sound exposure. Canadian Journal of Fisheries and Aquatic Science, 61, 1124–1134. Ross, D (2005) Ship sources of ambient noise. IEEE Journal of Oceanic Engineering, 30, 257–261. Salmon, M (1965) Waving display and sound production in the courtship behavior of Uca pugilator, with comparisons to U. minax and U. pugnax. Zoologica, 50, 123–150. Salmon, M (1967) Coastal distribution, display and sound production by Florida fiddler crabs (Genus Uca). Animal Behaviour, 15, 449−459. Salmon, M (1983) Acoustic “calling” by fiddler and ghost crabs. Australian Museum Memoir, 18, 63–76. Samson, JE, Mooney, TA, Gussekloo, SWS and Hanlon, RT (2014) Graded behavioural responses and habituation to sound in the common cuttlefish, Sepia officinalis. Journal of Experimental Biology, 217, 4347–4355. Sandström, A, Eriksson, BK, Karas, P, Isaeus, M and Schreiber, H (2005) Boating and navigation activities influence the recruitment of fish in a Baltic Sea archipelago area. Ambio, 34, 125–130. Santulli, A, Modica, A, Messina, C, et al. (1999) Biochemical responses of European sea bass (Dicentrarchus labrax L.) to the stress induced by off shore experimental seismic prospecting. Marine Pollution Bulletin, 38, 1105–1114. Sarà, G, Dean, JM, d’Amato, D, et al. (2007) Effect of shipping traffic on behaviour of bluefin tuna (Thunnus thynnus). Marine Ecology Progress Series, 331, 243–253. Saunders, JC, Cohen, YE and Szymko, YM (1991) The structural and functional consequences of acoustic injury in the cochlea and peripheral auditory system: a five‐year update. Journal of the Acoustical Society of America, 90, 147–155. Schmitz, B. (2002) Sound Production in Crustacea with Special Reference to the Alpheidae. The Crustacean Nervous System, Springer, Berlin, pp.536–547. Scholik, AR and Yan, HY (2001) Effects of underwater noise on auditory sensitivity of a cyprinid fish. Hearing Research, 152, 17–24. Schwarz, AL and Greer, GL (1984) Responses of Pacific herring, Clupea harengus pallasi, to some underwater sounds. Can. J. Fish. Aquat. Sci. , 41, 1183–1192. Skalski, JR, Pearson, WH and Malme, CI (1992) Effects of sounds from a geophysical survey device on catch‐per‐unit‐effort in a hook‐and‐line fishery for rockfish (Sebastes spp.). Canadian Journal of Fisheries and Aquatic Science, 49, 1357–1365. Skoglund, CR (1961) Functional analysis of swimbladder muscles engaged in sound production of the toadfish. Journal of Biophysical and Biochemical Cytology, 10, 187–200. Slotte, A, Hansen, K, Dalen, J and Ona, E (2004) Acoustic mapping of pelagic fish distribution and abundance in relation to a seismic shooting area off the Norwegian west coast. Fisheries Research, 67, 143–150.

4  The Role of Sound in the Aquatic Environment

Smith, ME, Kane, AS and Popper, AN (2004) Noise‐induced stress response and hearing loss in goldfish (Carassius auratus). Journal of Experimental Biology, 207, 427–435. Southall, BL, Schusterman, RJ and Kastak, D (2003) Acoustic communication ranges for northern elephant seals (Mirounga angustirostris). Aquatic Mammals, 29, 202–213. Sun, Y, Song, Y, Zhao, J, et al. (2001) Effect of drilling noise and vibration on growth of carp (Cyprinus carpio) by cut‐fin marking. Marine Fisheries Research, 22, 62–68. Thomas, JA, Kastelein, RA and Awbrey, FT (1990) Behaviour and blood catecholamines of captive belugas during playbacks of noise from an oil drilling platform. Zoo Biology, 9, 393–402. Tyack, PL (2008) Implications for marine mammals of large‐scale changes in the marine acoustic environment. Journal of Mammalogy, 89, 549–558. Tyack, PL, Zimmer, WM, Moretti, D, et al. (2011) Beaked whales respond to simulated and actual navy sonar. PLoS One, 6, e17009. Urick, RJ (1983) Principles of Underwater Sound, McGraw‐Hill, New York. Vasconcelos, RO, Amorim, MC and Ladich, F (2007) Effects of ship noise on the detectability of communication signals in the Lusitanian toadfish. Journal of Experimental Biology, 210, 2104–2112. Voellmy, IK, Purser, J, Flynn, D, Kennedy, P, Simpson, SD and Radford, AN (2014a) Acoustic noise reduces foraging success via different mechanisms in two sympatric fish species. Animal Behavior, 89, 191–198. Voellmy, IK, Purser, J, Simpson, SD and Radford, AN (2014b) Increased noise levels have different impacts on the antipredator behaviour of two sympatric fish species. PLoS One, 9, e102946. Wale, MA, Simpson, SD and Radford, AN (2013) Noise negatively affects foraging and antipredator behaviour in shore crabs, Animal Behavior, 86, 111–118. Wardle, CS, Carter, TJ, Urquhart, GG, et al. (2001) Effects of seismic air‐guns on marine fish. Continental Shelf Research, 21, 1005–1027. Watanabe, M, Sekine, M, Hamada, E, Ukita, M and Imai, T (2002) Monitoring of shallow sea environment by using snapping shrimps. Water Science and Technology, 46, 419–424. Winn, HE (1964) The biological significance of fish sounds, in Marine Bioacoustics (ed. WN Tavolga), Pergamon Press, New York, pp.213–231. Wysocki, LE and Ladich, F (2005) Hearing in fishes under noise conditions. Journal of the Association for Research in Otolaryngology, 6, 28–36. Wysocki, LE, Dittami, JP and Ladich, F (2006) Ship noise and cortisol secretion in European freshwater fishes. Biological Conservation, 128, 501–508.

79

81

5 The Acoustic Chorus and its Ecological Significance Almo Farina and Maria Ceraulo Department of Pure and Applied Sciences, Urbino University, Urbino, Italy

5.1 ­Introduction A chorus may be defined as the vocalization of animals which occurs at dawn and at dusk. This phenomenon, common in nature, has perplexed scientists for decades and it remains an enigma in terms of adaptation, fascinating generations of bioacousticians and ethologists (Allard 1930; Leopold and Eynon 1961; Shaver and Walker 1930; Wright 1912, 1913). Recently, ecoacousticians have added a rich literature on this subject (Farina 2014; Farina et al. 2015; Staicer et al. 1996). However, the study of these choruses has focused on individual species, and there is little information available at the community level of organization (Allard 1930; Allen 1913; Leopold 1961; Wright 1912). Moreover, few studies have considered the role of habitat and the different composition of the communities (Farina et  al. 2015; Hasan 2011; Keast 1994). For instance, Lindenmayer et al. (2004) found a decline in bird song activity when birds moved from a eucalyptus forest to a radiate pine plantation in south‐eastern Australia. A majority of vocal animals spend part of their time calling or singing together, particularly in spring at dawn, midday or dusk (Figure 5.1). This common exchange of vocal information seems to be a paradox according to the theory of communication, but chorusing is common in insects, shrimps, fishes, frogs, birds, and mammals and it occurs at every latitude, in terrestrial, in freshwater and marine biomes (Cato 1978; D’Spain & Batchelor 2006). Calling or singing in a group provides some advantages that may include an exchange of information about food resources or security during mating behavior. The calls and songs may create indecision in predators regarding prey detection, but other explanations are possible. For instance, predators seem less active when the intensity of sunlight is low. When a predator hears an interspecific chorus, it is difficult to distinguish and locate the individual and this appears to contradict the acoustic niche theory (Krause 1993) that predicts a frequency partitioning to reduce interspecific competition. Moreover, chorus activities result in the expenditure of considerable energy (Barnett and Briskie 2007). During the chorus period, individuals interact using sound to communicate in a persistent way and individuals alternatively broadcast and receive information (Burt and Vehrencamp 2005). Ecoacoustics: The Ecological Role of Sounds, First Edition. Edited by Almo Farina and Stuart H. Gage. © 2017 John Wiley & Sons Ltd. Published 2017 by John Wiley & Sons Ltd.

82

Ecoacoustics

T1

T2

T3

Dawn chorus

T4

T1

T2

T3

T4

Post chorus

Figure 5.1  Representation of a chorus and a postchorus period at the time T1, T2, T3 and T4. The small black circles represent singing species (dawn chorus) and the empty circles silent species (post chorus).

Figure 5.2  Spectrogram of a bird chorus (Madonna dei Colli, 25042016 at 04.53, 44°12′30″N,10°03′34″E, 250 m a.s.l.). Turdus merula and Sylvia atricapilla are the major contributors. (See color plate section for the color representation of this figure.)

In terrestrial tropical biomes, insects and frogs tend to dominate the choruses but in temperate and boreal biomes birds dominate the chorus (Figure 5.2). In marine systems, snapping shrimps are major agents of the chorus but some fishes and cetaceans contribute to this phenomenon (Au et al. 2000; Cato and McCauley 2006; McCauley 2012). Choruses have been used by ecologists to assess the abundance of populations. Pellet et al. (2007) estimated the numbers of European tree frog (Hyla arborea) by counting individuals that were acoustically active during chorus time.

5.2 ­Time of Chorus The time of chorus is strongly related to breeding periods or to feeding or territorial activities. Terrestrial choruses occur especially during the breeding season. In particular, choruses produced by birds appear twice a day, at sunrise and sunset (Wright 1912)

5  The Acoustic Chorus and its Ecological Significance

(the first seems more vigorous, at least in birds) during the breeding season. Gage and Axel (2013) also showed that there were two peaks in the amount of soundscape power greater than 2 kHz at dawn and to a lesser extent at dusk in a long‐term study in northern Michigan. There is also evidence that migratory birds perform a chorus at midday, as does the song thrush (Turdus philomelos) in winter in the Mediterranean region. Bird choruses investigated in this area have a length of about 50 minutes in the morning and reach their maximum during May and June (Farina et al. 2015). Their duration largely depends on the season. Weather conditions have a great influence on chorus activity of birds (Bruni et al. 2014; Farina et al. 2015; Robbins 1981). Low temperature, heavy rains, and strong winds can depress and reduce the intensity and length of choruses (Figure 5.3). Farina et al. (2015) compared the chorus at five localities, showing the effect of geographic location and habitat type on the intensity and length of the chorus. The dawn chorus showed an interesting pattern. At sunrise, there was a dramatic reduction in singing activity, but after a few minutes vocalizations increased. This phenomenon was described in eucalyptus forests of Australia (Keast 1994) and was explained by Hutchinson (2002) as a pattern connected to the replenishment of resources that consequently produces a gradual reduction in singing activity.

Figure 5.3  A comparison between a chorus performed in good weather conditions (Madonna dei Colli, 22042016 at 0543h, 44°12′30″N,10°03′34″E, 250 m a.s.l.).) and a chorus during rain at dawn (Madonna dei Colli, 26042016 at 0543h). (See color plate section for the color representation of this figure.)

83

84

Ecoacoustics

Acoustic activity of marine species is related to lunar phases and season, changing in accordance with feeding behavior and spawning activity, with high variability between species (Amorim and Clara 2006; Hildebrand 2009). In marine systems, choruses have been observed during night hours, reaching peak energy during dusk and sunset (Fish and Cummings 1972; McCauley 2012; Radford et al. 2008b), but chorusing also occurs during the evening. McCauley and Cato (2000) analyzed daily and season rhythms of the interspecific fish chorus along the Australian coast where the calling rates reached maximum level during the summer, although some choruses occurred during winter. The summer daily cycle is characterized by a temporal shift of acoustic activities of a different kind of chorus, that reduced overlapping signals with the consequent loss of information. Radford et al. (2008b) reported a similar trend, where fish chorusing activity increased by 20 dB during the evening. Erbe (2015) reported the presence of fish chorus activity every night all year long in the Perth Canyon, with an increase in activity during the winter. This different trend could be interpreted as an adaptation to a habitat such as the Perth Canyon. In cetaceans, the presence of chorus (defined as simultaneous vocalizations of many individuals) was reported for whales, spotted and spinner dolphins, and bottlenose dolphins, generally during the night (Au et  al. 2000; Fertl 1995; Lammers 2004; Norris 1994; Powell, 1966; Smolker and Pepper 1999). Crustaceans emit a noisy chorus that dominates the soundscape of coastal areas. Most of this chorus is due to acoustic activity of different species of “snapping shrimps” which emit broadband impulsive snaps with energy extending to 200 kHz. These signals occur all day long, but the number of signal/min increases during dusk and at night, and in temperate ecosystems, these signals occur during summer, coinciding with the new moon (Buscaino et al. 2016; Lammers et al. 2008; Radford et al. 2008b). The sea urchin, Evechinus chloroticus, choruses during feeding. These are nocturnal animals (Hereu 2005; Jones and Andrew 1990; Nelson and Vance 1979) and they increase their acoustic activity during the new moon (Radford et al. 2008b). They emit signals at a frequency of 800–2400 Hz and their sound is amplified by the ovoid calcareous skeleton that functions as a Helmholtz resonator (Radford et al. 2008a). To explain the origins and patterns of bird choruses, several explanations have been proposed by Burt and Vehrencamp (2005). Staicer et al. (1996) reviewed 12 hypotheses to explain the chorus phenomenon at the individual, population, and community level (Figure 5.4). It has been recognized that causal factors occur in three broad categories: intrinsic, environmental, and social. Intrinsic factors are dominated by circadian cycles of testosterone and the physiological needs of individuals. The amount of song produced at dawn has been correlated with the amount of fat reserve accumulated during the previous day in the common blackbird (Turdus merula), in the European robin (Erithachus rubecula) (Cuthill and Macdonald 1990; Thomas 1999; Thomas and Cuthill 2002), and in silvereyes (Zosterops lateralis) (Barnett and Briskie 2007). Environmental conditions are created by natural phenomena including rain, temperature, and wind (Henwood and Fabrick 1979). Among environmental factors that may be correlated with choruses, light intensity and air motion appear to be good candidates. For instance, when light intensity is low, predation risk is negligible because predator foraging activity at this time is particularly inefficient (Berg et al. 2006; Kacelnik 1979). This is one of the possible causes of chorus temporal pattern for marine animals

5  The Acoustic Chorus and its Ecological Significance

Circadian cycles Self-stimulation Mate attraction Mate stimulation Mate guarding Territory defence Social dynamics Handicap Low predation Acoustic transmission Inefficient foraging Unpredictable conditions

Figure 5.4  The hypothesis presented to explain choruses. Source: Adapted from Staicer et al. (1996).

(Radford et al. 2008b). Air turbulence may reduce the active distance of acoustic signals at sunrise and at sunset because wind is light and signals may be transmitted for a longer distance (Dabelsteen and Mathevon 2002; Henwood and Fabrick 1979; Wiley 1991; Wiley and Richards 1978). This is the time when birds vocalize for social communication (Burt and Vehrencamp 2005; Tobias et al. 2014). In great tits (Parus major), the dawn period is important for mate selection and intrapair communication (Gorissen and Eens 2004). Mate attraction, territory defense, and the resolution of social dynamics are factors that influence chorus dynamics (Greenfield 1994; Hoi‐Leitner et al. 1995; Morse 1989). Attitude of foraging position may have implications for chorus dynamics. Berg et  al. (2006) found a temporal trend in the chorus of neotropical passerines, observing that species which first perform a song are birds foraging in the canopy, followed by songs of species that forage on the ground. Considering that choruses in birds begin when light intensity is very low (predawn), and end at postdusk before darkness arrives, the first and the last species that sing appear to be those with the largest eyes (Thomas et al. 2002). Social factors appear to be extremely important in that they create a continuous exchange of information at intra‐ and interspecific levels. Every acoustic disturbance seems to influence dawn choruses (Gil et al. 2014). During the chorus event, it is possible to evaluate acoustic niche partitioning and this has been investigated in anurans by Drewry and Rand (1983) and Sinsch et al. (2012), in cricket assemblages by Schmidt et al. (2013), in birds by Seddon (2005), Luther (2009), and Malavasi and Farina (2013), and in fish by Buscaino et al. (2016) and McCauley and Cato (2000).

85

86

Ecoacoustics

5.3 ­The Chorus Hypothesis Many investigators have attempted to explain the chorus phenomenon but it still remains a mystery. Physical and biological factors have been hypothesized, tested, and empirically verified. The acoustic transmission hypothesis (ATH) is that birds sing most intensively when song is not distorted by air turbulence and not masked by anthropogenic noise. Light winds related to the thermal behavior of the Earth can mask acoustic signals in significant ways. Dawn and dusk are periods of wind neutrality and this temporal window of calm air can be utilized for optimal singing. Late in the day, especially in tropical regions, insect calls can be severe competitors of frog and bird vocalizations. During the day, the environment is rich with sounds, including the sounds of human activity. During this time, background noise increases, thus reducing the signal‐to‐noise ratio (SNR). Brown and Handford (2003) tested the ATH on the white‐throated sparrow (Zonotrichia albicollis) and the swamp sparrow (Melospiza georgiana) at dawn and midday without obtaining clear results in terms of absolute transmission, but found the quality of the signal was higher at dawn. The ATH remains controversial. Dabelsteen and Mathevon (2002) conducted experiments on the propagation of two types of songs of blackcaps (Sylvia atricapilla) at different times of day without finding significant differences. However, this could be because this species prefers to sing later in the day (Farina et al. 2015). The seasonal variability of background noise (wind, rain) has been documented by Farina et al. (2013) in the Mediterranean maquis which from March to September has a significantly higher SNR that is favorable for singing activity (at least for birds). The lack of light at dawn may be a factor that enables birds to sing without fear of predation. After a night of rest, birds could better communicate their health status as a signal of quality of the territory (Kacelnik 1979). According to Burt and Vehrencamp (2005), the dawn chorus represents a communication network, a true acoustic community. Different models can be considered. A broadcasting network is one in which one individual sends information to two receivers. An eavesdropping network is where two senders interact and a third individual eavesdrops in order to obtain information from the two acoustically active subjects. And finally, an interactive network is where all individuals (three in the proposed model) send information and in turn eavesdrop on the other two. We have to consider that individuals are not sedentary but move continuously throughout the habitat, reducing the possibility of characterizing this as a static model. In birds, the same species can produce a great masking effect when several individuals are singing simultaneously, and it may be that an active space is associated with the type of individual reply to acoustic cues. The study of the chorus phenomenon in marine environments in terms of functional adaptation is limited by insufficient knowledge, mostly about fish sounds (Hildebrand 2009). The major hypothesis used to explain the time distribution of the chorus in marine environment is to lower the risk of predation. The reduction of light during night and new moon periods may expose possible prey with strong acoustic activity for mating display or for feeding (Radford et  al. 2008b). Otherwise, the absence of light makes acoustic communication one of the most efficient ways to transmit information when visual displays cannot. For fishes, it is recognized that arginine vasotocin (AVT)

5  The Acoustic Chorus and its Ecological Significance

has a role as a vocal activity regulator (Goodson and Bass 2000) but there is no information about how this can regulate the chorus activity. In crustaceans, the regulation of acoustic activity is hypothesized to be related also to temperature, especially for snapping shrimps that are a poikilothermic species (Radford et al. 2008b). This author proposed that there is less predation pressure when there is a positive relationship between the acoustic activity of snapping shrimp due to temperature of the water and the reduction of light. In cetaceans, chorus activity is probably due to the need to organize group activity. In spotted and striped dolphins, the presence of simultaneous emissions has been recorded during feeding activity at night, probably to assist with recruitment of others to synchronize the hunt for prey (Lammers 2004; Norris 1994; Papale et al. 2013). Recently, Kremers (2014) reported decreased breathing rates after a peak of vocal activity in a bottlenose dolphin group. These authors suggested a coordination function of whistle choruses emitted before resting activity. In humpback whales, chorus do not appear to be related to feeding activity. Au et al. (2000) reported continuous, mostly nocturnal, acoustic activity of whales from January to April along the west coast of Maui. At lower latitudes, during winter, humpback wales are not known to feed, so probably the chorusing activity is related to sexual advertisement.

5.4 ­Choruses in Birds Songbirds are active participants in both dawn and dusk choruses. The function of these choruses is explained using energetic and behavioral hypotheses. Singing is an expensive use of energy because singing at dawn especially is energy depleting when fat reserves are low. The energetic level in the morning after an overnight rest probably has an important influence on the length and quantity of song produced (Thomas and Cuthill 2002). Evidence that chorus performance is related to fat reserves has been found by Barnett and Briskie (2007) in the silvereye (Zosterops lateralis), a small passerine bird distributed throughout Australia, New Zealand, and many South Pacific islands. Males of this species, when artificially nourished, were able to sing longer with a song repertoire of higher quality. The behavioral hypothesis considers that the chorus is a mechanism to create favorable conditions for mating or for defending a territory. Foote et al. (2010) have shown that the dawn chorus of the black‐capped chickadee (Poecile atricapillus) creates a communication network in which two or three males compete. In the winter wren (Troglodytes troglodytes), Amrhein and Erne (2006) observed an increase of dawn chorus song in males when the song of another male made on the previous day was played back to simulate a territorial intrusion. This has also been observed by Erne and Amrhein (2008) with other avian species and the dawn chorus seems to be a proclamation of territory by the owners. Probably males without a territory may receive an important signal of ownership at dawn and at dusk.

5.5 ­Choruses in Amphibians Choruses are extensively used by amphibians and the sonic network is more complex than simple interindividual acoustic competition. For species that breed in ephemeral ponds, the need to concentrate the entire reproduction period into a short time causes

87

88

Ecoacoustics

males and females to exhibit behavior that creates acoustic competition and masking. When a chorus occurs at high temporal resolution, the gaps in call activity of an individual are well calibrated to avoid temporal overlap with the closest male (see Grafe 2005 for a review). The capacity to adjust the timing of a call has been observed in the neotropical treefrog (Hyla microcephala) by Schwartz (1993). A strategy to reduce interindividual competition has been observed in Boophis madascariensis, a frog native to Madagascar that can make at least 28 distinct calls (Narins et al. 2000). The search for an acceptable explanation regarding the significance of the chorus in anurans is being actively pursued by the research community. It is reasonable to assume that a synchronized chorus may produce confusion among predators, thus reducing the potential to locate an individual. This also causes an increase in female detection because synchronization increases the acoustic peaks. From the point of view of a female, attraction seems a more efficient competitive strategy because leading calls are more attractive for females than two separate sounds; when presented with a brief delay in onset, the sound higher in amplitude is the first to be localized (Litovsky et al. 1999). The capacity of males to avoid competition, at least between two or three neighboring males, has been demonstrated in the Puerto Rican treefrog (Eleutherodactylus coqui) by Brush and Narins (1989). Females appear to have the capacity to eavesdrop on active males and to assess their reproductive capacity. Leading calls seem to be preferred although in some cases follower calls are preferred. Silent males may decide to participate in the chorus, attracted by males that are calling around a female. In this case, the silent male may have an advantage due to eavesdropping. The number of calls increases with the increase in density of males around a pond, as observed by Grafe (1997) in gray tree frogs, Hyla versicolor. The theory of the honest signal can be utilized to explain male activity at dawn as a mechanism to inform a female about the fitness of an individual. This hypothesis has been verified by Murphy et al. (2008) for Eastern kingbirds (Tyrannus tyrannus). This species utilizes a distinct song at dawn and males that sing early in the morning have more mating success with females. However, in nightingales (Luscinia megarhynchos), the intensity of the dawn chorus remains constant throughout the breeding season and it appears that the song is only utilized for defending territory (Kunc et al. 2005).

5.6 ­Choruses in the Marine Environment Although the role of chorus activity in marine animals is poorly understood, several researchers have emphasized the important role of this phenomenon as a source of information for different species. The velocity of sound transmission in water is a perfect vector of information at large scales of diffusion. The chorus of the reef community may have an important role for larvae orientation and the settlement of many species of fish and crustaceans. The sounds emitted by many species together can provide indirect information about the kind of habitat (Lillis et  al. 2013; Radford et  al. 2010, 2011; Staaterman et al. 2013). The quality of this (Piercy et al. 2014) and the time period are important (Buscaino et  al. 2016; Nedelec et  al. 2015; Radford et  al. 2008b). Simpson et al. (2008) showed that some families of reef fish species (Pomacentridae, Apogonidae, Lethrinidae, and Gobiidae) are attracted to high‐frequency sounds in the reef. Generally, they are broadband impulses produced by crustaceans or mollusks, that contribute to

5  The Acoustic Chorus and its Ecological Significance

create the dawn and dusk chorus. Responsiveness in terms of behavior, settlement, and metamorphism to reef sounds was also found for crab larvae (Stanley et  al. 2010). Therefore habitat selection seems to be driven by the perception of heterospecific sounds (Jeffs et al. 2003; Simpson et al. 2005; Tolimieri et al. 2000) but the chorus phenomenon is the most important expression.

5.7 ­Conclusions and Discussion Choruses remain an important subject of behavioral and ecological investigation. This phenomenon, which is common to many species and occurs in terrestrial and aquatic biomes, requires further investigation to clarify the mechanisms and evolutionary patterns. The vocal activity at dawn and dusk of so many animal types and species has a clear evolutionary advantage. Evidence shows that choruses are depressed by disturbance events like wind or rain and by human technophonies. This demonstrates that at least some constraints are in action during the chorus sessions but a comparative analysis based on the energetic availability could improve our understanding of this phenomenon.

­References Allard, HA (1930) The first morning song of some birds of Washington, D.C.: its relation to light. American Naturalist, 64(694), 436–469. Allen, FH (1913) More notes on the morning awakening Auk, 30, 229–235. Amorim, MCP and Clara, P (2006) Diversity of sound production in fish. Communication in Fishes, 1, 71–104. Amrhein, V and Erne, N (2006) Dawn singing reflects past territorial challenges in the winter wren. Animal Behaviour, 71, 1075–1080. Au, WW, Mobley, J, Burgess, WC, Lammers, MO and Nachtigall, PE (2000) Seasonal and diurnal trends of chorusing humpback whales wintering in waters off western Maui. Marine Mammal Science, 16(3), 530–544. Barnett, CA and Briskie, JW (2007) Energetic state and performance of dawn chorus in silvereyes (Zosterops lateralis). Behavioral Ecology and Sociobiology, 61, 579–587. Berg, KS, Brumfield, RT and Apanius, V (2006) Phylogenetic and ecological determinants of the neotrpical dawn chorus. Proceedings of the Royal Society B, 273, 999–1005. Brown, TJ and Handford, P (2003) Why birds sing at dawn: the role of consistent song transmission. Ibis, 145, 120–129. Bruni, A, Mennill, DJ and Foote, JR. (2014) Dawn chorus start time variation in a temperate bird community: relationships with seasonality, weather, and ambient light. Journal of Ornithology, 155(4), 877–890. Brush, J and Narins PM (1989) Chorus dynamics of a neotropical amphibian assemblage: comparison of computer simulation and natural behaviour. Animal Behaviour, 37, 33–44. Burt, JM and Vehrencamp, SL (2005) Dawn chorus as an interactive communication network, in Animal Communication Networks (ed. PK McGregor), Cambridge University Press, Cambridge.

89

90

Ecoacoustics

Buscaino, G, Ceraulo, M, Pieretti, N et al. (2016) Temporal patterns in the soundscape of the shallow waters of a Mediterranean Marine Protected Area. Scientific Reports, 6, 34230. Cato, DH (1978) Marine biological choruses observed in tropical waters near Australia. Journal of the Acoustical Society of America, 64(3), 736–743. Cato, DH and McCauley, RD (2002) Australian research in ambient sea noise. Acoustics Australia, 30, 13–20. Cuthill, IC and Macdonald, WA (1990) Experimental manipulation of the dawn and dusk chorus in the blackbird Turdus merula. Behavioral Ecology and Sociobiology, 26, 209–216. Dabelsteen, T and Mathevon, N (2002) Why do songbirds sing intensively at dawn? A test of the acoustic transmission hypothesis. Acta Ethologica, 4, 65–72. Drewry, GE and Rand, S (1983) Characteristics of an acoustic community: Puerto Rican frogs of the genus Eleutherodactylus. Copeia, 4, 941–953. D’Spain, GL and Batchelor, HH (2006) Observations of biological choruses in the southern California Bight: a chorus at mid frequencies. Journal of the Acoustical Society of America, 10(4), 1942–1956. Erbe, C, Verma, A, McCauley, R, Gavrilov, A and Parnum, I (2015) The marine soundscape of the Perth Canyon. Progress in Oceanography, 137, 38–51. Erne, N and Amrhein, V (2008) Long‐term influence of simulated territorial intrusions on dawn and dusk singing in the winter wren: spring versus autumn. Journal of Ornithology, 149, 479–486. Farina, A (2014) Soundscape Ecology, Springer, Dordrecht. Farina, A, Pieretti, N and Morganti, N (2013) Acoustic patterns of an invasive species: the Red‐billed Leiothrix (Leiothrix lutea Scopoli 1786) in a Mediterranean shrubland. Bioacoustics, 22(3), 175–194. Farina, A, Ceraulo, M, Bobryk, C, Pieretti, N, Quinci, E and Lattanzi, E (2015) Spatial and temporal variation of bird dawn chorus and successive acoustic morning activity in a Mediterranean landscape. Bioacoustics, 24(3), 269–288. Fertl, D and Würsig, B (1995) Coordinated feeding by Atlantic spotted dolphins (Stenella frontalis) in the Gulf of Mexico. Aquatic Mammals, 21, 3–3. Fish, JF and Cummings, WC (1972) A 50‐dB increase in sustained ambient noise from fish (Cynoscion xanthulus). Journal of the Acoustical Society of America, 52(4), 1266–1270. Foote, JR, Fitzsimmons, LP, Mennill, DJ and Ratcliffe, LM (2010) Black‐capped chickadee dawn choruses are interactive communication networks. Behaviour, 147, 1219–1248. Gage, SH and Axel, AC (2013) Visualization of temporal change in soundscape power of a Michigan lake habitat over a 4‐year period. Ecological Informatics, 21, 100–109. Gage, SH and Miller CA (1978) A Long‐Term Bird Census in Spruce Budworm‐Prone Balsam Fir Habitats in Northwestern New Brunswick. Fisheries and Environment Canada, Canadian Forest Service, Maritimes Forest Research Centre, Fredericton, NB. Gil, D, Honarmand, M, Pascual, J, Perez‐Mena, E and Macias Garcia, C (2014) Birds living near airports advance their dawn chorus and reduce overlap with aircraft noise. Behavioral Ecology. DOI: 10.1093/beheco/aru207 Goodson, JL and Bass, AH (2000) Vasotocin innervation and modulation of vocal‐acoustic circuitry in the teleost Porichthys notatus. Journal of Comparative Neurology, 422, 363.

5  The Acoustic Chorus and its Ecological Significance

Gorissen, L and Eens, M (2004) Interactive communication between male and female great tit (Parus major) during the dawn chorus. Auk, 121(1), 184–191. Grafe, TU (1997) Use of metabolic substrates in the gray treefrog, Hyla versicolor: implications for calling behavior. Copeia, 2, 356–362. Grafe, TU (2005) Anuran choruses as communication networks, in Animal Communication Networks (ed. PK McGregor), Cambridge University Press, Cambridge. Greenfield, MD (1994) Synchronous and alternating choruses in insects and anurans: common mechanisms and diverse functions. American Zoologist, 34, 605–615. Hasan, NM (2011) Effect of seasonal variations, altitude and geographical location on the onset of dawn chorus in three bird species in Middle East. Open Ornithological Journal, 4, 30–34. Henwood, K and Fabrick, A (1979) A quantitative analysis of the dawn chorus: temporal selection for communicatory optimization. American Naturalist, 114, 260–274. Hereu, B (2005) Movement patterns of the sea urchin Paracentrotus lividus in a marine reserve and an unprotected area in the NW Mediterranean. Marine Ecology, 26(1), 54–62. Hildebrand, JA (2009) Anthropogenic and natural sources of ambient noise in the ocean. Marine Ecology Progress Series, 395(5), 5–20. Hoi‐Leitner, M, Nechtelberg, H and Hoi, H (1995) Song rate as a signal for nest site quality in blackcaps (Sylvia atricapilla), Behavioural Ecology and Sociobiology, 37, 399–405. Hutchinson, JMC (2002) Two explanations of the dawn chorus compared: how monotonically changing light level favour a short break from singing. Animal Behaviour, 64, 527–539. Jeffs, A, Tolimieri, N and Montgomery, JC (2003) Crabs on cue for the coast: the use of underwater sound for orientation by pelagic crab stages. Marine and Freshwater Research, 54, 841–845. Jones, GP and Andrew, NL (1990) Herbivory and patch dynamics on rocky reefs in temperate Australasia: the roles of fish and sea urchins. Australian Journal of Ecology, 15(4), 505–520. Kacelnik, A (1979) The foraging efficiency of great tits (Parus major) in relation to light intensity. Animl Behavior, 27, 237–242. Keast, A (1994) The dawn chorus in a eucalypt forest bird community, seasonal shifts in timing and contribution of individual species. Corella, 18(5), 133–140. Krause, B (1993) The niche hypothesis. Soundscape Newsletter, 6, 6–10. Kremers, D, Briseño‐Jaramillo, M, Böye, M. Lemasson, A, and Hausberger, M (2014) Nocturnal vocal activity in captive bottlenose dolphins (Tursiops truncatus): could dolphins have presleep choruses? Animal Behavior and Cognition, 1, 464–469. Kunc, HP, Amrhein, V and Naguib, M (2005) Seasonal variation in dawn song characteristics in the common nightingale, Animal Behaviour, 70, 1265–1271. Lammers, MO (2004) Occurrence and behavior of Hawaiian spinner dolphins (Stenella longirostris) along Oahu’s leeward and south shores. Aquatic Mammals, 30, 237–250. Lammers, MO, Brainard, RE, Au, WW, Mooney, TA and Wong, KB (2008) An ecological acoustic recorder (EAR) for long‐term monitoring of biological and anthropogenic sounds on coral reefs and other marine habitats. Journal of the Acoustical Society of America, 123(3), 1720–1728.

91

92

Ecoacoustics

Lillis, A, Eggleston, DB and Bohnenstiehl, DR (2013) Oyster larvae settle in response to habitat‐associated underwater sounds PloS One, 8(10), e79337. Leopold, A and Eynon, AE (1961) Avian daybreak and evening song in relation to time and light intensity. Condor, 63(4), 269–293. Lindenmayer, DB, Cunningham, RB, Donnelly, CF, Nix, HA and Lindenmayer, BD (2004) Sound recording of bird vocalization in forests. II. Longitudinal profiles in vocal activity. Wildlife Research, 31, 209–217. Litovsky, R, Colburn, HS, Yost, WA and Guzman, SJ (1999) The precedence effect. Journal of the Acoustical Society of America, 4, 1633–1654. Luther, D (2009) The influence of the acoustic community on songs of birds in a neotropical rain forest. Behavioral Ecology, 20(4), 864–871. Malavasi, R and Farina, A (2013) Neighbours talk: interspecific choruses among songbirds. Bioacoustics, 22, 33–48. McCauley, RD (2012) Fish choruses from the Kimberley, seasonal and lunar links as determined by long term sea noise monitoring, in McMinn, T. (ed), Conference Proceedings of Acoustics 2012 Fremantle: Acoustics, Development and the Environment, Nov 21-23 2012. Fremantle, Western Australia: Acoustical Society of Australia. McCauley, RD and Cato, DH (2000) Patterns of fish calling in a nearshore environment in the Great Barrier Reef. Philosophical Transactions of the Royal Society of London B: Biological Sciences, 355(1401), 1289–1293. Morse, DH (1989) Song patters of warblers at dawn and dusk. Wilson Bulletin, 101(1), 26–35. Murphy, MT, Sexton, K, Dolan, AC and Redmond LJ (2008) Dawn song of the eastern kingbird: an honest signal of male quality? Animal Behaviour, 75, 1075–1084. Narins, PM, Lewis, ER and McClelland, BE (2000) Hyperextended call note repertoire of the endemic Madagascar treefrog Boophis madagascariensis (Rhacophoridae). Journal of Zoology, 250, 283–298. Nedelec, SL, Simpson, SD, Holderied, M et al. (2015) Soundscapes and living communities in coral reefs: temporal and spatial variation. Marine Ecology Progress Series, 525, 125–135. Nelson, B and Vance, RR (1979) Diel foraging patterns of the sea urchin Centrostephanus coronatus as a predator avoidance strategy. Marine Biology, 51(3), 251–258. Norris, KS, Wursig, B, Wells, RS and Wursig, M (1994) The Hawaiian Spinner Dolphin, University of California Press, Berkeley. Pellet, J, Helfer, V and Yannic, G (2007) Estimating population size in the European tree frog (Hyla arborea) using individual recognition and chorus counts. Amphibia–Reptilia, 28, 287–294. Piercy, JJ, Codling, EA, Hill, AJ, Smith, DJ and Simpson, SD (2014) Habitat quality affects sound production and likely distance of detection on coral reefs. Marine Ecology Progress Series, 516, 35–47. Powell, BA (1966) Periodicity of vocal activity of captive Atlantic bottlenose dolphins: Tursiops truncatus. Bulletin of the Southern California Academy of Sciences, 65(4), 237–244. Radford, C, Jeffs, A, Tindle, C and Montgomery, JC (2008a) Resonating sea urchin skeletons create coastal choruses. Marine Ecology Progress Series, 362, 37–43.

5  The Acoustic Chorus and its Ecological Significance

Radford, CA, Jeffs, AG, Tindle, CT and Montgomery, JC (2008b) Temporal patterns in ambient noise of biological origin from a shallow water temperate reef. Oecologia, 156(4), 921–929. Radford, CA, Stanley, JA, Tindle, CT, Montgomery, JC and Jeffs, AG (2010) Localised coastal habitats have distinct underwater sound signatures. Marine Ecology Progress Series, 401, 21–29. Radford, CA, Stanley, JA, Simpson, SD and Jeffs, AG (2011) Juvenile coral reef fish use sound to locate habitats. Coral Reefs, 30(2), 295–305. Robbins, CS (1981) Effect of time of day on bird activity. Studies in Avian Biology, 6, 275–286. Schmidt, AD, Romer, H and Riede, K (2013) Spectral niche segregation and community organization in a tropical cricket assemblage. Behavioral Ecology, 24(2), 470–480. Schwartz, JJ (1993) Male calling behavior, female discrimination and acoustic interference in the Neotropical treefrog Hyla microcephala under realistic acoustic conditions. Behavioral Ecology and Sociobiology, 32, 401–414. Seddon, N (2005) Ecological adaptation and species recognition drives vocal evolution in neotropical suboscine birds. Evolution, 59, 200–215. Shaver, JM and Walker, G (1930) A preliminary study of the effects of temperature on the time of ending of the evening song of the monckbird. Auk, 47, 385–396. Simpson, SD, Meekan, M, Montgomery, J, McCauley R and Jeffs, A (2005) Homeward sound. Science, 30, 221. Simpson, SD, Meekan, MG, Jeffs, A, Montgomery, JC and McCauley, RD (2008) Settlement‐stage coral reef fish prefer the higher‐frequency invertebrate generated audible component of reef noise. Animal Behaviour, 75, 1861–1868. Sinsch, U, Lumkemann, K, Rosar, K, Schwarz, C and Dehling, JM (2012) Acoustic niche partitioning in an anuran community inhabiting an afromontane wetland (Butare, Ruanda). African Zoology, 47, 60–73. Smolker, R and Pepper, JW (1999) Whistle convergence among allied male bottlenose dolphins (Delphinidae, Tursiops sp.). Ethology, 105(7), 595–617. Staaterman, E, Rice, AN, Mann, DA and Paris, CB (2013) Soundscapes from a tropical Eastern Pacific reef and a Caribbean Sea reef. Coral Reefs, 32(2), 553–557. Staicer, CA, Spector, DA and Horn, AG (1996) The dawn chorus and other diel patterns in acoustic signaling, in Ecology and Evolution of Acoustic Communication in Birds (eds DE Kroodsma and EH Miller), Cornell University Press, New York, pp.426–453. Stanley, J, Radford, C and Jeffs A (2010) Induction of settlement in crab megalopae by ambient underwater reef sound. Behavioral Ecology, 2, 113–120. Stanley, JA, Radford, CA and Jeffs, AG (2012) Location, location, location: finding a suitable home among the noise. Proceedings of the Royal Society B, 279, 3622–3631. Thomas, RJ (1999) The effect of variability in the food supply on the daily singing routines of European robin: a test of a stochastic dynamic programming model. Animal Behaviour, 57, 365–369. Thomas, RJ and Cuthill, IC (2002) Body mass regulation and the daily singing routines of European robins. Animal Behaviour, 63, 285–295. Thomas, RJ, Szekely, T, Cuthill, IC, et al. (2002) Eye size in birds and the timing of song at dawn. Proceedings of the Royal Society B, 269, 831–837.

93

94

Ecoacoustics

Tobias, JA, Planqué, R, Cram, DL and Seddon, N (2014) Species interactions and the structure of complex communication networks. Proceedings of the National Academy of Sciences, 111(3), 1020–1025. Tolimieri, N, Jeffs, A and Montgomery, JC (2000) Ambient sound as a cue for navigation by the pelagic larvae of reef fishes. Marine Ecology Progress Series, 207, 219–224. Wiley, RH (1981) Association of song properties with habitats for territorial oscine birds of eastern North America. American Naturalist, 138, 973–993. Wiley, RH and Richards, DG (1978) Physical constraints on acoustic communication in the atmosphere: implications for the evolution of animal vocalization. Behavioral Ecology and Sociobiology, 3, 69–94. Wright, HW (1912) Morning awakening and even‐song. Auk, 29(3), 307–327. Wright, HW (1913) Morning awakening and even‐song: second paper. Auk, 30(4), 512–537.

95

6 The Ecological Effects of Noise on Species and Communities Almo Farina Department of Pure and Applied Sciences, Urbino University, Urbino, Italy

6.1 ­Introduction The importance of the function of sounds in nature has been discussed by Sueur and Farina (2015) who have offered a new ecoacoustics perspective to analyze and interpret the role of sound in ecological processes. In fact, sounds are integral part of nature’s “metabolism” as the majority of sounds, particularly those appreciated by people for their melodies, are the result of the acoustic activity of animals creating complex acoustic systems that Krause (2012) imagined as a “great animal orchestra.” However, sounds in nature may also originate from geophonies (wind, rain, volcanoes, water movement, sea undertow, etc.) which are important components of the soundscape. With the advent of steam and the combustion engine, new technophonic sounds were introduced into nature, especially where people live in large communities (urban and metropolitan areas) (Pivato 2011). Some sounds are considered unwanted and unpleasant and are called noise, a term coming from the Latin nausea. Noise is a sound that may be characterized by high amplitude, such as thunder or waterfalls. The Canadian composer and scholar Barry Truax defines noise in several ways, as unwanted sound, unmusical sound, or any loud sound or disturbance in any communication system (Truax 1999). Noise may also be considered as energy that has no relevance in the communication between a signaler and a particular receiver (Luther and Gentry 2013), and often noise masks or impedes communication between organisms. Noise can be also defined as a sound from which is not possible to extract explicit information, like the buzz of a multitude of people (conversation in an overcrowded bar, the choruses during a football match, or applause after an artistic performance or a political assembly). The World Health Organization (WHO 1971) considers noise as a major threat to human well‐being. Noise of anthropogenic origin also has a great impact on animal communication with effects at individual, population, and community levels (Francis et al. 2009). Noise is difficult to manage, especially in urban areas and inside or around transportation hubs such as railway stations, harbors or airports. Noise is an important cause of human annoyance, and the negative effects of large and prolonged doses of noise on human behavior and health are well documented (Babisch et  al. 2005; Ecoacoustics: The Ecological Role of Sounds, First Edition. Edited by Almo Farina and Stuart H. Gage. © 2017 John Wiley & Sons Ltd. Published 2017 by John Wiley & Sons Ltd.

96

Ecoacoustics

Barber et al. 2009; Goines and Hagler 2007). Protecting human society from noise is not a simple matter and it requires a great deal of investment in mitigation structures (Bucur 2006).

6.2 ­The Nature of Noise Noise is a sound with little intrinsic information and cannot be used as a vehicle of communication. This characteristic is common to a biosemiotics, ecological, and behavioral point of view and the role of noise remains a subject of discussion. In animal communication studies, noise is defined as “any factor that reduces the ability of receiver to detect a signal or to discriminate one signal from another” (Brumm and Slabbekoorn 2005), but from a biosemiotics perspective, noise per se is a source of information and represents a type of ecological code (Farina and Pieretti 2014) and is often a relevant part of a sonotope (Farina 2014). Nevertheless, noise may be a proxy for environmental events (Farina et al. 2016) and may be itself an event that characterizes a landscape. For instance, a natural environment noise is often the result of rain or wind and these events change with the season and are connected to climate change (Krause and Farina 2016). Such events may have significant consequences with respect to acoustic activity of vocal animals, reducing their capacity for communication and forcing the vocal organism to use extra energy to overcome noisy sources. From an anthropogenic perspective, noise is considered an event that degrades acoustic quality, especially in natural environments, and may affect the distribution of animal species. The distribution of sounds in every landscape is not homogeneous in time and space and these acoustic patches are defined as sonotopes (Farina 2014). The heterogeneous distribution of sonotopes has an effect on species that have a different tolerance for noise. With the increase of human intrusion, in an era defined as the Anthropocene (Crutzen and Stoermer 2000), the increase of noise in the soundscape is becoming more hostile to vocal animals in addition to being unhealthy for humans. The need to preserve natural sounds and to reduce the effect of sound produced by machines (technophonies) is an important issue for policy makers, stakeholders, and citizens (Luther and Gentry 2013), especially in terms of ecological conservation (Brumm 2010; Ritts et al. 2016; Slabbekoorn and Ripmeester 2008).

6.3 ­Natural Sources of Noise In nature, noise originates from abiotic and biotic agents. Abiotic agents include wind, rain, rivers, and waterfalls. Some noise sources occur in one position in the landscape and are constant over time. Many birds and frogs can adapt to these constant noise sources. An example of species that can adapt to wind occurs in the maquis of the western Mediterranean, where bird species have songs and calls that are redundant. The repetition of an acoustic performance increases the probability that the message is transferred. This is particularly evident in grassland birds that have signals that are highly redundant so they can be detected and discriminated by receivers (Brown and Handford 1996).

6  The Ecological Effects of Noise on Species and Communities

Generally, the noise close to a waterfall is so strong that only very few species of vocal animals are not disturbed by the sounds of rushing water. This is the case with the dipper (Cinclus cinclus), a bird that has a high‐frequency song that overcomes the sounds of the water (Krause et al. 2011). Wind and rain may occur anywhere in a landscape, and the acoustic frequencies from these sources generally do not mask bird sounds at moderate levels of intensity but can have an influence on the energy budget of species that increase their vocalization during such events. However, when noise has high intensity, it reduces the communication of vocal animals (Bruni et  al. 2014; Robbins 1981). In addition, air turbulence can have a major impact on birds. In fact, many species of birds can sing during rain events but strong wind causes most birds to be silent. In tropical forests, wind has a significant effect on vocalization in the midcanopy. Wind is generally absent in the early morning and at dusk, and increases in velocity in the late morning and these events are important for signal transmission. One of the explanations of the dawn and dusk choruses is that they are due to the absence of wind at sunrise and sunset (see Chapter 5). Insects like orthopterans and cicadas can vocalize during dusk at frequencies of 3–12.5 kHz, especially in tropical regions. Colonial birds (penguins, swifts, seagulls, starlings) create noise close to their breeding or roosting areas. Frogs contribute to environmental noise during their dusk and nocturnal choruses and these choruses may be used by migratory birds to navigate at night (Griffin 1976).

6.4 ­Anthropogenic Sources of Noise Urbanization, transportation, industry, and energy production are increasing with the rise in human intrusion, becoming major agents of anthropogenic noise (technophonies). Highways, railways, harbors, and airports are important sources of noise, especially in the morning and during late afternoon due to increased commuter activity. An automobile operating at 80 km/h produces noise at 71 dB at a 3 m distance and a truck operating at the same speed creates sound levels of 85 dB. Trains are very noisy (80 dB) but are not as continuous as vehicles that use roads. In remote areas, gas compression stations (Bunkley et  al. 2015) and surface mines (Duarte et  al. 2015) are important sources of noise production.

6.5 ­Effects of Noise on the Animal World The relationship between noise and animal behavior has been the subject of intensive investigation (Klump 1996), and in more recent times, it has been the object of investigation in ecoacoustics (Slabbekoorn et al. 2010). Several studies of noise and vertebrates, in particular birds, show that they are sensitive to noise because they utilize acoustic cues to communicate and to defend their territories. There is growing evidence that noise also has an impact on several species of invertebrates (see Morley et al. 2014 for a review). There is evidence that high noise levels can produce physical damage to hearing. Loud noise is also responsible for causing stress, avoidance, and fright‐flight responses. Other behaviors, such as foraging frequency and changes in reproductive success, may also be affected. Alteration of vocal

97

98

Ecoacoustics

communication, masking the ability to hear predators and other strategic sounds for animal survival, and the potential of noise to cause changes in populations are observed in species that live in noisy environments such as urban systems (Ortega 2012) (Figure 6.1). Noise has an impact on the active space of species. Active acoustic space is defined as the distance at which a signal can be detected and correctly decoded (Brenowitz 1982; Klump 1996; Marten and Marler 1977). When a species is outside the active space of an emitter, due to physical degradation, the perceived sound does not have enough information in it to be properly decoded. The active space plays an important role in territorial species. For instance, in tawny owls (Strix aluco) the active space can be reduced 69‐fold (from 118 ha to 1.7 ha) during heavy rain (Lengagne and Slater 2002). The majority of vocal animals use acoustic cues to maintain relationship between individuals, groups, and the wider community (Malavasi and Farina 2013; McComb et al. 2003). If the environment is too noisy, it means that when the signal‐to‐noise ratio is too small, it is difficult for an animal to maintain social aggregation and this may prevent the exchange of strategic information such as the location of areas with abundant food (Slabbekoorn 2004). This ratio varies by time of day or season and is probably an important factor for mate attraction. An example of the important role of sound in the ecology and behavior of vocal species occurs during dawn and dusk bird choruses, where to be effective these performances require the absence of noise such as traffic or wind (Farina et al. 2015). The presence of noise in a natural environment due to human intrusion may conflict with adaptation of vocal animals to a specific habitat. In fact, according to the acoustic adaptation hypothesis (Morton 1975), species attempt to adapt their acoustic performance to the selected habitat to improve the efficiency of their transmission of acoustic signals. When a species shifts its sonic repertoire to higher frequencies in order to reduce the masking effect of a noise, this may reduce the capacity of the signal to be maintained at the same level of information at the same distance. For instance, some alarm calls during the disturbance of a noise are uttered at a higher frequency and at higher amplitude, and this could expose a species to a higher risk of predation because it can be more easily located. These effects have a fitness cost that are discussed by Read et al. (2013). Animals exposed to anthropogenic noise may suffer physiological stress. In female wood frogs (Lithobates sylvaticus) exposed to heavy traffic noise, Tennessen et al. (2014) found an increase in glucocorticoid hormone (plasma corticosterone). This substance can have a dramatic effect on survival and on reproductive capacity in this species. Creel et al. (2002) found an increase in fecal glucocorticoids (GC) in populations of elk (Cervus elaphus) living in Yellowstone National Park as a consequence of snowmobile intrusion during winter. The variability in response of animals to noise is probably the cause of unexpected behavior in some bird species. In complex communities where predation is an important factor regulating population density, predators that are less sensitive to noise exposure than their prey, considered noise tolerant, may be able to occupy areas with a high noise level, such as those that occur in urban parks (Francis et al. 2009). Brazilian free‐tailed bats (Tadarida brasiliensis) that use sophisticated echolocation to capture flying arthropods are very sensitive to background noise which can cause a decrease of 40% in the rate of prey capture (Bunkley et al. 2015).

Figure 6.1 Example of recording of an acoustic community of birds close to a traffic road (State road #63, Fivizzano, 0437‐10062016, 44°13’33”N, 10°07’00”E, 264 m a.s.l.). The noise is present for few seconds in the left part of the spectrogram. (See color plate section for the color representation of this figure.)

100

Ecoacoustics

6.6 ­How Animals Neutralize the Effect of Noise Animals use several strategies to neutralize the masking effect of noise acting on the change of amplitude, frequency, changing signal redundancy, and behavior (Figure 6.2). 6.6.1  Changing Amplitude

It is common for people to increase the amplitude of their voices during a conversation in a noisy environment in order to increase the intelligibility of a speech signal. This phenomenon was described by the French otolaryngologist Etienne Lombard in 1911, and this is known as the Lombard effect (Lombard 1911). Sixty years later, this effect was discovered in the Japanese quail (Coturnix japonica), when they were exposed to a loud white noise by Potash (1972). This effect was also observed in zebra finches (Taeniopygia guttata) by Cynx et al. (1998). More recently, in an urban area of Germany, Brumm and Todt (2003) found the Lombard effect in a population of nightingales (Luscinia megarhynchos). This phenomenon has also been confirmed by Holt et  al. (2008) in killer whales (Orcinus orca) that were exposed to environmental noise, with an increase of 1 dB of call amplitude for every 1 dB of increase in the background noise level. 6.6.2  Changing Frequency

In evolutionary time, species living close to natural and persistent noise sources emit sounds at higher frequencies (often ultrasonic) than do species living in a quieter environment. The concave‐eared torrent frog (Amolops tormotus), a arboreal frog species from central China, uses ultrasonic harmonics to communicate in noisy places. The black‐faced warbler (Abroscopus albogularis), an oscine songbird that lives close to torrents, also increases its song frequency to be heard (Narins et al. 2004). An increase in song frequency in the presence of background noise has been observed in many species of birds, such as the great tit (Parus major) in urban areas (Slabbekoorn and Peet 2003). Mockford and Marshall (2009), comparing two populations of great tits, one living in an urban area and the other in a rural area, found through the use of playbacks that there was a difference in spectral aspects of the song of the two populations. Along an urban gradient in the San Francisco peninsula in California, three different

Animals versus Noise Strategy

Goals

Changing Signal Amplitude :

to mask the noise

Changing Signal Frequency :

to distinguish from the noise

Changing Signal Redundancy:

to confirm the signal in a noisy context

Changing Behavior:

to escape a noise context

Figure 6.2  Different strategies can be adopted by animals to neutralize noise and its effects.

6  The Ecological Effects of Noise on Species and Communities

dialects were found in a population of the white‐crowned sparrow Zonotrichia leucophrys by Luther and Baptista (2010). The level of noise in vegetated areas near urban parks may be an important factor in regulating the diversity of birds. In Spain and Portugal, of the 91 species of birds living in urban areas, 10 species (Regulus regulus, Streptopelia turtur, Dendrocopos minor, Buteo buteo, Hirundo daurica, Corvus corax, Oriolus oriolus, Cettia cetti, Passer hispaniolensis, and Sylvia melanocephala) were affected by the noise level (Paton et al. 2012). 6.6.3  Changing Signal Redundancy

Another strategy, according to the theory of communication (Shannon 1948), involves bypassing the acoustic limit imposed by a noisy environment, where an organism can increase signal redundancy (Wiley 1994). This effect has been found in different species of birds, including Japanese quail (Coturnix japonica) (Potash 1972), European serin (Serinus serinus) (Diaz et al. 2011) and in singing male chaffinches (Fringilla coelebs) (Brumm and Slater 2006). 6.6.4  Changing Behavior

In environments in which noise events happen in an irregular way in space and time, evolutionary processes cannot operate because a species cannot overcome the masking effect of noise by changing its behavior due to the irregularity of noise occurrence. For instance, road traffic noise has been shown to affect the prairie dog, Cynomys ludovicianus, living on the border of a road with high traffic. In this situation, this species reduces its aboveground activity. In California ground squirrels (Spermophilus beecheyi) at the Altamont Pass Wind Resource Area in northern California, the presence of wind turbines masks alarm calls of this species in the presence of potential predators and thus forces the species to become more vigilant and causes a more frequent return to underground refuges immediately after an alarm call (Rabin et al. 2006). The European robin (Erithacus rubecula) avoids environments with high noise and reduces the complexity of its song (McLaughlin and Kunc 2012). Experiments involving song playback have demonstrated that toads (Bufo bufo) remain immobile in the presence of noise (Lupo et al. 1991). Similar behavior has been observed in shore crabs (Carcinus maenas) when ship noise is reproduced. This species stops feeding and retreats to shelter (Wale et  al. 2013). Ambient noise may have an impact on predatory risk in Caribbean hermit crabs (Coenobita clypeatus) that are more easily approached when exposed to boat noise playback (Chan et al. 2010).

6.7 ­Noise in Marine and Freshwater Systems The effect of noise in aquatic systems is distinct from terrestrial noise. In aquatic systems, sounds are transmitted five times faster (1484 m/s) than in the atmosphere (343 m/s) and sound travels further than in terrestrial systems due to the lack of obstacles such as vegetation. Aquatic life, including crustaceans, fish, and mammals, are affected by noise. The increase of global shipping and recreational motorboats (Whitefield and Becker 2014), and the use of larger ships result a growing intrusion in

101

102

Ecoacoustics

all the oceans (Frisk 2012), affecting pelagic and benthonic life forms. Wind farms in shallow waters (Wahlberg and Westerberg 2005), seismic exploration in deep sea, especially at higher latitudes encouraged by ice reduction (Klinck et al. 2012), and the emission of noise at medium and high frequencies mainly by warships (Koper and Plön 2012) are important sources of acoustic disturbance. A further and dramatic source of noise pollution in aquatic systems is caused by the acidification of water using carbonic acid which increases the distance to which sound propagates. It is estimated by Etter (2012), McDonald et  al. (2006) and Simpson et  al. (2011) that an increase of 2.5–3 dB per decade in the frequency 30–50 Hz is occurring in ocean systems. Tolimieri et al. (2000) and Vermeij et al. (2010) have shown that the sonic environment is important for aquatic life by demonstrating that the phonotaxis of fish and crustaceans in a coral reef during planktonic life changes in the presence of noise. The investigation of the effect of sound on aquatic life is just beginning but we suspect that sound has an enormous impact on the behavior of thousands of species of fish (Nedelec et al. 2016; Radford et al. 2014). The sensitivity of fish to noise was shown by McCauley et al. (2003) who studied the effect of noise on pink snappers (Pagrus auratus), in which air cells were damaged after exposure to airgun noise. Damage has also been found in bottlenose dolphins in the vicinity of the sounds of pile driving and the effect on their behavior was noted at a distance of 50 km (Bailey et al. 2010). Whales are especially vulnerable to noise during the singing season and boats in the vicinity of this activity can create serious disruptions in communication between individuals (Au and Green 2000), with a reduction in acoustic emissions (Parks and Clark 2007). The amount of background noise in marine systems is not easily assessed and this creates inefficiency of noise management in protected areas for species like fin, humpback, and killer whales, because whale sanctuaries are often located in areas with high noise levels (Williams et al. 2013).

6.8 ­Conclusions The reduction of noise to protect living organisms is a mandatory task for policy makers, stakeholders, and citizens. Noise in the environment is expected to grow in future due to the rapid expansion of the human population (Cohen 2003). The WHO (2016) estimates that urban areas will expand by 54%. The spread of urban settlements creates a new typology of landscapes in which many forms of life may disappear. However, often, paradoxically, these landscapes are important for the survival of many species. It is urgent that landscapes in which people live have acceptable rates of quietness, because this is important for the health and social interaction of billions of people. Noise pollution is a plague, and is just as important as air pollution (Farina et al. 2007). Improvement of the quality of the environment from a human perspective results in benefits for many other organisms, assuring a reasonable level of biodiversity. Living in quiet areas is an important attribute of every urban space and this quality is as important as the quality of food, drinkable water, and unpolluted air (Kaplan and Kaplan 1989). Quiet areas are also important for recreation, as demonstrated by tourist preferences for natural parks, remote environments, and heritage sites (O’Connor 2008).

6  The Ecological Effects of Noise on Species and Communities

People associate an additional economic value with such areas (Merchant et al. 2014) and laws to protects parks and natural areas are present in many national legislations although it is not easy to find a compromise between tourist expectation, logistic facilities, and biodiversity conservation.

­References Au, WWL and Green, M (2000) Acoustic interaction of humpback whales and whale‐ watching boats. Marine Environmental Research, 49, 469–481. Babisch, W, Beule, B, Schust, M, Kersten, N and Ising, H (2005) Traffic noise and risk of myocardial infarction. Epidemiology, 16, 33–40. Bailey, H, Senior, B, Simmons, D, Rusin, J, Picken, G and Thompson, PM (2010) Assessing underwater noise levels during pile‐driving at an offshore windfarm and its potential effects on marine mammals. Marine Pollution Bulletin, 60(6), 888–897. Barber, JR, Crooks, KR and Fristrup, KM (2009) The costs of chronic noise exposure for terrestrial organisms. Trends in Ecology and Evolution, 25(3), 180–189. Brenowitz, EA (1982) The active space of red‐winged blackbird song. Journal of Comparative Physiology, 147, 511–522. Brown, TJ and Handford, P (1996) Acoustical signal amplitude patterns: a computer simulation investigation of the acoustic adaptation hypothesis. Condor, 98, 608–623. Brumm, H. (2010) Anthropogenic noise: implications for conservation, in Encyclopedia of Animal Behavior (eds M Breed and J Moore), Oxford Academic Press, Oxford, pp.89–93. Brumm, H and Slabbekoorn, H (2005) Acoustic communication in noise, in Advances in the Study of Behavior, vol. 35 (eds PJB Slater, CT Snowdon, TJ Roper, HJ Brockmann and M Naguib), Academic Press, San Diego, pp.151–209. Brumm, H and Slater, PJB. (2006) Ambient noise, motor fatigue, and serial redundancy in chaffinch song. Behavioral Ecology and Sociobiology, 60, 475–481. Brumm, H and Todt, D (2003) Facing the rival: directional singing behaviour in nightingales. Behaviour, 140, 43–53. Bruni, A, Mennill, DJ and Foote, JR (2014) Dawn chorus start time variation in a temperate bird community: relationships with seasonality, weather, and ambient light. Journal of Ornithology, 155, 877. Bucur, V (2006) Urban Forest Acoustics, Springer, Berlin. Bunkley, JP, McClure, JW, Kleist, N, Francis, CD and Barber, JR (2015) Anthropogenic noise alters bat activity levels and echolocation calls. Global Ecology and Conservation, 3, 62–71. Chan, AAY, Giraldo‐Perez, P, Smith, S and Blumstein, DT (2010) Anthropogenic noise effects risk assessment and attention: the distracted prey hypothesis. Biology Letters, 6, 458–461. Cohen, J (2003) Human population: the next half century. Science, 302, 1172–1175. Creel, S, Fox, JE, Hardy, A, Sand, J, Garrot, B and Peterson, RO (2002) Snowmobile activity and glucocorticoid stress responses in wolves and elk. Conservation Biology, 16, 809–814. Crutzen, P and Stoermer, E (2000) The Anthropocene, global change. IGBP Newsletter, 41, 17–18.

103

104

Ecoacoustics

Cynx, J, Lewis, R, Tavel, B and Tse, H (1998) Amplitude regulation of vocalizations in noise by a songbird, Taeniopigya guttata. Animal Behaviour, 56, 107–113. Diaz, M, Parra, A and Gallardo, C (2011) Serins respond to anthropogenic noise by increasing vocal acivity. Behavioral Ecology, 22, 332–336. Duarte, MHL, Sousa‐Lima, RS, Young, RJ, et al. (2015) The impact of anthropogenic noise from open cast mining on Atlantic forest biophony. Biological Conservation, 191, 623–631. Etter, PC (2012) Advanced applications for underwater acoustic modeling. Advanced in Acoustics and Vibration, DOI: 10.1155/2012/214839 Farina, A (2014) Soundscape Ecology, Springer, Dordrecht. Farina, A and Pieretti, N (2014) Acoustic codes in action in a soundscape context. Biosemiotics, 7, 321–328. Farina, A, Scozzafava, S and Napoletano, B (2007) Therapeutic landscapes: paradigms and applications. Journal of Mediterranean Ecology, 8, 9–13. Farina, A, Ceraulo, M, Bobryk, C, Pieretti, N, Quinci, E and Lattanzi, E (2015) Spatial and temporal variation of bird dawn chorus and successive acoustic morning activity in a Mediterranean landscape. Bioacoustics, 24(3), 269–288. Farina, A, Pieretti, N, Salutari, P, Tognari, E and Lombardi, A (2016) The application of the Acoustic Complexity Indices (ACI) to ecoacoustic event detection and identification (EEDI) modeling. Biosemiotics, 9(2), 227–246. Francis, CD, Ortega, CP and Cruz, A (2009) Noise pollution changes avian communities and species interactions. Current Biology, 19, 1415–1419. Frisk, GV (2012) Noiseonomics: the relationship between ambient noise levels in the sea and global economic trends. Scientific Reports, 2, 437. Goines, L and Hagler, L (2007) Noise pollution: a modern plague. Southern Medical Journal, 100, 287–294. Griffin, DR (1976) The audibility of frog choruses to migrating birds. Animal Behaviour, 24, 421–427. Holt, MN, Noren, DP, Veirs, V, Emmons, CK and Veirs, S. (2008) Speaking up: killer whales (Orcinus orca) increase their call amplitude in response to vessel noise. Journal of the Acoustical Society of America, 125(1), EL27–EL32. Kaplan, S and Kaplan, R (1989) The Experience of Nature: A Psychological Perspective, Cambridge University Press, Cambridge. Klinck, H, Nieukirk, SL, Mellinger, DK, Klinck, K, Matsumoto, H and Dziak, RP (2012) Seasonal presence of cetaceans and ambient noise levels in polar waters of the North Atlantic. Journal of the Acoustical Society of America, 132(3), EL176–EL181. Klump, GM (1996) Bird communication in the noisy world, in Ecology and Evolution of Acoustic Communication in Birds (eds DE Kroodsma and EH Miller), Cornell University Press, New York, pp.321–338. Koper, RP and Plön, S (2012) The Potential Impacts of Anthropogenic Noise on Marine Animals and Recommendations for Research in South Africa. EWT Research and Technical Paper No. 1. Endangered Wildlife Trust, South Africa. Krause, B (2012) The Great Animal Orchestra, Little, Brown, New York. Krause, B and Farina, A (2016) Using ecoacoustic methods to survey the impacts of climate change on biodiversity. Biological Conservation, 195, 245–254. Krause, BL, Gage, SH and Joo, W (2011) Measuring and interpreting the temporal variability in the soundscape at four places in Sequoia National Park. Landscape Ecology, 26, 1247–1256.

6  The Ecological Effects of Noise on Species and Communities

Lengagne, T and Slater, PJB (2002) The effects of rain on acoustic communication: tawny owls have good reason for calling less in wet weather. Proceedings of the Royal Society B, 269, 2121–2125. Lombard, E (1911) Le signe de l’elevation de la voix. Annales des Maladies de l’Oreille e du Larynx, 37, 101–119. Lupo, C, Lodi, L, Paluffi, G and Viti, A (1991) Central and peripheral endocrine correlates of the immobility reaction in the toad Bufo bufo. Behavioural Process, 24, 1–7. Luther, D (2009) The influence of the acoustic community on songs of birds in a neotropical rain forest. Behavioral Ecology, 20, 864–871. Luther, D and Baptista, L (2010) Urban noise and the cultural evolution of bird songs. Proceedings of the Royal Society B, 277, 469–473. Luther, D and Gentry, K (2013) Sources of background noise and their influence on vertebrate acoustic communication. Behaviour, 150, 1045–1068. Malavasi, R and Farina, A (2013) Neighbours talk: interspecific choruses among songbirds. Bioacoustics, 22, 33–48. Marten, K and Marler, P (1977) Sound transmission and its significance for animal vocalization. Behavioral Ecology and Sociobiology, 2, 271–290. McCauley, RD, Fewtrell, J and Popper, AN (2003) High intensity anthropogenic sound damages fish ears. Journal of the Acoustical Society of America, 113(1), 638–642. McComb, K, Reby, D, Baker, L, Moss, C and Sayalel, S (2003) Long distance communication of acoustic cues to social identity in African elephants. Animal Behaviour, 65, 317–329. McDonald, MA, Hildebrand, JA and Wiggins, SM (2006) Increases in deep ocean ambient noise in the Northeast Pacific west of San Nicolas Island, California. Journal of the Acoustical Society of America, 120(2), 711–718. McLaughlin, KE and Kunc, HP (2012) Experimentally increased noise levels change spatial and singing behaviour. Biology Letters, 9, 20120771. Merchant, CI, Diaz‐Balteiro, L and Solino, M (2014) Noise pollution in national parks: soundscape and economic valuation. Landscape and Urban Planning, 123, 1–9. Mockford, EJ and Marshall, RC (2009) Effects of urban noise on song and response behaviour in great tits. Proceedings of the Royal Society B, 276, 2979–2985. Morley, EL, Jones, G, and Radford, A (2014) The importance of invertebrates when considering the impacts of anthropogenic noise. Proceedings of the Royal Society B, 281, 20132683. Morton, E (1975) Ecological sources of selection on avian sounds. American Naturalist, 109(965), 17–34. Narins, PM, Feng, AS, Lin, WY, et al. (2004) Old World frog and bird vocalizations contain prominent ultrasonic harmonics. Journal of the Acoustical Society of America, 115, 910–913. Nedelec, SL, Mills, SC, Lecchini, D, Nedelec, B, Simpson, SD and Radford, AN (2016) Repeated exposure to noise increases tolerance in a coral reef fish. Environmental Pollution, 216, 428–436. O’Connor, P (2008) Sound of silence: valuing acoustics in heritage conservation. Geographical Research, 46(3), 361–373. Ortega, CP (2012) Effects of noise pollution on birds. Ornithological Monographs, 74(1), 6–22. Parks, SE, Clark, CW and Tyack, PL (2007) Short and long‐term changes in right whale calling behavior: the potential effects of noise on acoustic communication. Journal of the Acoustical Society of America, 122(6), 3725–3731.

105

106

Ecoacoustics

Paton, L, Romero, F, Cuenca, J and Escudero, JC (2012) Tolerance to noise in 91 bird species from 27 urban gardens of Iberian peninsula. Landscape and Urban Planning, 104, 1–8. Pivato, S (2011) Il Secolo del Rumore, Il Mulino, Milan. Potash, LM (1972) Noise‐induced changes in calls of the Japanese quail. Psychonomic Science, 26, 252–254. Rabin, LA, Goss, RG and Owings, DH (2006) The effects of wind turbines on antipredator behavior in California ground squirrels (Spermophilus beecheyi). Biological Conservation, 131, 410–420. Radford, AN, Kerridge, E and Simpson, SD (2014) Acoustic communication in a noisy world: can fish compete with anthropogenic noise? Behavioral Ecology, 25(5), 1022–1030. Read, J, Jones, G and Radford, AN (2013) Fitness costs as well as benefits are important when considering responses to anthropogenic noise. Behavioral Ecology, 25(1), 4–7. Ritts, M, Gage, SH, Picard, CR, Dundas, E and Dundas, S (2016) Collaborative research praxis to establish baseline acoustic conditions in Gitgaat Territory. Global Ecology and Conservation, 7, 25–38. Robbins, CS (1981) Bird activity levels related to weather, in Estimating Numbers of Terrestrial Birds (eds J Ralph and MJ Scott), Studies in Avian Biology No. 6, Allen Press, Lawrence, Kansas, pp.301–310. Shannon, CE (1948) A mathematical theory of communication. Bell System Technical Journal, 27, 379–424. Simpson, SD, Mumday, PL, Wittenrich, ML, et al. (2011) Ocean acidification erodes crucial auditory behaviour in a marine fish. Biology Letters, 7, 917–920. Slabbekoorn, H (2004) Habitat‐dependent ambient noise: consistent spectral profiles in two African forest types. Journal of the Acoustical Society of America, 116(6), 3727–3733. Slabbekoorn , H and Peet, M (2003) Birds sing at a higher pitch in urban noise. Nature, 424, 267. Slabbekoorn, H and Ripmeester, EAP (2008) Birdsong and anthropogenic noise: implication and application for conservation. Molecular Ecology, 17, 72–83. Slabbekoorn, H, Bouton, N, van Opzeeland, I, et al. (2010) A noisy spring: the impact of globally rising underwater sound levels on fish. Trends in Ecology and Evolution, 25, 419–427. Sueur, J and Farina, A (2015) Ecoacoustics: the ecological investigation and interpretation of environment sound. Biosemiotics, 8(3), 493–502. Tennessen, JB, Parks, SE and Langkilde, T (2014) Traffic noise causes physiological stress and impairs breeding migration behaviour in frogs. Conservation Physiology, 2, 1–8. Tolimieri, N, Jeffs, A and Montgomery, JC (2000) Ambient sound as a cue for navigation by the pelagic larvae of reef fishes. Marine Ecology Progress Series, 207, 219–224. Truax, B (ed.) (1999) Handbook for Acoustic Ecology, , Cambridge Street Publishing, Cambridge. Van Kempen, E, van Kamp, I, Nilsson, M, et al. (2010) The role of annoyance in the relation between transportation noise and children’s health and cognition. Journal of the Acoustical Society of America, 128(5), 2817–2828. Vermeij, MJA, Marhaver, KL, Huijbers, CM, Nagelkerken, I and Simpson, SD (2010) Coral larvae move toward reef sounds. PLos ONE, 5(5), e10660.

6  The Ecological Effects of Noise on Species and Communities

Wahlberg, M and Westerberg, H (2005) Hearing in fish and their reactions to sounds from offshore wind farms. Marine Ecology Progress Series, 288, 295–309. Wale, MA, Simpson, SD and Radford, AN (2013) Noise negatively affects foraging and antipredator behaviour in shore crabs. Animal Behaviour, 86, 111–118. Whitefield, AK and Becker, A. (2014) Impacts of recreational motorboats on fishes: a review. Marine Pollution Bulletin, 83, 24–31. Wiley, RH (1994) Errors, exaggeration, and deception in animal communication, in Behavioral Mechanisms in Evolutionary Ecology (ed. LA Real), University of Chicago Press, Chicago, pp.157–189. Williams, R, Clark, CW, Ponirakis, D and Ashe, E (2013) Acoustic quality of critical habitats for three threatened whale populations. Animal Conservation, 17, 174–185. World Health Organization (1971) Development of Environmental Health Criteria for Urban Planning, World Health Organization Technical Report Series No. 511, World Health Organization, Geneva. World Health Organization (2016) Global Report on Urban Health: Equitable, Healthier Cities for Sustainable Development, World Health Organization, Geneva.

107

109

7 Biodiversity Assessment in Temperate Biomes using Ecoacoustics Almo Farina and Nadia Pieretti Department of Pure and Applied Sciences, University of Urbino, Urbino, Italy

7.1 ­Introduction Biodiversity is the number and heterogeneity of species of plants, animals, and microorganisms, but it also entails the diversity of genes, chemical diversity, and the diversity of ecosystems on the planet, regarding all levels of the hierarchy of life, from molecules to ecosystems (Gaston 2000). Today, we are experiencing a dramatic loss in biodiversity due to climate change and anthropogenic activities. Effective systems of biodiversity assessment are crucial in conservation management. The numerous challenges associated with traditional in‐field ­appraisals, including those which are excessively invasive, time‐consuming, labor‐­intensive, and costly, can lead to logistical and financial problems that, particularly for the most vulnerable and remote sites, can result in belated answers for managers and stakeholders ­leading to tardy mitigation activity. It follows that the measurement of biodiversity is extremely important since it is linked to the health of ecosystems and, consequently, to the functioning of human society. It is at the basis of our economic development and our biological and food resources. It is also the potential solution to many medical issues and relates directly to climate change. The richer the diversity of life, the greater the ecosystem productivity where even the smallest species has an important role to play (Tilman et al. 1996). Biodiversity assessment represents an important step toward a more ethical and efficient strategy to preserve the complexity of the environment threatened by changes in land use and the effects of human intrusion in climate dynamics (IPCC 2013). The present time is called the Anthropocene by Crutzen and Stoermer (2000). The speed with which these changes occur prevents species adaptimg, causing local and global extinctions (Carey 2009; Crick 2004; Llusia et al. 2013; Parmesan 2006). If human intrusion has a geographic pattern and some parts of the Earth can survive such contamination, at least in a semi‐natural condition, climate change has no limit or geographical boundary and represents the largest threat to the integrity and survival of several ecosystems at a planetary scale. For this reason, intensification of climate change studies is needed to gain an idea of the magnitude of climate change impact on biomes, habitats, and species distribution (IPCC 2014). There are complex interactions between Ecoacoustics: The Ecological Role of Sounds, First Edition. Edited by Almo Farina and Stuart H. Gage. © 2017 John Wiley & Sons Ltd. Published 2017 by John Wiley & Sons Ltd.

110

Ecoacoustics

the energetic environment, animal biomass, and the structure of social interactions (Brackenbury 1979; Wallschager 1980). It is disputed that extreme events associated with climate change are the major factors responsible for irreversible transformation of habitats and the dismantling of complex communities (Alois and Cheng 2007; Easterling et al. 2000), with interruption of seasonal dynamics such as the migration of stopover species (Gordo 2007; Huntley et al. 2006). The relationship between environmental energy and biophonic sounds represents an important factor which needs further investigation (Gillooly and Ophir 2010). A variety of ecological “services” essential to human well‐being and sustainable development derive from the normal functioning of ecosystems. In the last two decades, the connection between biodiversity and ecosystem functioning has emerged as a central issue in ecological and environmental sciences (Millennium Ecosystem Assessment 2005). Recirculation of nutrients, fertile soils, provision of food, clean air and water, productive seas, and other ecosystem services are expressions of a functional diversity level. Current biodiversity loss caused by the growing impact of human activities can produce large‐scale ecological consequences and also affect human well‐being. Consequently, monitoring the health of the environment by measuring its biodiversity level is essential. Biodiversity is usually assessed by calculating a series of indices, which are mathematical measures of species diversity in a community. Richness (S, or the number of species in an area; Whittaker 1972); the Berger–Parker Index (the proportion of the most common species in the population; Berger and Parker 1970), Shannon Diversity Index (H, an entropy measure, assuming all species are represented in a sample and randomly sampled; Shannon 1948), and Simpson’s Index (D, the probability that two individuals randomly selected from a sample will belong to the same species; Simpson 1949) are the commonly used diversity indices in ecology. Diversity indices provide fundamental information about rarity and commonness of species in a community. Nevertheless, quantifying biodiversity continues to be challenging and there is no single index that can provide a complete measurement (Hurlbert 1971; Purvis and Hector 2000). Moreover, all these indices are usually applied to species counts or appraisals performed directly in the field, which can be extremely time‐consuming and present strong geographical and temporal limits due to the heterogeneity of aspect, vegetation cover, and the availability of field technicians and funds.

7.2 ­Sound as Proxy for Biodiversity Sound is used by many animals as a carrier of meaning to maintain intra‐ and interspecific communication, to incorporate into a cognitive process their acoustic habitat or umwelt (von Uexküll 1982), and to facilitate navigation of organisms across their individual‐based cognitive landscape (Farina 2010, p.19). Only recently sound of biological and physical origin has been recognized as an important component of ecological investigation, discovering, for instance, that early signs of animal stress connected to climate change are represented by the changing sound performance of several species that use sound to communicate, to mate, and to regulate social disputes (Krause and Farina 2016; Sueur and Farina 2015).

7  Biodiversity Assessment in Temperate Biomes using Ecoacoustics

Moreover, using an ecoacoustic approach, it is possible to investigate the composition and dynamics of acoustic communities (Sueur et al. 2008), acoustic diversity (Gasc et al. 2013), acoustic interactions (Tobias et al. 2014) and the relationship between the ecoacoustic community and the landscape (Farina et al. 2010; Fuller et al. 2015; Joo et al. 2011; Tucker et al. 2014). Acoustic communication has proved to be related to animal metabolism, creating an energetic constraint and producing a great variety of sounds. For this reason, the relationship between climate that modifies the energetic field in which organisms operate and the results of acoustic emissions is strong (Gillooly and Ophir 2010). Biodiversity assessment using an ecoacoustic approach allows use of very common signals, including the biophonic expression of vocal animals. New and advanced digital recorders and new metrics allow exploration of the world of sound from a long‐term perspective. Using passive acoustic procedures requires special consideration. One of these is the range of perceived sounds (Merchant et al. 2015) that can change depending on the species. Moreover, sound changes according to the physical environment. For instance, on steep slopes, sound is transmitted differently than in a flat region (Hunter 1989). In addition, the distribution of vegetation and morphological aspects are factors that can change sound propagation.

7.3 ­Methods and Application of Ecoacoustics To investigate the richness and diversity of vertebrates such as frogs and birds, there are different methodologies based on song, call, and visual counts along transects (Bystrak 1991; Crump and Scott 1994) or in spot counts (Bibby et al. 1992; Parker 1991; Ralph and Scott 1981; Ralph et al. 1993; Verner 1985). However, all these methods are dependent on time and field expertise. Using passive acoustic survey technology is an alternative method that offers several opportunities to identify some groups and to investigate the biology and ecology of individual species and communities. This methodology has been extensively utilized to identify groups of species like baleen whales (Mellinger and Clark 1997) and bats (Henriquez et al. 2014). Moskwik et al. (2013) used passive acoustic survey in the search for the ivory‐billed woodpecker (Campephilus principalis), a species living in old growth forests of the south‐eastern United States and considered critically endangered or possibly extinct (BirdLife International 2013; Fitzpatrick et al. 2005). This methodology can also be used to monitor habitats at large spatial temporal scales in both terrestrial and marine environments (André et  al. 2011; Boebel et  al. 2006; Merchant et al. 2015). Passive acoustic monitoring is powered by a new generation of digital recorders able to capture the sounds of nature at different sampling frequencies and at different times of day. Programmability, waterproof, long‐life batteries, high storage capacities (e.g. 32–64 Gb), and transmission of data via global system for mobile communication (GSM) technology are some of these new monitoring capacities. In particular, the commercial market now offers a broad range of microphones, including sophisticated and expensive ones (Monacchi 2014) or less expensive ones that capture acoustic information at close distances (Farina et al. 2014). Moreover, new metrics allow the processing of acoustic files, returning data‐rich and meaningful information (Bystrack

111

112

Ecoacoustics

et  al. 2010; Gage and Axel 2014; Kasten et  al. 2010; Sueur et  al. 2014; Towsey et  al. 2014b). Several acoustic indices have been developed to infer information about species diversity. Sueur et al. (2014) recently provided a review of these indices. As for traditional biodiversity indices, they can be divided into two macro‐groups, the α diversity and β diversity indices, assessing respectively within‐group and between‐group diversity (Sueur et al. 2014). Earlier methods used in acoustic studies were based on a simple measure of the amplitude or sound pressure level (SPL) (the energy of the sound, usually expressed in dB or root‐mean‐square (RMS)). Successively, acoustic indices have focused on other parameters in recorded sounds, from addressing the complexity or entropy of the acoustic emissions to consecutively considering time, frequency, and amplitude, or by calculating the contribution of each soundscape component (biophony, geophony, and technophony) (Sueur et al. 2014). A series of indices assume that the more species and individuals are found in a community, the higher will be the acoustic output of that community. Others suggest finding and separating biological from anthropogenic sounds. A few of the most employed indices in scientific studies include the Acoustic Complexity Index (ACI) (Pieretti et al. 2011), the Acoustic Entropy Index (H) (Sueur et al. 2008), the Acoustic Richness Index (AR) (Depraetere et al. 2012) and the Normalized Difference Soundscape Index (NDSI) (Kasten et al. 2012). The Acoustic Entropy Index (H) (Sueur et al. 2008) was one of the first acoustic indices to make biodiversity appraisals and it is commonly used in diversity assessments. Adapted from Shannon’s Index, it is a spectral and temporal entropy measure. It was designed to reflect the even distribution of a signal’s amplitude over time and frequencies. A single pure tone gives a value near zero, while a community composed of many vocalizing species will have an index number closer to 1, which represents the maximum level of entropy. It was shown that H increased with species richness on a logarithmic model (Sueur et al. 2008) for both simulated choruses and field recordings in tropical forests with high animal acoustic activity and low background noise. However, when noise or broadband signals (e.g. rain or cicadas) were higher, spectral entropy measures risked to provide counterintuitive results close to 1. Modeled after H, the Acoustic Richness Index (AR) weights the overall amplitude of the signal to account for background noise (Depraetere et al. 2012). This was intended especially for temperate habitat investigations, where the acoustic output of a community is not as intense as those in tropical environments. The ACI focuses on the intrinsic variability of sounds, by measuring variations in intensity within single frequency intervals (Farina et al. 2011). It relies on the assumption that biological sounds have an inner variability that can distinguish them from most human‐made noises. The algorithm can indeed be used to filter out most technophonies, such as trains, cars or airplane noise. The ACI is shown to correlate with the number of bird vocalizations (Pieretti et  al. 2011), the sounds emitted by fishes and sounds made by snapping shrimps (McWilliam and Hawkins 2013). The ACI has been useful for analyzing recordings in noisy areas (Duarte et al. 2015; Pieretti and Farina 2013) or for tracking changes in behavior and composition of vocalizing communities. Linked to principles of soundscape ecology, the Normalized Difference Soundscape Index (NDSI) is obtained by calculating the ratio (biophony – technophony)/(biophony +

7  Biodiversity Assessment in Temperate Biomes using Ecoacoustics

technophony), considering the summed power spectral densities (PSD) for the two categories. Kasten et al. (2012) suggest that biophony occurrs mostly in the 2–8 kHz frequency range and that technophony (sounds made by machines) occurs in the 1–2 kHz frequency range. The range in frequency values has since been modified where biophony is 2–11 kHz and technophony is 1–2 kHz (Gage and Axel 2014; Fuller et al. 2015). The NDSI has been shown to be associated with seasonal and diurnal variation of a landscape and to help in the detection of long‐term interactions between communities and human society (Fuller et al. 2015; Kasten et al. 2012).

7.4 ­Acoustic Communities as a Proxy for Biodiversity The community is the level of ecological organization at which biodiversity is conceptually connected and calculated. Defined in classic ecology as a group of actually or potentially interacting species living in the same space (Morin 1999), animal and plant communities are described with direct and indirect counts of species and their abundance in a delimited space and within a temporal interval (Mueller‐Dombois and Ellenberg 1974). The fundamentals that assure the exchange of information within communities may be represented by biological molecules, energy such as light or as acoustic waves. The interaction between species may be of trophic or behavioral type. Among the behavioral types, the acoustic interactions in birds represent an important mechanism of cohesion of individuals and species (Malavasi and Farina 2013) and a fundamental element of evolutionary mechanisms. Accordingly, the acoustic interaction between animal species is an important process necessary to maintain interspecific relationships between the components of a community, and to calibrate and regulate several behavioral processes such as territorial patrolling, resource partitioning or predator prevention. When birds are singing during the day they create sonotopes (Farina 2014) that when separated from the contribution of nonbiophonic sounds are considered as an acoustic community (Farina and James 2016). The existence and maintenance of an acoustic community are assured by specific semethic processes that can be characterized by biophysical or cybernetic mechanisms (Hoffmeyer 2008). An acoustic community is an aggregation of species that produces sound by using internal or extrabody sound‐producing tools (Farina and James 2016). An acoustic community is the result of an exchange of acoustic information between individuals and species, and occurs in aquatic (freshwater and marine) and terrestrial environments. Three broad types of acoustic communities can be distinguished: infrasonic (e.g. whales (Cetacea) 20 000 Hz (e.g. bats (Chiroptera), dolphins (Cetacea), some frogs and some insects). Acoustic communities have peculiar acoustic signatures that depend on the assemblage of species in a specific habitat or part of the habitat (Bormpoudakis et al. 2013). An acoustic signature is the result of the distribution of frequency categories of sounds used by the acoustically active species (Farina and Pieretti 2014a; Gage and Axel 2014, Malavasi et al. 2014). The acoustic signature of each species is used to measure the acoustic niche overlap and breadth of the entire acoustic community (Sinsch et al. 2012), where the niche overlap measures the degree of potential competition between two or more species, and niche

113

114

Ecoacoustics

breadth is an indicator of species richness in an acoustic community. In fact, the more species that are singing in an acoustic community, the larger its niche breadth. This metric enables a comparison of different acoustic communities.

7.5 ­Problems and Open Questions It is possible to record the singing activity of species over time and after downloading the acoustic file, it is possible to identify the individual species present. This approach discloses only a small part of the ecological information that an acoustic community offers. However, an acoustic community, especially in bird communities, is the result of temporary sonic contributions of different individuals belonging to different species. Thus it is challenging due to the dynamics of bird vocalizations in space and time. In fact, acoustic communities are characterized by a great variability of sounds due to the different physiological dynamics of the composing species. In the same location at the same time of the day, there is little coincidence in terms of acoustic community patterns. Furthermore, during the migratory periods of the year there can be great variation in the acoustic community patterns due to the arrival of migratory species that may leave the area after only a few hours of stopping to rest and feed. Despite this temporal variability, the acoustic signature has a higher conservative character and is distinctive of a location and of the habitat. Within an acoustic community, the distribution of frequencies is not random but is the result of frequency partitioning and, based on the niche theory of Hutchinson (1957, 1978), every species tends to reduce the competition with other species within the community. It has been shown that the species that sing at the same time try to partition the frequency to reduce possible competition and the masking effect (Sinsch et al. 2012; Sueur 2002; Malavasi and Farina 2013). The acoustic niche hypothesis (ANH) formulated by Krause (1993) is still the subject of debate (Planqué and Slabbekoorn 2008; Tobias et al. 2014) but there is empirical evidence of such partitioning, especially in tropical areas where many species of different groups of animals are calling or singing together. Although sounds from nature may be represented by an orchestra metaphor (Krause 1993), there is no evidence of a common theme for such an orchestra as there is not a fixed sheet of music. This gives rise to several issues, to measure or model an acoustic community that could be likened to a free jazz model! In fact, sound from different species may enter the symphony with irregularity. The population density of different species living in a habitat may be relatively constant during a season (e.g. birds), but their acoustic activity varies according to the phenological traits of each species, especially during the breeding season when acoustic signals are the dominant life trait for the majority of species. Outside the breeding season, at least in temperate biomes, the majority of species are silent and the acoustic communities are created by a small number of species (e.g. European robin, song thrush, carrion crow, jay, green woodpecker). Their sonority is very different in terms of intensity and repetition. Often subsongs and alarm and social calls are the only performances and their distribution during the day largely depends on weather conditions and seasonal patterns.

7  Biodiversity Assessment in Temperate Biomes using Ecoacoustics

Event b: Mosaic 1

Event a: Dawn Chorus

Time 0

Time 1

Event c: Mosaic 2

Time 2

Figure 7.1  Representation of three acoustic events. Event a – dawn chorus (all the species are singing at the same time), time 0. Event b – a mosaic of acoustic communities, and time 1. Event c – a mosaic of acoustic communities, at time 2 at dawn chorus; all the individuals are singing at the same time. In the successive time period the number of individuals acoustically active is variable in space and time. (See color plate section for the color representation of this figure.)

When an acoustic sensor is deployed in one location, it records an assemblage of species living there, but the sensor probably records sonic patterns created by different acoustic communities like interacting clusters that emerge around the sensor throughout time (Figure 7.1). Moreover, the sensor collects acoustic information of the surroundings according to a variable circle and this variability is caused by the internal technical characteristics of the device (microphone type, sensitivity, and gain) and by the distance and intensity at which organisms produce sounds. Changing microphone types and input gains results in significant differences in the data collected. To bypass these difficulties, an approach is proposed based on the consideration that around a sensor there are different acoustic communities that are continuously reshaping according to behavioral and environmental factors. This continuous reshaping ­creates a flux of information that may be intercepted by the sensors with an intrinsic distortion that prevents the real spatial patterns of an acoustic community being represented. Internal factors connected to the phenological status of individual species may

115

116

Ecoacoustics

contribute to such variability. Often, extrapair singers that are engaged in temporary performances with no fixed territory may contribute to the uncertain distribution of an acoustic community. Further variability can be introduced by external factors like the presence of wind, rain, and technophonic intrusion. In birds, the same species can interact acoustically in a different manner according to season and time of day, where a species can maintain acoustic activity during the day but with a different individual participation. Species generalists in terms of habitat types such as Turdus merula, Erithachus rubecula, and Sylvia atricapilla may have a different song activity in different habitats and a different ecological role. The variability of acoustic performances in an acoustic community creates a challenge to consistently collect and interpret quantitative data about the distribution of biophonies. The use of a matrix of recorders is one method to intercept the dynamics of the acoustic community but the methodology is time‐consuming and costly due to the number of sensors required, difficulties in timing synchronization between sensors and data‐processing complexities (Farina and Pieretti 2014b). In addition, the distance between sensors adds difficulties to the challenge because of sound duplication.

7.6 ­Ecoacoustic Events: Concepts and Procedures The amount of information in acoustic recordings is high, and this can be an obstacle for the use of continuous passive acoustic monitoring over long time periods. This approach requires a great deal of computational time, which forces investigators to sample the acoustic commmunity rather than to record continuously (Pieretti et al. 2015; Winner et al. 2013). Furthermore, the acoustic information contained in an acoustic recording is not entirely interpreted by ecoacoustic metrics, thus losing important details about the acoustic patterns. To interpret this information, more efficient models are needed. For analysis of the acoustic community, the identification of individual species is important, but it is also useful to capture the entire sonic context in which individual sounds are embedded. In this way, it is possible to distinguish functional entities, which may be performances of an entire community, like choruses, or an environmental event like thunder, wind or rain. A new and efficient way to investigate acoustic communities is based on the principle of ecoacoustic events (Farina et al. 2016). An acoustic event is defined as an emergent sonic pattern resulting from individual geophonies, biophonies or technophonies or their combination. Usually, an event appears suddenly, thus modifying the acoustic scenario and influencing the behavior of species accordingly. Event detection is commonly used in various areas of theoretical and applied acoustics (Heittola et al. 2011; Kasten et al. 2011; Mesaros et al. 2010; Zhuang et al. 2010). An investigation based on events has been proposed by Farina et al. (2016) using the Ecoacoustic Event Detection and Identification model (EEDI). This model, mainly tested on birds, is based on five assumptions. 1) Acoustic communities can be detected from the surroundings. 2) Only during the chorus time all acoustic communities are active and this means that all species are singing (at least during the breeding period). 3) After the chorus period, some communities may have a more limited activity.

7  Biodiversity Assessment in Temperate Biomes using Ecoacoustics

4) Activity of acoustics communities outside the dawn and dusk chorus periods may be discriminated based on a criterion of distance (far versus close, only far, only close, a mix of the two) and by a criterion of frequency diversity. 5) The ratio between an event and a dawn chorus (at its maximum) represents the distance between all the communities in action and the communities active at different times of the day. Bird species (the majority are singing males) sing at dawn, with a second singing peak during dusk (Farina et al. 2015) although this occurs only before sunrise and after sunset and this pattern is seasonally dependent. This point is extremely important to enable application of the theory of the acoustic community to soundscape evaluation because it is possible to determine an empirical and objective point of the overall activity, at least during the breeding season. The philosophy of EEDI analysis of sounds in a recording is based on determination of complex acoustic configurations by the exclusion of all conditions that do not have explicit significance. Ecoacoustics events are not simply the result of some phenomenon but are the environment from which species receive important information. Figure 7.2 shows a spectrum in which heavy rain is mixed with a blackcap (Sylvia atricapilla) song. This recording might be discarded using traditional analysis because biophonies are too disturbed by rain. However, this is a special event in which two different sources of sound overlap. The blackcap is singing despite the heavy rain and its voice is modified by the presence of the rain. This event can happen several times during a breeding season and may have an important impact on mate selection, assuming that only a healthy blackcap would have the acoustic energy to compete with the sound of the rain. These competing events have importance because they help us to understand the relationship between meteorological conditions and biophonic activity. An ecoacoustic event can be interpreted by the biosemiotic model of the (acoustic) eco-field (Farina and Belgrano 2004, 2006). This model uses the triadic mechanism of the Peirce semiosis (Atkin 2013) by considering a sound collection (summation of individual sounds) as a spatial configuration carrier of meaning, the “representamen” that is used by a “representant” (the function) (Farina and Belgrano 2004, 2006) to recognize

Figure 7.2  An acoustic event is often the combination of different acoustic contests. In this case, heavy rain and a blackcap song overlap (Agnino, time 06:55, date 03062016, 44°14’12”N, 10°04’16”E, 254 m a.s.l.).

117

118

Ecoacoustics

the object that is requested to locate resources (territory, nesting site, roosting, presence of predators, and location of food resources) (Farina 2012). Broad categories of sounds, such as thunder, blasting, rifle shots, horn blows, wind, rain, alarm calls of birds, morning and dusk animal choruses, silence, human voices, road traffic and their combinations, create ecoacoustic events. The categories used to describe ecoacoustic events are not fixed and can change because the category to which an event is assigned is related to the goal of the classification method. From a human perspective, ecoacoustic events may include most of the events connected to human activity and describe human dynamics. For instance, consider an acoustic event produced by the human voice inside a football stadium. Also consider an event consisting of the chorus of starlings (Sturnus vulgaris) at an urban roost or a chorus of thrushes at midday in the Mediterranean during a migration stopover. Using a biosemiotic model, it is possible to isolate different events produced by either natural or human processes. There can also be a mix of acoustic events like a church bell that tolls at noon along with the song of nearby birds. Due to methodological constraints, the spatial distribution of an acoustic community is often excluded from a research investigation. Thresholds are used to filter conditions which are difficult to interpret or when a decision is made to exclude a specific category of events, such as geophonies (e.g. rain, wind) or technophonies (e.g. urban traffic, airplanes). The EEDI model can restrict the analysis to the processes of interest. Two fundamentals that can be assessed are time and the acoustic signature. Time is how long and when an acoustic community is active. The acoustic signature characterizes the acoustic community according to its frequency ranges. The EEDI model is based on the use of ACI metrics: ACIf, ACIfe, ACIt, and ACIte (Farina et al. 2016) (Figure 7.3). ACIf measures the distribution of information in time when the differences in acoustic amplitude between frequencies are evaluated. ACIfe (evenness) measures how ACIf is distributed along a temporal step (e.g. one minute). ACIt is the measure of differences between different amplitudes along each frequency bin and represents the importance of each frequency at which an acoustic event occurrs. ACIt is used to describe the acoustic Numerical analysis

Coding process

Sound survey

Event Detection

Event Identification

Acoustic patterns

Unclassified event

Classified event

Event’s library

Figure 7.3  EEDI procedure. The numerical analysis selects the potential events according to parameters such as environmental variables and ACI metrics thresholds. The coding process produces the event identification according to a selection of known events from an event’s library.

7  Biodiversity Assessment in Temperate Biomes using Ecoacoustics

I

III Hz

(a)

(b) 0

ACIte 1

(c) 1

a

(d) ACIfe

0

Time

e

e

e

e

c

0

(e)

b

d ACIf

1

II

Figure 7.4  Five possible temporal patterns of signals according to a spectrographic representation. I: (a) A low‐intensity signal, uniform along time (low ACIf and max ACIfe). (b) A high‐intensity signal, uniform along time (high ACIf and max ACIfe). (c) A low‐intensity signal, heterogeneous along time (low ACIf and min ACIfe). (d) A high‐intensity signal, heterogeneous along time (high ACIf and min ACIfe). (e) An irregular intensity signal, heterogeneous along time (medium ACIf and medium ACIfe). II: Distribution of the five conditions according to a matrix created by ordering ACIf and ACIfe according to four percentiles. Both ACIf and ACIfe are expressed in standard format (from 0 to 1). III: ACIte represents the distribution of intensity along frequencies. This parameter ranges as all the evenness from 0 (one frequency bin represented) to 1 (all frequency bins represented). (See color plate section for the color representation of this figure.)

signature. ACIte (evenness) measures how different frequencies present in an acoustic community are distributed along the total frequency spectrum (Figure 7.4). The software SoundscapeMeter 2.0 performs the calculations (Farina and Salutari 2016). Three steps are considered in this procedure. In the first step, environmental variables including time, temperature, light, and humidity are selected according to an empirical threshold. In the second step, events are detected after the comparison between ACIf and ACIfe and ACIte according to an empirical threshold. The plot of ACIf × ACIte × ACIfe creates an ecoacoustic event space (EES). This space may be divided into four nominal quadrants or regions, where each quadrant is characterized by a different distribution of ACIf, ACIfe, and ACIte. The first quadrant is characterized by low ACIf and high ACIfe (e.g. biophonies at low amplitude but at high evenness). The second quadrant is characterized by high ACIf and high ACIfe. The third quadrant is characterized by high ACIf and low ACIfe. The fourth quadrant is characterized by low ACIf and low ACIfe. During the third step, an event is identified after the computation of the level of correlation/similarity between the detected event(s) and the acoustic signatures (ACIt) in the library. Event identification is based on the statistical comparison between the identified events and specific events that have been selected from the library. The comparison is carried out by applying the Pearson

119

Ecoacoustics

6 4 2

October

November

December

February

March

51 41

71 61

91 81

121 111 101

141 131

January

April

1

11

September

31 21

J…

211 201

M…

191 181

S… D…

171 161 151

0

241 231 221

120

May

June

July

Figure 7.5  Annual distribution (September 2015 to July 2016) of acoustic events at Carpaneta (44°13’34”N, 10°07’16”E, 290 m a.s.l., Fivizzano, Italy) based on a model of 250 potential events created by the combination of 10 categories of ACIf, five categories of ACIfe, and five categories of ACIte.

(a)

(b)

(c)

(d)

Figure 7.6  Example of four one‐minute events. (a) 1055 (Heavy rain), (b) 753 (Dawn chorus), (c) 525 (European robin song), (d) 323 (blackbird isolated alarm call) according to the classification from Table 7.1.

correlation, the chord distance (Orloci 1967), or Whittaker’s Index of Association (Whittaker  1952). In recent research, Farina used 250 EEDI categories of events. This detailed classification allows better discrimination of the complex interactions between geophonies, techonophonies, and biophonies (Figures 7.5 and 7.6; Table 7.1).

1055

1054

1053

1052

1045

1044

1043

1035

1034

1033

954

953

952

945

944

943

942

935

934

854

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

Code

1

#

20

1

4

2

5

12

11

32

72

15

1

2

4

5

17

26

23

236

163

51

Tot

40

39

38

37

36

35

34

33

32

31

30

29

28

27

26

25

24

23

22

21

#

653

654

732

733

734

742

743

744

745

752

753

754

833

835

842

843

844

845

852

853

Code

266

27

1

2

3

10

33

47

3

107

183

25

1

3

2

10

23

7

51

105

Tot

60

59

58

57

56

55

54

53

52

51

50

49

48

47

46

45

44

43

42

41

#

534

541

542

543

544

551

552

553

554

632

633

634

635

641

642

643

644

645

651

652

Code

4

1

50

251

49

145

565

466

14

1

8

8

1

1

23

95

40

1

8

237

Tot

80

79

78

77

76

75

74

73

72

71

70

69

68

67

66

65

64

63

62

61

#

351

352

353

422

423

424

432

433

434

441

442

443

444

451

452

453

454

525

532

533

Code

1878

2098

319

1

1

2

2

37

14

11

153

511

37

480

1127

513

1

1

1

10

Tot

100

99

98

97

96

95

94

93

92

91

90

89

88

87

86

85

84

83

82

81

#

231

232

233

234

241

242

243

251

252

253

322

323

324

332

333

334

341

342

343

344

Code

10

86

96

1

312

1539

423

8454

3318

63

2

4

2

19

115

16

47

653

859

3

Tot

115

114

113

112

111

110

109

108

107

106

105

104

103

102

101

#

121

122

131

132

133

141

142

143

151

152

212

221

222

223

224

Code

5

5

46

45

3

833

263

2

25321

394

1

5

9

12

1

Tot

Table 7.1  Annual distribution (September 2015 to July 2016) of acoustic events at Carpaneta (44°13’34”N, 10°07’16”E, 290 m a.s.l., Fivizzano, Italy), only categories >0 where considered and sorted according to the code. #, number of events; code, event code; Tot, total number of events. From this table four examples are reported in Figure 7.6.

122

Ecoacoustics

7.7 ­Conclusion Ecoacoustics offers rapid analysis of sounds produced by animal communities using a cost‐effective methodology for monitoring the environment and providing a rapid biodiversity assessment. Acoustic monitoring is possible due to the availability of low‐cost digital recorders, powerful metrics, and new semi‐automatic procedures of event detection and identification. The ecoacoustics approach can also provide metrics other than biodiversity. Acoustic indices have recently proven to be b ­ eneficial in characterizing disturbance and landscape characteristics in terrestrial habitats (Fuller et al. 2015; Gage and Axel 2014; Sueur and Farina 2015) and in marine environments (Harris et al. 2016) as noninvasive, effective ­diversity measures, by providing detailed information about species community composition and tracking their acoustic dynamics through different seasons and in diverse environments. There is a growing body of work on the ecological uses of acoustic indices. The use of sound is now being considered as a valid tool for ecologists trying to evaluate the complexity of any animal community. Nevertheless, numerous challenges are associated with measuring acoustic diversity, such as the almost omnipresent noise produced by human activity or the influence of adverse weather conditions (Pieretti et al. 2015). An ecoacoustic approach allows the collection of important information about the physiology and ecology of sound‐producing species in terrestrial and aquatic environments. The acoustic cues from animals are full of meaningful information because acoustic signals are true ecological codes (Barbieri 2015), used to communicate position in the habitat, health status, resource access, and social dominance. Acoustic behavior is a vital trait that is highly plastic and is one of the first characteristics that changes under environmental pressure. Passive recording should be considered in conjunction with satellite imagery. Acoustic information is collected in close proximity to the sensor and provides details that satellite images cannot. Often biodiversity assessment requires collection of field data within a narrow timeframe and this is possible using a set of acoustic sensors deployed in a regular grid. A matrix of sensors may cope with landscape heterogeneity and can describe with great accuracy the distribution of acoustic communities (Bardeli et  al. 2010) and indirectly the distribution of resources. Animals that communicate using acoustics in terrestrial habitats are often well known and their calls or songs can be easily identified. In tropical areas, there is lack of taxonomic knowledge. Species identification in aquatic systems is more problematic and will require more time to standardize the procedures of individual species identification (Desjonquères et al. 2015). Passive recording has produced a shift in methodologies for conducting field surveys because sensors are relatively inexpensive, do not require the presence of experts, and do not disturb the environment. These multiple attributes allow the design and implementation of sensor networks for long‐term monitoring projects in remote and hostile regions as well as in coastal and deep seas.

­References Aide, TM, Corrada‐Bravo, C, Campos‐Cerqueira, M, Milan, C, Vega, G and Alvarez, R (2013) Real‐time bioacoustics monitoring and automated species identification. Peer Journal, 1, e103.

7  Biodiversity Assessment in Temperate Biomes using Ecoacoustics

Alois, P and Cheng, V (2007) Keystone species extinction overview. Available at: www. arlingtoninstitute.org/wbp/species‐extinction/443 (accessed 14 December 2016). André, M, der Schaar, M, Zaugg, S, Houégnigan, L, Sànchez, AM and Castell, JV (2011) Listening to the deep: live monitoring of ocean noise and cetacean acoustic signals. Marine Pollution Bulletin, 63, 18–26. Atkin, A (2013) Peirce’s Theory of Signs. Stanford Encyclopedia of Philosophy. Available at: https://plato.stanford.edu/entries/peirce‐semiotics/ (accessed 14 December 2016). Barbieri, M (2015) Code Biology, Springer International, New York. Bardeli, R, Wolff, D, Kurth, F, Koch, M and Frommolt, KH (2010) Detecting bird sounds in a complex acoustic environment and application to bioacoustic monitoring. Pattern Recognition Letters, 31, 1524–1534. Berger, WH and Parker, FL (1970) Diversity of planktonic Foraminifera in deep sea sediments. Science, 168, 1345–1347. Bibby, CJ, Burgess, ND and Hill, DA (1992) Bird Census Techniques, Academic Press, London. BirdLife International (2013) Campephilus principalis, IUCN Red List of Threatened Species, Version 2013.2. Boebel, O, Kindermann, L, Klinck, H, et al. (2006) Real‐time underwater sounds from the Southern Ocean. EOS, 87(36), 361–372. Bormpoudakis, D, Sueur, J and Pantis, J (2013) Spatial heterogeneity of ambient sound at the habitat type level: ecological implications and applications. Landscape Ecology, 28, 495–506. Brackenbury, JH (1979) Power capabilities of the avian sound‐producing system. Journal of Experimental Biology, 78, 163–166. Brandes, TS (2008) Automated sound recording and analysis techniques for bird surveys and conservation. Bird Conservation International, 18, S163–S173. Bystrak, D (1981) The North American breeding bird survey. Studies in Avian Biology, 6, 34–41. Carey, C (2009) The impacts of climate change on the annual cycles of birds. Philosophical Transactions of the Royal Society B, 364, 3321–3330. Crick, HQP (2004) The impact of climate change on birds. Ibis, 146(Suppl. 1), 48–56. Crump, ML and Scott, NJ Jr (1994) Visual encounter surveys, in Measuring and Monitoring Biological Diversity: Standard Methods for Amphibians (eds WR Heyer, MA Donnelly, RW McDiarmid, LC Hayek and MS Foster), Smithsonian Institution Press, Washington, DC, pp.84–92. Crutzen, P and Stoermer, E (2000) The Anthropocene, global change. IGBP Newsletter, 41,17–18. Depraetere, M, Pavoine, S, Jiguet, F, Gasc, A, Duvail, S and Sueur, J (2012) Monitoring animal diversity using acoustic indices: implementation in a temperate woodland. Ecological Indicators, 13, 46–54. Desjonquères, C, Rybak, F, Depraetere, M, et al. (2015) First description of underwater acoustic diversity in three temperate ponds. Peer Journal, 3, 1393. Duarte, MHL, Sousa‐Lima, RS, Young, RJ, et al. (2015) The impact of noise from open‐cast mining on Atlantic forest biophony. Biological Conservation,191, 623–631. Easterling, DR, Meehl, GA, Parmesa, n C, Changnon, SA, Karl, TR and Mearns, LO (2000) Climate extremes: observations, modeling, and impacts. Science, 289, 2068–2074.

123

124

Ecoacoustics

Farina, A, (2010) Ecology, Cognition and Landscape. Linking Natural and Social Systems, Springer, Dordrecht. Farina, A (2012) A biosemiotic perspective of the resource criterion: toward a general theory of resources. Biosemiotics, 5, 17–32. Farina, A (2014) Soundscape Ecology, Springer, Dordrecht. Farina, A and Belgrano, A (2004) The eco‐field: a new paradigm for landscape ecology. Ecological Research, 19, 107–110. Farina, A and Belgrano, A (2006) The eco‐field hypothesis: toward a cognitive landscape. Landscape Ecology, 21, 5–17. Farina, A and James, P (2016) Acoustic community structure and dynamics: a fundamental component of ecoacoustics. Biosystems, 147, 11–20. Farina, A and Pieretti, N (2014a) Acoustic codes in action in a soundscape context. Biosemiotics, 7(2), 321–328. Farina, A and Pieretti, N (2014b) Sonic environment and vegetation structure: a methodological approach for a soundscape analysis of a Mediterranean maqui. Ecological Informatics, 21, 120–132. Farina, A and Salutari, P (2016a) Applying the Ecoacoustic Event Detection and Identification (EEDI) model to the analysis of acoustic complexity. Journal of Mediterranean Ecology, 14, 13–42. Farina, A, Pieretti, N and Piccioli, L (2011) The soundscape methodology for long‐term bird monitoring: a Mediterranean Europe case‐study. Ecological Informatics, 6, 354–363. Farina, A, James, P, Bobryk, C, Pieretti, N, Lattanzi, E and McWilliam, J (2014) Low cost (audio) recording (LCR) for advancing soundscape ecology towards the conservation of sonic complexity and biodiversity in natural and urban landscapes. Urban Ecosystems, 17, 923–944. Farina, A, Ceraulo, M, Bobryk, C, Pieretti, N, Quinci, E and Lattanzi, E (2015) Spatial and temporal variation of bird dawn chorus and successive acoustic morning activity in a Mediterranean landscape. Bioacoustics, 24(3), 269–288. Farina, A, Pieretti, N, Salutari, P, Tognari, E and Lombardi, A (2016) The application of the acoustic complexity indices (ACI) to ecoacoustic event detection and identification (EDDI) modeling. Biosemiotics, 9(2), 227–246. Fitzpatrick, JW, Lammertink, M, Luneau Jr, MD, et al. (2005) Ivory‐billed woodpecker (Campephilus principalis) persists in continental North America. Science, 308(5727), 1460–1462. Fuller, S, Axel, AC, Tucker, D and Gage, SH (2015) Connecting soundscape to landscape: which acoustic index best describes landscape configuration? Ecological Indicators, 58, 207–215. Gage, SH and Axel, AC (2014) Visualization of temporal change in soundscape power of a Michigan lake habitat over a 4‐year period. Ecological Informatics, 21, 100–109. Gasc, A, Sueur, J, Jiguet, F, et al. (2013) Assessing biodiversity with sound: do acoustic diversity indices reflect phylogenetic and functional diversities of bird communities? Ecological Indicators, 25, 279–287. Gaston, KJ (2000) Global patterns in biodiversity. Nature, 405, 220–227. Gillooly, JF and Ophir, AG (2010) The energetic basis of acoustic communication. Proceedings of the Royal Society B, 277, 1325–1331. Gordo, O (2007) Why are bird migration dates shifting? A review of weather and climate effects on avian migratory phenology. Climate Research, 35, 37–58.

7  Biodiversity Assessment in Temperate Biomes using Ecoacoustics

Harris, S, Shears, NT and Radford, CA (2016) Ecoacoustic indices as proxies for biodiversity on temperate reefs. Methods in Ecology and Evolution, 7, 713–724. Heittola, T, Mesaros, A, Virtanen, T and Eronen, A (2011) Sound event detection and context recognition. CHiME 2011 Workshop on Machine Listening in Multisource Environments. Henriquez, A, Alonso, JB, Travieso, CM, et al. (2014) An automatic acoustic bat identification system based on the audible spectrum. Expert Systems with Applications, 41, 5451–5465. Hoffmeyer, J (2008) Biosemiotics. An examination into the Signs of Life and the Life of Signs, Scranton University Press, Scranton. Hunter, M Jr (1989) Himalayan birds face uphill while singing. Auk, 106, 728–729. Huntley, B, Collingham, YC, Green, RE, Hilton, GM, Rahbek, C and Willis, SG (2006) Potential impacts of climatic change upon geographical distributions of birds. Ibis, 148, 8–28. Hurlbert, SH (1971) The nonconcept of species diversity: a critique and alternative parameters. Ecology, 52(4), 577–585. Hutchinson, GE (1957) Concluding remarks. Cold Spring Harbor Symposia on Quantitative Biology, 22, 415–427. Hutchinson, GE (1978) An Introduction to Population Ecology, Yale University Press, New Haven. Inouye, DW, Barr, B, Armitage, KB and Inouye, BD (2000) Climate change affecting altitudinal migrants and hibernating species. Proceedings of the National Academy of Sciences, 97(4), 1630–1633. IPCC (2013) Summary for Policymakers, in Climate Change 2013: The Physical Science Basis, Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Available at: www.ipcc.ch IPCC (2014) https://ipcc‐wg2.gov/AR5/report/ Joo, W, Gage, SH and Kasten, EP (2011) Analysis and interpretation of variability in soundscapes along an urban–rural gradient. Landscape and Urban Planning, 103, 259–276. Kasten, EP, McKinley, PK and Gage, SH (2010) Ensemble extraction for classification and detection of bird species. Ecological Informatics, 5(3), 153–166. Kasten, E, Gage, SH, Fox, J and Joo, W (2012) The remote environmental assessment laboratory’s acoustic library: an archive for studying soundscape ecology. Ecological Informatics, 12, 50–67. Krause, B (1987) Bioacoustics, habitat ambience in ecological balance. Whole Earth Review, 57, 14–18. Krause, B (1993) The niche hypothesis. Soundscape Newsletter, 6, 6–10. Krause, B and Farina, A (2016) Using ecoacoustic methods to survey the impacts of climate change on biodiversity. Biological Conservation, 195, 245–253. Llusia, D, Marquez, R, Beltran, JF, Benitez, M and Do Amara, J (2013) Calling behavior under climatic change: geographical and seasonal variation of calling temperatures in ectotherms. Global Change Biology, 19, 2655–2674. Lomolino, MV, Pijanowski, BC and Gasc, A (2015) The silence of biogeography. Journal of Biogeography, 42, 1187–1196. Malavasi, R and Farina, A (2013) Neighbors talk: interspecific choruses among songbirds. Bioacoustics, 22(1), 33–48.

125

126

Ecoacoustics

Malavasi , R, Kull, K and Farina A (2014) The acoustic codes: how animal sign processes create sound‐topes an consortia via conflict avoidance. Biosemiotics, 7, 89–95. McWilliam, JN and Hawkins, AD (2013) A comparison of inshore marine soundscapes. Journal of Experimental Marine Biology and Ecology, 446, 166–176. Mellinger, DK and Clark, CW (1997) Methods for automatic detection of mysticete sounds. Marine and Freshwater Behaviour and Physiology, 29, 163–181. Merchant, ND, Fristrup, KM, Johnson, MP, et al. (2015) Measuring acoustic habitats. Methods in Ecology and Evolution, 6, 257–265. Mesaros, A, Heittola, T, Eronen, A and Virtanen, T (2010) Acoustic event detection in real‐life recordings. 18th European Signal Processing Conference (EUSIPCO), Aalborg, Denmark, pp.1267–1271. Millennium Ecosystem Assessment (2005) Ecosystems and Human Well‐being: Synthesis, Island Press, Washington, DC. Monacchi, D (2014) Fragments of extinction. Leonardo Music Journal, 23. Morin, PJ (1999) Community Ecology, Wiley‐Blackwell, Oxford. Moskwik, M, Thom, T, Barnhill, LM, et al. (2013) Search efforts for ivory‐billed woodpecker in South Carolina. Southern Naturalist, 12(1), 73–84. Mueller‐Dombois, D and Ellenberg, H (1974) Aims and Methods of Vegetation Ecology, Wiley, New York. Naeem, S (2002) Ecosystem consequences of biodiversity loss: the evolution of a paradigm. Ecology, 83(6), 1537–1552. Orloci, L (1967) An agglomerative method for classification of plant communities. Journal of Ecology, 55, 193–205. Parker, TA III (1991) On the use of tape recorders in avifaunal surveys. Auk, 108, 443–444. Parmesan, C (2006) Ecological and evolutionary responses to recent climate change. Annual Review of Ecology, Evolution and Systematics, 37, 637–69. Pieretti, N (2015) The impact of noise from open‐cast mining on Atlantic forest biophony. Biological Conservation, 191, 623–631. Pieretti, N, Farina, A and Morri, D (2011) A new methodology to infer the singing activity of an avian community: the Acoustic Complexity Index (ACI). Ecological Indicators, 11, 868–873. Pieretti, N, Duarte, MHL, Sousa‐Lima, RS, Rodrigues, M, Young, RJ and Farina, A (2015) Determining temporal sampling schemes for passive acoustic studies in different tropical ecosystems. Tropical Conservation Science, 8(1), 215–234. Planqué, R and Slabbekoorn, H (2008) Spectral overlap in songs and temporal avoidance in a Peruvian bird assemblage. Ethology, 114, 262–271. Purvis, A and Hector, A (2000) Getting the measure of biodiversity. Nature, 405, 212–219. Ralph, CJ and Scott, JM (1981) Estimating Numbers of Terrestrial Birds, Studies in Avian Biology No. 6, Allen Press, Lawrence. Ralph, CJ, Guepel, GR, Pyle, P, Martin, TE and DeSante, DF (1993) Handbook of Field Methods for Monitoring Landbirds, Pacific Southwest Research Station, US Forest Service, Albany. Shannon, CE (1948) The mathematical theory of communication, in The Mathematical Theory of Communication (eds CE Shannon and W Weaver), University of Illinois Press, Urbana, pp.29–125. Simpson, EH (1949) Measurement of diversity. Nature, 163, 688.

7  Biodiversity Assessment in Temperate Biomes using Ecoacoustics

Sinsch, U, Lumkemann, K and Rosar, K (2012) Acoustic niche partitioning in an anuran community inhabiting an Afromontane wetland (Butare, Rwanda). African Zoology, 47(1), 60–73. Sueur, J (2002) Cicada acoustic communication: potential sound partitioning in a multispecies community from Mexico (Hemiptera: Cicadomorpha: Cicadidae). Biological Journal of the Linnean Society, 75, 379–394. Sueur, J and Farina, A (2015) Ecoacoustics: the ecological investigation and interpretation of environmental sound. Biosemiotics, 8, 493–502. Sueur, J, Pavoine, S, Hamerlynck, O and Duvail, S (2008) Rapid acoustic survey for biodiversity appraisal. PLoS One, 3, 1–9. Sueur, J, Farina, A, Gasc, A, Pieretti, N and Pavoine, S (2014) Acoustic indices for biodiversity assessment and landscape investigation. Acta Acustica united with Acustica, 100, 772–781. Tilman, D, Wedin, D and Knops, J (1996) Productivity and sustainability influenced by biodiversity in grassland ecosystems. Nature, 379, 718–720. Tobias, JA, Planqué, R, Cram, DL and Seddon, N (2014) Species interactions and the structure of complex communication networks. Proceedings of the National Academy of Sciences, 111(3), 1020–1025. Towsey, M, Wimmer, J, Williamson, I, Roe, P (2014a) The use of acoustic indices to determine avian species richness in audio‐recordings of the environment. Ecological Informatics, 21, 110–119. Towsey, M, Zhang, L, Cottman‐Fields, M, Wimmer, J, Zhang, J, Roe, P (2014b) Visualization of long‐duration acoustic recordings of the environment. Proceedia Computer Science, 29, 703–712. Tucker, D, Gage, SH, Williamson, I and Fuller, S (2014) Linking ecological condition and the soundscape in fragmented Australian forests. Landscape Ecology, 29, 745–758. Verner, J (1985) Assessment of counting techniques. Current Ornithology, 2, 247–302. Von Uexküll, J (1982) The theory of meaning. Semiotica, 42(1), 25–82. Wallschager, D (1980) Correlation of song frequency and body weight in passerine birds. Experientia, 36, 412. Whittaker, RH (1952) A study of the summer foliage insect communities in the Great Smoky Mountains. Ecological Monographs, 22, 1–44. Whittaker, RH (1972) Evolution and measurement of species diversity. Taxon, 21, 213–251. Winner, J, Towsey, M, Roe, P and Williamson, I (2013) Sampling environmental acoustic recordings to determine bird species richness. Ecological Applications, 23(6), 1419–1428. Zhuang, X, Zhiu, X, Hasegawa‐Johnson, MA and Huang, TS (2010) Real‐world acoustic event detection. Pattern Recognition Letters, 31, 1543–1551.

127

129

8 Biodiversity Assessment in Tropical Biomes using Ecoacoustics: Linking Soundscape to Forest Structure in a Human-dominated Tropical Dry Forest in  Southern Madagascar Lyndsay Rankin1 and Anne C. Axel2 1 2

Department of Biological Sciences, Northern Illinois University, DeKalb, USA Department of Biological Sciences, Marshall University, Huntington, USA

8.1 ­Introduction Tropical biomes contain some of the most biodiverse ecoregions in the world (Myers et al. 2000). Many of these regions are greatly impacted by humans, with one, the tropical dry forest, experiencing the greatest loss of habitat worldwide (Hoekstra et al. 2005). As of 2000, the tropical dry forest covered 1 048 700 km2 within the Americas, Africa, Eurasia, and Australasia (Miles et al. 2006) – some 20% of global forest area (Hansen et  al. 2010). By 2005, nearly 3% of that had been removed (Hansen et  al. 2010). The hospitable climate and forest structure of tropical dry forest provide favorable conditions for human uses, including agriculture, forest product extraction, and livestock grazing (Murphy and Lugo 1986). Additional threats to tropical dry forests include climate change, fire, forest fragmentation, and human population growth (Miles et al. 2006). In Africa, 70–80% of all forest cover is defined as tropical dry forest (Murphy and Lugo 1986) and approximately half of the continent’s population relies on its resources (Chidumayo and Gumbo 2010). In Madagascar, tropical dry forest is found in areas with warm year-round climates, brief periods of variable rainfall, and long dry seasons (Mooney et al. 1995; Murphy and Lugo 1986). As a biodiversity hotspot (Myers et al. 2000), the forests of Madagascar are home to more than 90% of the island endemics (Dufils 2003), yet there was a 40% loss in forest cover from 1950 to 2000 (Harper et al. 2007). The tropical dry forest is particularly important as in some regions it represents the country’s highest plant endemism at 95% (Koechlin 1972), and it is experiencing a higher rate of deforestation than that of Malagasy rainforests (Sussman and Rakotozafy 1994; Waeber et al. 2015). The majority of the remaining tropical dry forest is disturbed to some degree, yet it still supports some of Madagascar’s flagship animal species, including radiated tortoise (Astrochelys radiata), subdesert mesite (Monias benschi), long-tailed ground-roller (Uratelornis chimaera), and ring-tailed lemur (Lemur catta). Small isolated forest patches in southern Madagascar may still be capable of sustaining wildlife (Bodin et al. 2006) since the collective area of suitable habitat patches in close proximity to one another may remain functional to certain species (Andrén 1994). Ecoacoustics: The Ecological Role of Sounds, First Edition. Edited by Almo Farina and Stuart H. Gage. © 2017 John Wiley & Sons Ltd. Published 2017 by John Wiley & Sons Ltd.

130

Ecoacoustics

A foundational principle of ecology is the habitat heterogeneity hypothesis, which assumes that habitats with the most structural complexity will provide more niches, leading to more species diversity (MacArthur and MacArthur 1961). A metaanalysis of research on the subject found that an overwhelming majority of papers reported a positive relationship between some measure of habitat heterogeneity (i.e. habitat diversity, habitat complexity, structural diversity, structural complexity, spatial complexity, etc.) and species diversity (Tews et  al. 2004). Managers and researchers are frequently charged with characterizing human impacts on landscapes. Such impacts often result in habitat structural changes with subsequent loss of species abundance (Pardini et  al. 2005) and diversity (Soulé 1991), as well as negative effects on ecosystem function (Chapin et al. 2000). The biodiversity field survey is a primary means of characterizing ecosystem health and function in both intact and disturbed landscapes, yet these are costly and time-intensive (Gardner et al. 2008). Soundscape monitoring is a relatively new cost- and time-efficient method of characterizing ecosystem condition (Tucker et al. 2014) and assessing biodiversity (Sueur et al. 2014). Soundscape ecology is the study of the relationship between the acoustic properties of an environment and its physical and behavioral characteristics (Pijanowski et al. 2011a,b; Schafer 1994). A soundscape consists of all acoustic energy produced by an ecosystem, including biological, anthropogenic, and geophysical sounds (Qi et al. 2008; Pijanowski et  al. 2011a; Schafer 1994; Truax 1978). Anthrophony is all the mechanical sounds produced by humans, such as vehicles and machines. Biophony represents all the sounds produced by animals in their environment. And the collection of sounds produced by the Earth’s physical processes, such as wind, rain, and running water, is known as geophony. Soundscape analysis provides a novel means of assessing impacts of human disturbance on landscapes. Acoustic sound analysis has been used to characterize the biological diversity of soundscapes (Sueur et al. 2008a), connect soundscape to ecological condition (Tucker et al. 2014), visualize changes in soundscapes over time (Gage and Axel 2014), and assess the effects of human disturbances on a landscape (Joo et al. 2011). Multiple acoustic indices have been developed to highlight different aspects of a soundscape. Studies have shown that more biodiverse habitats will emit more biophony (wildlife sounds) (Gage et al. 2001; Qi et al. 2008), and biophonic sounds can be quantified in a variety of ways using acoustic metrics. Acoustic diversity indices have been correlated with traditional diversity indices (Gage et al. 2001) and used as a method of rapid biodiversity assessment (Sueur et al. 2008b, 2012). To examine the connection between soundscape and forest structure in a humandominated landscape, we characterized the forest structure in two distinct forest classes in a highly seasonal tropical dry forest and then investigated the relationship between the soundscape and forest measurements. We collected soundscape samples over a period of one year, and then computed the Bioacoustic Index (Boelman et al. 2007) acoustic metric for each sample. We expected greater biodiversity in areas of the landscape with the most structure. In this seasonal landscape, forest structure changes dramatically with seasons, so we modeled relationships by season. We expected to find the greatest species diversity, as indicated by the Bioacoustic Index (BIO), in the wet season when the forest canopy is at its fullest, and the least diversity during the dry season. Furthermore, we expected the Bioacoustic Index to be higher in areas with the most forest structure.

8  Biodiversity Assessment in Tropical Biomes using Ecoacoustics

8.2 ­Methods 8.2.1  Study Area

The study occurred at the Beza Mahafaly Special Reserve (–23.6505, 44.631931) located in the tropical dry forest of south-west Madagascar (Figure 8.1) about 35 kilometers NE of Betioky, capital of the Betioky Sud district. The climate here is highly seasonal, with distinct rainy (November to March) and dry (April to October) seasons. The reserve, established in 1986, consists of two original discontinuous parcels (600 ha) and an additional 4000 ha “extension” added in 2007 that encompasses areas between and beyond the two parcels. Three classes of tropical dry forest are found in the reserve situated along a water gradient. Gallery forest is located along the Sakamena River; this then transitions to dry deciduous forest about 0.5 km from water. Spiny thicket is found on very dry soils in areas most distant (~3–4 km) from water. Grazing is the largest human impact on all forest classes in Beza Mahafaly Special Reserve. The major grazing impact in gallery forest is coppicing which results in loss of canopy cover. The major grazing impacts in dry deciduous forest are cutting of small-stemmed tree species, resulting in loss of biomass, and soil trampling, leading to soil compaction. Grazing is prohibited in the 80 ha Parcel 1, but unrestricted within Parcel 2 and the extension. This creates a mixture of moderately to heavily grazed areas outside Parcel 1. The wild organisms that predominate in this soundscape are birds, insects, lemurs (nocturnal and diurnal), and bats. There are only a handful of amphibian species at the site so, unlike many tropical forests, amphibians do not contribute substantially to the soundscape. There are many reptiles at the site, but the only vocalizing species are lizards (including geckos), and their vocalizations are largely overpowered by insects and birds.

Legend Roads Beza Mahafaly Special Reserve Sakamena River Study area 3

4

Manasoa Mahazoarivo

Kilometers

Parcel I

Sakamena River

00.51 2

Onilahy River

Analafaly

Camp Parcel II Sarodrano

Road to Betioky

Ampitanabo N

Miary Boribery

Figure 8.1  Map of the Beza Mahafaly Special Reserve with study area in gray and reference to location in Madagascar.

131

132

Ecoacoustics

Study area

N

Dry deciduous Gallery Recorders

0

0.75

1.5 km

Figure 8.2  Map of sampling sites established within the study area at the Beza Mahafaly Special Reserve.

Six sites, three in gallery forest and three in dry deciduous, were selected within the study area from areas experiencing different intensities of grazing (Figure 8.2). Forest class was determined via a land cover classification from multitemporal satellite imagery of the study area (Axel 2011) and by on-the-ground assessment. 8.2.2  Forest Sampling

Forest structure sampling was conducted in June and July 2013. A grid of nine circular plots, each with a radius of 10 m spaced 30 m apart, was located at each of the six sampling sites to measure canopy characteristics. Variables sampled within each plot included tree species richness; number of kily (Tamarindus indica); total kily diameter at breast height (DBH, m); basal area (m2/ha); canopy height (m); and percent canopy cover. The tamarind tree, known locally as kily, is a defining species of gallery forest (Jolly 1966; Sussman and Rakotozafy 1994), as well as a keystone species for many wildlife, including the ring-tailed lemur (Lemur catta) (Jolly 1966; Sauther 1998; Sussman and Rakotozafy 1994). Three 1 sq m subplots were randomly placed within each circular plot to measure understory characteristics, including seedling species richness (tree and vine species less than 1 m in height/length), total number of seedlings, seedling species diversity, percent seedling cover, soil hardness (pound force, lbf ), leaf litter thickness (cm), and percent grass cover. On average, forest structure canopy and understory variables were all significantly higher in gallery forest, except for soil hardness, which was significantly higher in dry deciduous forest, and tree species richness, which was not significantly different between classes (Rankin 2014). However, with grazing disturbance, it is not always the

8  Biodiversity Assessment in Tropical Biomes using Ecoacoustics

case, for instance, that a gallery forest site will have more canopy cover than a dry forest site. For this reason, we used forest structure variables, and not forest class factor, as predictors of forest structure. 8.2.3  Soundscape Survey

Wildlife Acoustics Song Meter (Wildlife Acoustics 2012) autonomous recorders were installed at each of the six sampling sites in January 2013. Each sensor was installed in a tree roughly 2–3 m above ground at the center of each site. Recorders were deployed for 12 months, between January and December 2013, taking one-minute sound samples at an interval of 15 minutes. There were some data gaps through the sampling period. Infrequently, recorders skipped a quarterly-hour session, resulting in only 2–3 recordings per hour. Additionally, battery failures accounted for some loss of data. Notably, one recorder failed across the month-long “leanest season” (defined below) in the dry deciduous forest. Acoustic samples were recorded in the waveform audio file format (WAV) at a frequency of 24 000 Hz with a recorded frequency range up to 11 kHz. All recordings were resampled to 22 050 Hz to conform to parameters of code obtained through the R Project using tuneR (Ligges et  al. 2015) and Soundecology (Villanueva-Rivera and Pijanowski 2015) packages. Roughly 8.5% of the files were removed from the analyses as they were either corrupted due to recorder malfunction or overwhelmed with cicada calls or hard rainfall that masked sounds of all other biological diversity (as described below). We chose to remove all samples where the bulk of the frequency range was overwhelmed by cicadas, as there is evidence that bird species avoid vocalizing during bouts of cicada signaling (Hart et al. 2015), and it is reasonable to assume that other vocalizing species would avoid spectral overlap as well. 8.2.4  Acoustic Index

The Bioacoustic Index was calculated for each quarterly-hour sound recording using the R statistical computing environment (R Core Team 2015). The BIO (Boelman et al. 2007) was calculated on sounds from 1000 to 11 000 Hz using default parameters in the Soundecology package (Villanueva Rivera and Pijanowski 2015). Given that the study area did not contain mechanical sounds associated with anthrophony, the bioacoustic range was extended below the default minimum frequency of 2000 Hz. The maximum frequency was also raised from 8000 to 11 000 to detect acoustic activity of insects and nocturnal mammals known to exist within this range in the study area. Monthly averages of the BIO were computed and graphed to represent seasonal phenology of the soundscape. Sound-truth evaluation of our data indicates that very small BIO values correspond to samples with an absence of sound activity, while the largest BIO values (>90) reliably correspond to samples with dense broadband sound structure typical of hard rain and/ or cicada chorusing. BIO values between 60 and 90 with high acoustic activity in the 7000–11 000 Hz range also corresponded to hard rain and cicada chorusing. Values approaching 90 without rain and cicada chorus are the most acoustically rich, with loud acoustic activity across most frequencies. We attributed all calls with BIO >90 and 60–90 with high activity in the higher frequency range to cicadas and hard rainfall and removed those from the dataset. The BIO is a particularly satisfying acoustic index due to its ease in interpretability. After clipping cicadas and heavy rain, large values represent sound samples composed

133

134

Ecoacoustics

of loud vocalizations emanating from a large range of frequencies. This scenario should occur in areas of high community richness with high abundance. In contrast, small values represent areas with little or no acoustic activity, typical of sites with few vocalizing individuals or species. The algorithm converts a sound sample to a normalized frequency-amplitude plot (a plot of dB versus frequency) and then draws an envelope along the curve. The value of the index is the area under this curve. Sound samples with very loud sounds, but which fall within only a small range of frequency, will have a flat curve with a single steep peak. Sound samples which span the entire frequency range at low amplitudes will have a fairly flat curve across the entire range. Neither will result in a great deal of area under the curve. Those sound samples with loud sounds (close to the recorder) that span a larger frequency range (typically reflective of multiple species) will have a tall curve across the entire frequency range resulting in a large area and, thereby, a large BIO value. The BIO was found to correlate strongly with avifauna counts from direct bird surveys (Boelman et  al. 2007); higher BIO values corresponded to higher abundance of all target species observed in the field. One expects to find more community members – not just more species, but also more individuals – in areas with high BIO values. This should be indicative of quality habitat for vocalizing species. 8.2.5  Mixed Model Analysis

A “full” mixed model with all forest structure variables was run on the entire year-long acoustic data set. In addition, the 12-month dataset was subset into three representative one-month seasons: sampling, greenest, and least green (“leanest”). The forest sampling season occurred between June 15 and July 15 when forest measurements were collected within the study area. At this time, differences in canopy between the two forest class canopies are most extreme as kily in the gallery forest are still quite leafy, while many trees in dry deciduous forest are in leaf-off condition by this time. The greenest and leanest seasons for 2013 were determined from a time series of weekly mean Normalized Difference Vegetation Index (NDVI) values of the study area calculated from MODIS Terra (MOD13Q1) and Aqua (MYD13Q1) satellite images. Using “composite day of the year” information accompanying the MODIS NDVI, month-long timeframes with the highest and lowest NDVI values were assigned as the greenest and leanest months respectively. The greenest month-long range was February 14 to March 14 and was characterized by a closed canopy of thick foliage in gallery forest and a full leaf-on canopy in the relatively more open dry deciduous forest. The leanest month-long range was September 15 to October 15. This is the time of year when the gallery forest is most open as kily trees tend to lose their leaves en masse in October. Understory vegetation is more prevalent in gallery forest during this time when sun is finally able to reach the forest floor. The dry deciduous forest canopy is at this time very open and the understory is relatively dry and deplenished. Given the lag in vegetation response to precipitation, the greenest season was at the end of the rainy season and the leanest season at the end of the dry season. The sampling season took place in the middle of these two extremes. Acoustic metric values (BIO) were averaged by hour, resulting in 24 hourly values per day per site. Hourly acoustic metrics were treated as repeated samples within sites rather than independent values (Gutzwiller and Riffell 2007; Schielzeth and Forstmeier 2009). We modeled relationships between BIO and forest class and BIO and forest structure using a random intercepts model in a mixed model analysis (Pinheiro and

8  Biodiversity Assessment in Tropical Biomes using Ecoacoustics

Bates 2000). Sites were modeled as a random effect (nested within forest class) using a covariance structure to account for potential temporal autocorrelation. This is the preferred method for handling potentially autocorrelated data instead of using repeatedmeasures analysis of variance which requires independence between samples (Gutzwiller and Riffell 2007). We selected forest structure variables for each seasonal model that we felt best reflected the structure of that season. We selected the following forest structure variables for modeling: canopy cover, tree basal area, kily DBH, and grass cover. Canopy cover estimates were best modeled in the sampling season model. We felt canopy cover sampled in July was not representative of differences in forest class during other seasons. While grass cover does vary by season, unlike canopy cover, relative proportions of grass cover will remain fairly constant outside of the sampling season. For instance, areas of high grass cover in July would have relatively more grass cover in other seasons than those sites with low grass cover in July. (This reasoning does not hold true with canopy cover due to the nature of unique pattern of leaf-out and leaf loss in kily trees in gallery forest that differs from leafing timing patterns in the dry deciduous forest.) Mixed-effects models relating BIO to forest class (Y = Xβ + Zγ + ε) were fit using restricted maximum likelihood estimation (REML) with and without covariance structures using the nlme package in the R Statistical Program (Bates et al. 2014; R Core Team 2015). Fixed effects included the acoustic metric, BIO (Y), and forest class. Random effects included sites and season or month (Z) (if the dataset spanned multiple seasons). Mixed-effects models relating BIO to forest structure (Y = Xβ + Zγ + ε) were also fit using restricted maximum likelihood estimation (REML) using nlme package with and without covariance structures. Fixed effects included BIO (Y) and forest structure variable(s) including canopy cover, tree basal area, kily basal area, and grass cover. Random effects included sites nested within forest class and season or month (Z) (if the dataset spanned multiple seasons). In the case of both forest class and forest structure variable models, the fixed intercept effect (β) was common among sites while random intercepts (γ) allowed for site- or season/month-specific variation in the models (Mikkonen et  al. 2008). Residual covariance structures (ε) were built into models to account for possible temporal autocorrelation. Candidate covariance structures included autoregressive (corAR1), autoregressive moving average (corARMA), and autoregressive with heterogeneous variances (corARH). Candidate models were compared using Akaike’s Information Criterion (AIC). Models with the smallest AIC or largest ΔAIC were selected as the best fit model for each analysis. Mixed models were run over the following time periods: full year; combined greenest, leanest, and sampling seasons; combined greenest and leanest seasons; greenest only; leanest only; and forest sampling season only.

8.3 ­Results 8.3.1  Acoustic Index by Season

Average monthly BIO values were relatively high at the beginning of the year (in the middle of the rainy season). Values then steadily declined into the dry season, reaching a low point in September. There followed a very steep increase between September and December (the peak value) from late in the dry season to early in the rainy season (Figure 8.3).

135

20

Mean Hourly BIO 40 60

Ecoacoustics

1

2

3

4

5

6 7 Month

8

9

10

11

12

Figure 8.3  Boxplots of Bioacoustic Index (BIO) by month across the year-long sampling period. Black line is median and boxes represent the upper and lower quartiles.

30

BIO

136

20

10 2

4

6 Month

8

10

12

Figure 8.4  Bioacoustic Index (BIO) averaged by month across the year-long sampling period. Solid line is dry deciduous forest and dashed line is gallery forest. Error bars are standard errors.

Monthly averages of the BIO were plotted by forest class over the sampling year (Figure 8.4). Both gallery and dry deciduous forest sites followed the same general pattern in monthly mean BIO values except during April, when gallery forest values continued to climb and deciduous dry forest values held steady, and in June, when gallery forest values remained steady (or slightly increased) and dry deciduous forest values continued to decline. During the transitional period between the extremes of the wet and dry season (mid-April to mid-May and June–July), differences in BIO between forest classes are at their maximum.

8  Biodiversity Assessment in Tropical Biomes using Ecoacoustics

8.3.2  Mixed Model Analyses

There was no evidence for a significant relationship between BIO and forest class in any of the seasonal models (Table 8.1). While the two forest classes are significantly different for every measured forest structure variable, there is still high variability for each measure within each forest class. Forest structure variables provided a more direct measurement of the landscape, and thus were found to be more suitable for detecting relationships between landscape and soundscape. Using model selection methods outlined in Zuur et al. (2009), we found BIO was significantly related to forest structure variables, although not every seasonal model showed significance (see Table 8.1). The BIO was significantly related to tree canopy cover and grass cover. All best fit models contain either autoregressive (corAR1), autoregressive with heterogeneous variance (corAR1H), or autoregressive moving average (corARMA) covariance structures. Best fit models during the full year, greenest, leanest, and sampling seasons included random intercepts by site. Best fit models for combined season models included random intercepts by site and season. There was a significantly positive relationship during the sampling season between mean hourly BIO and canopy cover (P = 0.004), and a significantly negative relationship during the leanest season between mean hourly BIO and grass cover (P = 0.04) (Table 8.2).

8.4 ­Discussion Soundscape analyses have the potential to be an effective and widespread approach to biodiversity and landscape assessment. While previous studies have examined the relationship between acoustic metrics and landscape structure (Fuller et  al. 2015; Tucker et al. 2014) or species diversity (Sueur et al. 2014) in diverse ecosystems and at varying timeframes, few have explored differences in these relationships in different seasons. Our study is one of the first to examine how structural differences related to seasonality in the landscape are reflected in the soundscape. This is also the first extensive soundscape analysis of a highly seasonal forest. As expected, soundscape activity, as measured by the Bioacoustic Index, is at its maximum during the beginning of the rainy season – a time when many organisms are entering their mating season – and it is at its minimum during the dry season, a time when most food resources and protective cover are at their minimum. The similarity in hourly means of BIO between forest classes during the bulk of the rainy and dry seasons suggests wildlife is utilizing these forests fairly equally during seasonal extremes. This may be due, in part, to the spatial proximity of forest classes. Agricultural activities have greatly reduced the amount of gallery forest in the southern section of the study area where patches of gallery forest are practically embedded in areas of dry deciduous forest. Recorders were situated so that sounds would only be detected from gallery forest, but given the short distance to dry deciduous forest, vocalizing organisms that travel a distance greater than about 200 m probably make use of both dry deciduous and gallery forest to some extent. While spatial proximity may help explain similarity in BIO between forest classes at seasonal extremes, this similarity is more likely due to the homogenizing effect of plant water availability. During the rainy season, new growth can be found in both forest classes throughout the landscape, and during the dry season, animals rely

137

2.115

Forest class

Intercept

Forest class

greenest, leanest

Greenest season

Sampling season

Leanest season

22.279

Intercept

Combined seasons:

11.024 2.658

Forest class

-2.681

Forest class

Intercept

18.717

Intercept

–0.587

20.892

0.914

Forest class

17.194

Intercept

greenest, leanest, sampling

1.461

Forest class

Combined seasons:

19.514

Intercept

Full year

Estimate

Estimate

Model

1.528

1.076

2.490

1.934

3.573

2.490

3.061

2.259

2.853

2.074

2.539

1.835

Standard error

4

3966

3

3157

4

2718

9

6033

15

9999

4

32875

d.f.

1.739

10.243

–1.077

9.677

0.592

8.948

–0.192

9.248

0.753

8.289

0.575

10.632

t-Value

Table 8.1  Best fit linear mixed models between BIO and forest class during each seasonal period.

0.157

0.000

0.360

0.000

0.586

0.000

0.852

0.000

0.753

0.000

0.596

0.000

p-Value

AR(1) H

AR(1) H

AR(1) H

AR(1)

AR(1)

AR(1)

Covariance structure

0.47

0.43

0.72

0.59

0.58

0.58

Phi ϕ

–1619

–907

–2346

–3533

–7719

–11201

Δ AIC

–0.787

Intercept

Basal area

Grass cover

Combined seasons:

greenest, leanest

Sampling season

Leanest season

Greenest season

2.427

Grass cover

sampling

-0.707

Grass cover

Canopy cover

0.237

–2.885

1.379

Intercept

36.677

–0.759

Grass cover

Basal area

3.238

Basal area

Intercept

30.264

Intercept

35.047

–0.558

2.209

24.131

Intercept

Basal area

–0.212

Grass cover

Combined seasons:

0.639

Basal area

greenest, leanest,

24.316

Intercept

Full year

Estimate

Estimate

Model

0.039

2.538

0.139

0.533

3.025

0.618

2.497

11.871

0.513

2.018

10.457

0.515

2.054

10.252

0.515

2.062

10.150

  Standard error

4

3966

2

2

3157

3

3

2718

8

8

6033

14

14

9999

3

3

32875

d.f.

6.012

–1.136

–5.075

2.590

12.125

–1.230

1.297

2.549

–1.535

1.202

3.352

–1.083

1.075

2.354

–0.412

0.310

2.396

t-Value

0.004

0.256

0.037

0.122

0.000

0.307

0.285

0.011

0.163

0.264

 0.0008

0.297

0.300

0.019

0.708

0.777

0.017

p-Value

Table 8.2  Best fit linear mixed models between BIO and forest structure variable(s) during each seasonal period.

AR(1) H

AR(1) H

AR(1) H

AR(1)

AR(1)

AR(1) H

Covariance structure

0.47

0.43

0.72

0.59

0.58

0.58

Phi ϕ

–1122

–771

–2299

–3300

–7531

–31330

Δ AIC

140

Ecoacoustics

more on foods found in the understory of both forest classes. Moderate grazing alters the structure of gallery forest such that its structural properties – not tree species composition – resemble dry deciduous forest in the dry season. It is during the transitional period between the extremes of the wet and dry season that differences in soundscape activity between forest classes are at their maximum, suggesting that organisms might be shifting or concentrating resources during this period. From mid-April to mid-May, soundscape activity remains high in the dry deciduous forest, while activity continues to decline in gallery forest. Following a long period of leaf-on condition, there would, at this time, be much higher concentration of understory material available in the more open deciduous dry forest. Just a matter of weeks later, during June and July, there is a shift in maximum soundscape activity to gallery forest when BIO continues to decline in the deciduous dry forest but slightly increases in gallery forest. By this time, temperatures are at their minimum in the forest and the dry season is well under way. Animals may favor the warmer temperatures provided by the fuller canopy in the gallery forest; the kily trees afford a large number of sleeping spots in the still dense canopy that provides cover during the cold cloudless nights. Results from the mixed models support this explanation. Forest sampling was conducted during the June–July period, so all forest structural variables have the potential to be significantly related to BIO during this time frame. However, canopy cover was the only significant predictor in June–July. Outside this time period, there was no other canopy structural variable significantly related to BIO. And just several weeks later, at the height of the dry season, it is grass cover, an understory variable, that is significantly negatively related to BIO. Furthermore, grass cover is strongly positively correlated with canopy cover, yet there is no longer a significant relationship between BIO and canopy cover. Low grass cover in the leanest season (lowest NDVI across the site) corresponds to areas of forest without leaves much of the year. Even though the gallery forest is opening up during the lean season, there would still be relatively less grass there since large openings would be present for such a short window of time. If low grass cover was acting as a proxy for gallery forest sites, leaf litter cover (which is not significantly related to BIO) would be a much better predictor of highquality gallery forest than grass cover. It seems more likely that low grass cover in the lean season is a proxy for moderately disturbed gallery forest (due to grazing impacts) where some steady light had reached the canopy floor to support herb and shrub growth but not enough steady light throughout the year to support a lot of grass. Herbs and shrubs are a staple resource for lemurs (and presumably other vocalizing creatures) during this time of the year when so many trees are in leaf-off condition. Our results support the habitat heterogeneity–animal species diversity relationship; during the leanest periods of the year, organisms appear to be selecting on habitat structure more than habitat type. Grazing disturbance across both forest classes creates heterogeneity within and between classes. A complicating factor is that an intact forest in this landscape does not necessarily confer the greatest structural complexity. Although this is not the first comparison study of soundscapes of different habitat types (Krause et al. 2011; Tucker et al. 2014), it is the first to compare two terrestrial habitats that are so proximal, as well as so structurally similar. In a comparison study of proximal habitats having vastly different structural differences (such as short-grass prairie and forest) one would expect to find large structural differences throughout the year, despite

8  Biodiversity Assessment in Tropical Biomes using Ecoacoustics

phenological changes. In that case, the timing of acoustic monitoring may not be so critical, as structural differences should reflect acoustic differences throughout the year. In our case, there is heterogeneity in structure across the landscape throughout periods of the year. With a year-long acoustic time series, we were able to examine differences in both structural and acoustic diversity throughout the year and capture differences in habitat usage. In the wet season, resources are plentiful and insects, birds, and mammals are maximizing their movements throughout the landscape  –  even if they are territorial. At this time, soundscapes of both forest classes are essentially the same, because in this typically dry environment, all hospitable environments become more suitable during the wet season. In the dry season, however, structural differences in the forest classes are amplified and acoustic activity reflects this change. Our ability to detect a relationship between forest structure and soundscape activity in this heterogeneous study area strongly supports the theory that soundscapes may indeed serve as a suitable proxy for more expensive traditional biodiversity surveys. In addition, the BIO performed well in showing a significant relationship between forest structure and the soundscape in a disturbed, seasonal landscape. In comparison studies of soundscapes by habitat type, care should be taken to sample the soundscape when differences in habitat structure are most pronounced, as this will maximize the ability to detect a relationship between structure and soundscape activity. In the case of this highly seasonal tropical dry forest, the best time to sample is during the dry season when differences in structure between the two forest classes are at its greatest. In the sampling season, deciduous dry forest is well into leaf-off condition while gallery forest still has full canopy cover. In late dry season, the understory is dramatically different between forest classes, with a great deal of grass cover in the dry deciduous forest and comparatively less in the gallery. One of the benefits of remote acoustic sampling is the flexibility in sampling design. Autonomous recorders are typically deployed during times best suited for researchers’ schedules, as opposed to during the ideal timing for acoustic sampling. When possible, it is best to deploy autonomous recorders over multiple seasons in comparison studies so that the ideal acoustic sampling period may be identified and sampled. While not all tropical biomes are as seasonal as the tropical dry forest, seasonality appears to play an important role in detecting connections between acoustic activity and forest structure. Acknowledgments We would like to thank the staff of the Madagascar Institut pour la Conservation des Ecosystèmes Tropicaux (MICET) in Antananarivo for logistical support and securing of research permission. Thanks to the staff at the Beza Mahafaly Special Reserve, especially Efitiria and Enafa for their field assistance, botanical knowledge, and management of equipment throughout the year. Permission to conduct research was given by Association pour la Gestation et Aires Protégées, the Department des Eaux et Forêts, and the Université d’Antananarivo in Madagascar. Funding for the research was generously provided by the NASA West Virginia Space Grant Consortium Graduate Fellowship and Marshall University.

141

142

Ecoacoustics

References Andrén, H (1994) Effects of habitat fragmentation on birds and mammals in landscapes with different proportions of suitable habitat: a review. Oikos, 71, 355–366. Axel, AC (2011) Lemur–environment relationships in in altered and unaltered tropical dry forests in southern Madagascar. PhD dissertation, Michigan State University, East Lansing. Bates, D, Maechler, M, Bolker, B and Walker, S (2014) lme4: Linear Mixed-Effects Models Using Eigen and S4, R Package Version 1, 1-7. Available at: http://CRAN.R-project.org/ package=lme4 Bodin, O, Tengö, M, Norman, A, Lundberg, T and Elmqvist, T (2006) The value of small size: loss of forest patches and ecological thresholds in southern Madagascar. Ecological Applications, 16, 440–451. Boelman, NT, Asner, GP, Hart, PJ and Martin, RE (2007) Multi-trophic invasion resistance in Hawaii: bioacoustics, field surveys and airborne remote sensing. Ecological Applications, 17, 2137–2144. Chapin, III, FS, Zavaleta, ES, Eviner, VT, et al. (2000) Consequences of changing biodiversity. Nature, 405, 234–242. Chidumayo, EN and Gumbo, DJ (2010) The Dry Forests and Woodlands of Africa: Managing for Products and Services, Earthscan, London. Dufils, JM (2003) Remaining forest cover, in The Natural History of Madagascar (ed. SM Goodman and JP Benstead), University of Chicago Press, Chicago, pp.88–96. Fuller, S, Axel, AC, Tucker, D and Gage, SH (2015) Connecting soundscape to landscape: which acoustic index best describes landscape configuration? Ecological Indicators, 58, 207–215. Gage, SH and Axel, AC (2013) Visualization of temporal change in soundscape power of a Michigan lake habitat over a 4-year period. Ecological Informatics, 21, 100–109. Gage, SH, Napoletano, B and Cooper, M (2001) Assessment of ecosystem biodiversity by acoustic diversity indices. Journal of the Acoustic Society of America, 109, 2430. Gardner, TA, Barlow, J, Araujo, IS, et al. (2008) The cost-effectiveness of biodiversity surveys in tropical forests. Ecology Letters, 11, 139–150. Gutzwiller, KJ and Riffell, SK (2007) Using statistical models to study temporal dynamics of animal–landscape relations, in Temporal Dimensions of Landscape Ecology: Wildlife Responses to Variable Resources (ed. J Bissonette and I Storch), Springer, New York, pp.93–118. Hansen, MC, Stehman, SV and Potapov, PV (2010) Quantification of global gross forest cover loss. Proceedings of the National Academy of Sciences USA, 107, 8650–8655. Harper, G, Steininger, MK, Tucker, CJ, Juhn, D and Hawkins, F (2007) Fifty years of deforestation and forest fragmentation in Madagascar. Environmental Conservation, 34, 1–9. Hart, PJ, Hall, R, Ray, W, Beck, A and Zook, J (2015) Cicadas impact bird communication in a noisy tropical rainforest. Behavioral Ecology, DOI: 10.1093/beheco/arv018 Hoekstra, JM, Boucher, TM, Ricketts, TH and Roberts, C (2005) Confronting a biome crisis: global disparities of habitat loss and protection. Ecology Letters, 8, 23–29. Jolly, A (1966) Lemur Behavior, University of Chicago Press, Chicago. Joo, W, Gage, SH and Kasten, EP (2011) Analysis and interpretation of variability in soundscapes along an urban-rural gradient. Landscape and Urban Planning, 103, 259–276.

8  Biodiversity Assessment in Tropical Biomes using Ecoacoustics

Koechlin, J (1972) Flora and vegetation of Madagascar, in Biogeography and Ecology in Madagascar (ed. R Battistini and G Richard-Vinard), Dr W Junk B.V. Publishers, The Hague, pp.145–190. Krause, B, Gage, SH and Joo, W (2011) Measuring and interpreting the temporal variability in the soundscape at four places in Sequoia National Park. Landscape Ecology, 26, 1247–1256. Ligges, U, Preusser, A, Thieler, A and Weihs, C (2015) Package tuneR. Available at: https:// cran.r-project.org/web/packages/tuneR/tuneR.pdf (accessed 15 December 2016). MacArthur, RH and MacArthur, JW (1961) On bird species diversity. Ecology, 42, 594–598. Mikkonen, S, Rahikainen, M, Virtanen, J, Lehtonen, R, Kuikka, S and Ahvonen, A (2008) A linear mixed model with temporal covariance structures in modelling catch per unit effort of Baltic herring. ICES Journal of Marine Science, 65, 1645–1654. Miles, L, Newton, AC, DeFries, RS, et al. (2006) A global overview of the conservation status of tropical dry forests. Journal of Biogeography, 33, 491–505. Mooney, HA, Bullock, SH and Medina, E (1995) Introduction, in Seasonally Dry Tropical Forests (ed. SH Bullock, HA Mooney and E Medina), Cambridge University Press, Cambridge, pp.1–8. Murphy, PG and Lugo, AE (1985) Ecology of tropical dry forest. Annual Review Ecology and Systematics, 17, 67–88. Myers, N, Mittermeier, RA, Mittermeier, CG, Da Fonseca, GA and Kent, J (2000) Biodiversity hotspots for conservation priorities. Nature, 403, 853–858. Pardini, R, de Souza, SM, Braga-Neto, R and Metzger, JP (2005) The role of forest structure, fragment size and corridors in maintaining small mammal abundance and diversity in an Atlantic forest landscape. Biological Conservation, 124, 253–266. Pijanowski, BC, Farina, A, Gage, SH, Dumyahn, SL and Krause, BL (2011a) What is soundscape ecology? An introduction and overview of an emerging new science. Landscape Ecology, 26, 1213–1232. Pijanowski, BC, Villanueva-Rivera, LJ, Dumyahn SL, et al. (2011b) Soundscape ecology: the science of sound in the landscape. Bioscience, 61, 203–216. Pinheiro, JC and Bates, D (2000) Mixed-Effects Models in S and S-PLUS, Springer, New York. Qi, J, Gage, SH, Joo, W, Napoletano, B and Biswas, S (2008) Soundscape characteristics of an environment: a new ecological indicator of ecosystem health, in Wetland and Water Resource Modeling and Assessment (ed. W Ji), CRC Press, New York, pp.201–211. R Core Team (2015) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. Available at: www.R-project.org/ (accessed 15 December 2016). Rankin, LL (2014) Assessing grazing impacts on a tropical dry forest system in Madagascar through vegetation, satellite image, lemur occupancy, and acoustic analysis. MS thesis, Marshall University, Huntington. Sauther, M (1998) Interplay of phenology and reproduction in ring-tailed lemurs: implications for ring-tailed lemur conservation. Folia Primatologica, 69, 309–320. Schafer, RM (1994) The Soundscape: Our Sonic Environment and the Soundscape and the Tuning Of the World, Destiny Books, Rochester. Schielzeth, H and Forstmeier, W (2009) Conclusions beyond support: overconfident estimates in mixed models. Behavioral Ecology, 20, 416–420.

143

144

Ecoacoustics

Soulé, ME (1991) Conservation: tactics for a constant crisis. Science (Washington), 253, 744–750. Sueur, J, Aubin, T and Simonis, C (2008a) Seewave: a free modular tool for sound analysis and synthesis. Bioacoustics, 18, 213–226. Sueur, J, Pavoine, S, Amerlynck, O and Duvail, S (2008b) Rapid acoustic survey for biodiversity appraisal. PLoS ONE, 3, e4065. Sueur, J, Gasc, A, Grandcolas, P and Pavoine, S (2012) Global estimation of animal diversity using automatic acoustic sensors, in Sensors for Ecology: Towards Integrated Knowledge of Ecosystems (eds JF Le Galliard, JM Guarini and F Gaill), CNRS Editions, Paris, pp. 101–119. Sueur, J, Farina, A, Gasc, A, Pieretti, N and Pavoine, S (2014) Acoustic indices for biodiversity assessment and landscape investigation. Acta Acustica united with Acustica, 100, 772–781. Sussman, RW and Rakotozafy, A (1994) Plant diversity and structural analysis of a tropical dry forest in southwestern Madagascar. Biotropica, 26, 241–254. Tews, J, Brose, U, Grimm, V, et al. (2004) Animal species diversity driven by habitat heterogeneity/diversity: the importance of keystone structures. Journal of Biogeography, 31, 79–92. Truax, B (1978) Handbook for Acoustic Ecology, ARC Publications, Vancouver. Tucker, D, Gage, SH, Williamson, I and Fuller, S (2014) Linking ecological condition and the soundscape in fragmented Australian forests. Landscape Ecology, 29, 745–758. Villanueva Rivera, LJ and Pijanowski, BC (2015) Package soundecology. Available at: https://cran.r-project.org/web/packages/soundecology/soundecology.pdf (accessed 15 December 2016). Waeber, PO, Wilme, L, Ramamonjisoa, B, et al. (2015) Dry forests in Madagascar: neglected and under pressure. International Forestry Review, 17, 127–148. Zuur, AF, Ieno, EN and Saveliev, AA (2009) Mixed Effects Models and Extensions in Ecology with R, Springer, New York.

145

9 Biodiversity Assessment and Environmental Monitoring in Freshwater and Marine Biomes using Ecoacoustics Denise Risch1 and Susan E. Parks2 1 2

Scottish Association for Marine Science (SAMS), Oban, Scotland, UK Department of Biology, Syracuse University, New York, USA

9.1 ­Introduction In terrestrial ecosystems, spatial heterogeneity in landscape features drives species dispersal and distribution through local and regional processes (Cornell and Harrison 2013). Since a large number of species actively produce sounds in a variety of behavioral contexts, it has been shown that the resulting local ecological communities can be characterized by their acoustic signatures (Depraetere et al. 2012; Gasc et al. 2013; Sueur et al. 2008). In contrast, the spatiotemporal distribution of resources in aquatic ecosystems depends on much more dynamic physical processes and is often more ephemeral. However, here too processes such as oceanographic fronts, upwellings, eddies, and currents can aggregate prey and structure local communities with specific sound signatures (Nedelec et  al. 2015; Parks et  al. 2014; Ruppé et  al. 2015). Thus, similarly to terrestrial acoustic habitats, underwater soundscapes are driven by species presence (biotic), physical habitat characteristics (abiotic), and anthropogenic noise contributions (Staaterman et al. 2014). The unique properties of sound propagation under water, where sound travels about five times faster than in air, have long been understood and studied. First efforts of modern ocean acoustic research started after the sinking of the Titanic in 1912 and continued during and after World War I, both events spurring the development of active and passive systems to detect icebergs and submarines (Lurton 2002). By the early 1960s global ocean soundscapes had been assessed for the first time (Wenz 1962). While due to military and economic interests, underwater acoustics developed most rapidly in a marine context, freshwater acoustic research has started to expand in more recent years (Desjonquères et al. 2015; Lugli and Fine 2007; Wysocki et al. 2007). Since the earliest studies, the field of underwater acoustics has developed in many different directions, from the description of different ambient noise sources, such as icebergs, wind, waves, and rain (Nystuen et al. 2000; Royer et al. 2015; Shaw et al. 1978), to the study of animal sound production and reception on different organizational levels from individuals (Parks et al. 2011; Risch et al. 2014b) to taxonomic groups (Tyack and Clark 2000). On an ecosystem level, combining the study of ambient noise and Ecoacoustics: The Ecological Role of Sounds, First Edition. Edited by Almo Farina and Stuart H. Gage. © 2017 John Wiley & Sons Ltd. Published 2017 by John Wiley & Sons Ltd.

146

Ecoacoustics

bioacoustics, underwater habitat acoustics may answer questions about the evolution of signal characteristics (Kyhn et  al. 2013, Ladegaard et  al. 2015) or determine the importance of habitat‐specific soundscapes as cues for animal orientation and navigation (Piercy et al. 2016; Simpson et al. 2010). There have been numerous calls for marine and freshwater biodiversity conservation and observation networks to include acoustic sensors (Duffy et  al. 2013; Scholes et  al. 2012). However, although active and passive acoustic monitoring has been used widely in the past to census different species groups, from zooplankton (Powell and Ohman 2015; Warren et al. 2016) to vertebrates (Van Parijs et al. 2009), biodiversity assessment based on autonomous passive acoustic recordings has so far been applied mostly in a terrestrial context (Sueur and Farina 2015; Sueur et al. 2008, 2014). With the exception of efforts to assess coral reef biodiversity (Kaplan et al. 2015; Kennedy et al. 2010; Staaterman et al. 2013), comparatively few studies to date have investigated aquatic biodiversity on an ecosystem or community level using ecoacoustics. Like many terrestrial ecological systems, global marine and freshwater ecosystems are experiencing an unprecedented loss and redistribution of biodiversity and species richness, due to the large‐scale and far‐reaching effects of human activities, such as accelerated climate change (Dudgeon et al. 2006; Halpern et al. 2008; García Molinos et al. 2016; Strayer and Dudgeon 2010). Such changes in aquatic diversity patterns will not just affect the provision of important marine and freshwater ecosystem services, such as coastal zone protection, provision of drinking water, recreational and commercial fisheries or tourism (Stendera et  al. 2012; Worm et  al. 2006), but will lead to shifting baselines with respect to species richness and distribution, which need to be monitored and taken into account in future conservation planning (Elliott et al. 2015). Due to its applicability in surveying remote areas over extended timescales (Van Parijs et al. 2015; Van Parijs et al. 2009) (Figure 9.1), ecoacoustics can play a vital part in monitoring such large‐scale changes in aquatic biodiversity.

(a)

(b)

Figure 9.1  The range of passive acoustic technologies (PAM), including bottom‐mounted archival recorders, animal‐borne acoustic recording tags, acoustic arrays, autonomous underwater vehicles (a) to survey marine soundscapes consisting of biotic, abiotic, and anthropogenic contributions (b). Source: Van Parijs et al. (2015). Reproduced with permission of Marine Technology Society. (See color plate section for the color representation of this figure.)

9  Biodiversity Assessment and Environmental Monitoring in Freshwater and Marine Biomes

Concerns about the increasing impact of anthropogenic noise, particularly of long‐range and ubiquitous noise sources such as global shipping traffic or seismic surveys, have also led to calls for a more holistic approach to monitoring aquatic soundscapes, away from an entirely species‐specific focus (Clark et al. 2009; Hatch and Fristrup 2009). From a conservation perspective, such larger scale monitoring is fundamental to the full characterization of human noise sources and their spatiotemporal effects on aquatic species and ecosystems (Clark et al. 2009; McWilliam and Hawkins 2013; Merchant et  al. 2014; Slabbekoorn et  al. 2010). In addition, underwater noise can degrade and change the natural acoustic signature of a habitat (for example, through species displacement and loss of biodiversity), a process which  can also be documented using ecoacoustic approaches (Butler et  al. 2016; Piercy et al. 2014). This chapter reviews the use of traditional and newly developed ecoacoustic approaches to monitor the three soundscape components (biotic, abiotic, and anthropogenic) in different freshwater and marine environments, with the aim to identify gaps in knowledge, and provide recommendations for future applications of ecoacoustic tools to aid in the conservation of freshwater and marine biodiversity.

9.2 ­Freshwater Habitats 9.2.1 Rivers

From the longest and widest rivers to the smallest streams, the Earth’s river systems comprise about 2000 km3, which represents less than 0.1% of the total global water resource (Gleick 1993). Despite being restricted in surface area, river systems and freshwater bodies as a whole are hotspots for biodiversity, including 40% of the global fish diversity (Dudgeon et al. 2006). Many species are yet to be described; this is particularly true for invertebrates of the tropics but also for some species of fish. At the same time, rivers and freshwater systems in general also comprise some of the ecosystems with the highest decline of biodiversity (Dudgeon et al. 2006; Pracheil et al. 2013). Many of the largest river systems are heavily used and have been altered for human transport and power production as well as food and water supply. About 1 million dams are fragmenting rivers worldwide (Jackson et  al. 2001) and loadings of nutrients and toxins have poisoned many parts of rivers to an extent that they are effectively ecological dead zones. Some of the largest rivers are drying up due to the construction of large‐scale reservoirs and humans capturing over 50% of global freshwater runoff (Jackson et al. 2001; Postel 2000). Given this crisis, assessing biodiversity, documenting change, and driving effective conservation of river systems are urgent and ecoacoustics may add another tool to bridge the apparent gap between freshwater ecology and conservation biology (Strayer and Dudgeon 2010). 9.2.1.1  Remote Monitoring of Biotic Signals in the Environment

Rivers and streams are naturally noisy habitats, due to flow noise and frequent entrapment of air, and thus represent a challenging environment for acoustically active species. However, several species of aquatic insects, teleost fish, and cetaceans have adapted to communicate in these environments. For example, male water boatmen (Micronecta scholtzi) sing at very high source levels and can dominate soundscapes, especially in

147

148

Ecoacoustics

slow‐flowing rivers (Sueur et al. 2011; Wysocki et al. 2007). Acoustic monitoring in a variety of river systems has revealed that several species of teleost fish are exploiting low‐frequency, quiet noise windows for species‐specific communication (Lugli 2010, 2015). Similarly, river dolphins have adapted to their habitat by producing short, broadband echolocation signals of lower source level to de-emphasize the effects of reverberation and clutter (Jensen et  al. 2013; Ladegaard et  al. 2015). Passive acoustic monitoring of these signals has been used to determine the seasonal and spatial distribution of endangered species (Wang et  al. 2015; Yamamoto et  al. 2016) and, sadly, helped to confirm the first human‐caused extinction of a cetacean species, the Yangtze river dolphin (Lipotes vexillifer) (Turvey et al. 2007). 9.2.1.2  Remote Monitoring of the Environment Using Sound in River Habitats

The description and study of ambient riverine soundscapes largely lag behind other aquatic habitats. Rivers and shallow streams are acoustically complex habitats. Variability in recorded ambient noise levels is influenced by turbulent flow, surface roughness, sediment transport, and discharge (Tonolla et al. 2011). While higher flow levels generally result in higher sound pressure levels over all frequency bands, turbulence caused by surface roughness affects midrange frequencies up to 1 kHz and sediment transport affects higher frequencies and the temporal variability in signal structure (Tonolla et al. 2010, 2011). Other environmental factors, such as wind affecting surface roughness, also influence ambient noise in rivers, but with a lesser relative influence on the overall soundscape than in lakes and oceans (Vračar and Mijić 2011). Ambient noise monitoring can be used to characterize heterogeneity of river segments and acoustic habitats. This information may then be used to elucidate the potential for masking and evolution of various species‐specific communication signals (Lugli and Fine 2007; Wysocki et al. 2007). Ambient noise monitoring in rivers can also be used to rapidly map sediment transport patterns during disturbance events such as river bank erosion or bed scour (Lorang and Tonolla 2014). 9.2.1.3  Anthropogenic Sources of Noise in River Systems

Although anthropogenic contributions to underwater noise have been monitored in most aquatic habitats, few studies of human‐generated noise in rivers exist. Recent baseline recordings in the Hudson River, New York, show spatiotemporal variability with boat density and traffic noise emanating from bridges (Martin and Popper 2016). Bridge traffic noise is a common source of river noise pollution, especially in urban areas, and may mask acoustic signals of freshwater stream fish (Holt and Johnston 2015). Future hydrokinetic turbine developments will be an additional source of noise pollution in large river systems although noise contributions produced by these devices will likely be close to or below natural ambient noise levels in high‐flow environments (Bevelhimer et al. 2016). 9.2.2  Lakes and Ponds

Similar to the great rivers, freshwater lakes and ponds are hotspots of rapidly declining biodiversity (Dudgeon et  al. 2006; Strayer and Dudgeon 2010). As naturally isolated ecosystems, many freshwater lakes show a high degree of endemic species inhabiting small geographic areas that are especially vulnerable to human impact (Strayer and

9  Biodiversity Assessment and Environmental Monitoring in Freshwater and Marine Biomes

Dudgeon 2010). Although recognized as a threat to freshwater species (Wysocki et al. 2006) in much the same way as for marine species, underwater noise from commercial and recreational shipping and other human activities has received dramatically less attention in freshwater environments than in the marine context. Apart from a few key taxa and species, acoustic recordings of freshwater species and soundscapes are also only slowly emerging in recent years. 9.2.2.1  Remote Monitoring of Biotic Signals in the Environment

Acoustics of sound‐producing species in freshwater environments have concentrated on only a few taxa, some of which have been studied extensively. Although not always strictly aquatic, sounds of anuran breeding assemblages have been studied in the laboratory as well as in the field (Bleach et al. 2015; James et al. 2015; Platz 1993; Protázio et al. 2014). Sound production has also been studied in several species of freshwater fish, particularly in an evolutionary and behavioral context (Amorim et  al. 2015; Maruska et al. 2012; Pisanski et al. 2015). Underwater sounds produced by a variety of stridulating aquatic insects have been recorded, but rarely quantified (Aiken 1985; Sueur et al. 2011; Wilson et al. 2015). On an ecosystem level, only one study has investigated the acoustic diversity of freshwater ponds (Desjonquères et al. 2015). Soundscape comparisons of three different pond environments showed variability between and within ponds and suggested the applicability of the Acoustic Richness (AR) Index (Desjonqueres et al. 2015) to study pond biodiversity acoustically. However, more data on the communities of sound‐producing freshwater species and natural background noise levels in different habitats are necessary to determine the usefulness of ecoacoustics indexes for freshwater biodiversity assessments. 9.2.2.2  Remote Monitoring of the Environment Using Sound in Lakes and Ponds

Studies of ambient soundscapes in freshwater environments are rare. One of the few studies that do exist quantified and compared noise levels from a variety of freshwater habitats (Wysocki et al. 2007). Environmental ambient noise was mostly concentrated in frequencies below 500 Hz and, not surprisingly, stagnant water bodies such as lakes were generally quieter than fast‐flowing water (Wysocki et  al. 2007). A recent study conducted in the Northwest Territories of Canada described the under‐ice acoustic environment of a freshwater lake. Seasonal and diel variations in the recorded soundscapes could here be related to biotic and abiotic sources as well as originating from anthropogenic impacts (Martin and Cott 2015). 9.2.2.3  Anthropogenic Sources of Noise in Lakes and Ponds

Similar to the general lack of acoustic data from freshwater environments, few studies have investigated anthropogenic noise pollution in lakes and ponds. However, the increasing influence of commercial and recreational boat noise can be observed and measured in freshwater lakes (Amoser et  al. 2004; Bolgan et  al. 2016; Seppänen and Nieminen 2004) but has seldom be quantified on longer timescales. One study found underwater noise levels to be strongly positively correlated with a metric of urbanization (Kuehne et  al. 2013). Despite increased recognition of the need to regulate and manage increasing anthropogenic noise in terrestrial and marine environments (Agardy et al. 2007; Blickley and Patricelli 2010; Hatch and Fristrup 2009), similar efforts have not been made to conserve freshwater soundscapes. And regulative instruments for

149

150

Ecoacoustics

freshwater conservation like the European Water Framework Directive (WFD 2000/60/ EC) do not currently mention or provide for the management of underwater noise. This should urgently be addressed given the imperiled status of freshwater biodiversity and the known effects of anthropogenic noise on many taxa (Jacobsen et al. 2014; Radford et al. 2014a; Slabbekoorn et al. 2010; Voellmy et al. 2014).

9.3 ­Marine Neritic Habitats 9.3.1  Estuaries and Coastal Habitats

Human populations are concentrated along the world’s coastlines. Even more so than other marine habitats, coastal environments are thus facing a multitude of anthropogenic pressures, including coastal shipping, harbour development, pollution, fishing and aquaculture, and increasingly offshore renewable energy development. Important coastal habitats such as seagrass ecosystems or coastal lagoons are changing rapidly, accelerated by continued population growth and climate change (Cloern et al. 2016). Although more accessible than offshore habitats, at present relatively little is known about coastal soundscapes. However, a basic understanding of the ambient noise environment in these heavily used areas is of importance for marine spatial planning and in order to assess the potential impacts of new developments. For example, collision risk between tidal turbines and marine mammals may depend on the detectability of these devices above ambient noise (Carter 2013; Wilson et al. 2014). 9.3.1.1  Remote Monitoring of Biotic Signals in the Environment

Several different types of nearshore underwater soundscapes have been characterized in recent years. Oyster reef environments have been shown to exhibit specific soundscape characteristics which are likely heavily influenced by benthic invertebrate species and soniferous fish and could act as settlement cues for planktonic larvae (Eggleston et al. 2016; Lillis et al. 2014a,b). Similarly, maerl beds can be characterized by sound‐ producing benthic species such as feeding sea urchins (Echinus esculentus), spider crabs (Maja brachydactyla) or snapping shrimp (Athanas nitescens) (Coquereau et al. 2016). Snapping shrimp sounds in the frequency range of 2–5 kHz in particular are characteristic for many inshore marine habitats worldwide, ranging from estuarine habitats off North Carolina and the Florida Keys to nearshore habitats off New Zealand or Irish sea lochs (Butler et al. 2016; McWilliam and Hawkins 2013; Pine et al. 2015; Radford et al. 2008; Ricci et al. 2016). An ecoacoustic study of fish communities in coastal waters off South Africa has recently highlighted the partitioning of the soundscape by different fish species particularly at night to avoid acoustic interference (Ruppé et al. 2015). Fish choruses are strongly influencing soundscapes in many nearshore habitats and can often be linked with a variety of environmental variables such as temperature or tidal and lunar cycles (Guan et al. 2015b; Rowell et al. 2012; Staaterman et al. 2014) (Figure 9.2). 9.3.1.2  Remote Monitoring of the Environment Using Sound in Estuarine and Coastal Habitats

In some shallow sea locations, ambient noise levels in low‐frequency bands (2–7 kHz) have been shown to be related to rainfall (Ashokan et al. 2015). However, in general, studies of ambient noise in coastal waters have revealed that sound levels vary greatly

(a)

Sand Island ACI

720

Frequency (Hz)

(b)

700 680 2000 1000

0 (c)

7

21

28

21

28

Pickles

760 ACI

14

740 720

Frequency (Hz)

700 (d)

2000 1000

0

7

14 Time (days)

Figure 9.2 (a,c) Acoustic Complexity Index (ACI) and spectrogram (FFT: 3509 pt, overlap: 50%) (b,d) for two different reef environments, showing the relationship between diurnal and lunar cycles and acoustic energy in the lower frequency bands over a period of one month. Source: Staaterman et al. (2014). Reproduced with permission of Marine Ecology Progress Series. (See color plate section for the color representation of this figure.)

152

Ecoacoustics

depending on location, season, and biodiversity. More variable sound speeds, complex bottom topography, and increased noise levels from anthropogenic activities mean that ambient noise levels in nearshore environments are generally less dependent on geophysical factors such as wind speed or rainfall than those in deeper open ocean environments (Mathias et  al. 2016; Ramji et  al. 2008). Coastal soundscapes are also heavily influenced by the dynamics of moving bodies of water and sometimes considerable variations in tidal flow (Guan et al. 2015b; Willis et al. 2013; Wilson et al. 2014). 9.3.1.3  Anthropogenic Sources of Noise in Estuarine and Coastal Habitats

Due to the multitude of anthropogenic pressures which concentrate along global coastlines, these habitats are also often prone to very high levels of noise pollution. One major contribution to ambient noise levels in many nearshore habitats originates from intense shipping traffic, which may vary seasonally if dominated by recreational traffic (Rako et al. 2012) or be more constant in areas close to harbours and major shipping lanes (Bittencourt et al. 2014; Codarin and Picciulin 2015; Erbe et al. 2014; Hatch et al. 2008). Other major contributors to coastal noise pollution include seismic surveys, construction and resource extraction activities, military exercises, and explosions (Estabrook et al. 2016; Hildebrand 2009). New industries such as offshore aquaculture and marine renewable energy may also heavily shape coastal soundscapes (Götz and Janik 2013; Madsen et al. 2006; Pine et al. 2014). The combination of all these different sources of noise has been shown to result in short‐ and long‐term consequences for different species at all ecosystem levels and will ultimately dramatically change coastal soundscapes (Hawkins and Popper 2014; Morley et al. 2013; Shannon et al. 2015; Slabbekoorn et al. 2010). 9.3.2  Coral Reefs

Coral reefs are the most biodiverse ecosystems of the world’s oceans, providing habitat for over 1 million species (Knowlton et al. 2010). However, global coral reefs are in serious decline due to a diverse array of factors, including overfishing, pollution, ocean acidification, and climate change (Bellwood et al. 2004; De’ath et al. 2012). This decline has serious consequences not only for reef‐associated biodiversity but also ecosystem services which the reefs provide, such as fisheries, recreation or coastal protection (Graham 2014; Graham and Nash 2013). Monitoring changes in coral reef ecosystems and associated biodiversity is thus of high priority in order to inform and drive conservation planning. Several ecological metrics have been used to assess reef condition on a local scale (Bellwood et al. 2004). Exploiting recent advances in technology, high‐resolution satellite imagery can provide a larger scale picture (Madin and Madin 2015), including the assessment of deepwater reefs which can be difficult to survey by other methods. In addition, many coral reef fish and invertebrates produce sounds which, together with habitat characteristics, create distinct sound signatures of the local reef environment. Ecoacoustics thus present another set of tools to survey biodiversity on coral reefs over longer timescales and at remote, deepwater sites and lower costs than traditional survey methods. Passive acoustic studies at a variety of different sites are starting to explore this opportunity (Harris et al. 2015; Kaplan et al. 2015; Nedelec et al. 2015). 9.3.2.1  Remote Monitoring of Biotic Signals in the Environment

Coral reef soundscapes are as diverse as the biological communities that inhabit them. They change on diurnal, lunar or seasonal scales, as a function of the acoustic activity

9  Biodiversity Assessment and Environmental Monitoring in Freshwater and Marine Biomes

of reef fishes and invertebrates (Bertucci et al. 2015; Kaplan et al. 2015; Nedelec et al. 2015). In many subtropical and tropical coral reef environments, these temporal changes in frequencies above 1–2 kHz are dominated by the short and impulsive sounds produced by snapping shrimp (Au et al. 2012; Bohnenstiehl et al. 2016; Lillis et al. 2014b). In the lower frequencies, coral reef fish assemblages influence soundscape variability (Kaplan et al. 2015; Kennedy et al. 2010; Staaterman et al. 2013). Several studies have shown these soundscapes to play an important part in larval recruitment of fishes and invertebrates to reef habitats (Montgomery et al. 2006; Piercy et al. 2016; Simpson et al. 2005, 2008). Thus, with its importance for population density and diversity, biological reef noise is also an important indicator of coral reef health. It has been shown that degraded reef habitats are significantly quieter and acoustically less diverse than healthy reefs, which may further affect their ability to attract settlement‐stage larvae and may further decrease reef quality (Piercy et  al. 2014). Ecoacoustics can be used to detect such changes in coral reef status and several authors have successfully correlated acoustic spatiotemporal variability with fish and benthic species density and diversity (Bertucci et al. 2015; Harris et al. 2015; Radford et al. 2014b; Tricas and Boyle 2014). 9.3.2.2  Remote Monitoring of the Environment Using Sound in Coral Reef Environments

Similar to other environments, wind‐generated noise has been shown to influence reef soundscapes (Nedelec et  al. 2015). However, in contrast to open water environments, weather‐related noise seems to be of relatively lesser importance than biological sounds in driving changes in coral reef ambient noise conditions (Cagua et al. 2013). At the same time, many biological processes are in turn driven by environmental factors. For example, spectral levels driven by snapping shrimp sounds are closely correlated to changes in temperature (Bohnenstiehl et al. 2016), which could thus be tracked indirectly by recording sound pressure levels in snapping shrimp‐dominated frequency bands. 9.3.2.3  Anthropogenic Sources of Noise in Coral Reef Environments

The importance of acoustic orientation for many coral reef fish and benthic species highlights their vulnerability to changes of natural reef soundscapes due to human impacts. Coral reef habitats may be impacted through general habitat degradation and associated reduced detection ranges of less diverse reef soundscapes (Piercy et al. 2014), as well as the presence of anthropogenic noise from local and distant boat traffic (Kaplan and Mooney 2015) and other noise sources such as distant seismic surveys. Boat noise has been shown to affect settlement behavior of coral reef fishes (Holles et  al. 2013; Simpson et al. 2016) which, coupled with habitat degradation, may have serious consequences for coral reef biodiversity.

9.4 ­Marine Oceanic Habitats 9.4.1  Open Ocean and Deep Sea Habitats

The open ocean and deep seafloor habitats, beyond the 200 m continental shelf depths, make up over 70% of the total surface area of our planet, but many of these areas remain relatively poorly studied due to their inaccessibility far from land, with extreme depths

153

154

Ecoacoustics

exceeding 10 000 m in some areas of the deep ocean. A large number of pelagic species, ranging from microscopic bacteria and planktonic species to the largest marine vertebrates, can be found in this realm but research has historically been limited to vessel‐ based direct surveys of small areas. With recent technological advances, tagging technology, autonomous vehicles, and passive acoustic recordings, remote sensing of biological activity in this environment is now possible (Eriksen et al. 2001; Block et al. 2011; McDonald and Fox 1999). A large number of these species produce sounds, and acoustic recordings have been used to detect, and even track, species or individuals over long periods of time or over large spatial scales (Mellinger et al. 2007a; Tyack and Clark 2000). 9.4.1.1  Remote Monitoring of Biotic Signals in the Environment

Given the enormous scale and relative inaccessibility of the open ocean and deep sea habitats, biodiversity surveys can be challenging. Indeed, recently a 10‐year international effort focused on cataloging biodiversity in the sea, the Census of Marine Life, attempted to increase our understanding of organisms in the ocean, leading to the discovery of numerous new species (Snelgrove 2010). Passive acoustic monitoring in the open ocean and deep sea has led to improved understanding of the seasonal distributions of a number of marine mammals (Mellinger et al. 2007b; Risch et al. 2014a), but relatively little is known regarding fish or invertebrate species in these habitats (Rountree et al. 2006; Wall et al. 2014b). To date, soundscape analyses with open ocean or deep sea recordings have been limited (Erbe et al. 2015; Parks et al. 2014), though it is expected that increasing information will become available as more datasets are examined. 9.4.1.2  Remote Monitoring of the Environment Using Sound in the Open Ocean

It has been evident from some of the earliest acoustic measurements of the open ocean environment that environmental factors, particularly wind‐driven waves and precipitation, can dominate some parts of the acoustic spectrum (Wenz 1962). Acoustic measurements can be used to accurately track wind speed in the open ocean (Shaw et al. 1978). Distinct acoustic patterns allow for measurements of precipitation events that are consistent with nearshore radar measurements (Nystuen and Farmer 1989), making it possible to monitor precipitation events in remote ocean waters where radar detection is not possible (Nystuen et al. 2000). 9.4.1.3  Anthropogenic Sources of Noise in the Open Ocean

Several long‐term monitoring studies have documented increases in the ambient low‐ frequency sound levels in the open ocean (Andrew et  al. 2011; Chapman and Price 2011). These increases have been tied to an increase in commercial shipping traffic (Frisk 2012; Hildebrand 2009). Another major source of human‐generated sounds in the open ocean is seismic geophysical surveys, which can be detected from thousands of miles away (Nieukirk et al. 2004, 2012), and can dominate the acoustic soundscapes at low frequencies in some regions (Parks et al. 2014). Other more regional or occasional sources of noise in the open ocean can be linked to military and commercial sonar systems, which are generally higher in frequency (Hildebrand 2009). Several studies have explored the effects of different types of human signals on marine life (Nowacek et al. 2007; Slabbekoorn et al. 2010).

9  Biodiversity Assessment and Environmental Monitoring in Freshwater and Marine Biomes

9.4.2  Polar Oceans

Polar marine environments, in both the Arctic and Antarctic, are full of a cacophony of sound. In both habitats, the seasonal absence of sunlight and related variations in ice coverage have led to complex and highly variable marine acoustic environments. Acoustic soundscape measurements have been used to monitor the environment in polar regions, with specific information relating to ice cover, wind, and precipitation. Further work through acoustic recordings can be used to assess the biological activity and distribution of acoustically active marine species, that are challenging or impossible to detect through more traditional visual survey methods in the harsh polar habitats. Acoustic soundscape monitoring has also focused on the advent and introduction of human sounds into even these remote marine environments, with seismic exploration, oil and gas construction, and increasing vessel traffic leading to man‐made changes in the ambient soundscapes. Given rapid climate change, particularly in polar regions, rapid changes in the seasonality of ice cover are anticipated to have a cascading effect on seasonality and distribution of marine life in these habitats, as well as leading to increased human use of previously ice‐covered habitats (Hoegh‐Guldberg and Bruno 2010). 9.4.2.1  Remote Monitoring of Biotic Signals in the Environment

Given the remote, harsh conditions in many polar habitats, access to visually assess species distribution and abundance is limited. Therefore, the utility of acoustic recordings to remotely assess biodiversity in polar habitats was realized over 40 years ago. Early records identified a diversity of biological sounds from polar marine species in underwater acoustic recordings (Cummings et al. 1989; Kibblewhite and Jones 1976). The biological signals in both the Arctic and Antarctic seasonally dominate the soundscape, due to the acoustic reproductive advertisement signals of both cetaceans and pinnipeds (Kibblewhite and Jones 1976). Acoustic monitoring capabilities in the remote, polar environments have increased from short‐term deployments spanning multiple days to year‐round monitoring through subsampling recorders in the Arctic (Miksis‐ Olds et  al. 2010) or cabled observatories such as PALAOA (PerreniAL Acoustic Observatory in the Antarctic) (Boebel et al. 2008; Klinck et al. 2016). Acoustic recording tools are now applied extensively to long‐term monitoring of the seasonal presence of marine species in polar regions (Mellinger et  al. 2007a), including pinnipeds (Van Opzeeland et al. 2010; Van Parijs et al. 2009) and cetaceans (Miksis‐Olds et al. 2010; Širović et al. 2004). Long‐term acoustic monitoring can be used for ecosystem and biodiversity assessment in remote polar marine habitats, yet only limited information directly related to ecosystem biodiversity has been reported for these habitats using acoustic data. 9.4.2.2  Remote Monitoring of the Environment with Sound in Polar Regions

The earliest marine soundscape monitoring work in the polar marine regions focused on detecting and quantifying abiotic environmental parameters. Ice, in particular, creates a wide variety of acoustic signals, ranging from dampening effects in extremely heavy, stable ice coverage, leading to noise levels below Knudsen’s sea state zero (Macpherson 1962; Kibblewhite and Jones 1976), to screeching high‐pitched tonal signals for shifting and moving platforms of ice (Kinda 2013; Milne and Ganton 1964).

155

Ecoacoustics

The sounds of ice are highly variable, with variations due to external weather ­conditions, including wind, precipitation, and temperature (Cummings et  al. 1989; Ganton and Milne 1965). The sounds of icebergs calving, colliding, and rubbing are extremely loud (213–245 dB re μPa p‐p at 1 m) and can be detected at long distances in the southern hemisphere, several thousand kilometers away from their source (Royer et  al. 2015). Acoustic recordings can provide fine‐scale direct evidence for ice coverage in specific locations in the Arctic. The melting of sea ice, either seasonally (Kinda 2013) or during short‐term partial retreats, results in increased ambient background noise conditions due to wind effects on water (Miksis‐Olds et al. 2013). 9.4.2.3  Anthropogenic Sources of Noise in the Polar Regions

There is urgent need to establish permanent long‐term acoustic monitoring systems to aid in using soundscape analyses to assess climate change impacts including abiotic changes in ice cover and precipitation; changes in the distribution and biodiversity of the marine biological ecosystem; and anticipated increased human use of this habitat (Mikhalevsky et al. 2015). Several studies in the Arctic have documented anthropogenic noise from oil and gas exploration, construction, and shipping activities and their potential impact on marine mammal species (Ljungblad et al. 1988; Richardson et al. 1995). Long‐term acoustic recordings spanning the migratory range of different species are providing year‐round perspectives of the acoustic environment in the Arctic (Clark et al. 2015). Some of these long‐term studies have detected high levels of anthropogenic noise from seismic surveys, that can dominate the soundscapes when they are present (Guan et al. 2015a; Klinck et al. 2012) (Figure 9.3).

9.5 ­Summary and Future Directions This chapter has highlighted the many applications of traditional underwater acoustics as well as newly developed ecoacoustic approaches to monitor aquatic soundscapes. One of the most evident advantages of ecoacoustics in the aquatic realm is the potential for SPL (dB re 1 μPa)

July 30, 2012 160 140 120 100 80 12 frequency (kHz)

156

~ 2 km 13.50

14.00

14.10

10

80

8

60

6

40

4

20

2 0

100

13.50:00

14.00:00

time

14.10:00

0

Figure 9.3  Example spectrogram and sound pressure levels (SPLs) during seismic airgun activity (640 in3 airgun array firing at a rate of 60 s before 14:00:00 on July 30th 2012 and at a rate of 10 s after 14:00:00 on the same day) in polar waters, measured at about 2 km distance from source. Vertical scale bar to the right of the spectrogram represents SPL in dB re 1 μPa2/Hz. Source: Guan et al. (2015a). Reproduced with permission of AIP Scitation. (See color plate section for the color representation of this figure.)

9  Biodiversity Assessment and Environmental Monitoring in Freshwater and Marine Biomes

remote sensing in an environment that is not amenable to sampling through some of the currently applied tools in terrestrial ecology, such as Light Detection and Ranging (LiDAR) technology (Diaz et al. 2013; Pettorelli et al. 2014). Ecoacoustics may provide a comparable window into ecosystem‐level changes, broadening the current focus on single species acoustic monitoring to a more holistic view of the aquatic environment. Aquatic ecoacoustics is a field that is expanding rapidly and developing new technologies, including gliders and real‐time passive acoustic buoys (Van Parijs et al. 2015), as well as analytical approaches which may include documentation of the addition or loss of single species, through the assessment of the entire sound‐producing biosphere, depending on the metrics used (Depraetere et al. 2012; Pieretti et al. 2011; Sueur et al. 2008). Ecoacoustics can also assist in monitoring changes in abiotic environmental factors, including precipitation and wind events. Such integrated, landscape‐scale ecological monitoring concepts using acoustic methods are much further along in the terrestrial realm, where landscape ecology has long been recognized as an independent field of study (Wu and Hobbs 2002). However, the value of acoustics for large‐scale ecosystem monitoring is increasingly being recognized in aquatic ecology as well. This chapter also brought to light the many uses of aquatic ecoacoustics for answering questions of habitat use, connectivity, resource partitioning, as well as animal movement, distribution, and relative abundance. Carried out over the long term, acoustic monitoring can be used to track changes in near real time, potentially allowing for the identification of areas undergoing rapid degradation that would be missed through traditional sampling methods (Gage and Axel 2013). By recording changes in vocalization patterns or spectral properties of a soundscape in relation to changing environmental parameters, such as elevated CO2 levels as a result of ocean acidification (Rossi et al. 2016) or harmful algae blooms (Wall et al. 2014a), ecosystem health can be monitored. In addition, long‐ term acoustic monitoring of critical habitat, including breeding or spawning grounds, migration routes or reef habitats, may reveal changes in important ecological processes such as predator–prey relationships, larvae settlement or reproductive behavior as a result of environmental change. If the main contributors to a particular soundscape are known, ecoacoustic approaches may therefore allow a quicker appreciation of possible changes in the overall health status of aquatic ecosystems. Ecoacoustics also provide a way to quantify anthropogenic activity over long timescales and large geographic areas in order to assess the effects of noise on spatiotemporal relevant scales (Van Parijs et al. 2009). It is now widely recognized that in order to fully understand the effects of noise on aquatic ecosystems, we need to move away from single source assessments and investigate the cumulative impacts of noise (Ellison et al. 2016). Longer term ecoacoustic studies provide the means to do this and also allow the assessment of chronic noise pollution and the development of methods for monitoring population‐ and ecosystem‐scale responses to anthropogenic underwater noise. Such ecosystem‐scale approaches to aquatic soundscape monitoring are also increasingly sought by international regulators dealing with underwater noise. In Europe, Descriptor 11 (emission of energy, including underwater noise) of the Marine Strategy Framework Directive (MSFD) requires member states to report long‐term trends in underwater spectrum levels with the aim of ensuring that the collective introduction of noise by all ongoing activities is kept within levels to ensure Good Environmental Status (GES) (Maccarrone et al. 2015). Similarly, the SoundMap Project of the National Oceanic and Atmospheric Administration (NOAA) is developing tools to predict the contribution of

157

158

Ecoacoustics

human activities to underwater noise in the US Exclusive Economic Zone (EEZ) (Gedamke et al. 2016). Although similar regulatory initiatives are currently missing for the introduction of underwater noise in freshwater biomes, it is clear that ecoacoustics can play a major role in assessing the impacts of noise and other human activity on aquatic biodiversity and highlight meaningful conservation measures. Similar to terrestrial ecoacoustics, studies of aquatic soundscapes may play a central role in assessing biodiversity and monitoring changes over time and space. However, there are some notable gaps in the development of aquatic ecoacoustics. Most striking is the difference in ecoacoustic studies carried out in marine compared to freshwater habitats. Studies of acoustic communication, biodiversity assessment or effects of noise in rivers and lakes have so far largely been neglected (Gammell and O’Brien 2013). Future studies in aquatic ecoacoustics should include these environments, in particular in the context of noise impact assessments. Another bias exists in the taxonomic groups currently being studied acoustically, with most studies carried out on marine mammals, while studies on amphibians, fish, and invertebrates are generally underrepresented in the literature (Shannon et al. 2015, Williams et al. 2015). Future studies of aquatic ecoacoustics should aim to fill these gaps and continue the development of analytical tools for the assessment of ecosystem‐scale changes in biodiversity using acoustics.

­References Agardy, T, Aguilar, N, Cañadas, A, et al. (2007) A Global Scientific Workshop on Spatio‐ Temporal Management of Noise. Available at: www.cetaceanhabitat.org/pdf_bin/ dokumentedesmeeres_report_eng_final_amended_v2.pdf (accessed 15 December 2016). Aiken, RB (1985) Sound production by aquatic insects. Biological Reviews, 60, 163–211. Amorim, MCP, Vasconcelos, RO and Fonseca, PJ (2015) Fish sounds and mate choice, in Sound Communication in Fishes, Springer, New York, pp.1–33. Amoser, S, Wysocki, LE and Ladich, F (2004) Noise emission during the first powerboat race in an Alpine lake and potential impact on fish communities. Journal of the Acoustical Society of America, 116, 3789. Andrew, RK, Howe, BM and Mercer, J (2011) Long‐time trends in ship traffic noise for four sites off the North American West Coast. Journal of the Acoustical Society of America, 129, 642–651. Ashokan, M, Latha, G and Ramesh, R (2015) Analysis of shallow water ambient noise due to rain and derivation of rain parameters. Applied Acoustics, 88, 114–122. Au, WWL, Richlen, M and Lammers, MO (2012) Soundscape of a nearshore coral reef near an urban center, in Advances in Experimental Medicine and Biology, Springer New York, pp.345–351. Bellwood, DR, Hughes, TP, Folke, C and Nyström, M (2004) Confronting the coral reef crisis. Nature, 429, 827–833. Bertucci, F, Parmentier, E, Berten, L, Brooker, RM and Lecchini, D (2015) Temporal and spatial comparisons of underwater sound signatures of different reef habitats in Moorea Island, French Polynesia. PLoS One, 10, e0135733–e0135733.

9  Biodiversity Assessment and Environmental Monitoring in Freshwater and Marine Biomes

Bevelhimer, MS, Deng, ZD and Scherelis, C (2016) Characterizing large river sounds: providing context for understanding the environmental effects of noise produced by hydrokinetic turbines. Journal of the Acoustical Society of America, 139, 85–92. Bittencourt, L, Carvalho, RR, Lailson‐Brito, J and Azevedo, AF (2014) Underwater noise pollution in a coastal tropical environment. Marine Pollution Bulletin, 83(1), 331–336. Bleach, IT, Beckmann, C, Both, C, Brown, GP and Shine, R (2015) Noisy neighbours at the frog pond: effects of invasive cane toads on the calling behaviour of native Australian frogs. Behavioral Ecology and Sociobiology, 69, 675–683. Blickley, JL and Patricelli, GL (2010) Impacts of anthropogenic noise on wildlife: research priorities for the development of standards and mitigation. Journal of International Wildlife Law and Policy, 13, 274–292. Block, BA, Jonsen, ID, Jorgensen, SJ, et al. (2011) Tracking apex marine predator movements in a dynamic ocean. Nature, 475, 86–90. Boebel, O, Klinck, H, Kindermann, L and El Naggar, S (2008) PALAOA: broadband recordings of the Antarctic coastal soundscape. Bioacoustics, 17, 18–21. Bohnenstiehl, DR, Lillis, A and Eggleston, DB (2016) The curious acoustic behavior of estuarine snapping shrimp: temporal patterns of snapping shrimp sound in sub‐tidal oyster reef habitat. PLoS One, 11, e0143691–e0143691. Bolgan, M, Chorazyczewska, E, Winfield, IJ, Codarin, A, O’Brien, J, and Gammell, M (2016) First observations of anthropogenic underwater noise in a large multi‐use lake. Journal of Limnology, 75(3), DOI: org/10.4081/jlimnol.2016.1405. Butler, J, Stanley, JA and Butler, MJ (2016) Underwater soundscapes in near‐shore tropical habitats and the effects of environmental degradation and habitat restoration. Journal of Experimental Marine Biology and Ecology, 479, 89–96. Cagua, EF, Berumen, ML and Tyler, EHM (2013) Topography and biological noise determine acoustic detectability on coral reefs. Coral Reefs, 32, 1123–1134. Carter, C (2013) Tidal Energy, Underwater Noise and Marine Mammals, PhD Thesis, University of Aberdeen. Chapman, NR and Price A (2011) Low frequency deep ocean ambient noise trend in the Northeast Pacific Ocean. Journal of the Acoustical Society of America, 129, EL161–5. Clark, CW, Ellison, WT, Southall, BL, et al. (2009) Acoustic masking in marine ecosystems: intuitions, analysis, and implication. Marine Ecology Progress Series, 395, 201–222. Clark, CW, Berchok, CL, Blackwell, SB, et al. (2015) A year in the acoustic world of bowhead whales in the Bering, Chukchi and Beaufort seas. Progress in Oceanography, 136, 223–240. Cloern, JE, Abreu, PC, Carstensen, J, et al. (2016) Human activities and climate variability drive fast‐paced change across the worlds estuarine–coastal ecosystems. Global Change Biology, 22, 513–529. Codarin, A and Picciulin, M (2015) Underwater noise assessment in the Gulf of Trieste (Northern Adriatic Sea, Italy) using an MSFD approach. Marine Pollution Bulletin, 101, 694–700. Coquereau, L, Grall, J, Chauvaud, L, et al. (2016) Sound production and associated behaviours of benthic invertebrates from a coastal habitat in the north‐east Atlantic. Marine Biology, 163, 1–13. Cornell, HV and Harrison, SP (2013) Regional effects as important determinants of local diversity in both marine and terrestrial systems. Oikos, 122, 288–297.

159

160

Ecoacoustics

Cummings, WC, Diachok, OI and Shaffer, JD (1989) Acoustic Transients of the Marginal Sea Ice Zone: A Provisional Catalog, Naval Research Lab, Washington, DC. De’ath, G, Fabricius, KE, Sweatman, H and Puotinen, M (2012) The 27‐year decline of coral cover on the Great Barrier Reef and its causes. Proceedings of the National Academy of Sciences USA, 109, 17995–17999. Depraetere, M, Pavoine, S, Jiguet, F, Gasc, A, Duvail, S and Sueur, J (2012) Monitoring animal diversity using acoustic indices: implementation in a temperate woodland. Ecological Indicators, 13, 46–54. Desjonquères, C, Rybak, F, Depraetere, M, et al. (2015) First description of underwater acoustic diversity in three temperate ponds. Peer Journal, 3, e1393–e1393. Diaz, JCF, Carter, WE, Shrestha, RL and Glennie, CL (2013) Lidar remote sensing, in Handbook of Satellite Applications, Springer, New York, pp.757–808. Dudgeon, D, Arthington, AH, Gessner, MO, et al. (2006) Freshwater biodiversity: importance, threats, status and conservation challenges. Biological Reviews, 81, 163–182. Duffy, JE, Amaral‐Zettler, LA, Fautin, DG, et al. (2013) Envisioning a marine biodiversity observation network. Bioscience, 63, 350–361. Eggleston, DB, Lillis, A and Bohnenstiehl, DR (2016) Soundscapes and larval settlement: larval bivalve responses to habitat‐associated underwater sounds, in Advances in Experimental Medicine and Biology, Springer, New York, pp.255–263. Elliott, M, Borja, Á, McQuatters‐Gollop, A, et al. (2015) Force majeure: will climate change affect our ability to attain Good Environmental Status for marine biodiversity? Marine Pollution Bulletin, 95, 7–27. Ellison, WT, Racca, R, Clark, CW, et al. (2016) Modeling the aggregated exposure and responses of bowhead whales Balaena mysticetus to multiple sources of anthropogenic underwater sound. Endangered Species Research, 30, 95–108. Erbe, C, Williams, R, Sandilands, D and Ashe, E (2014) Identifying modeled ship noise hotspots for marine mammals of Canada’s Pacific Region. PLoS One, 9, e89820–e89820. Erbe, C, Verma, A, McCauley, R, Gavrilov, A and Parnum, I (2015) The marine soundscape of the Perth Canyon. Progress in Oceanography, 137(A), 38–51. Eriksen, CC, Osse, TJ, Light, RD, et al. (2001) Seaglider: a long‐range autonomous underwater vehicle for oceanographic research. IEEE Journal of Oceanic Engineering, 26, 424–436. Estabrook, BJ, Ponirakis, DW, Clark, CW and Rice, AN (2016) Widespread spatial and temporal extent of anthropogenic noise across the northeastern Gulf of Mexico shelf ecosystem. Endangered Species Research, 30, 267–282. Frisk, GV (2012) Noiseonomics: the relationship between ambient noise levels in the sea and global economic trends. Scientific Reports, 2, 437. Gage, SH and Axel, AC (2013) Visualization of temporal change in soundscape power of a Michigan lake habitat over a 4‐year period. Ecological Informatics, 2, 100–109. Gammell, MP and O’Brien, JM (2013) Acoustic communication in aquatic animals: all quiet on the freshwater front? Aquatic Conservation: Marine and Freshwater Ecosystems, 23, 363–365. Ganton, JH and Milne, AR (1965) Temperature‐and wind‐dependent ambient noise under midwinter pack ice. Journal of the Acoustical Society of America, 38, 406–411. García Molinos, J, Halpern, BS, Schoeman, DS, et al. (2016) Climate velocity and the future global redistribution of marine biodiversity. Nature Climate Change, 6, 83–88.

9  Biodiversity Assessment and Environmental Monitoring in Freshwater and Marine Biomes

Gasc, A, Sueur, J, Jiguet, F, et al. (2013) Assessing biodiversity with sound: do acoustic diversity indices reflect phylogenetic and functional diversities of bird communities? Ecological Indicators, 25, 279–287. Gedamke, J, Ferguson, M, Harrison, J, et al. (2016) Predicting anthropogenic noise contributions to US waters, in Advances in Experimental Medicine and Biology, Springer, New York, pp.341–347. Gleick, PH (1993) Water in Crisis: A Guide to the World’s Fresh Water Resources, Oxford University Press, New York. Götz, T and Janik, VM (2013) Acoustic deterrent devices to prevent pinniped depredation: efficiency, conservation concerns and possible solutions. Marine Ecology Progress Series, 492, 285–302. Graham, NAJ (2014) Habitat complexity: coral structural loss leads to fisheries declines. Current Biology, 24, R359–R361. Graham, NAJ and Nash, KL (2013) The importance of structural complexity in coral reef ecosystems. Coral Reefs, 32, 315–326. Guan, S, Vignola, J, Judge, J and Turo, D (2015a) Airgun inter‐pulse noise field during a seismic survey in an Arctic ultra shallow marine environment. Journal of the Acoustical Society of America, 138, 3447–3457. Guan, S, Lin, TH, Chou, LS, Vignola, J, Judge, J and Turo, D (2015b) Dynamics of soundscape in a shallow water marine environment: a study of the habitat of the Indo‐Pacific humpback dolphin. Journal of the Acoustical Society of America, 137, 2939–2949. Halpern, BS, Walbridge, S, Selkoe, K, et al. (2008) A global map of human impact on marine ecosystems. Science, 319, 948–952. Harris, SA, Shears, NT and Radford, CA (2015) Ecoacoustic indices as proxies for biodiversity on temperate reef. Methods in Ecology and Evolution, 7(6), 713–724. Hatch, L, Clark, C, Merrick, R, eet al. (2008) Characterizing the relative contributions of large vessels to total ocean noise fields: a case study using the Gerry E. Studds Stellwagen Bank National Marine Sanctuary. Environmental Management, 42, 735–752. Hatch, L and Fristrup, K (2009) No barrier at the boundaries: implementing regional frameworks for noise management in protected natural areas. Marine Ecology Progress Series, 395, 223–244. Hawkins, AD and Popper, AN (2014) Assessing the impacts of underwater sounds on fishes and other forms of marine life. Acoustics Today, 10, 30–41. Hildebrand, JA (2009) Anthropogenic and natural sources of ambient noise in the ocean. Marine Ecology Progress Series, 395, 5–20. Hoegh‐Guldberg, O and Bruno, JF (2010) The impact of climate change on the world’s marine ecosystems. Science, 328, 1523–1528. Holles, S, Simpson, SD, Radford, AN, Berten, L and Lecchini, D (2013) Boat noise disrupts orientation behaviour in a coral reef fish. Marine Ecology Progress Series, 485, 295–300. Holt, DE and Johnston, CE (2015) Traffic noise masks acoustic signals of freshwater stream fish. Biological Conservation, 187, 27–33. Jackson, RB, Carpenter, SR, Dahm, CN, et al. (2001) Water in a changing world. Ecological Applications, 11, 1027–1045. Jacobsen, L, Baktoft, H, Jepsen, N, Aarestrup, K, Berg, S and Skov, C (2014) Effect of boat noise and angling on lake fish behaviour. Journal of Fish Biology, 84, 1768–1780.

161

162

Ecoacoustics

James, MS, Stockwell, MP, Clulow, J, Clulow, S and Mahony, MJ (2015) Investigating behaviour for conservation goals: conspecific call playback can be used to alter amphibian distributions within ponds. Biological Conservation, 192, 287–293. Jensen, FH, Rocco, A, Mansur, RM, Smith, BD, Janik, VM and Madsen, PT (2013) Clicking in shallow rivers: short‐range echolocation of Irrawaddy and Ganges River dolphins in a shallow, acoustically complex habitat. PLoS One, 8, e59284–e59284. Kaplan, MB and Mooney, TA (2015) Ambient noise and temporal patterns of boat activity in the US Virgin Islands National Park. Marine Pollution Bulletin, 98(1‐2), 221–228. Kaplan, MB, Mooney, TA, Partan, J and Solow, AR (2015) Coral reef species assemblages are associated with ambient soundscapes. Marine Ecology Progress Series, 533, 93–107. Kennedy, EV, Holderied, MW, Mair, JM, Guzman, HM and Simpson, SD (2010) Spatial patterns in reef‐generated noise relate to habitats and communities: evidence from a Panamanian case study. Journal of Experimental Marine Biology and Ecology, 395, 85–92. Kibblewhite, AC and Jones, DA (1976) Ambient noise under Antarctic sea ice. Journal of the Acoustical Society of America, 59, 790–798. Kinda, B (2013) Acoustic remote sensing of Arctic Sea Ice from long term soundscape measurements. Signal and Image processing. PhD thesis, University de Grenoble. Klinck, H, Nieukirk, SL, Mellinger, DK, Klinck, K, Matsumoto, H and Dziak, RP (2012) Seasonal presence of cetaceans and ambient noise levels in polar waters of the North Atlantic. Journal of the Acoustical Society of America, 132, EL176–EL176. Klinck, H, Kindermann, L and Boebel, O (2016) PALAOA: The Perennial Acoustic Observatory in the Antarctic Ocean – real‐time eavesdropping on the Antarctic underwater soundscape, in Listening in the Ocean, Springer, New York, pp.207–219. Knowlton, N, Brainard, RE, Fisher, R, Moews, M, Plaisance, L and Caley, MJ (2010) Coral reef biodiversity, in Life in the World’s Oceans, Wiley‐Blackwell, Chichester, pp.65–78. Krause, B (2016) Wild Soundscapes: Discovering the Voice of the Natural World, Yale University Press, New Haven. Kuehne, LM, Padgham, BL and Olden, JD (2013) The soundscapes of lakes across an urbanization gradient. PLoS One, 8, e55661–e55661. Kyhn, LA, Tougaard, J, Beedholm, K, et al. (2013) Clicking in a killer whale habitat: narrow‐band, high‐frequency biosonar clicks of harbour porpoise (Phocoena phocoena) and Dall’s porpoise (Phocoenoides dalli). PLoS One, 8, e63763–e63763. Ladegaard, M, Jensen, FH, de Freitas, M, da Silva, VMF and Madsen, PT (2015) Amazon river dolphins (Inia geoffrensis) use a high‐frequency short‐range biosonar. Journal of Experimental Biology, 218, 3091–3101. Lillis, A, Eggleston, DB and Bohnenstiehl, DR (2014a) Soundscape variation from a larval perspective: the case for habitat‐associated sound as a settlement cue for weakly swimming estuarine larvae. Marine Ecology Progress Series, 509, 57–70. Lillis, A, Eggleston, DB and Bohnenstiehl, DR (2014b) Estuarine soundscapes: distinct acoustic characteristics of oyster reefs compared to soft‐bottom habitats. Marine Ecology Progress Series, 505, 1–17. Ljungblad, DK, Würsig, B, Swartz, SL and Keene, JM (1988) Observations on the behavioral responses of bowhead whales (Balaena mysticetus) to active geophysical vessels in the Alaskan Beaufort Sea. Arctic, 41(3), 183–194.

9  Biodiversity Assessment and Environmental Monitoring in Freshwater and Marine Biomes

Lorang, MS, and Tonolla, D (2014) Combining active and passive hydroacoustic techniques during flood events for rapid spatial mapping of bedload transport patterns in gravel‐ bed rivers. Fundamental Applied Limnology, 184, 231–246. Lugli, M (2010) Sounds of shallow water fishes pitch within the quiet window of the habitat ambient noise. Journal of Comparative Physiology A, 196, 439–451. Lugli, M (2015) Habitat acoustics and the low‐frequency communication of shallow water fishes, in Animal Signals and Communication, Springer, Vienna, pp.175–206. Lugli, M and Fine, ML (2007) Stream ambient noise, spectrum and propagation of sounds in the goby Padogobius martensii: sound pressure and particle velocity. Journal of the Acoustical Society of America, 122, 2881–2892. Lurton, X (2002) An Introduction to Underwater Acoustics: Principles And Applications, Springer Science and Business Media, New York. Maccarrone, V, Filiciotto, F, de Vincenzi, G, Mazzola, S and Buscaino, G (2015) An Italian proposal on the monitoring of underwater noise: relationship between the EU Marine Strategy Framework Directive (MSFD) and marine spatial planning directive (MSP). Ocean and Coastal Management, 118, 215–224. Macpherson, JD (1962) Some under‐ice acoustic ambient noise measurements. Journal of the Acoustical Society of America, 34, 1149–1150. Madin, JS and Madin, EMP (2015) The full extent of the global coral reef crisis. Conservation Biology, 29, 1724–1726. Madsen, PT, Wahlberg, M, Tougaard, J, Lucke, K and Tyack, P (2006) Wind turbine underwater noise and marine mammals: implications of current knowledge and data needs. Marine Ecology Progress Series, 30, 279–295. Martin, SB and Cott, PA (2015) The under‐ice soundscape in Great Slave Lake near the city of Yellowknife, Northwest Territories, Canada. Journal of Great Lakes Research, 42(2), 248–255. Martin, SB and Popper, AN (2016) Short‐and long‐term monitoring of underwater sound levels in the Hudson River (New York, USA). Journal of the Acoustical Society of America, 139, 1886–1897. Maruska, KP, Ung, US and Fernald, RD (2012) The African cichlid fish Astatotilapia burtoni uses acoustic communication for reproduction: sound production, hearing, and behavioral significance. PLoS One, 7, e37612. Mathias, D, Gervaise, C and di Iorio, L (2016) Wind dependence of ambient noise in a biologically rich coastal area. Journal of the Acoustical Society of America, 139, 839–850. McDonald, MA and Fox, CG (1999) Passive acoustic methods applied to fin whale population density estimation. Journal of the Acoustical Society of America, 105, 2643. McWilliam, JN and Hawkins, AD (2013) A comparison of inshore marine soundscapes. Journal of Experimental Marine Biology and Ecology, 446, 166–176. Mellinger, DK, Nieukirk, SL, Matsumoto, H, et al. (2007a) Seasonal occurrence of North Atlantic right whale (Eubalaena glacialis) vocalizations at two sites on the Scotian Shelf. Marine Mammal Science, 23, 856–867. Mellinger, DK, Stafford, KM, Moore, SE, Dziak, RP and Matsumoto, H (2007b) An overview of fixed passive acoustic observation methods for cetaceans. Oceanography, 20, 36–45. Merchant, ND, Pirotta, E, Barton, TR and Thompson, PM (2014) Monitoring ship noise to assess the impact of coastal developments on marine mammals. Marine Pollution Bulletin, 78, 85–95.

163

164

Ecoacoustics

Mikhalevsky, PN, Sagen, H, Worcester, PF, et al. (2015) Multipurpose acoustic networks in the Integrated Arctic Ocean Observing System. Arctic, 68, 17. Miksis‐Olds, JL, Nystuen, JA and Parks, SE (2010) Detecting marine mammals with an adaptive sub‐sampling recorder in the Bering Sea. Applied Acoustics, 71, 1087–1092. Miksis‐Olds, JL, Stabeno, PJ, Napp, JM, et al. (2013) Ecosystem response to a temporary sea ice retreat in the Bering Sea: Winter 2009. Progress in Oceanography, 111, 38–51. Milne, AR and Ganton, JH (1964) Temperature‐ and wind‐dependent ambient noise under midwinter pack ice. Journal of the Acoustical Society of America, 36, 1994. Montgomery, JC, Jeffs, A, Simpson, SD, Meekan, M and Tindle, C (2006) Sound as an orientation cue for the pelagic larvae of reef fishes and decapod crustaceans. Advances in Marine Biology, 51, 143–196. Morley, EL, Jones, G and Radford, AN (2013) The importance of invertebrates when considering the impacts of anthropogenic noise. Proceedings of the Royal Society B, 281, 20132683. Nedelec, SL, Simpson, SD, Holderied, M, et al. (2015) Soundscapes and living communities in coral reefs: temporal and spatial variation. Marine Ecology Progress Series, 524, 125–135. Nieukirk, SL, Stafford, KM, Mellinger, DK, Dziak, RP and Fox, CG (2004) Low‐frequency whale and seismic airgun sounds recorded in the mid‐Atlantic Ocean. Journal of the Acoustical Society of America, 115, 1832–1843. Nieukirk, SL, Mellinger, DK, Moore, SE, Klinck, K, Dziak, RP and Goslin, J (2012) Sounds from airguns and fin whales recorded in the mid‐Atlantic Ocean, 1999–2009. Journal of the Acoustical Society of America, 131, 1102–1112. Nowacek, DP, Thorne, LH, Johnston, DW and Tyack, PL (2007) Responses of cetaceans to anthropogenic noise. Mammal Review, 37, 81–115. Nystuen, JA and Farmer, DM (1989) Precipitation in the Canadian Atlantic Storms Program: measurements of the acoustic signature. Atmosphere‐Ocean, 27, 237–257. Nystuen, JA, McPhaden, MJ and Freitag, HP (2000) Surface measurements of precipitation from an ocean mooring: the underwater acoustic log from the South China Sea. Journal of Applied Meteorology, 39, 2182–2197. Parks, SE, Johnson, M, Nowacek, D and Tyack, PL (2011) Individual right whales call louder in increased environmental noise. Biology Letters, 7, 33–35. Parks, SE, Miksis‐Olds, JL and Denes, SL (2014) Assessing marine ecosystem acoustic diversity across ocean basins. Ecological Informatics, 21, 81–88. Pettorelli, N, Laurance, WF, O’Brien, TG, Wegmann, M, Nagendra, H and Turner, W (2014) Satellite remote sensing for applied ecologists: opportunities and challenges. Journal of Applied Ecology, 51, 839–848. Piercy, JJB, Codling, EA, Hill, AJ, Smith, DJ and Simpson, SD (2014) Habitat quality affects sound production and likely distance of detection on coral reefs. Marine Ecology Progress Series, 516, 35–47. Piercy, JJB, Smith, DJ, Codling, EA, Hill, AJ and Simpson, SD (2016) The good, the bad, and the distant: soundscape cues for larval fish, in Advances in Experimental Medicine and Biology, Springer, New York, pp.829–837. Pieretti, N, Farina, A and Morri, D (2011) A new methodology to infer the singing activity of an avian community: the Acoustic Complexity Index (ACI). Ecological Indicators, 11, 868–873.

9  Biodiversity Assessment and Environmental Monitoring in Freshwater and Marine Biomes

Pine, MK, Jeffs, AG and Radford, CA (2014) The cumulative effect on sound levels from multiple underwater anthropogenic sound sources in shallow coastal waters. Journal of Applied Ecology, 51, 23–30. Pine, MK, Radford, CA and Jeffs, AG (2015) Eavesdropping on the Kaipara Harbour: characterising underwater soundscapes within a seagrass bed and a subtidal mudflat. New Zealand Journal of Marine and Freshwater Research, 49, 247–258. Pisanski, K, Marsh‐Rollo, SE and Balshine, S (2015) Courting and fighting quietly: a lack of acoustic signals in a cooperative Tanganyikan cichlid fish. Hydrobiologia, 748, 87–97. Platz, JE (1993) Rana subaquavocalis, a remarkable new species of leopard frog (Rana pipiens Complex) from southeastern Arizona that calls under water. Journal of Herpetology, 27(2), 154–162. Postel, SL (2000) Entering an era of water scarcity: the challenges ahead. Ecological Applications, 10, 941–948. Powell, JR and Ohman, MD (2015) Changes in zooplankton habitat, behavior, and acoustic scattering characteristics across glider‐resolved fronts in the Southern California Current System. Progress in Oceanography, 134, 77–92. Pracheil, BM, McIntyre, PB and Lyons, JD (2013) Enhancing conservation of large‐river biodiversity by accounting for tributaries. Frontiers in Ecology and the Environment, 11, 124–128. Protázio, A dos S, Albuquerque, RL, Falkenberg, LM and Mesquita, DO (2014) Acoustic ecology of an anuran assemblage in the arid Caatinga of northeastern Brazil. Journal of Natural History, 49, 957–976. Radford, CA, Jeffs, AG, Tindle, CT and Montgomery, JC (2008) Temporal patterns in ambient noise of biological origin from a shallow water temperate reef. Oecologia, 156, 921–929. Radford, AN, Kerridge, E and Simpson, SD (2014a) Acoustic communication in a noisy world: can fish compete with anthropogenic noise? Behavioral Ecology, 25(5), 1022–1030. Radford, CA, Stanley, JA and Jeffs, AG (2014b) Adjacent coral reef habitats produce different underwater sound signatures. Marine Ecology Progress Series, 505, 19–28. Rako, N, Picciulin, M, Vilibić, I and Fortuna, CM (2012) Spatial and temporal variability of sea ambient noise as an anthropogenic pressure index: the case of the Cres‐Lošinj archipelago, Croatia. Journal of the Marine Biology Association UK, 93, 27–36. Ramji, S, Latha, G, Rajendran, V and Ramakrishnan, S (2008) Wind dependence of ambient noise in shallow water of Bay of Bengal. Applied Acoustics, 69, 1294–1298. Ricci, SW, Eggleston, DB, Bohnenstiehl, DR and Lillis, A (2016) Temporal soundscape patterns and processes in an estuarine reserve. Marine Ecology Progress Series, 550, 25–38. Richardson, WJ, Greene, CR, Malme, CI and Thomson, DH (1995) Marine Mammals and Noise, Academic Press, San Diego. Risch, D, Castellote, M, Clark, CW, et al. (2014a) Seasonal migrations of North Atlantic minke whales: novel insights from large‐scale passive acoustic monitoring networks. Movement Ecology, 2, 24. Risch, D, Siebert, U and Van Parijs, SM (2014b) Individual calling behaviour and movements of North Atlantic minke whales (Balaenoptera acutorostrata). Behaviour, 15, 1335–1360. Rossi, T, Connell, SD and Nagelkerken, I (2016) Silent oceans: ocean acidification impoverishes natural soundscapes by altering sound production of the world’s noisiest marine invertebrate. Proceedings of the Royal Society B, 283(1826), 20153046.

165

166

Ecoacoustics

Rountree, RA, Gilmore, RG, Goudey, CA, Hawkins, AD, Luczkovich, JJ and Mann, D (2006) Listening to fish: applications of passive acoustics to fisheries science. Fisheries, 3, 433–446. Rowell, TJ, Schärer, MT, Appeldoorn, RS, Nemeth, MI, Mann, DA and Rivera, JA (2012) Sound production as an indicator of red hind density at a spawning aggregation. Marine Ecology Progress Series, 462, 241–250. Royer, JY, Chateau, R, Dziak, RP and Bohnenstiehl, DR (2015) Seafloor seismicity, Antarctic ice‐sounds, cetacean vocalizations and long‐term ambient sound in the Indian Ocean basin. Geophysics Journal International, 202, 748–762. Ruppé, L, Clément, G, Herrel, A, et al. (2015) Environmental constraints drive the partitioning of the soundscape in fishes. Proceedings of the National Academy of Sciences USA, 112, 201424667. Scholes, RJ, Walters, M, Turak, E, et al. (2012) Building a global observing system for biodiversity. Current Opinion in Environmental Sustainability, 4, 139–146. Seppänen, J and Nieminen, M (2004) Measurements and descriptions of underwater noise in Finland. Geophysicam, 40, 23–38. Shannon, G, Mckenna, MF, Angeloni, LM, et al. (2015) A synthesis of two decades of research documenting the effects of noise on wildlife. Biological Reviews, 91(4), 982–1005. Shaw, PT, Watts, DR and Rossby, HT (1978) On the estimation of oceanic wind speed and stress from ambient noise measurements. Deep Sea Research, 25, 1225–1233. Simpson, SD, Meekan, M, Montgomery, J, McCauley, R and Jeffs, A (2005) Homeward sound. Science, 308(5719), 221. Simpson, SD, Meekan, MG, Larsen, NJ, McCauley, RD and Jeffs, A (2010) Behavioral plasticity in larval reef fish: orientation is influenced by recent acoustic experiences. Behavioral Ecology, 21, 1098–1105. Simpson, SD, Radford, AN, Holles, S, et al. (2016) Small‐boat noise impacts natural settlement behavior of coral reef fish larvae, in The Effects of Noise on Aquatic Life II, Springer, New York, pp.1041–1048. Simpson, V, Meekan, M, Jeffs, A, Montgomery, J and McCauley, R (2008) Settlement‐stage coral reef fish prefer the higher‐frequency invertebrate‐generated audible component of reef noise. Animal Behaviour, 75, 1861–1868. Širović, A, Hildebrand, JA, Wiggins, SM, McDonald, MA, Moore, SE and Thiele, D (2004) Seasonality of blue and fin whale calls and the influence of sea ice in the Western Antarctic Peninsula. Deep Sea Research Part II Topical Studies in Oceanography, 51, 2327–2344. Slabbekoorn, H, Bouton, N, van Opzeeland, I, Coers, A, ten Cate, C and Popper, AN (2010) A noisy spring: the impact of globally rising underwater sound levels on fish. Trends in Ecology and Evolution, 25, 419–427. Snelgrove, PVR (2010) Discoveries of the Census of Marine Life, Cambridge University Press, Cambridge. Staaterman, E, Rice AN, Mann, DA and Paris, CB (2013) Soundscapes from a Tropical Eastern Pacific reef and a Caribbean Sea reef. Coral Reefs, 32, 553–557. Staaterman, E, Paris, CB, de Ferrari, HA, Mann, DA, Rice, AN and d’Alessandro, EK (2014) Celestial patterns in marine soundscapes. Marine Ecology Progress Series, 508, 17–32. Stendera, S, Adrian, R, Bonada, N, et al. (2012) Drivers and stressors of freshwater biodiversity patterns across different ecosystems and scales: a review. Hydrobiologia, 696, 1–28.

9  Biodiversity Assessment and Environmental Monitoring in Freshwater and Marine Biomes

Strayer, DL and Dudgeon, D (2010) Freshwater biodiversity conservation: recent progress and future challenges. Journal of the North American Benthological Society, 29, 344–358. Sueur, J and Farina, A (2015) Ecoacoustics: the ecological investigation and interpretation of environmental sound. Biosemiotics, 8, 493–502. Sueur, J, Pavoine, S, Hamerlynck, O and Duvail, S (2008) Rapid acoustic survey for biodiversity appraisal. PLoS One, 3, e4065–e4065. Sueur, J, Mackie, D and Windmill, JFC (2011) So small, so loud: extremely high sound pressure level from a pygmy aquatic insect (Corixidae, Micronectinae). PLoS One, 6, e21089–e21089. Sueur, J, Farina, A, Gasc, A, Pieretti, N and Pavoine, S (2014) Acoustic indices for biodiversity assessment and landscape investigation. Acta Acustica united with Acustica, 100, 772–781. Tonolla, D, Acuña, V, Lorang, MS, Heutschi, K and Tockner, K (2010) A field‐based investigation to examine underwater soundscapes of five common river habitats. Hydrological Processes, 24, 3146–3156. Tonolla, D, Lorang, MS, Heutschi, K, Gotschalk, CC and Tockner, K (2011) Characterization of spatial heterogeneity in underwater soundscapes at the river segment scale. Limnology and Oceanography, 56, 2319–2333. Tricas, TC and Boyle, KS (2014) Acoustic behaviors in Hawaiian coral reef fish communities. Marine Ecology Progress Series, 511, 1–16. Turvey, ST, Pitman, RL, Taylor, BL, et al. (2007) First human‐caused extinction of a cetacean species? Biology Letters, 3, 537–540. Tyack, PL and Clark, CW (2000) Communication and acoustic behavior of dolphins and whales, in Hearing by Whales and Dolphins (eds WWL Au, NS Popper and RR Fay), Springer New York, pp.156–224. Van Opzeeland, I, Van Parijs, S, Bornemann, H, et al. (2010) Acoustic ecology of Antarctic pinnipeds. Marine Ecology Progress Series, 414, 267–291. Van Parijs, SM, Baumgartner, M, Cholewiak, D, Davis, G, Gedamke, J, Gerlach, D, Haver, S, Hatch, J, Hatch, L, Hotchkin, C, and Izzi, A (2015) NEPAN: A US northeast passive acoustic sensing network for monitoring, reducing threats and the conservation of marine animals. Marine Technology Society Journal, 49(2), pp. 70–86. Van Parijs, SM, Clark, CW, Sousa‐Lima, RW, et al. (2009) Management and research applications of real‐time and archival passive acoustic sensors over varying temporal and spatial scales. Marine Ecology Progress Series, 395, 21–36. Voellmy, IK, Purser, J, Flynn, D, Kennedy, P, Simpson, SD and Radford, AN (2014) Acoustic noise reduces foraging success in two sympatric fish species via different mechanisms. Animal Behaviour, 89, 191–198. Vračar, MS and Mijić, M (2011) Ambient noise in large rivers (L). Journal of the Acoustical Society of America, 130, 1787. Wall, CC, Lembke, C, Hu, C and Mann, DA (2014a) Fish sound production in the presence of harmful algal blooms in the eastern Gulf of Mexico. PLoS One, 9, e114893–e114893. Wall, CC, Rountree, RA, Pomerleau, C and Juanes, F (2014b) An exploration for deep‐sea fish sounds off Vancouver Island from the NEPTUNE Canada ocean observing system. Deep Sea Research Part I, 83, 57–64. Wang, Z, Akamatsu, T, Mei, Z, et al. (2015) Frequent and prolonged nocturnal occupation of port areas by Yangtze finless porpoises (Neophocaena asiaeorientalis): forced choice for feeding? Integrative Zoology, 10, 122–132.

167

168

Ecoacoustics

Warren, JD, Leach, TH and Williamson, CE (2016) Measuring the distribution, abundance, and biovolume of zooplankton in an oligotrophic freshwater lake with a 710 kHz scientific echosounder. Limnology and Oceanography: Methods, 14(4), 231–244. Wenz, GM (1962) Acoustic ambient noise in the ocean: spectra and sources. Journal of the Acoustical Society of America, 34, 1936–1956. Williams, R, Wright, AJ, Ashe, E, Blight, LK, Bruintjes, R, Canessa, R, Clark, CW, CullisSuzuki, S, Dakin, DT, Erbe, C, Hammond, PS, Merchant, ND, O’Hara, PD, Purser, J, Radford, AN, Simpson, SD, Thomas, L, Wale, MA, Impacts of anthropogenic noise on marine life: Publication patterns, new discoveries, and future directions in research and management, Ocean & Coastal Management, 115, October 2015, 17–24. Willis, MR, Broudic, M, Haywood, C, Masters, I and Thomas, S (2013) Measuring underwater background noise in high tidal flow environments. Renewable Energy, 49, 255–258. Wilson, B, Lepper, PA, Carter, C and Robinson, SP (2014) Rethinking underwater sound‐recording methods to work at tidal‐stream and wave‐energy sites, in Marine Renewable Energy Technology and Environmental Interactions SE ‐ 9 (eds MA Shields and AIL Payne), Springer, Dordrecht, pp.111–126. Wilson, N, Flinn, MB, West, B and Hereford, J (2015) Identification of sound‐producing hydrophilid beetles (Coleoptera: Hydrophilidae) in underwater recordings using digital signal processing, Coleoptera Bulletin, 69, 305–315. Worm, B, Barbier, EB, Beaumont, N, et al. (2006) Impacts of biodiversity loss on ocean ecosystem services. Science, 314, 787–790. Wu, J and Hobbs, R (2002) Key issues and research priorities in landscape ecology: an idiosyncratic synthesis. Landscape Ecology, 17, 355–365. Wysocki, LE, Dittami, JP and Ladich, F (2006) Ship noise and cortisol secretion in European freshwater fishes. Biological Conservation, 128, 501–508. Wysocki, LE, Amoser, S and Ladich, F (2007) Diversity in ambient noise in European freshwater habitats: noise levels, spectral profiles, and impact on fishes. Journal of the Acoustical Society of America, 121, 2559–2566. Yamamoto, Y, Akamatsu, T, da Silva, VMF, Kohshima S (2016) Local habitat use by botos (Amazon river dolphins, Inia geoffrensis) using passive acoustic methods. Marine Mammal Science, 32, 220–240.

169

10 Integrating Biophony into Biodiversity Measurement and Assessment Brian Michael Napoletano Centro de Investigaciones en Geografía Ambiental, Universidad Nacional Autónoma de México, Morelia, México

The sounds of the living are uttered only within a thin shell around the earth’s surface – much less than 1 per cent of its radius in width. They are confined to the land surface, the sea a few score fathoms below the surface and the air immediately above. But within this relatively small area the diversity of sounds produced by living organisms is bewilderingly complex. (R. Murray Schafer 1994, p.36)

10.1 ­Introduction The soundscape carries significant volumes of information about the number, identity, life history, and interactions of the organisms using it that would be useful to the study and measurement of biodiversity. As advances in digital technology continue to improve the storage and operational capacity of field sensors, more of this information is becoming accessible. The evolution of the study of ecoacoustics provides a theoretical foundation from which we may begin examining socio‐ecological dynamics at higher spatiotemporal resolutions and developing new variables and suites of system indicators of current and desired biophysical conditions. This chapter primarily focuses on potential and actual contributions of ecoacoustics to the study and assessment of biodiversity. It begins with a review of the types of biological information in the soundscape and how such information could benefit biodiversity studies, and then briefly reviews the technical aspects of sampling the soundscape and utilizing sound information in conventional biodiversity assessments. The chapter concludes with a short reflection on how ecoacoustics may encourage greater collaboration between the natural and social sciences. The underlying premise is that although significant theoretical and technical advances are needed before many are realized (Brandes 2008; Porter et al. 2005), the potential contribution of ecoacoustics to our understanding and assessment of biodiversity is significant enough to merit substantial investment into research along this trajectory.

Ecoacoustics: The Ecological Role of Sounds, First Edition. Edited by Almo Farina and Stuart H. Gage. © 2017 John Wiley & Sons Ltd. Published 2017 by John Wiley & Sons Ltd.

170

Ecoacoustics

10.1.1  Biodiversity and its Parameterization

The potential role of biophony in the study of biodiversity depends largely on how the latter is defined. The most widely accepted definition is that of the United Nations’ 1992 Convention on Biological Diversity, which defines the concept as “the variability among living organisms from all sources including, inter alia, terrestrial, marine and other aquatic ecosystems and the ecological complexes of which they are part; this includes diversity within species, between species and of ecosystems” (UNCBD 1992). This definition is commonly posited to delineate three dimensions of biodiversity: the number of different types of components present, the number of individuals of each type of component, and the degree of differentiation within and between each type of component (Figure 10.1). The second part of the definition identifies three common levels of differentiation between types of components: (1) intraspecific, (2) specific, and (3) ecosystemic (Figure 10.2). Although this remains one of the most widely accepted definitions of biodiversity (Lévêque and Mounolou 2003), its parameterization remains inconsistent, particularly as related to spatial scale (Whittaker et al. 2001), and it tends to overlook the historical contextualization of biodiversity as a dynamic evolutionary process (Gaston and Spicer 1998). Because the number of different objects present is generally the easiest of the three parameters to measure, some measure of species richness is the most frequently

Biodiversity

Operationalization of “variability among living organisms” into quantifiable parameters

Numbers

Evenness

Difference

Counts and estimates of quantities of unique classes of individuals

Extent to which differences are distributed among individual units

Extent to which species, alleles, populations, etc. diverge, e.g., structure or function

Species richness

Number of higher taxa

Number of different species per sample site, habitat, clade, etc.

Numbers in broader taxonomic groups

e.g., Orders, Families

e.g., Aves, Insecta

Figure 10.1  A tiered representation of three dimensions of biodiversity identified in the Rio Convention. (1) Number of different units, (2) distribution of the differences among the units, and (3) extent of differences between units. Of these, the number of different units is most readily and frequently measured, with species designation used as the differentiation criterion. Source: Purvis and Hector (2000). Reproduced with permission of Springer Nature.

10  Integrating Biophony into Biodiversity Measurement and Assessment

Biodiversity Ecosystem diversity

Interspecific diversity

Intraspecific diversity

Figure 10.2  Hierarchical relationships between the three levels of biodiversity identified in the Rio Convention. Source: After Lévêque and Mounolou (2003) and di Castri and Younès (1996).

employed surrogate for biodiversity (Gaston and Spicer 1998), with more complex estimates of relative richness also indicating the distribution of the number of individuals of each species (Table 10.1). Even at this level of simplification, however, the number of species present across all taxa is impossible to measure, so that estimates typically reflect the number of species present in a particular group or clade. Groups of insects (e.g. butterflies), birds, amphibians, mammals, and plants are all frequently employed in biodiversity assessment (Pereira and Cooper 2006). Diversity at higher taxonomic levels can also serve as a fairly reliable predictor of species richness, but this relationship weakens as taxonomic level increases (Gaston and Williams 1993). To the extent that taxonomic classifications reflect phylogenetic divergence, species richness can also offer some insight into the other two dimensions of biodiversity mentioned in the Rio Convention, particularly when such phylogenetic divergence is relatively constant between species (Purvis and Hector 2000). In addition to providing recordings in which the number and identity of sources of sounds in the biophony can be calculated to supplement measures of species richness (Celis‐Murillo et al. 2009; Riede 1993) (see section 10.3), ecoacoustic studies can yield more detailed biological information and elucidate systemic characteristics that offer significantly more detailed information about the diversity of organisms present in the landscape and some of the evolutionary forces that shape these spatiotemporal patterns of biodiversity (Mazaris et al. 2009).

10.2 ­Biological Information in the Soundscape Aspects of animal vocalizations have frequently been studied by biologists, physiologists, behavioral ecologists, and taxonomists to advance human understanding of the evolution of communication and hearing, how vocal communication is incorporated into basic life functions, the operation of auditory systems, and the diversity of sounds that organisms produce, and these studies have helped build a large body of knowledge

171

172

Ecoacoustics

Table 10.1  Mathematical tools used to estimate diversity parameters from species surveys. “Importance” refers to the relative dominance or prevalence of a species. 1. Species diversity S = number of species in a sample



d=

S S or d = ln A ln N

d = species area/sample‐weighted richness A = sample area N = number of individuals sampled 2. Relative species richness



S S n  C = ∑ i =1 pi2 = ∑ i =1  i  N 

2

C = Simpson index N = total of importance values for all species ni = importance value of species i pi = importance value of species i as a fraction of the total



H ′ = ∑ i =1 pi ln pi S

H' = Shannon‐Wiener information index S Ec = (ln nmax − ln nmin )



Ec = species per log cycle index nmax = importance value of the most important species nmin = importance value of the least important species 3. Similarity between samples



Sim =

2S A∩ B S A + SB

Dif =

ST S

Sim = Similarity Index Dif = Differentiation SA = number of species in sample A SB = number of species in sample B SA∩B = number of species common to both A and B ST = number of species in all samples S = mean number of species per sample Source: After Whittaker (1975) and Odum (1971).

10  Integrating Biophony into Biodiversity Measurement and Assessment

Physiology

Communication

Sound production sound perception discrimination speciation

advertisement identification warning information

Soundscape Coordination

Adaptation

community composition competition modulation acoustic niche

interference startle response displacment silencing

Figure 10.3  Categories of information in the soundscape relevant to the study of biodiversity.

about the role of sound in the biosphere (Pijanowski et al. 2011). If this information can be incorporated into the study of biodiversity and biogeography and the technological resources to sample soundscapes at multiple spatial and temporal scales, current understanding of the dynamics of biodiversity could be improved considerably, and conservation and restoration efforts could be rendered more effective. The extensive body of work in bioacoustics therefore serves as the logical foundation on which to begin integrating ecoacoustic studies into biodiversity assessment. I have chosen to divide this review of biological information in the soundscape into four general categories (Figure 10.3): ●● ●● ●● ●●

physiology communication coordination adaptation.

The category “physiology” refers to the organs and neurological processes that allow organisms to interact with the soundscape, and how these features reflect adaptations to an organism’s habitat. “Communication” refers primarily to the type of information carried in intentional interactions between organisms in the soundscape, in the form of both intraspecies and interspecies signaling. “Coordination” refers to both unintentional aspects of organisms’ interactions within the soundscape and how they respond to these interactions in the effort to communicate. Finally, “adaptation” discusses how organisms are responding to the abrupt changes to soundscapes that have accompanied the rise and expansion of industrial capitalism. Information in each of these categories relates directly or indirectly to the study and assessment of biodiversity.

173

174

Ecoacoustics

10.2.1  Physiology: Sound Production and Detection

Building on von Uexküll’s 1909 principle of umwelt (von Uexküll 2001), Farina and Belgrano (2004) propose the concept of the eco-field to describe an organism’s cognition of its environment in terms of the spatial configuration of available resources as detected by its sensory organs. This dynamic field can theoretically be generalized for a given species by identifying the needs common to that species and the resources in the study area capable of fulfilling them. As a carrier of information, the soundscape is an important element of the eco-field (Marler 1967), but any given organism’s use of the soundscape is mediated by its physiology and acoustic cognition. A given nonhuman organism’s perception of the soundscape may differ significantly from that of humans, and thus understanding how acoustic information is obtained and processed by various organisms is central to interpreting biological information in the soundscape. The field of bioacoustics offers a significant body of knowledge on the hearing ­mechanisms, cognition, and sound‐producing capacities of many organisms, the majority of which differ significantly from that of humans. Dooling et al. (1999), for instance, report that budgerigars (Melopsittacus undulatus) possess a much higher degree of temporal resolving power than do humans and other mammals, lending support to the belief that many avian species are more sensitive than humans to the temporal aspects of complex signals. Dooling et al. (2002) reinforce these finding in a similar study of three different species’ abilities to discriminate the fine temporal structures in complex waveforms. If the sampling rate of monitoring equipment is set too low to capture these structures, then the information will be lost to the observers. Some frogs also rely heavily on temporal variations; Allan and Simmons (1994) report that the temporal aspect of amplitude modulation in green treefrogs (Hyla cinerea) is likely a significant component in their call recognition. Hillery (1984) confirms the role that amplitude modulation plays in signal identification by examining the nerve action potential from the auditory nerve and midbrain averaged evoked potential in Hyla chrysoscelis and Hyla versicolor. Although no evidence of frequency processing was found in these sections of the brain, an experimental analysis of male tundra frog (Physalaemus pustulosus) responses to conspecific whines by Bosch et al. (2000) found that vocal competition tends to escalate when competitors have whines near the frogs’ own frequencies, but not when at the mean or higher frequency for the population, indicating that frequency processing may be conducted in another portion of the brain. When examining the neurological responses of the downy woodpecker (Picoides pubescens), Carolina chickadee (Poecile carolinensis), tufted titmouse (Baeolophus bicolor), white‐breasted nuthatch (Sitta carolinensis), house sparrow (Passer domesticus), and European starling (Sturnus vulgaris) to broadband click stimuli, Lucas et al. (2002) found that relationships between the species’ responses in the auditory brainstem differ substantially between spring and winter, adding a seasonal component to auditory reception. In a subsequent analysis measuring the neurological responses of the Carolina chickadee, tufted titmouse, and white‐breasted nuthatch (Lucas et al. 2007), they found that the seasonal changes differ between the three species, with the changes in the chickadees generally trending in the opposite direction to those in the nuthatches. The manner in which pileated woodpeckers (Dryocopus pileatus) adjust their vocalizations seasonally is similar to that of several passerines, reinforcing the concept of seasonal variation in auditory attributes. This reinforces the hypothesis that calls and

10  Integrating Biophony into Biodiversity Measurement and Assessment

drumming serve as the counterparts to passerine song for pileated woodpeckers (Tremain et  al. 2008). Temporal patterns of vocalizations can also be important at a diurnal scale. In a study of temporal variation in the calls of frogs in South Carolina, for instance, Bridges and Dorcas (2000) found that southern leopard frogs (Rana sphenocephala) consistently call outside the timeframe of most call survey protocols. Lohr et  al.’s (2003) derivation of the active space of three different species’ communication signals provides an example of how knowledge of physiological attributes may be combined with experimental tools (in this case, operant conditioning) to calculate perceptual variables. These authors conclude that measures of peak sound pressure level combined with the spectrum level of noise within the frequency band in which the most signal power is concentrated relative to background noise provides the most consistent model of active space. A red‐winged blackbird (Agelaius phoeniceus), Brenowitz (1982) found, has a maximum active space of approximately 189 meters, which allows an individual male to advertise his presence throughout the extent of his neighbor’s territories. The results of this analysis also indicate that the red‐winged blackbird is capable of discriminating a song with a signal‐to‐noise ratio of only 3 dB. Even before computational advances, Dooling (1978) demonstrated that the wealth of data generated by a decade of research can be used to characterize entire groups of organisms across multiple parameters. He maintained that the region of maximum sensitivity in the hearing of birds across a wide range of species is 1–5 kHz, generally placing them between mammals and reptiles. The addition of nearly four more decades of research may allow these generalizations to be refined and extended to multiple groups of organisms, thereby facilitating the application of bioacoustic monitoring to measure aspects of nonavian biodiversity. In another study, Dooling et  al. (1992) found that budgerigars, canaries, and zebra finches (Poephila guttata) all exhibit an enhanced ability to discriminate between conspecific and heterospecific calls. European starlings are also able to discriminate the three different specific calls, although they discriminate most effectively among canary calls. Greenwood (1996) compared the cochlear maps of a human (Homo sapiens), cat (Felis catus), guinea pig (Cavia porcellus), chinchilla (Chinchilla lanigera), and monkey and found that mechanical and physiological data regarding the cochlea can be effectively modeled by an almost exponential frequency‐position function, and that these organisms all have similar optimal frequency detection ranges. In another group of mammals – the domestic rabbit (Oryctolagus cuniculus), cotton rat (Sigmodon hispidus), house mouse (Mus musculus), and kangaroo rat (Dipodomys merriami) – Heffner and Masterton (1980) found a significantly different range of frequency perception, with low‐frequency limits from 0.05 to 2.3 kHz and high‐frequency limits from 49 to 90 kHz. They also found that interaural distance is strongly and inversely correlated with high‐frequency hearing ability and good high‐frequency hearing is incompatible with good low‐frequency hearing in most terrestrial mammals. In their analysis of masking by harmonic complexes, Lauer et al. (2006) found that birds have a reduced cochlear delay compared to that of humans. In a study comparing human hearing to that of hooded crows (Corvus corone), Jensen and Klokker (2006) found that the crows’ hearing sensitivity is similar to that of humans below 5.6 kHz. A  middle region of 5–6 dB lower than average critical ratios from 0.5 to 2 kHz was found, however, whereas humans’ critical ratios at that range are 3–6 dB lower than those of the crows.

175

176

Ecoacoustics

Another important dimension of organisms’ perception of the soundscape is their ability to separate sounds that occur simultaneously in different frequency bands, conducting what is called an “auditory scene analysis.” MacDougall‐Shackleton et al. (1998) found that European starlings are highly capable of such analysis, and suggest that this ability might be widespread among species that rely on information in the soundscape. The unique physiology of song production is also important to understanding the information that bioacoustic signals carry about the organisms that produce them. After examining the calls and context of free‐ranging male baboons (Papio cynocephalus), Fischer et  al. (2002) found that calls that served significantly different functions consisted essentially of variations on a single call type, indicating that nonhuman primates may have evolutionarily constrained vocal production capacities. While examining the anatomy of a species of cicada (Cyclochila australasiae), Bennet‐Clark (1997) found that the initial pulses in the sound produced by this insect are caused by an inward movement of its tymbal plate, while the last pulse is produced by the outward movement. The masses involved in moving the plate in and out differ, which may be what allows the insect to maintain a coherent resonance of its initial pulse. In a subsequent study, Bennet‐ Clark (1998) related the sound production strategies of different types of small insects back to their physiological constraints and attempted to determine how the strategies employed, such as pure tone components and pulses in airborne insects and substrate vibration in others, optimize communication distance. Beaver (1977) conducted spectrographic analyses on snow goose (Chen caerulescens) vocalizations that suggest that this species attempts to produce songs that resonate within its trachea, and that many other species may also employ this technique. The relationship between body size and song frequency was also confirmed by Bertelli and Tubaro (2002) who, in a study of song properties and habitat attributes of tinamous (Tinamiformes), found that frequency of their song is negatively correlated to their body size, indicating that songs are at least partially determined by physiology. They also found that songs of species in open habitats exhibit wider bandwidths than those in closed habitats, suggesting that habitat influences song development. A comparison between song power and body sizes of 32 species of birds and the sizes of the areas occupied by individuals of 86 species by Calder (1990) found that both song power and territory size tend to scale to body size. Consequently, any census or measurement that relies on auditory detection can be heavily biased against smaller species if its spatial resolution is inadequate. The extent to which physiology influences bird song allows it to serve as a criterion for speciation in some cases (Hamao et al. 2008; Leger and Mountjoy 2003). The songs of insects and anurans are also frequently studied to gain insight into the organisms’ phylogenetic relationships and evolutionary history (Gerhardt and Huber 2002). Thus, the physiological factors of vocal organisms are important considerations when examining the soundscape for information regarding biodiversity. 10.2.2  Communication: Medium and Context

Among terrestrial organisms, many insects and virtually all vertebrates make extensive use of vocal communication, as the physical properties of sound signals allow more information to be transmitted over larger distances than in the case of olfactory or visual communication, and sound signals do not require a direct line of sight (Marler 1967). The information carried in such acoustic signals primarily depends on the source

10  Integrating Biophony into Biodiversity Measurement and Assessment

(see section 10.2.1), the medium, and the context in which the signal is given (Pierce 1980), although the function of many vocalizations is still unknown or uncertain (Gill 2007). In the case of animal communication, the medium may be important in terms of both its immediate effects on sound propagation and its long‐term influence on the evolution of species’ vocalizations (Endler 1992; Marten et al. 1977; Slabbekoorn et al. 2007). Morton (1975), for instance, proposes an acoustic adaptation hypothesis to explain how the dominant frequency of signals by birds in forest and grassland habitat is related to the physical influence of the habitat on sound propagation. Ryan and Brenowitz (1985) also included ambient noise in these habitats and body size as confounding factors, and propose that the selection for low‐frequency sounds among birds in forest and grassland habitats is related to the relatively lower concentration of noise in that region of the acoustic spectrum. In a study of the songs of 51 species of North American wood warblers (Parulidae), Buskirk (1997) found that habitat attributes such as the type, density, and moisture of canopy foliage correspond to the frequency, number, and arrangement of notes in the songs, but not the structure of the notes. Among primates, de la Torre and Snowdon (2002) found that pygmy marmosets (Cebuella pygmaea) select calls from their repertoire that are suited to the acoustic attributes of their habitats, and Brown et al. (1995) found that the calls of two species of rainforest primates and two species of savannah primates each propagate better in the primates’ native habitats than those of the others. Wang et  al. (2007) reported that, among Chinese alligators (Alligator sinensis), the dense vegetation prevalent in their habitat has prompted them to vocalize at lower frequencies than most other organisms. On the other hand, some studies have not found evidence of a significant influence of habitat on vocalizations (Daniel and Blumstein 1998; Penna et al. 2006), suggesting that, in some cases, other factors may be more important than maximizing signal propagation. In other cases, organisms have been observed altering their habitat to maximize sound transmission, such as when small Australian burrowing crickets (Rufocephalus) construct burrows that resonate with the sound‐producing mechanisms on their wings (Bailey et al. 2001). Apart from that of humans, communication in birds, insects, anurans, and nonhuman primates appears to be the most commonly studied (among terrestrial organisms). In birds, mate attraction and territorial defense are considered the two primary communication functions (Collins 2004), with other functions being alarm calls, mobbing calls, distress calls, contact calls, separation calls, flight calls, roosting calls, food calls, begging calls, and agonistic calls (Marler 2004). Insects and anurans are believed to primarily vocalize mating or advertisement calls, aggressive signals, encounter calls, and distress calls (Gerhardt and Huber 2002). Nonhuman primates are believed to generally communicate the same information as do birds, but insufficient studies are available to indicate whether this applies to nonprimate mammals as well (Seyfarth and Cheney 2010). In addition to these general contexts, the complexity of auditory signals allows more specific information to be transmitted. Black‐capped chickadees (Poecile atricapilla), for instance, encode information regarding the size and risk of a predator in their alarm calls (Templeton et al. 2005), and red‐breasted nuthatches (Sitta canadensis) recognize and also use this information (Templeton and Greene 2007). Interspecies communication appears particularly common in birds, where alarm calls are also used deceptively to access food (Munn 1986; Ridley et al. 2007), reinforcing the importance of context in

177

178

Ecoacoustics

interpreting the information carried in a biological signal. Møller (1988) observed subordinate great tits (Parus major) using a similar strategy to reclaim a feeding site from a more dominant conspecific, indicating that auditory deception occurs within as well as between species. In many cases, animal vocalizations have been found to contain life history and identity information about the individual vocalizing. The mating calls of black‐capped chickadees, for instance, contain information identifying the gender and social group of the caller. Developmental stress has been observed to influence the size and quality of the repertoires of both European starlings (Buchanan et  al. 2003) and zebra finches (Spencer et al. 2003), and body condition to influence the songs of male barn swallows (Hirundo rustica) (Galeotti et al. 1997). Dickcissels (Spiza americana) exhibit a degree of song sharing with their neighbors (Schook et al. 2008), and the common loon can identify neighbors and determine level of aggression and physical condition based on the species’ yodel call (Mager et al. 2010). Although the technical challenges at this level of signal identification remain significant, the ability to extract, directly from the soundscape, information on the number of individuals, their physical health, and life histories would obviously be useful to biodiversity studies. 10.2.3  Coordination: Evolution of the Biophony

In addition to the physical properties of the habitat, the presence of other vocal organisms also influences an individual’s ability to communicate. Particularly if the habitat structure favors only a relatively narrow band of frequencies, that band could easily be flooded by multiple overlapping signals. Krause (1987) proposes that the concept of the ecological niche applies to the soundscape, where signals of different species evolve to occupy different time or frequency ranges so that each organism present can vocalize with minimal interference from other species. It is this complex arrangement of signals partitioned into different acoustic “niches” that Krause (1998) defines as the biophony. Although this “acoustic niche hypothesis” cannot be tested directly according to the dominant scientific paradigm, it is partially supported by various field studies. In some cases, organisms appear to consciously adjust their vocalizations when they encounter species with similar sound structures. Ficken et  al. (1974), for instance, observed that when the red‐eyed vireo (Vireo olivaceus) and the least flycatcher (Empidonax minimus) share a habitat, the flycatcher, which has the song more susceptible to total masking, adjusts the timing of its song to avoid interference from the vireo. Similarly, the common nightingale (Luscinia megarhynchos), when presented with overlapping vocalizations from other species, adjusts the timing of its songs to fit the silent intervals between heterospecific signals (Brumm 2006). Conversely, the black‐capped chickadee, when engaged in a territorial contest, is believed to match the frequency of its call to that of its rival to signal its readiness for a physical confrontation (Otter et al. 2002). Indications of niche partitioning have also been found in the time and frequency of cicada signals in Mexico (Sueur 2002), and in some, but not all, vocalizations of multispecies frog assemblages (Chek et al. 2003). Statistical analyses of multiple vocalizations also offer partial support for the acoustic niche hypothesis. In a study of 20 species of Peruvian rainforest birds, Planqué and Slabbekoorn (2008) found significantly less overlap at the most commonly used frequencies than would be expected from a random collection of songs, although they

High

10  Integrating Biophony into Biodiversity Measurement and Assessment

A) Signals frequently overlap, biophony exhibits little complexity

Signal Overlap

g

nin

tio r ti

e ich

Pa

N

B) Signals carefully partitioned, biophony exhibits significant complexity

Low

Transition Period

Dispersal-Assembly

Niche-Assembly

Temporal Scale Figure 10.4  A hypothetical model of the evolution of the biophony as an indicator of the transition of a community from dispersal‐assembly to niche‐assembly rules as the temporal scale increases to reflect evolutionary time. If such a relationship exists, instances where the biophony reflects a transition (i.e. a mixture of overlapping and partitioned signals) may be observable.

did not find evidence of efficient shifts in song properties (i.e. either shifting time or frequency instead of both). A more complex study of 11 attributes of the songs of 82 birds of the Amazonian rainforest by Luther (2009) similarly found that the songs of species inhabiting the same stratum exhibit the furthest Euclidean distance from each other. If an evolution tendency to coordinate signals exists within the soundscape, it may be useful in examining instances where niche‐assembly or dispersal‐assembly rules of species composition dominate (Hubbell 2001; Lévêque and Mounolou 2003). Because vocalizations tend to be more plastic in a number of passerines (Gill 2007), active space may be one of the earlier attributes that organisms adjust as they disperse into new habitats. In instances where niche‐assembly dominates, therefore, sound signals should already be shifting to occupy different acoustic spaces, and the arrangement of biological signals would be expected to demonstrate a relatively high degree of complexity and a low degree of signal overlap (Krause 1998). In instances where dispersal‐assembly dominates, the arrangement of biological signals would be expected to exhibit relatively low complexity and a high degree of signal overlap, particularly if the species present also have other overlapping niche dimensions (MacArthur and Wilson 1967; Whittaker 1975). Although Krause’s acoustic niche hypothesis suggests that all communities tend towards a higher degree of signal coordination, intervening factors, such as intermediate disturbance (Connell 1978; Hutchinson 1961), may counteract this tendency and allow the partitioning process to be observed in periodically disturbed habitats (Figure 10.4). Temporal dynamics of patterns of biological activity in the soundscape may themselves significantly expand our understanding of biodiversity as a dynamic process.

179

180

Ecoacoustics

At a higher level of conceptualization, this biophony itself may be regarded as another dimension of biodiversity, and species metrics such as richness, dominance, rarity, and endemism (Gaston and Spicer 1998) could be translated directly to the composition of biological signals in the soundscape (Sueur et al. 2014). Metrics derived from information theory and applied in landscape ecology, such as relative richness, connectivity, and fragmentation (Farina 1998), could likewise be applied to the arrangement of biological signals within the soundscape as a whole (Farina et  al. 2011). Because sound signals are highly temporal, changes in the biophony may respond more rapidly and be more sensitive to environmental change (Mazaris et al. 2009) and other socio‐ecological issues such as the rapid alteration of the planetary climate (IPCC 2008), land use change (Riede 1993), and rapidly increasing rates of species extinctions (Dirzo and Raven 2003, to cite one of many reports). As the title suggests, the underlying theme of Silent Spring, for instance, is the rapid disappearance of multiple bird songs from local soundscapes due to the secondary effects of the dispersion of chemical pesticides (Carson 1962). Should present trends continue, recordings may be all that remains of most contemporary soundscapes. 10.2.4  Adaptation: Mechanization of the Soundscape

This last aspect of biological information in the soundscape considers how industrial capitalism’s transformation of the soundscape affects the (human and nonhuman) organisms that use it. Mechanical systems are largely powered by hydrocarbons, making their signals evolutionarily distinct in both their perpetuity (i.e. their tendency to become constant or periodic components of the soundscape) and the large amount of acoustic energy they splash across the acoustic spectrum (Schafer 1994). The potential effect of this radical reconstruction of the soundscape on biodiversity is a key issue in ecoacoustics and conservation (Farina 2014). Studies of the effects of mechanical sounds on nonhumans have raised numerous concerns about observed changes in reproductive success, foraging, predator evasion, population dynamics, and community composition in response to such sounds (Barber et al. 2010). This poses a particular problem to conservation, as transportation corridors and other infrastructure frequently saturate the soundscapes of protected areas with mechanical sounds (Barber et al. 2011). Aircraft overflights, particularly by military aircraft, have frequently been implicated in instances of increased mortality of wild and domestic animals (Austin et al. 1972; Bel 1972; Krause 2002). As with other disturbances, the intrusion of mechanical sounds into the soundscape affects individual and groups of organisms differently depending on their physiology, vocal plasticity, physical and mental condition, and other contingent factors, as well as the characteristics of the introduced sounds themselves. In Puerto Rico, for instance, Herrera‐Montes and Aide (2011) found that avian species community composition varies significantly with proximity to roads, whereas the composition of anuran communities does not. A study of breeding bird populations in woodlands by Reijnen et al. (1995) similarly found reduced densities with closer proximities to roads in 26 of the 43 species surveyed, with traffic noise identified as the single most statistically significant explanatory variable. Beyond its contribution to increased mortality, the primary concern regarding the effects of mechanical sounds on nonhuman organisms is the extent to which such signals reduce the signal‐to‐noise ratio (Shannon 1948) of organisms attempting to

10  Integrating Biophony into Biodiversity Measurement and Assessment

Table 10.2  Possible responses to interference from mechanical sounds and their associated trade‐offs. Adjustment

Advantages

Drawbacks

No response

No adjustment required

Communication impaired

Relocation

No adjustment required

Loss of current resources Constrained by mobility Availability of territory

Amplitude Timing

Active space restored

Significant energy investment

Probable conspecific recognition

More audible to predators

Active space restored

May disrupt other functions

Lower energy demand

Conspecifics must also adjust Noise may be constant

Frequency

Active space restored

Limited by physiology

Lower energy demand

Conspecifics must also adjust Noise may be broadband

communicate, which in turn may impair mating success and other vital life functions that require vocal communication (Barber et al. 2010). If the signal‐to‐noise ratio drops below an organism’s detection threshold, then the signal is effectively masked from the organism, and communication is interrupted. Even above the detection threshold, excessive noise within the communication channel can distort or disrupt the signal enough to prevent the message from being received or understood. Lower population densities of bird species that vocalize at lower frequencies nearer to roads, for instance, may indicate that traffic sounds are masking their vocalizations (Rheindt 2003). If timing, frequency, and amplitude are taken as the three primary dimensions of a sound signal, then an organism finding its signal masked by increased noise in its communication channel may either adjust one or more of these dimensions of its signal or move to an area where the signal‐to‐noise ratio is once again high enough to permit communication (Table 10.2), although each of these strategies involves different trade‐ offs (Patricelli and Blickley 2006). In cases where other factors prevent alternative responses, the ability of organisms to discriminate multiple signals within masking noise (Appeltants et al. 2005) may allow them to tolerate interference as long as it does not lower the signal‐to‐noise ratio below the detection threshold. Increasing the amplitude of a vocalization may be the most physiologically practical response to masking or interference, and has been observed in nightingales singing in locations with higher levels of traffic noise in Berlin, Germany (Brumm 2004). The distribution of energy within the signal may also be adjusted, such as when California ground squirrels (Spermaphilus beecheyi) shift the majority of the sound energy in their vocalizations to the second or third harmonic when ambient noise masks the first (Rabin et al. 2003). In addition to increasing the risk of inadvertently advertising one’s presence to a predator, this strategy increases energy demands as more energy is required to produce signals at higher amplitudes, and the maximum

181

182

Ecoacoustics

sound energy any individual organism can produce is unlikely to surpass that of most mechanical sources. In situations where amplitude modulation is insufficient, impractical, or outweighed by its drawbacks, organisms may respond by adjusting the timing of their signals or aspects of their temporal structure such as length, the number of notes, or spacing between notes. European robins (Erithacus rubecula) in urban areas in Sheffield, England, for instance, have been observed singing at night when daytime conditions are noisy (Fuller et al. 2007). Such adjustments only work if the mechanical signal is not continuous, however, and are constrained by the physiology and vocal plasticity of the organism. The trade‐offs to shifting the primary frequency of a vocalization or altering its spectral structure (Wood and Yezerinac 2006) are similar to those associated with temporal shifts. In many cases, organisms have been observed adjusting two or more of these parameters simultaneously (Fernañdez‐Juricic et al. 2005; Slabbekoorn and Peet 2003), and the efficacy of these adaptations is difficult to evaluate (Patricelli and Blickley 2006). Studies of the effects of mechanical sounds on humans focus primarily on health and quality‐of‐life consequences of exposure to high‐intensity sounds, and the effectiveness of noise abatement legislation and technological innovations to eliminate or control this unwanted noise (USNAE 2010). As both human populations and sources of mechanical sound are more heavily concentrated in urban areas, such studies are frequently undertaken in these areas, with a focus on the acoustic features of the mechanical signals present rather than on other biological signals or how these groups of signals interact (Martínez Suárez and Moreno Jiménez 2005; Yepes et al. 2009). The presence of mechanical signals in recreational areas, at least in the case of National Parks within the USA, has also become a prominent issue, particularly in terms of conflicts between visitors seeking respite from mechanical sounds and visitors seeking to use recreational vehicles that generate such sounds (Miller 2008). In studies of both human and nonhuman animals, mechanical sounds are frequently treated as a problem of unwanted “noise pollution” to be resolved through legislation and technological fixes (USNAE 2010), rather than as symptoms of underlying social conflicts and contradictions (Schafer 1994).

10.3 ­Ecoacoustics in Biodiversity Assessment In addition to capturing the biological information described above, the ability to couple a comprehensive soundscape monitoring network with a system capable of automated signal and pattern identification could significantly increase the spatial and temporal coverage of conventional biodiversity assessment (Acevedo et al. 2009). The relative ease with which sound signals propagate suggests that they may be of use in situations where more direct observations are unavailable or impractical (Riede 1998). Moreover, the ability to obtain repeated samples and automate signal classification increases the temporal resolution of otherwise labor‐intensive field surveys while simultaneously resolving observer bias (Celis‐Murillo et al. 2009; Emlen and DeJong 1992). 10.3.1  Developing a Soundscape Monitoring Network

The development of a sensor network capable of sampling the soundscape at multiple locations and relaying the data to an accessible storage system remains a significant

10  Integrating Biophony into Biodiversity Measurement and Assessment

technological challenge, as sensors within such a network must optimize power consumption to facilitate long‐term deployment in areas without access to an electrical grid, on the one hand, but repeatedly undertake the power‐intensive tasks of recording and transmitting large quantities of data. An ideal soundscape monitoring network would consist of acoustic sensors capable of logging large volumes of data, transmitting this data quickly and efficiently, and operating for extended periods without an external power supply, yet remaining affordable enough to allow deployment at relatively high resolution over large areas. Several attempts have been made to optimize power consumption by shifting the signal‐processing step to the sensors themselves, which can then limit their transmissions to the results (e.g. the species to which a signal is assigned, the number of different signals or species detected, and the time of each detection) (Wang et  al. 2003, 2005). A closely related variant on this approach that generally performs better in simulations of power consumption (Chen et al. 2006) implements a two‐tiered system wherein relatively inexpensive clusters of sensors transmit sound recordings to nearby microprocessors that then perform the signal processing and relay the results (Hu et al. 2005). Such hierarchical network topologies offer significant power savings (Heinzelman et  al. 2000), but with the loss of the sound recordings themselves. In addition to preventing comparison between automated processing and human interpretation of the recordings, this limit constrains the sensor network to sampling only the soundscape parameters specified at the deployment phase (Gage et al. 2015). As the need to increase data throughput capacity in sensor networks is already recognized as an important factor when using wireless sensor networks in environmental monitoring (Estrin et al. 2003), other approaches to balancing power consumption and data throughput are being examined. Proposals include implementation of a coding scheme that compresses oversampled or redundant data (Roy and Vetterli 2006), dynamic power management systems that minimize power consumption whenever a sensor is inactive (Sinha and Chandrakasan 2001), and various other optimization strategies (Porter et al. 2005). While these advances make the ideal network described above more feasible, even short‐term recording systems that require regular visits can be used to extend the spatiotemporal coverage of biodiversity assessments (Celis‐Murillo et al. 2009; Joo et al. 2011). 10.3.2  Acoustic Data Processing and Management

The corollary to collecting large quantities of acoustic data is the ability to process, store, and access them quickly and efficiently. Signal classification and data management are the two key components of this process. Among the more advanced approaches to these challenges are the Pumilio system designed by Villanueva‐Rivera and Pijanowski (2012) at Purdue University and the Remote Environmental and Assessment Laboratory’s digital library developed by Kasten et al. (2012) at Michigan State University, both of which combine the two tasks and integrate them into a web interface accessible via the internet. Much of the interest in automating the analysis of sound recordings relates to the ability to automate species identification in sound recordings. This entails identifying the features of animal vocalizations most likely to facilitate accurate classification and developing the algorithms necessary to undertake it (Brandes 2008). Several feasibility studies have successfully classified the signals of limited sets of species from soundscape recordings (Bardeli et  al. 2010; Härmä 2003), but the distinguishing features vary

183

184

Ecoacoustics

significantly with the type of vocalization and the organism producing it. For instance, identification of bat signals frequently uses amplitude information and disregards frequency, although Obrist (1995) has demonstrated that call frequencies vary between individuals, and Harris and Skowronski (2006) found that incorporating such information and applying speech identification algorithms to its analysis significantly increases detection accuracy. Given their ability to dynamically adjust both the classification criteria and the algorithms employed, artificial neural networks (ANNs) appear to offer the most potential for automated signal classification (McIlraith and Card 1997), and have been successfully used to classify vocalizations of birds (Kogan and Margoliash 1998; Selouani et al. 2005) and insects (Chesmore 2004; Dietrich et al. 2004). In addition to identifying and classifying vocalizations by species, ANNs have been used to identify and compare individual call components (Dawson et al. 2006). In each study, however, different signal parameters and classification criteria are employed, implying that the development of a system capable of identifying and classifying all the signals present in the soundscape remains a significant challenge. That no such system yet exists reinforces the need for a soundscape monitoring network capable of recording and transmitting the actual sound recordings from its sensors rather than just the results of onboard analysis.

10.4 ­Conclusion While the technological and analytical challenges facing the field of soundscape ecology remain daunting, progress is being made, and the wealth of biological information in the soundscape could substantially enhance our understanding of biodiversity and benefit related studies in theoretical ecology, biogeography (Cadotte 2007), and conservation biology (Hubbell 2001). The higher degrees of spatiotemporal variability in the soundscape than in the landscape may also facilitate the biodiversity studies across scales (Storch et al. 2007), and may indicate changes triggered by other socio‐ ecological issues more quickly than conventional indicators. At a more conceptual level, the study of ecoacoustics offers an opportunity to better integrate the natural and social sciences, particularly in natural and human geography (Lave et al. 2014). The conventional division of sounds into biophony versus anthrophony is somewhat problematic here, as most contemporary soundscapes, like their counterpart landscapes, are “co‐produced” by human and nonhuman organisms and biophysical processes (Joo et al. 2011). Refining this division by situating human communication within the biophony and distinguishing it from the “cacophony” of various incidental signals or “noise” emitted by mechanical systems could be a first step towards linking ecoacoustics to noise studies and maps in urban areas associated with human geography (Martínez Suárez and Moreno Jiménez 2005; Yepes et al. 2009), and strengthening linkages between ecoacoustics and acoustic ecology. Calls from within both these fields to move beyond negative noise abatement measures to a more proactive and holistic treatment of the soundscape as a common‐pool resource suggest that both fields have already begun moving in this direction (Dumyahn and Pijanowski 2011a,b; Karlsson 2000; Sueur and Farina 2015), and further collaboration will become increasingly important as the mechanization of the soundscape under industrial capitalism continues to intensify.

10  Integrating Biophony into Biodiversity Measurement and Assessment

References Acevedo, MA, Corrada‐Bravo, CJ, Corrada‐Bravo, H, Villanueva‐Rivera, LJ and Aide, TM (2009) Automated classification of bird and amphibian calls using machine learning: a comparison of methods. Ecological Informatics, 4(4), 206–214. Allan, SE and Simmons, AM (1994) Temporal features mediating call recognition in the Green Treefrog, Hyla cinerea: amplitude modulation. Animal Behaviour, 47(5, 1073. Appeltants, D, Gentner, TQ, Hulse, SH, Balthazart, J and Ball, GF (2005) The effect of auditory distractors on song discrimination in male Canaries (Serinus canaria). Behavioural Processes, 69(3), 331–341. Austin, OL Jr, Robertson, WB Jr and Woolfenden, GE (1972) Mass hatching failure in Dry Tortugas sooty terns. Proceedings of the International Ornithological Conference, Volume 627, Netherlands. Bailey, WJ, Bennet‐Clark, H and Fletcher, N (2001) Acoustics of a small Australian burrowing cricket: the control of low‐frequency pure‐tone songs. Journal of Experimental Biology, 204, 2827–2841. Barber, JR, Crooks, KR and Fristrup, KM (2010) The costs of chronic noise exposure for terrestrial organisms. Trends in Ecology and Evolution, 25(3), 180–189. Barber, JR, Burdett, CL, Reed, SE, et al. (2011) Anthropogenic noise exposure in protected natural areas: estimating the scale of ecological consequences. Landscape Ecology, 26(9), 1281–1295. Bardeli, R, Wolff, D, Kurth, F, Koch, M, Tauchert, KH and Frommolt, KH (2010) Detecting bird sounds in a complex acoustic environment and application to bioacoustic monitoring. Pattern Recognition Letters, 31(12), 1524–1534. Beaver, JC (1977) Some unexplained properties of bird song. Journal of the Acoustical Society of America, 62(S1), S89–S89. Bell, WB (1972) Animal response to sonic booms. Journal of the Acoustical Society of America, 51(2C), 758–765. Bennet‐Clark, H (1997) Tymbal mechanics and the control of song frequency in the Cicada Cyclochila australasiae. Journal of Experimental Biology, 200(11), 1681–1694. Bennet‐Clark, H (1998) Size and scale effects as constraints in insect sound communication. Philosophical Transactions of the Royal Society B, 353(1367), 407–419. Bertelli, S and Tubaro, PL (2002) Body mass and habitat correlates of song structure in a primitive group of birds. Biological Journal of the Linnean Society, 77(4), 423–430. Bosch, J, Rand, A and Ryan, M (2000) Acoustic competition in Physalaemus pustulosus, a differential response to calls of relative frequency. Ethology, 106, 865–871. Brandes, TS (2008) Automated sound recording and analysis techniques for bird surveys and conservation. Bird Conservation International, 18, 163–173. Brenowitz, EA (1982) The active space of red‐winged blackbird song. Journal of Comparative Physiology A, 147(4), 511–522. Bridges, AS and Dorcas, ME (2000) Temporal variation in anuran calling behavior: implications for surveys and monitoring programs. Copeia, 2000(2), 587–592. Brown, CH, Gomez, R and Waser, PM (1995) Old World Monkey vocalizations: adaptation to the local habitat? Animal Behaviour, 50(4), 945–961. Brumm, H (2004) The impact of environmental noise on song amplitude in a territorial bird. Journal of Animal Ecology, 73(3), 434–440.

185

186

Ecoacoustics

Brumm, H (2006) Signalling through acoustic windows: nightingales avoid interspecific competition by short‐term adjustment of song timing. Journal of Comparative Physiology A, 192(12), 1279–1285. Buchanan, KL, Spencer, KA, Goldsmith, AR and Catchpole, CK (2003) Song as an honest signal of past developmental stress in the European Starling (Sturnus vulgaris). Proceedings of the Royal Society B, 270, 1149–1156. Buskirk, J (1997) Independent evolution of song structure and note structure in American Wood Warblers. Proceedings of the Royal Society B, 264(1382), 755–761. Cadotte, MW (2007) Concurrent niche and neutral processes in the competition‐ colonization model of species coexistence. Proceedings of the Royal Society B, 274(1626), 2739–2744. Calder, WA (1990) The scaling of sound output and territory size: are they matched? Ecology, 71(5), 1810–1816. Carson, R (1962) Silent Spring, Houghton Mifflin, New York. Celis‐Murillo, A Deppe, JL and Allen, MF (2009) Using soundscape recordings to estimate bird species abundance, richness, and composition. Journal of Field Ornithology, 80(1), 64–78. Chek, AA, Bogart, JP and Lougheed, SC (2003) Mating signal partitioning in multi‐species assemblages: a null model test using frogs. Ecology Letters 6(3), 235–247. Chen, J, Kher, S and Somani, A (2006) Energy efficient model for data gathering in structured multi‐ clustered wireless sensor network. Paper presented at the Performance, Computing, and Communications Conference, IPCCC, pp. 381–388. Chesmore, D (2004) Automated bioacoustic identification of species. Anais da Academia Brasileira de Ciancias, 76(2), 436–440. Collins, S (2004) Vocal fighting and flirting: the functions of birdsong, in Nature’s Music: The Science of Birdsong (eds P Marler and H Slabbekoorn), Elsevier, Oxford, pp.39–78. Connell, JH (1978) Diversity in tropical rain forests and coral reefs. Science, 199(4335), 1302–1310. Daniel, JC and Blumstein, DT (1998) A test of the acoustic adaptation hypothesis in four species of marmots. Animal Behaviour, 56(6), 1517–1528. Dawson, MRW, Charrier, I and Sturdy, CB (2006) Using an artificial neural network to classify black‐capped chickadee (Poecile atricapillus) call note types. Journal of the Acoustical Society of America, 119(5), 3161–3172. de la Torre, S and Snowdon, CT (2002) Environmental correlates of vocal communication of wild Pygmy Marmosets (Cebuella pygmaea). Animal Behaviour, 63(5), 847–856. di Castri, F and Younes T (eds) (1996) Biodiversity, Science and Development: Towards a New Partnership, CAB International, Wallingford. Dietrich, C, Palm, G, Riede, K and Schwenker, F (2004) Classification of bioacoustic time series based on the combination of global and local decisions. Pattern Recognition, 37(12), 2293–2305. Dirzo, R and Raven, PH (2003) Global state of biodiversity and loss. Annual Review of Environment and Resources, 28(1), 137–167. Dooling, R (1978) Behavior and psychophysics of hearing in birds. Journal of the Acoustical Society of America, 64(S1), S4–S4. Dooling, RJ, Brown, SD, Klump, GM and Okanoya, K (1992) Auditory perception of conspecific and heterospecific vocalizations in birds: evidence for special processes. Journal of Comparative Psychology, 106(1), (20–28.

10  Integrating Biophony into Biodiversity Measurement and Assessment

Dooling, RJ, Leek, MR and Dent, ML (1999) Temporal resolution in birds and the perception of complex sounds. Journal of the Acoustical Society of America, 105(2), 1109–1109. Dooling, RJ, Leek, MR, Gleich, O and Dent, ML (2002) Auditory temporal resolution in birds: discrimination of harmonic complexes. Journal of the Acoustical Society of America, 112(2), 748–759. Dumyahn, SL and Pijanowski, BC (2011a) Beyond noise mitigation: managing soundscapes as common‐ pool resources. Landscape Ecology, 26(9), 1311–1326. Dumyahn, SL and Pijanowski, BC (2011b) Soundscape conservation. Landscape Ecology, 26(9), 1327–1344. Emlen, JT and DeJong, MJ (1992) Counting birds: the problem of variable hearing abilities. Journal of Field Ornithology, 63(1), 26–31. Endler, JA (1992) Signals, signal conditions, and the direction of evolution. American Naturalist, 139, S125–S153. Estrin, D, Michener, W and Bonito, G (2003) Environmental Cyberinfrastructure Needs for Distributed Sensor Networks. A report from a National Science Foundation workshop, Scripps Institute of Oceanography. Farina, A (1998) Principles and Methods in Landscape Ecology: Towards a Science of Landscape, 2nd edn, Springer, Dordrecht, Farina, A (2014) Soundscape Ecology: Principles, Patterns, Methods and Applications, Springer, Dordrecht. Farina, A and Belgrano, A (2004) The eco‐field: a new paradigm for landscape ecology. Ecological Research, 19(1), 107–110. Farina, A, Lattanzi, E, Malavasi, R, Pieretti, N and Piccioli, L (2011) Avian soundscapes and cognitive landscapes: theory, application and ecological perspectives. Landscape Ecology, 26(9), 1257–1267. Fernandez‐Juricic, E, Poston, R, de Collibus, K, et al. (2005) Microhabitat selection and singing behavior patterns of male House Finches (Carpodacus mexicanus) in urban parks in a heavily urbanized landscape in the western U.S. Urban Habitats, 3(1), 4–24. Ficken, RW, Ficken, MS and Hailman, JP (1974) Temporal pattern shifts to avoid acoustic interference in singing birds. Science, 183(4126), 762–763. Fischer, J, Hammerschmidt, K, Cheney, DL and Seyfarth, RM (2002) Acoustic features of male Baboon loud calls: influences of context, age, and individuality. Journal of the Acoustical Society of America, 111(3), 1465–1474. Fuller, RA, Warren, PH and Gaston, KJ (2007) Daytime noise predicts nocturnal singing in urban robins, Biology Letters Online, 1–3. Gage, SH, Joo, W, Kasten, EP, Fox J, and Biswas, S (2015) Acoustic observations in agricultural landscapes, in The Ecology of Agricultural Ecosystems: Long‐Term Research on the Path to Sustainability (eds SK Hamilton, JE Doll and GP Robertson, Oxford University Press, New York, pp.360–377. Galeotti, P, Saino, N, Sacchi, R and Moller, A (1997) Song correlates with social context, testosterone and body condition in male Barn Swallows. Animal Behaviour, 53(4), 687–700. Gaston, KJ and Spicer, JI (1998) Biodiversity: An Introduction, Blackwell Science, Oxford. Gaston, KJ and Williams, PH (1993) Mapping the world’s species – the higher taxon approach. Biodiversity Letters, 1(1), 2–8. Gerhardt, HC and Huber, F (2002) Acoustic Communication in Insects and Anurans, University of Chicago Press, Chicago.

187

188

Ecoacoustics

Gill, FB (2007) Ornithology, 3rd edn, WH Freeman, New York. Greenwood, DD (1996) Comparing octaves, frequency ranges, and cochlear‐map curvature across species. Hearing Research, 94(1‐2), 157–162. Hamao, S, Veluz, MJS, Saitoh, T and Nishiumi, I (2008) Phylogenetic relationship and song differences between closely related Bush Warblers (Cettia seebohmi and C. diphone). Wilson Journal of Ornithology, 120(2), 268–276. Harma, A (2003) Automatic Identification of Bird Species Based on Sinusoidal Modeling of Syllables. Paper presented at the IEEE International Conference on Acoustics, Speech, and Signal Processing, Volume 5, pp.545–548. Harris, JG and Skowronski, MD (2006) Automatic Speech Processing Methods for Bioacoustic Signal Analysis: A Case Study of Cross‐Disciplinary Acoustic Research. Paper presented at the IEEE International Conference on Acoustics, Speech, and Signal Processing, Volume 5, pp.793–796. Heffner, H. and Masterton, B (1980) Hearing in glires: Domestic Rabbit, Cotton Rat, feral House Mouse, and Kangaroo Rat. Journal of the Acoustical Society of America, 68(6), 1584–1599. Heinzelman, WR, Chandrakasan, A and Balakrishnan, H (2000) Energy‐Efficient Communication Protocol for Wireless Microsensor Networks. Proceedings of the 33rd Hawaii International Conference on System Sciences, Volume 8, Washington, DC, pp.1–10. Herrera‐Montes, MI and Aide, T (2011) Impacts of traffic noise on anuran and bird communities. Urban Ecosystems, 14, 415. Hillery, CM (1984) Detection of amplitude‐modulated tones by frogs: implications for temporal processing mechanisms. Hearing Research, 14(2), 129–143. Hu, W, Bulusu, N, Chou, CT, Jha, S and Taylor, A (2005) The Design and Evaluation of a Hybrid Sensor Network for Cane‐Toad Monitoring. Proceedings of the 4th International Symposium on Information Processing in Sensor Networks, Los Angeles, California, pp.71–77. Hubbell, SP (2001) The Unified Neutral Theory of Biodiversity and Biogeography, Princeton University Press, Princeton. Hutchinson, GE (1961) The paradox of the plankton. American Naturalist, 95(882), 137–145. IPCC (2008) Climate Change 2007 – Synthesis Report: Contribution of Working Groups I, II and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. United Nations Environment Programme, Intergovernmental Panel on Climate Change, Geneva. Jensen, KK and Klokker, S (2006) Hearing sensitivity and critical ratios of Hooded Crows (Corvus corone cornix). Journal of the Acoustical Society of America, 119(2), 1269–1276. Joo, W, Gage, SH and Kasten, EP (2011) Analysis and interpretation of variability in soundscapes along an urban‐rural gradient. Landscape and Urban Planning, 103(3‐4), 259–276. Karlsson, H (2000) The Acoustic Environment as a Public Domain. Paper presented at the International Conference on Acoustic Ecology, Volume 1, pp.10–13. Kasten, EP, Gage, SH, Fox, J and Joo, W (2012) The Remote Environmental Assessment Laboratory’s acoustic library: an archive for studying soundscape ecology. Ecological Informatics, 12, 50–67.

10  Integrating Biophony into Biodiversity Measurement and Assessment

Kogan, JA and Margoliash, D (1998) Automated recognition of bird song elements from continuous recordings using dynamic time warping and hidden Markov models: a comparative study. Journal of the Acoustical Society of America, 103(4), 2185–2196. Krause, B (1987) Bioacoustics, habitat ambience and ecological balance. Whole Earth Review, 57. Krause, B (1998) Into a Wild Sanctuary: A Life in Music and Natural Sound, Heyday Books, Berkeley. Krause, B (2002) Wild Soundscapes: Discovering the Voice of the Natural World, Wilderness Press, Berkeley. Lauer, AM, Dooling, RJ, Leek, MR and Lentz, JJ (2006) Phase effects in masking by harmonic complexes in birds. Journal of the Acoustical Society of America, 119(2), 1251–1259. Lave, R, Wilson, MW, Barron, ES, et al. (2014) Intervention: critical physical geography. Canadian Geographer, 58(1), 1–10. Leger, DW and Mountjoy, DJ (2003) Geographic variation in song of the Bright‐Rumped Attila (Tyrannidae: Attila spadiceus): implications for species status. Auk, 120(1), 6974. Lévêque, C and Mounolou, JC (2003) Biodiversity, Wiley, Chichester. Lohr, B, Wright, TF and Dooling, RJ (2003) Detection and discrimination of natural calls in masking noise by birds: estimating the active space of a signal. Animal Behaviour, 65(4), 763–777. Lucas, J, Freeberg, T, Krishnan, A and Long, G (2002) A comparative study of avian auditory brainstem responses: correlations with phylogeny and vocal complexity, and seasonal effects. Journal of Comparative Physiology A, 188(11), 981–992. Lucas, J, Freeberg, T, Long, G and Krishnan, A (2007) Seasonal variation in avian auditory evoked responses to tones: a comparative analysis of Carolina Chickadees, Tufted Titmice, and White‐Breasted Nuthatches. Journal of Comparative Physiology A, 193(2), (201–215. Luther, D (2009) The influence of the acoustic community on songs of birds in a neotropical rain forest. Behavioral Ecology, 20(4), 864–871. MacArthur, RH and Wilson, EO (1967) The Theory of Island Biogeography, Princeton University Press, Princeton. MacDougall‐Shackleton, SA, Hulse, SH, Gentner, TQ and White, W (1998) Auditory scene analysis by European Starlings (Sturnus vulgaris): perceptual segregation of tone sequences. Journal of the Acoustical Society of America, 103(6), 3581–3587. Mager, JN III, Walcott, C and Piper, WH (2010) Common Loons can differentiate models of neighboring and non‐neighboring conspecifics. Journal of Field Ornithology, 81(4), 392–401. Marler, P (1967) Animal communication signals. Science, 157(3790), 769–774. Marler, P (2004) Bird calls: a cornucopia for communication, in Nature’s Music: The Science of Birdsong (eds P Marler and H Slabbekoorn), Elsevier, Oxford, pp.132–176. Marten, K, Quine, D and Marler, P (1977) Sound transmission and its significance for animal vocalization: II. tropical forest habitats. Behavioral Ecology and Sociobiology, 2(3), 291–302. Martinez Suarez, P and Moreno Jimenez, A (2005) El ruido ambiental urbano de Madrid: caracterizacion y evaluacion cuantitativa de la poblacion potencialmente afectable. Boletın de la Asociacion de Geografos Espanoles, 40, 153–180.

189

190

Ecoacoustics

Mazaris, A, Kallimanis, A, Chatzigianidis, G, Papadimitriou, K and Pantis, J (2009) Spatiotemporal analysis of an acoustic environment: interactions between landscape features and sounds. Landscape Ecology, 24(6), 817–831. McIlraith, A and Card, H (1997) Birdsong recognition using backpropagation and multivariate statistics. Signal Processing, 45(11), 2740–2748. Miller, NP (2008) US National Parks and management of park soundscapes: a review. Applied Acoustics, 69(2), 77–92. Møller, AP (1988) False alarm calls as a means of resource usurpation in the Great Tit Parus major. Ethology, 79(1), 25–30. Morton, ES (1975) Ecological sources of selection on avian sounds. American Naturalist, 109(965), 17–34. Munn, CA (1986) Birds that cry wolf. Nature, 319(6049), 143–145. Obrist, MK (1995) Flexible bat echolocation: the influence of individual, habitat and conspecifics on sonar signal design. Behavioral Ecology and Sociobiology, 36(3), 207–219. Odum, EP (1971) Fundamentals of Ecology, 3rd edn, Saunders, Philadelphia. Otter, KA, Ratcliffe, L, Njegovan, M and Fotheringham, J (2002) Importance of frequency and temporal song matching in Black‐Capped Chickadees: evidence from interactive playback. Ethology, 108(2), 181. Patricelli, GL and Blickley, JL (2006) Avian communication in urban noise: causes and consequences of vocal adjustment. Auk, 123(3), 639–649. Penna, M, Marquez, R, Bosch, J and Crespo, EG (2006) Nonoptimal propagation of advertisement calls of Midwife Toads in Iberian habitats. Journal of the Acoustical Society of America, 119(2), 1227–1237. Pereira, HM and Cooper, HD (2006) Towards the global monitoring of biodiversity change. Trends in Ecology and Evolution, 21(3), 123–129. Pierce, JR (1980) An Introduction to Information Theory: Symbols, Signals and Noise, 2nd edn, Dover Publications, New York. Pijanowski, BC, Villanueva‐Rivera, LJ, Dumyahn, S, et al. (2011) Soundscape ecology: the science of sound in the landscape. BioScience, 61(3), 203–216. Planqué, R and Slabbekoorn, H (2008) Spectral overlap in songs and temporal avoidance in a Peruvian bird assemblage. Ethology, 114(3), 262–271. Porter, J, Arzberger, P, Braun, HW, et al. (2005) Wireless sensor networks for ecology. Bioscience, 55(7), 561–572. Purvis, A and Hector, A (2000) Getting the measure of biodiversity. Nature, 405(6783), 212–219. Rabin, LA, McCowan, B, Hooper, SL and Owings, DH (2003) Anthropogenic noise and its effect on animal communication: an interface between comparative psychology and conservation biology. International Journal of Comparative Psychology, 16(2), 172–192. Reijnen, R, Foppen, R, Braak, CT and Thissen, J (1995) The effects of car traffic on breeding bird populations in woodland. III. Reduction of density in relation to the proximity of main roads. Journal of Applied Ecology, 32(1), 187–202. Rheindt, FE (2003) The impact of roads on birds: does song frequency play a role in determining susceptibility to noise pollution? Journal fur Ornithologie, 144(3), 295–306.

10  Integrating Biophony into Biodiversity Measurement and Assessment

Ridley, AR, Child, MF and Bell, MB (2007) Interspecific audience effects on the alarm‐ calling behaviour of a kleptoparasitic bird. Biology Letters, 3(6), 589–591. Riede, K (1993) Monitoring biodiversity: analysis of Amazonian rainforest sounds. Ambio, 22(8), 546–549. Riede, K (1998) Acoustic monitoring of Orthoptera and its potential for conservation. Journal of Insect Conservation, 2(3–4), 217–223. Roy, O and Vetterli, M (2006) Distributed Compression in Acoustic Sensor Networks Using Oversampled A/D Conversion. Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing, Volume 4, pp.165–168. Ryan, MJ and Brenowitz, EA (1985) The role of body size, phylogeny, and ambient noise in the evolution of bird song. American Naturalist, 126(1), 87–100. Schafer, RM (1994) The Soundscape: Our Sonic Environment and the Tuning of the World, Destiny Books, Rochester. Schook, DM, Collins, MD, Jensen, WE, Williams, PJ, Bader, NE and Parker, TH (2008) Geographic patterns of song similarity in the Dickcissel (Spiza americana). Auk, 125(4), 953–964. Selouani, SA, Kardouchi, M, Hervet, E and Roy, D (2005) Automatic Birdsong Recognition Based on Autoregressive Time‐Delay Neural Networks. Proceedings of the ICSC Congress onComputational Intelligence Methods and Applications, pp. 1–6. Seyfarth, RM and Cheney, DL (2010) Production, usage, and comprehension in animal vocalizations. Brain and Language, 115, 92–100. Shannon, C (1948) A mathematical theory of communication. Bell System Technical Journal, 27, 379–423, 623–656. Sinha, A and Chandrakasan, A (2001) Dynamic power management in wireless sensor networks. Design and Test of Computers, 18(2), 62–74. Slabbekoorn, H and Peet, M (2003) Ecology: birds sing at a higher pitch in urban noise. Nature, 424(6946), 267–267. Slabbekoorn, H, Yeh, P and Hunt, K (2007) Sound transmission and song divergence: a comparison of urban and forest acoustics. Condor, 109(1), 67–78. Smith, N (2008) Uneven Development: Nature, Capital, and the Production of Space, 3rd edn, University of Georgia Press, Atlanta. Spencer, KA, Buchanan, KL, Goldsmith, AR and Catchpole, CK (2003) Song as an honest signal of developmental stress in the Zebra Finch (Taeniopygia guttata). Hormones and Behavior, 44(2), 132–139. Storch, D, Marquet, PA and Brown, JH (2007) Introduction: Scaling biodiversity – what is the problem? in Scaling Biodiversity (eds D Storch, P Marquet and J Brown), Cambridge University Press, Cambridge, pp.1–11. Sueur, J (2002) Cicada acoustic communication: potential sound partitioning in a multispecies community from Mexico (Hemiptera: Cicadomorpha: Cicadidae). Biological Journal of the Linnean Society, 75(3), 379–394. Sueur, J and Farina, A (2015) Ecoacoustics: the ecological investigation and interpretation of environmental sound. Biosemiotics, 8(3), 493‐502. Sueur, J, Farina, A, Gasc, A, Pieretti, N and Pavoine, S (2014) Acoustic indices for biodiversity assessment and landscape investigation. Acta Acustica united with Acustica, 100(4), 772–781.

191

192

Ecoacoustics

Templeton, C and Greene, E (2007) Nuthatches eavesdrop on variations in heterospecific chickadee mobbing alarm calls. Proceedings of the National Academy of Sciences USA, 104(13), 5479–5482. Templeton, CN, Greene, E and Davis, K (2005) Allometry of alarm calls: black‐capped chickadees encode information about predator size, Science, 308(5730), 1934–1937. Tremain, SB, Swiston, KA, and Mennill, DJ (2008) Seasonal variation in acoustic signals of Pileated Woodpeckers. Wilson Journal of Ornithology, 120(3), 499–504. UNCBD (1992) Rio Convention on Biological Diversity. United Nations, Rio de Janeiro. USNAE (2010) Technology for a Quieter America, National Academies Press, Washington, DC. Villanueva‐Rivera, LJ and Pijanowski, BC (2012) Pumilio: a web‐based management system for ecological recordings. Bulletin of the Ecological Society of America, 93(1), 71–81. von Uexküll, J (2001) An introduction to umwelt. Semiotica, 134, 107–110. Wang, H, Elson, J, Girod, L, Estrin, D and Yao, K (2003) Target Classification and Localization in Habitat Monitoring, Computer Science Department , UCLA, Los Angeles. Wang, H, Chen, C, Al, A, et al. (2005) Acoustic sensor networks for woodpecker localization, in Advanced Signal Processing Algorithms, Architectures, and Implementations XV (ed. FT Luk), Volume 5910, pp.80–91. Wang, X, Wang, D, Wu, X, Wang, R and Wang, C (2007) Acoustic signals of Chinese Alligators (Alligator sinensis): social communication. Journal of the Acoustical Society of America, 121(5), 2984–2989. Whittaker, RH (1975) Communities and Ecosystems, 2nd edn, Macmillan, New York. Whittaker, RJ, Willis, KJ and Field, R (2001) Scale and species richness: towards a general, hierarchical theory of species diversity. Journal of Biogeography, 28(4), 453–470. Wood, WE and Yezerinac, SM (2006) Song Sparrow (Melospiza melodia) song varies with urban noise. Auk, 123(3), 650–659. Yepes, DL, Gomez, M, Sanchez, L and Jaramillo, S (2009) Metodologia de elaboracion de mapas acusticos como herramienta de gestion del ruido urbano – caso Medellin. Dyna, 76(158), 29–40.

193

11 Landscape Patterns and Soundscape Processes Almo Farina1 and Susan Fuller 2 1 2

Department of Pure and Applied Sciences, Urbino University, Urbino, Italy Queensland University of Technology, Brisbane, Australia

11.1 ­An Introduction to Landscape Ecology (Theories and Applications) A landscape consists of a mosaic of patches and corridors within a background matrix (Forman 1995; Forman and Godron 1986). These landscape elements include a mix of ecosystem or land use types that may be randomly distributed or regularly repeated. Structural patterns within a landscape may be the result of a range of natural processes such as gradients in physical parameters, including sunlight, temperature, rainfall, and substrate, and biotic features like species interactions, dispersal, succession, and natural disturbance regimes. Anthropogenic processes of development, agriculture, hydrological modifications, and climate change also create landscape patterns. Boundaries between landscape elements may be abrupt or diffuse, depending on the type(s) of processes that have created them and feedback loops involved. Landscape pattern is the primary factor determining functional flows and movements and there is a feedback where flows create pattern by changing processes over time (Forman 1995). As landscape ecology integrates the science of ecology with spatial pattern, there is much ecological theory that provides a core foundation for understanding landscape patterns and patch‐matrix characteristics. The following section will explore this theory and provide relevant examples which demonstrate the important implications for the number of species that can persist within increasingly fragmented habitats and the value of maintaining (and restoring) connectivity for conservation of biodiversity. Importantly, it is necessary to understand flows between remnant patches as well as the surrounding matrix if we are to manage the impact of human activities and landscape change. 11.1.1  Patch Size, Shape, and Isolation

One of the cornerstones of ecological theory is that a positive correlation exists between species richness and area. This relationship was first mathematically described through the power function model or power curve (S = CAz) by Arrhenius in 1921. The species Ecoacoustics: The Ecological Role of Sounds, First Edition. Edited by Almo Farina and Stuart H. Gage. © 2017 John Wiley & Sons Ltd. Published 2017 by John Wiley & Sons Ltd.

194

Ecoacoustics

richness of plants was determined in quadrants of variable size, and shown to increase in an approximately logarithmic manner as the area increased (Arrhenius 1921). One explanation for this relationship is the habitat diversity hypothesis (Williams 1943). As area increases, there is a greater chance that a diversity of habitats will be captured within the area. Increased habitat diversity will result in more niches that can support the co‐existence of greater number of species in an area (Hortal et al. 2009). Island biogeography theory (MacArthur and Wilson 1963) incorporates the effect of isolation and area of the island to provide an explanation for the species–area relationship. These authors proposed that the number of species on an island was the result of a dynamic equilibrium between the influx of new species to the island through colonization and the loss of species through stochastic extinction. They hypothesized that the rate of colonization of new species is negatively related to increasing isolation. However, differences in colonization rates of species to islands may not be due solely to isolation. To some degree, species behavior can dictate the direction and choice of migration destination (Lomolino 1990; Stracey and Pimm 2009). Patch shape can also influence immigration. Patches with narrow lobes may enhance dispersal between more distant patches in the landscape and enable recolonization following extinction. The orientation of the long axis of a patch relative to the orientation of a flow may enhance its use as a corridor. Migrating birds have been found to more often utilize forest patches oriented in the direction of their migration route (Gutzwiller and Anderson 1992). Island biogeography theory also predicts that island size or area is negatively related to the extinction rate of species (MacArthur and Wilson 1963). Compared with a small area, a larger area experiences a lower rate of extinction because the individual populations are assumed to be larger. Large populations have a lower risk of genetic, environmental, and demographic stochasticity. Again, patch shape may also influence extinction rate. A round form minimizes edge effects (discussed later) relative to a more elongated form and curvilinear boundaries may enhance survival as a species may be able to more effectively escape predation. However, irrespective of patch size and shape, high rates of immigration from a source area (Pulliam 1988) may lower local extinction rates on an island (the rescue hypothesis) (Brown and Kodric‐Brown 1977), a notion that has since contributed to the development of metapopulation theory (Hanski 1998). 11.1.2  Patch‐Matrix Context

To gain a more realistic perspective of landscape patterns and flows, it is necessary to move beyond a patch‐centric analysis and consider the background matrix which may, or may not, be inhospitable and ecologically irrelevant to the species in the constituent patches. Successful migration of individuals between patches and through the matrix is an important factor influencing population viability and therefore species richness (Vandermeer and Carvajal 2001). Metapopulation theory attempts to predict the dynamic consequences of migration of individuals in spatially structured populations of unstable subpopulations (Hanski 1998). Metapopulation persistence is reliant on asynchronous extinction of subpopulations and subsequent recolonization. If increased isolation (for example, through patch loss) or matrix impermeability impedes recolonization, there will be a higher chance of total population extinction (Hanski 1999). Matrix type and permeability can have profound effects on animal movement and dispersal ability (Bender and Fahrig 2005; Hein et al. 2003; Ricketts 2001). In general, if

11  Landscape Patterns and Soundscape Processes

the matrix resembles the patch habitat, a species may be more likely to traverse it, therefore positively influencing population viability and species diversity (Prevedello and Vieira 2010). The survival rate of dispersing individuals will also depend on the characteristics of the matrix (Anderson et al. 2007). For example, human infrastructure, mainly roads, has been identified as having a major impact on dispersal, through mortality (Trombulak and Frissell 2000). McAlpine et al. (2006) concluded that road density was a far greater threat than isolation (distance to nearest patch) for koalas. While matrix effects on dispersal are highly species specific (Anderson et al. 2007), it appears that an urban matrix may impede fauna movement to a greater extent than other types of matrix, for instance rural. Irrespective of this, if only a limited number of species choose (and are able) to disperse to and from a patch, there may be a negative impact on species richness in the patch. Matrix type may also influence the levels of a potential resource, thus influencing population viability and species richness. A number of studies have shown that survivorship is significantly higher in species that utilize features of the matrix as a resource (Laurance 1991; Viveiros de Castro and Fernandez 2004). Tubelis et al. (2007) concluded that birds were utilizing the pine forest matrix surrounding native forest patches as a means of supplementing their food resources. However, species that are more adept at utilizing matrix habitat are more likely to be generalists (Westerberg et al. 2005). This bias may result in a decrease in biodiversity because generalists may dominate matrix and edge environments, whereas specialist species may be disadvantaged by edge effects and even become lost from a patch (Winfree et al. 2011). An edge is defined as the boundary between a patch and the matrix or between patches of differing characteristics (Ries et al. 2004). Edges are characterized as having different biotic features (vegetation structure and species composition), as well as abiotic differences (wind, light, etc.) compared with the center of the patch and as a function of distance from the edge (Harper et al. 2005). The permeability of an edge governs the rate at which energy, matter, and organisms are able to penetrate the edge (Stamps et al. 1987). Whether an edge acts as a complete barrier or a semi‐permeable membrane depends on its structure and what is passing through it (López‐Barrera et al. 2007). The contrast between two patches also influences the permeability of an edge. High‐­contrast edges are those between two very different patch types (e.g. forest and agriculture), whereas low‐contrast edges occur between patches that are structurally and functionally similar (e.g. two different forest types). Edge effects are likely to be more pronounced in high‐ than low‐contrast edges.

11.2 ­Relationship Between Landscape Ecology and Soundscape Ecology: A Semantic Approach Landscape ecology deals with the description and analysis of complex patterns created by vegetation successional dynamics, natural disturbance regimes, and human activity that mainly occur across different geographical and temporal scales (Forman and Godron 1981). The study of landscape ecology combines well‐grounded ecological theory with spatial (geographical) analysis of the processes, patterns, and flows across a  landscape (Farina 2006). In a similar manner, we can use ecological theory to

195

196

Ecoacoustics

understand the processes (biological, physical, and anthropogenic) that create sound patterns and flows across a soundscape. 11.2.1  The Contribution of Landscape Ecology to the Development of Ecoacoustics

The landscape represents a physical container of relevant processes that have an explicit representation across a geographical space (Forman and Godron 1981, 1986). But at the same time, a landscape is also a cognitive dimension in which every animal interacts visually with the environment (Farina 2010; Farina and Napoletano 2010), continuously engaged in tracking the resources which are necessary for it to accomplish vital functions (Farina 2012). As the landscape can be perceived from an individual point of view, this fact changes the focus to an individual during its movements, and in the same fashion a soundscape is the individual perception of sounds that change according to the position and speed of movement of a listener. Such cognitive and dynamic views of landscapes and soundscapes show that common cognitive mechanisms exist that are used by species to capture information from the surrounding world using visual and acoustic signals. The theory of landscape ecology has underpinned the development of the field of soundscape ecology and ecoacoustics. In fact, as the landscape is the emergent property of a world that embraces and surrounds every species, the soundscape represents the emergent property of all the acoustic sounds that are produced by abiotic (geophonies) and biotic agents (biophonies) and by human activity (technophonies) within and across a landscape. The shape and dimension of constituent patches in a landscape and the distance between other landscape objects, such as ecotones or corridors, represent important fundamentals that characterize the specific habitats of different species. The distribution of patches in the landscape and their typology and morphology, the distribution of water bodies, and the extent of natural and man‐made barriers are all important elements that affect animal movements (Johnson et  al. 1992) as well as sound diffusion (Morton 1975). Biophonic sources are extremely sensitive to the geography of a location (altitude, aspect, slope, etc.) and weather conditions (Snell‐Rood and Badayaev 2008). For instance, a high mountain landscape is more influenced by wind regimes than a lowland area, and this has a strong effect on typology and duration of song performances of residential birds. A landscape scale is appropriate for the description of acoustic processes that emerge from the blending of geophonies, technophonies, and biophonies (Pijanowski et  al. 2011a,b). The soundscape is the acoustic dimension at which the landscape features are related and soundscape ecology is a subdiscipline of ecoacoustics in which such relationships are investigated (Sueur and Farina 2015). As in a landscape, there are different soundscape scales, from a few meters to several kilometers. At every scale, there is a different expression of the soundscape and the physical and/or biological constraints may change. In addition, it is evident that at the same spatial scale, soundscapes exhibit variability of the sonic objects at a finer spatial resolution than at the landscape scale. Furthermore, the temporal resolution and variability of a soundscape can be measured in seconds, while landscape change occurs according to disturbance regimes that take place over a longer period, and thus the temporal resolution of landscape change may be months to centuries.

11  Landscape Patterns and Soundscape Processes

11.2.2  Acoustic Heterogeneity in a Landscape Across Space and Time

The physical attributes of sound allow investigation of individual sound sources when acoustic sensors are located near the acoustic agents, or investigation of emergent patterns when the sources are more distant from the listener. In fact, a listener which moves across a landscape can perceive a large amount of acoustic heterogeneity from its surroundings. However, this heterogeneity is reduced when the perception is concentrated on distant signals. Sounds change every few meters and this increases heterogeneity. Acoustic heterogeneity creates difficulties for a listener moving in a uniform manner across a landscape, preventing the creation of accurate and persistent maps of the soundscape. Soundscapes are highly dynamic and acoustic heterogeneity varies across different scales. Close sounds are perceived as being more heterogeneous in space and time than distant sounds, which are perceived as more constant in terms of intensity and frequencies. For instance, the sound from a city located some distance away is indistinct because it is the summation of an extraordinary quantity of sounds, originating in different ways, at different times and belonging to different acoustic frequencies. This type of acoustic entity is considered to be noise or unwanted sound (Schafer 1977), and it can have serious impacts on biodiversity. The majority of species try to avoid noisy conditions or modify their behavior in response to the noise (Brumm 2004; Brumm et al. 2005). Distance reduces the quality of the acoustic signal because sound degrades as it moves through the atmosphere. This is known as sound attenuation. We can distinguish a far acoustic field, expressed by lower frequencies that persist over long distances than those with higher frequencies, and this represents a background reference sound. As we approach a specific source of sound, its distinctness increases, especially if of anthropogenic or biophonic origin. Every organism that uses sound signals to communicate and/ or explore the environment has the capacity to estimate the distance over which the sounds are traveling. This capacity is called “ranging” (Wiley 1998) and it is probably enhanced by variation in temporal heterogeneity. The sound around an individual varies according to the intensity and temporal variation of pitch. At least two levels of a soundscape may be distinguished: high intensity (close) and low intensity (far). Intermediate levels can be distinguished according to the sensitivity of individual species and their specific functions (Farina and Belgrano 2004, 2006). It is likely that animals explore their surroundings using different behavioral responses at different soundscape levels. Nearby sounds require a quick reaction (e.g. antipredation, territory defense, pair selection, etc.), while distant sounds are more likely used to select the acoustic habitat (Merchant et al. 2015). Farina (2014) has distinguished different acoustic entities within a soundscape: sonotopes, soundtopes, sonotones, and acoustic communities (Figure 11.1). Each of these entities has a function and an ecological meaning that is related to the structure and dynamic of the landscape. Sonotopes are the result of all the acoustic interactions between geophonies, technophonies, and biophonies and represent the reference sound. The combination of these three components creates a specific sonotope for each area with a distinct acoustic signature. Similar to patches within a landscape mosaic, sonotopes can be considered as sonic patches within the acoustic mosaic. The three different components of the soundscape will vary in importance depending on a range of variables such as weather conditions or the extent of human activities. In particular,

197

198

Ecoacoustics

Landscape

Soundscape

Sonotope

Soundtope

(Geophonies+Biophonies+Technophonies)

(Biophonies)

Sonotones

Acoustic communities mosaic (Interacting biophonies)

Figure 11.1  Relationship between a landscape and the acoustic objects (sonotopes, soundtopes, sonotones and acoustic communities). Source: Farina (2014).

the weather may have a great influence on the quality of the sonotope. For instance, when heavy rains or strong winds occur, they cover the majority of biophonic frequencies, reducing or impeding interindividual communication. A landscape mosaic is by definition heterogeneous and the differences between surrounding patches produce ecotonal effects in the sonotopes. Such an acoustic effect is called an acoustic ecotone or sonotone (Farina 2014). Sonotones are tension zones where the overlap of co‐occurring sonotopes creates complex acoustic configurations. We suspect that sonotones have an additional energy investment required for a species to maintain its acoustic performance at an efficient rate. For example, at road edges with high vehicle traffic, technophonies invade the margin of vegetation and many vocal species, in particular including birds, mammals, frogs, and insects, react to these technophonies by defending their territories with increased acoustic activity (Duarte et al. 2015). This phenomenon has also been investigated from other perspectives by considering such intrusion as noise (Farina 2015). From an ecoacoustic perspective, species at the sonotone experience a considerable sonic disturbance regime that creates behavioral reactions, including an intensification

11  Landscape Patterns and Soundscape Processes

Figure 11.2  Example of a interaction of a blackcap (Sylvia atricapilla) near a busy road (State Road #63) (4°13’33N, 10°07’00E, 264 m a.s.l.), 0655‐19062016. Visible in the spectrogram are the frequencies created by passing cars and the blackcap song. Source: Pieretti, N and Farina, A. Effect of traffic on the acoustic activity of birds. Manuscript in preparation. (See color plate section for the color representation of this figure.)

of acoustic signals, both calls and songs, which cause an increase in energy expenditure (Figure 11.2). This may produce a sink effect (Pulliam 1988). The presence of masking sounds can produce effects similar to those created by the abundance of food in ecotones where an increase in the density of certain species is followed by an increase in their predators (Gage et al. 1970). Traffic sounds may attract species that want to defend their territory from such an acoustic menace by concentrating singing individuals along the road edge and falsifying the territorial strategy to protect the resources present into every patrolled territory. Ecotones are often considered to be ecological traps that attract prey and predators, with a negative balance of the prey (Farina 2006, p.213), and probably sonotones play a similar role.

11.3 ­Acoustic Community and Landscape Mosaics According to Farina (2014), the soundscape is composed of different perceived objects: sonotopes (acoustic patches) and soundtopes (the biophonic component of a sonotope). Soundtopes can be further divided into subunits called acoustic communities. Recently, Farina and James (2016) defined an acoustic community as “as an aggregation of species that produces sound by using internal or extra‐body sound‐producing tools. Such communities occur in aquatic (freshwater and marine) as well as in terrestrial environments.” The mosaic of acoustic communities represents the functional biophonic components of a soundtope. The relationship between an acoustic community and a landscape is neither immediate nor direct. In fact, a landscape is composed of functional units (patches) that are aggregated geographically based on vegetation patterns and/or human use of resources. Distinct units within the landscape mosaic may be stable in the short term, but with ecological succession they will change over the medium to long term. This description does not consider the dynamics that occur within the patches and can be the complex processes that link autotrophic and heterotrophic organisms. The geographical approach to describing a landscape is in conflict with the ephemeral processes that create an acoustic community.

199

200

Ecoacoustics

A recent methodology that detects and identifies ecoacoustic events (Ecoacoustic Event Detection and Identification, EEDI) (Farina and Salutari 2016; Farina et al. 2016) has been used to investigate acoustic communities. Ecoacoustic events are considered functional units of the acoustic environment and their classification represents an important contribution to interpreting ecological complexity (for more details see Chapter 7). Moreover, ecoacoustic events are not simply a distinguishable phenomenon but represent a biosemiotic acoustic eco-field, a carrier of meaning to intercept specific resources (Farina and Belgrano 2004, 2006). According to the eco-field theory, in order to track a resource, it is necessary that a biosemiotic pathway is present that connects resources that are heterogeneous in space and time, and often cryptic, with a function that is activated by internal needs. A cognitive template creates the conditions to visualize in the landscape the eco-field associated with a specific resource. An acoustic eco-field is the result of the activity of different species that may create the conditions to assess the quality of habitat. According to this model, the entire collection of different acoustic eco-fields represents the acoustic habitat of a species. Acoustic communities are highly dynamic in space and time (Figure 11.3), and are the result of interactions between individuals of different species that spatially or temporally share a common acoustic space. This functional model cannot be represented using a geographic‐based approach, but rather by a probabilistic approach (Farina and James 2016). In other words, we have evidence that different acoustic communities are present at the same time in a landscape, but their physical (spatial) delimitation is often approximate or not possible to identify. Because of the technical constraint associated with our inability to simultaneously locate all the interacting individuals for each community, we have to be content with using a probabilistic model. This constraint hinders the identification of a direct relationship between landscape structure and animal sounds produced by different species. To overcome this difficulty, we have to consider that in any location, an individual acoustic community does not exist but is a mosaic of communities. For birds, such a mosaic is most active during the dawn chorus. Outside this period, the number and the extent of acoustic communities vary considerably. The majority of bird species are active at dawn, and this phenomenon is observable at every latitude during the breeding season. Different reasons why birds call more frequently during the dawn chorus have been proposed (Staicer et  al. 1996) and while there is little consensus on the cause, there is agreement about the commonness of the phenomenon in terrestrial and aquatic systems (Farina et al. 2015) (see Chapter 5). This fact represents the reference point for the successive estimation of the percentage of acoustic community mosaic activity or ACMA. The ACMA represents a realistic model that can be processed by a simple metric, at least for birds during the breeding season, to estimate the activity of the acoustic community mosaic around a specific sensor. However, the use of a single sensor does not allow the collection of spatial information about animals that are acoustically active at a precise moment. However, it is not a trivial exercise to deploy a matrix of acoustic sensors to establish the limits of the clusters of interacting animals. This requires a great deal of technology and field work, with many sensors set in accurate locations, calibrated and synchronized (Celis‐Murillo et al. 2009). While data collected from a single sensor may have limited spatial utility, it can be a useful tool for estimating the overall acoustic activity. To increase the reliability of this approach, it is necessary to be able to distinguish a far acoustic field from a near one. Again, we have to simplify the reality and assume that individuals are using the same vocal energy to patrol and defend a territory. A low‐intensity pattern will then indicate low activity of the community mosaic.

23 June

1

1

1

1

0.8

0.8

0.8

0.8

0.6

0.6

0.6

0.6

0.4

0.4

0.4

0.4

0.2

0.2

0.2

0 1.54

24 June

21.54

41.54

61.54

81.54

0 1.77

6.77

11.77

16.77

21.77

0.2

0 1.69

2.69

3.69

4.69

0 1.47

1

1

1

1

0.8

0.8

0.8

0.8

0.6

0.6

0.6

0.6

0.4

0.4

0.4

0.4

0.2

0.2

0.2

0 1.55

6.55

11.55

Set #1

16.55

0 1.76

3.76

5.76

SET#2

0 1.44

21.47

41.47

61.47

81.47

0.2 3.44

5.44

7.44

SET#3

9.44

11.44

0 1.45

3.45

5.45

7.45

SET#4

ACIT evenness