Quantitative Conservation of Vertebrates E-Book

Michael J. Conroy

• Herausgeber: John Wiley & Sons
• Kategorie: Wissenschaft und neue Technologien
• Sprache: Englisch

Seitenzahl: 592

Veröffentlichungsjahr: 2011

Beschreibung

This book provides a hands-on introduction to the construction and application of models to studies of vertebrate distribution, abundance, and habitat. The book is aimed at field biologists, conservation planners, and advanced undergraduate and postgraduate students who are involved with planning and analyzing conservation studies, and applying the results to conservation decisions. The book also acts as a bridge to more advanced and mathematically challenging coverage in the wider literature. Part I provides a basic background in population and community modeling. It introduces statistical models, and familiarizes the reader with important concepts in the design of monitoring and research programs. These programs provide the essential data that guide conservation decision making. Part II covers the principal methods used to estimate abundance, occupancy, demographic parameters, and community parameters, including occupancy sampling, sample counts, distance sampling, and capture-mark-recapture (for both closed and open populations). Emphasis is placed on practical aspects of designing and implementing field studies, and the proper analysis of data. Part III introduces structured decision making and adaptive management, in which predictive models are used to inform conservation decision makers on appropriate decisions in the face of uncertainty--with the goal of reducing uncertainty through monitoring and research. A detailed case study is used to illustrate each of these themes. Numerous worked examples and accompanying electronic material (on a website - href="http://www.blackwellpublishing.com/conroy">http://www.blackwellpublishing.com/conroy - and accompanying CD) provide the details of model construction and application, and data analysis.

Leseprobe

Contents

Preface

Acknowledgements

Companion website and CD-ROM

1 Introduction: the role of science in conservation

Evolution of conservation science

Conservation advocacy versus science

Good science is cost effective

Conservation under fire

Structure of the book

Part I Basic concepts in scientific investigations for conservation7

2 Using models in conservation biology

Types of models

Basic principles of modeling

3 Models of population dynamics

Balance equation of population growth

Density-independent (geometric) population growth

Density-dependent population growth

Populations: age-structured growth

Stochastic (random) effects on population growth

Spatial structure and population growth

Models for two or more species

4 Applying population models to conservation

Harvest models

Small population models

Modeling genetic effects

Including habitat or other factors in models

Use of “canned” modeling software

5 Basics of study design and analysis

Statistical models

Sampling designs

Comparison of time and space (monitoring)

Accounting for incomplete detection

Summary

Part II Conservation studies and monitoring programs

6 General principles of estimation

Census versus estimate

Indices

General relationship between counts and abundance or occupancy

Count data as indices

“Closed” versus “open” estimation

Summary

7 Occupancy (presence–absence) analysis

Design and analysis of occupancy studies

Occupancy estimation

Application and programs

Summary

8 Sample counts for abundance estimation

Complete detection assumed

Incomplete detection assumed

Study design

9 Distance sampling for estimating density and abundance

Line transect sampling

Point transect sampling

10 Capture–mark–recapture studies for estimating abundance and density

Basics of capture–mark–recapture

Two-sample (Lincoln–Petersen)

Multiple sample CMR methods (Schnabel census)

Removal sampling

Estimation of density with CMR

Other methods

Summary

11 Estimation of survival from radiotelemetry, nesting success studies, and age distributions

Basic estimation problem

“Known fate” radiotelemetry studies

Nesting success studies

Other approaches to estimating reproduction rates

Analysis of age frequency data

Modeling variation in survival and reproduction rates

Summary

12 Markߝrecapture for estimating survival, recruitment, abundance, and movement rates

Basic estimation problem

Survival estimation with tag (ring, band) recovery data

Survival estimation with capture–recapture (CJS models)

Survival, abundance, and recruitment estimation with Jolly–Seber

Combined open and closed models: the Robust Design

Estimation of movement rates from recovery and capture–recapture data

Summary

13 Analysis of habitat

Sample units and scale

Assessing habitat from Design I studies

Assessing habitat using radiotagged animals from Design II, III, and IV studies

Sampling in radiotelemetry studies

Statistical methods for assessing habitat use from radiotagged animals

Future directions

14 Estimation of species richness and other community parameters

Detection and communities

Estimation of species richness

Community richness using occupancy models

Estimating parameters of community dynamics

Summary

Part III Integrating modeling and monitoring for conservation

15 Elements of conservation decision making

Elements of decision making

Steps in decision making

Example – reserve construction

Example – habitat management for two species

Example – sustainable harvest

Summary

16 Accounting for uncertainty in conservation decisions

Sources of uncertainty

Probability as a measure of uncertainty

Expected values of outcomes

Using probability and expected value in decision making

Summary

17 Learning and adaptive management

Prediction and monitoring

Constructing the models

Updating belief: Bayes’ Theorem

Adaptive management

The value of information

Summary

18 Case study: decision modeling and adaptive management for declining grassland birds in the southeastern USA

Elements of a BQI decision model

Dealing with uncertainty

Monitoring of response following management

Adaptive management

Summary

19 Summary and recommendations

Use of models

Proper design of field studies

Proper analysis of data

Systematic approaches to decision making

Literature cited

Glossary

Appendix A: Statistical and modeling programs available on the worldwide web

Appendix B: Other internet resources

Appendix C: Modeling and statistical notation

Appendix D: Key to abundance and parameter estimation

Index

An electronic companion to the book is available in two formats: hypertext on CD, and a webpage maintained at the University of Georgia, accessible through

www.blackwellpublishing.com/conroy

See p. x for further details.

This edition first published 2009

© 2009 by Michael J. Conroy and John P. Carroll

Blackwell Publishing was acquired by John Wiley & Sons in February 2007. Blackwell’s publishing program has been merged with Wiley’s global Scientific, Technical and Medical business to form Wiley-Blackwell.

Registered office

John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK

Editorial offices

9600 Garsington Road, Oxford, OX4 2DQ, UK

The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK

111 River Street, Hoboken, NJ 07030-5774, USA

For details of our global editorial offices, for customer services and for information about how to apply for permission to reuse the copyright material in this book please see our website at www.wiley.com/wiley-blackwell.

The right of the author to be identified as the author of this work has been asserted in accordance with the Copyright, Designs and Patents Act 1988.

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 the UK Copyright, Designs and Patents Act 1988, without the prior permission of the publisher.

Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books.

Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The publisher is not associated with any product or vendor mentioned in this book. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold on the understanding that the publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional should be sought.

Library of Congress Cataloguing-in-Publication DataConroy, Michael J. (Michael James), 1952-Quantitative conservation of vertebrates / Michael J. Conroy and John P. Carroll.p. cm.Includes bibliographical references.ISBN: 978-1-4051-9098-5 (hardback : alk. paper)ISBN: 978-1-4051-8228-7 (paperback : alk. paper)1. Vertebrate populations–Mathematical models. 2. Conservation biology–Mathematical models. I. Carroll, John P. II. Title.QL607.6.C66 2009639.901′5118–dc222008029041

A catalogue record for this book is available from the British Library.

Preface

This book is intended for use by field biologists and others, including future field biologists who might be in a university course, engaged in the day-to-day study and conservation of vertebrate animals. Our goal is that conservation biologists use this book as a (with apologies to our colleague Evan Cooch) “gentle introduction” to the field of quantitative ecology. We hope to convince readers that the methods and approaches within are not the domain of mathematicians, statisticians, and computer programers, but in fact are essential tools to doing the job of conservation in the twenty-first century.

We intend this book to be used. Read it, mark it up, and take it into the field. Consult it before collecting field data, to motivate and design your monitoring and research programs (Part I), and afterwards to properly analyze and interpret the data that (hopefully) have now been properly gathered (Part II). Especially in Part II we hope that this introduction gives field biologists the basic tools and confidence to tackle the specialized books covering many of the field techniques that we cite in this book.

We draw particular attention to Part III of the book, which deals with structured decision making and adaptive management. We would actually encourage readers to skip ahead to this section and get to grips with the essential ideas of applying models to conservation decisions. Then return to Parts I and II for the details of how to build models and collect and analyze field data. This will help keep the focus where we want it – on the practical application of these methods to solving real conservation problems.

As noted on page ii, we have provided a website for the book and an accompanying CD. The website will be updated (and corrected) periodically as new developments occur and as (inevitably) mistakes are found. The CD contains all the material on the website but cannot, of course, be kept up to date as easily as the website. Both provide ready access to all the worked examples in the boxes in the book, with much more detail than allowed for in the printed book. To that end, we have provided links to software and other resources, most of it freely available on the internet. The website and CD are essential resources for the book, so we strongly encourage readers to use them to repeat for themselves the analyses performed in the book, and as a template for performing analyses of their own data.

Finally, this book is not intended as a substitute for other, more comprehensive books, notably Williams et al. (2002). To keep this book to a reasonable length, and in order to remain accessible to a less mathematically oriented audience, we chose not to cover all the methods available, and have not provided the sort of depth that more advanced references such as Williams et al. (2002) provide. So, for example, we have only briefly described such important approaches as the Pradel temporal symmetry model, Barker/Burnham joint recapture–recover models, and multi-state models (Chapter 12), but have instead provided a context in which readers might assess whether these models could be useful in their applications, and have pointed the reader to the appropriate in-depth background and relevant software. Likewise, we have given only a “barebones” treatment of the Robust Design (Chapter 12), but we have provided an example that should give readers a good idea of just how powerful an approach this is. We have, rather, emphasized practical applications and examples – which is why we again exhort readers to use the CD and website to full advantage.

Acknowledgments

This book was made possible by the support and help of many people. M.J.C. thanks Liz, Mary,and Laura for putting up with another book project; his current and past graduate students for keeping him relatively honest; and colleagues and occasional partners in crime including Richard Barker, Bob Cooper, Chris Fonnesbeck, Clint Moore, Jim Nichols, Bill Palmer, Jim Peterson, Jon Runge, Juan Carlos Senar, Jeff Thompson, Ken Williams, and of course his coauthor John Carroll, to name a few. J.P.C. thanks Eileen, Caitlin, and Sean for their patience and understanding especially during periods of frustration and chaos. Also, to his partner in this venture, Mike Conroy, who provided lots of interesting times during our tenure teaching Applied Population Dynamics together and then the Galliformes Shortcourse. J.P.C. particularly thanks the ‘UGA Gamebird Posse,’ especially Brant Faircloth, Theron Terhune, and Jeff Thompson. We all upgraded our quantitative skills together, although J.P.C. now lags well behind the rest. It was 5 or 6 years of APD students who put the authors on the course for writing this book and they deserve our thanks. As a student once wrote in a course evaluation, ‘Dr. Conroy needs to learn how to dumb down and Dr. Carroll needs to get on Ritalin.’ This pretty much sums up our working relationship. Cheers!

Companion website and CD-ROM

We have provided an electronic companion to the book in two formats: hypertext on CD, and a webpage maintained at the University of Georgia available via www.blackwellpublishing.com/conroy

The companion contains the details of all the Box examples in the book, including data input and program output where specialty programs (such as MARK or DISTANCE) are used, or in many simpler cases, spreadsheets in Microsoft Excel format.

All software (except Microsoft and other standard proprietary products) referenced herein can be obtained, usually free of charge, via the internet.

We have provided links to these programs, as well as to other modeling and statistical software that, while not directly referenced, may be useful to readers. Readers should always obtain the most up-to-date versions of these programs.

Finally, we have provided links to advanced undergraduate and graduate courses that we and colleagues have taught at the University of Georgia, as well as to short courses and workshops on topics covered in the book.

We encourage readers to periodically consult the webpage and check for updates to this material, as well as to report to us any errors that they may find.

We trust that readers will find this material useful, and suggest that it is mainly by applying the concepts in the book to real examples that readers will most benefit from this material.

1

Introduction: the role of sciencein conservation

The impetus for this book began as the result of a rather fortunate convergence of the careers of the authors at the University of Georgia. Although we were educated in traditional wildlife management programs during the 1970s and 1980s, we both developed an interest in what is now better defined as conservation biology. Interestingly, we underwent an evolution in our thinking, leading to similar ideas relative to what we perceived as weaknesses in our own profession and to how the creation of conservation biology as a profession, while addressing some of these weaknesses, fell short in many areas. We have also become increasingly involved in international issues in wildlife conservation, leading to further career intersections with other collaborators. Indeed, we have discovered that our interest in mixing conservation and science transcends political boundaries and sub-disciplines.

Evolution of conservation science

The integration of science and conservation of wildlife has quite a long history and is found in many forms. Game management in Europe and North America is based on the fundamentals of agricultural management and animal husbandry. This form of conservation biology is essentially the treatment of stocks of wild animals as domestic livestock and has evolved over hundreds of years. In both North America and Europe, wildlife management as a profession developed over much of the twentieth century following a somewhat parallel course that focused on particular species or groups of species and their management. The resulting body of literature and understanding of the population dynamics of those species and their management is enormous and some of the best information is available on vertebrates.

A second development occurred during the latter part of the twentieth century as interest and concern for the diversity of wildlife in mainly tropical parts of the world moved to the forefront. Scientists who worked predominantly in the area of ecology theory began several attempts to integrate ecology as a science with biological conservation. Driven in large part by North American and Australian scientists and coming to fruition by the late 1980s, we see the wholesale movement of scientists who traditionally dealt with empirical questions in ecology adopt an additional strategy concerned with the conservation of biological diversity.

The above developments resulted in several scientific disciplines, each with different strengths, converging to form scientific conservation biology. We believe that each discipline brings different strengths to conservation science. For example, wildlife management in North America has an excellent track record of applying scientific research to management and policy making. By contrast, the discipline of conservation biology has generally excelled at integrating ecological principles and conservation. The third important component here is the popularization of conservation among the general public which has resulted in an enormous influence of popular culture and activism on conservation and biodiversity management.

These developments then leave us with two scientific disciplines – wildlife (and/or game) management and its sister profession conservation biology. These disciplines can aptly be described by the general heading applied ecology, and are driven in part by non-scientific goals. This creates an interesting and sometimes complex series of relationships that can affect the ability of professional “applied ecologists” to strive toward their scientific objectives of obtaining reliable knowledge. We encounter several issues that are critically important at this juncture. First, as with any applied or endpoint-driven research, we must be particularly careful that our research does not simply become a series of self-fulfilling prophecies. Just as in theoretical–ecological research, our preconceptions about how systems operate must not cloud our ability to undertake objective research. In many ways the goal objectivity is easier to attain in theoretical research, because the results of theoretical1 research might only involve individual egos and career development, rather than ecological systems and biodiversity that we as individuals and conservation biologists hold dear to us. Over the course of history in scientific endeavors someone who develops some “new” theory would be under some pressure to defend the theory and other scientists might strive to find evidence to falsify it. These traditional scientific tensions are also important in applied research; however, there are now the added pressures created by outside forces from those having a stake in the outcome of research. This is because conservation scientists operate within a socio-economic-political “real world” that includes other values and tradeoffs. Even with a sympathetic public, conservation scientists and managers must act responsibly to allow policy makers to make the best decisions possible, often with limited resources and competing demands.

Conservation advocacy versus science

We distinguish between conservation advocacy – where conservationists become directly involved in promoting policies relative to biological diversity – and conservation science – which uses science to help society make more informed decisions. The latter is the target of this book. We believe that by adopting a scientific approach, not only is science better served, but also in the long-term conservation will be better served. The task is to simultaneously increase our understanding of systems in a dynamic world and to provide decision makers with the necessary information. This is why we believe that modeling approaches and adaptive management are critical components of the conservation research “system.”

Fig. 1.1Classification of the relationship between data collection and understanding of systems. In theory we would like all of our conservation questions and issues to move into Box 3, where we have good data and good understanding of the system. Box 1 represents poor use of conservation effort and money. Boxes 2 and 4 represent the place where conservation biologists are starting their research on a particular issue.

In this book we will often use models to summarize how we think a particular population or other resource behaves and might respond to various factors, including management. Our models will involve a combination of (i) our understanding of basic biological processes, and (ii) data available to support the various model components (Figure 1.1, from Williams et al. 2002, p. 112). Typically we will need to increase each of these two factors, as represented by the x and y axes of the graph (Figure 1.1). On the x axis, we wish to move to the right, which represents increasing understanding of the mechanisms driving a particular system. On the y axis, we move vertically as we increase the amount of data available on the system or parts of it. The box itself represents general areas of data quantity and system understanding. Ideally we should be in Box 3, where we have good amounts of empirical data and also good understanding of how particular systems operate. This is why we can use our understanding of gravity, material properties, and other components of physics to build bridges – which generally do not fail. Although this works well in many of the physical sciences, the complexity of combining biological and social/political issues makes this a difficult direction to move in for conservation questions or issues. More likely we are working in Boxes 2 or 4, where we lack data, knowledge, or both. In fact, in dealing with some conservation questions and species outside of the charismatic megafauna or well-studied game species, we are often starting in Box 4 with the proverbial “blank slate.” Making matters worse is the fact that biologists are often in the position where policy and/or management recommendations are expected after a single inadequately funded study. Although never ideal, this is political reality. Beginning in Box 2 is slightly better because we can use our base knowledge of similar systems to hopefully give us a stronger starting point. In both of these scenarios (Boxes 2 and 4) modeling approaches combined with good data collection will be useful. In wildlife conservation and tradition game or wildlife management in North America and Europe we may also be operating in Box 1. We may have high-quality and long-term biological data, but in many cases our understanding of the mechanisms driving systems for issues we are interested in tackling is still lacking. The integration of science in conservation management will ultimately provide the foundation for more informed conservation decisions and management.

Good science is cost effective

One of the common issues in conservation biology is the problem of inadequate funding. Outside of a few areas of conservation that can garner large amounts of money, most conservation biologists are faced with enormous questions and tasks, but limited time and financial resources. Even in programs that are relatively well funded, such as game management in North America and Europe, the scale at which many biological questions should be addressed and the resources that are available often are quite disparate. Conservation research competes for limited funding with conservation implementation (“management”). This means that if funding for research increases then funding for management decreases and vice versa. The only way to “win” at this zero-sum game is to improve efficiency. As we will argue in a number of places in this book, bad research is almost worse than no research. It can lead to wrong conclusions and wrong management.

In this book we argue that poorly designed conservation research projects also steal resources from conservation. That is, spending money on bad science not only wastes that money, but also takes money from good science and good management. Thus good science combined with improved efficiency will yield better conservation.

Conservation under fire

More and more frequently we are faced with skeptical and even hostile groups, who demand that conservationists “prove” claims of adverse impacts of desired development, ecological benefits of restrictions of forest logging or other resource consumption, proposed reserve systems, or declaring a species endangered, to name a few. While it can never be possible to “prove” (in the logical sense) such assertions, it is possible to collect and analyze data in such a way that the evidence so provided is repeatable and defensible. Conversely, data collected or analyzed in an unscientific way lead to conclusions that, while perhaps intuitively reasonable, are not repeatable, and will not stand up to scientific scrutiny. Increasingly the opponents of conservation projects are technically informed, and will eagerly reveal conclusions made by the conservation community that are based on flawed approaches. Here we emphasize that ethical, scientific conservation includes the honest reporting of study and data flaws, so that results may be appropriately interpreted. Improper reporting of results, especially to exaggerate certitude of estimates or effects, is both unethical and, in the long term, counter-productive, because when (inevitably) discovered the resulting loss of credibility can be devastating (Beier et al. 2006; Conroy et al. 2006).

Structure of the book

We envision this book as a practical and hands-on resource for field biologists. This book should be analogous to the field identification guides. It is the book you take with you and use all the time, but is not the one where you go to obtain the in-depth theory or mathematical derivations. We hope it complements some recent volumes, such as Williams et al. (2002), in assisting practitioners and students. We also envision the book being used in short courses for field conservationists. In fact, the impetus for this book came as a result of the participation of J.P.C. and M.J.C. in development of a week-long short course following the main conference in each of the last three International Galliformes Symposia.

Part I covers mainly the background we believe all biologists should review when presented with a conservation problem or question and asked to develop a research program. Chapter 2 provides some basic concepts in modeling. This is not “ugly” and complex modeling that most field biologists fear, but practical modeling that assists us in problem solving. Chapter 3 is a review and application of some basic population models. Chapter 4 deals with the issues of applying models to conservation questions. Chapter 5 provides a basic review of study design. Again, this part of the book is setting the stage for couching conservation questions in a way that makes our research more scientifically sound, economically efficient, and defendable.

Part II moves on to those topics of most importance to field biologists in collecting appropriate data in answering conservation questions. In Chapter 6, we begin with the general principles of estimation. Chapter 7 is a basic overview of occupancy studies. We believe that occupancy research is underutilized, but will eventually be viewed as one of the most important techniques in conservation. Chapter 8 covers the estimation of abundance from sample counts, and introduced the importance issue of incomplete detection, a recurrent theme through the book. Chapters 9 covers the basic principles of distance sampling, including line transect and point counts (the latter are also now called point transects). Chapter 10 provides background on mark-recapture (re-sighting) and mark-removal sampling in abundance estimation. Chapter 11 focuses on the estimation of demographic rates using data from radio-telemetry, nesting success, and age distributions. Chapter 12 expands on the issues of demographic parameters by incorporating some aspects of Chapters 10 and 11. Chapter 13 deals with the issue of habitat use and selection. Finally, in Chapter 14 we touch on some sampling and estimation issues for wildlife communities.

In Part III we begin to apply modeling and estimation tools to conservation decision making. In Chapter 15 we describe how conservation goals can be combined with predictive models and used as tools for decision making. In Chapter 16 we deal with issues of uncertainty in research and conservation decision making. We remind readers that in the real world we are faced with profound uncertainties, in part because nature cannot be controlled, but also because of our incomplete understanding of how ecological systems work. This leads on to Chapter 17, in which we show how monitoring information can be integrated into decision making, leading to adaptive management. In Chapter 18 we illustrate many of the principles of the book via an example of conservation of grassland birds in North America. Chapter 19 provides a short summary of the book.

We also provide several appendices that we hope readers will find useful. Because many readers will be familiar with some but not all the terminology we use, in the Glossary we provide a comprehensive list of terms. See p. ii of this book for numerical examples in electronic form with a detailed accompanying narrative. In Appendices A and B we provide links to sites where software and other resources can be obtained, much of it at no cost. In Appendix C we provide a comprehensive explanation and cross-referencing for modeling and statistical notation. Finally, in Appendix D we provide a dichotomous key for abundance and parameter estimation that can be used to assist in identifying appropriate estimation techniques, in much the same way that taxonomic keys are used to aid in animal or plant identification.

We especially hope that the chapters in this book give field conservationists the courage to tackle some new ways of viewing conservation problems. This is where we believe this book is most useful – in taking the fear out of quantitative and modeling approaches to conservation, and making field conservationists realize they are not “black boxes” that are to be relegated to “systems ecologists” locked away in an office somewhere.

1Interestingly, theoretical biologists now find that outside influences are very much invading their realm, including recent debates involving religious organizations in the USA and other countries over evolutionary theory and natural selection.

Part I

Basic concepts in scientific investigations for conservation

3

Models of population dynamics

In this chapter, we apply the general ideas of modeling developed in Chapter 2 to the specific problem of modeling population dynamics. Here, our important state variables will usually be summary statistics of some kind, typically of the population at a particular time and location. By dynamics, we simply mean how abundance, age structure, distribution, or other state variables change through time. Our model will then describe the rates at which these state variables change through time, and will include such quantities as birth, death, and movement rates.