Bird Census Techniques PDF
Preview Bird Census Techniques
Bird Census Techniques Colin J. Bibby, Neil D. Burgess Royal Society for the Protection of Birds The Lodge Sandy Bedfordshire and David A. Hill British Trust for Ornithology The Nunnery Nunnery Place Thetford Norfolk Illustrated by Sandra Lambton, RSPB Published for the British Trust for Ornithology and the Royal Society British Trust for Ornithology for the Protection of Birds ACADEMIC PRESS Harcourt Brace & Company, Publishers London · San Diego · New York Boston · Sydney · Tokyo · Toronto ACADEMIC PRESS LIMITED 24-28 Oval Road London, NW1 7DX US edition published by ACADEMIC PRESS INC. San Diego, CA 92101 Copyright © 1992 by ACADEMIC PRESS LIMITED Third printing 1993 All Rights Reserved No part of this book may be reproduced in any form, by photostat, microfilm, or any other means, without written permission from the publishers This book is printed on acid-free paper A catalogue record for this book is available from the British Library ISBN 0-12-095830-9 Typeset by Paston Press Ltd, Loddon, Norfolk Printed in Great Britain by The University Press, Cambridge Acknowledgements The authors would like to thank in particular Dr Rob Fuller of the BTO for critically reading the whole book, Mrs Sandra Lambton of the RSPB for drawing the majority of the figures and Dr Jeremy Greenwood of the BTO for reading the proofs. The following critically read drafts of various chap ters: Dr Adrian del Nevo (RSPB), Dr James Cadbury (RSPB), Mr Jeff Kirby (Wildfowl and Wetlands Trust), Mr Leslie Street (RSPB), Mr John Mar- chant (BTO), Mr Dave Allen (RSPB), Mr Geoff Welch (RSPB), Mr John Wilson (RSPB) and Mrs Diana Ward (RSPB). We thank them for their improving comments, however, any errors remain our own. Mrs Anita McClune and Mrs Helen Morrow (RSPB), and Dr Rowena Langston (BTO) assisted with administrative aspects of the production. Mr Ian Dawson and Mrs Lynn Giddings (RSPB library) tracked down many references and figures. Neil Burgess and Colin Bibby acknowledge the institutional support of the RSPB and David Hill acknowledges the institu tional support of the BTO. The following also provided advice and criticism: Dr Gareth Thomas (RSPB), Dr Graham Hirons (RSPB), Dr Ceri Evans (RSPB), Dr John Cayford (RSPB), Mr Chris Mead (BTO), Dr Will Peach (BTO), Dr Steven Carter (BTO), Dr Paul Green (BTO), Dr Peter Robertson (Game Conser vancy Trust), Dr Nigel Clark (BTO), Dr Steve Tapper (BTO), Dr Dick Potts (Game Conservancy Trust) and Mr Robert Petley-Jones (English Nature). The following publishers gave permission to reproduce figures: Academic Press Ltd, American Ornithologists Union, British Trust for Ornithology, Blackwell Scientific Publications Ltd, British Birds Ltd, British Ecological Society, Cambridge University Press, Canadian Wildlife Service, Chapman and Hall, Cooper Ornithological Society, Devon Birdwatching and Preser vation Society, Game Conservancy Trust, Gauthier-Villars, Harcourt Brace Jovanovich, Institute of Terrestrial Ecology, Macmillan Magazines Ltd, Ornis Scandinavica, T & AD Poyser Ltd, Royal Society for the Protection of Birds, The Wildlife Society, Wildfowl and Wetlands Trust, US Fish and Wildlife Service. V About the Authors Colin Bibby is Director of Research at the International Council for Bird Preservation where he is interested in the role of birds as indicators for global biodiversity conservation. While Head of Conservation Science at the Royal Society for the Protection of Birds, he was a co-author of Red Data Birds in Britain. In both capacities, he has been struck by the small number of birds of conservation concern which have been counted adequately. He has counted birds in Britain and Europe, as a professional, as an amateur participant, and as an organiser of surveys for the British Trust for Ornithol ogy. He was motivated to start this book by the belief that bird-watchers would contribute more to conservation if they put more effort into counting birds, but lack of guidance on methods was a handicap. Neil Burgess graduated in Botany from the University of Bristol in 1983. He undertook a PhD on the evolution of the earliest land plants (400-430 million years before present) at the Natural History Museum in London, and Cardiff University in Wales from 1983 to 1987. He worked in Ecology, Research and Advisory Departments at the RSPB until 1991, producing botanical surveys of several RSPB reserves, and case studies of the methods and results of management of British habitats for birds (particularly wetland and heathland habitats). Since 1989 he has also been involved in scientific and conservation work in East Africa and since 1991 he has been employed by the RSPB International Department to manage conservation projects in Africa. This work forms part of the International Council for Bird Preser vation's Africa Programme. David Hill received his doctorate on the population ecology of wildfowl from the Edward Grey Institute of Field Ornithology at Oxford in 1982 and moved to the Game Conservancy as Head of Pheasant Research. Much of the work involved studies of marked birds, their behaviour and habitat preferences in relation to land use, largely farming, forestry and conser vation. In 1987 he joined the Royal Society for the Protection of Birds as Senior Ecologist, a post which involved setting up monitoring and experi- VI About the Authors VII ments on reserves, bringing him into close contact with sampling different bird species and habitats. In 1989 he joined the British Trust for Ornithology and established their Research, Development and Advisory Service, with responsibility for setting up and running research contracts on estuaries, farmland, woodland and uplands, using the methods outlined in this book. He is a council member of the British Ecological Society and a member of the Scientific Advisory Committee of the Wildfowl and Wetlands Trust. He has been interested in wildlife since as early as he can remember and has published scientific papers and books on the ecology of gamebirds, wildfowl, estuary waders, bird communities of native pine forest and grazed forests, insects in coppiced woodland, population dynamics, population and resource modelling, and bird distributions in relation to land cover determined by satellite imagery. Since leaving the BTO in 1992, he now runs an ecological consultancy, Ecoscope Applied Ecologists. The idea for this book came from a general perception amongst field workers of the British Trust for Ornithology (BTO) and Royal Society for the Protection of Birds (RSPB), that the methodology for various bird census methods was scattered throughout the scientific literature and difficult to access. Indeed many field-staff or volunteers were only familiar with one or two methods and there was often little attempt to standardise methods or to check that the counting was being undertaken in a systematic manner over wide geographical areas. It was also evident that there was a need amongst volunteers, junior researchers, students and scientists in the developing world for a practical guide synthesising all aspects of the various methods of counting birds, their uses, and the things which ought to be considered before, during and after counting birds. Methods described in the literature for counting birds are many and various. In this book we have devoted whole chapters to the most widely used and most suitable methods, and attempted to amalgamate other count ing methodologies into major groups. Examples of the use of methods are provided wherever possible and the relative value of various approaches for answering specific questions is also addressed. We have not attempted to obtain and review all the available literature on bird counting. Instead we have extracted examples in the hope of providing guidance and making workers more aware of the problems involved in bird counting. XVII 1 Purpose and Design in Counting Birds Introduction Birds are counted for a wide variety of reasons by a bewildering range of methods. This guide is offered largely to people lacking the time or facilities to read and assess the extensive and often conflicting bird census literature. For a critical review of methods, the reader is referred to Verner (1985). Ralph and Scott's (1981) book contains a wide collection of papers on many aspects of the problems of counting birds, and Wiens (1989) provides abundant illustration of the extent to which methodology is critical to the valid interpretation of the results of bird counts. A pessimist reading some of this material would conclude that bird counting is so unreliable as to be of limited value. We have tried to take a more positive view. There are practical reasons for counting birds and the pitfalls in methods and study design can be avoided. It is hence possible, with adequate care and awareness of the possible limitations of the methods used, to produce valuable results. Appropriate methods for a particular study become more obvious if there is a clear purpose, specified in advance. A recurrent theme of good study design is that methods must be tailored to aims. It is a mistake to rush to the field and start counting without prior thought. Many bird counts have no doubt been done at considerable cost of time in the field but have turned out to be just a waste of time. Results may fail to meet the aims if these are not properly understood until it comes to analysis and writing. Writing, in the sense of communicating the results to someone else, should be the final intention of any study. Because appropriate census methods are so much related to aims, it is worth giving some thought to the classes of possible aims. Questions about numbers of birds vary greatly in scale. Are the results intended to apply to a wide geographical area or to a single site? Are many species involved or just one? Are accurate counts needed or will relative counts or presence and absence data suffice? The available effort for counting is usually limited but may be adequate for studies of a single site or species which can be more intensive than is possible if many species or large areas are involved. Accurate counts are often very difficult to obtain but for many purposes are not really needed. The key to a good study lies in recognising what kinds of data are required and understanding the pitfalls of the possible 1 2 Bird Census Techniques counting methods. It is easy to agonise over errors of 10-20% when it may often be enough that counts are right within 100-200%. Questions about accuracy are so important that the next chapter is devoted to them. In truth, the perfect bird count probably does not exist, but this need not prevent the extraction of useful results from a good study. For a review of data analysis and the design of experiments in ornithology the reader is referred to James and McCulloch (1985) and Hairston (1989). Stock-taking The simplest aims are of a kind that ask what birds occur here. Such studies might be used for evaluating poorly known sites or for setting a baseline prior to fuller study or experiment. An answer that is little more than a species list, perhaps with gross approximations of numbers between common and rare, may be sufficient. Elaborate methods may not be necessary and it is surpris ing how informative such studies can be. On a wide scale, such are the aims and methods of atlas studies, but similar work is possible at a more local level with a finer grid (Chapter 9). The general presumption of such an approach is that presence or absence can be detected with some reliability given enough effort. Consideration might be given to problems of counting particu lar bird species (Chapter 7) or dealing with large numbers (Chapter 8). It will almost never be possible to do equally well for all species. Indeed, such an aim is almost hopeless; some birds are very difficult to detect, let alone count. Stock-taking aims are very often appropriate for studying single species on an extensive scale. The methods may be fairly crude but nonetheless valu able, especially if it is reasonable to expect to count the whole population of a wide area. Many surveys of single species requiring help from a large number of people have taken such an approach. It is very difficult to know how accurate the results are because both gaps in the coverage and flaws in the methods may be poorly known. A common assumption is that features of interest will be clear enough to survive some deficiencies of method. Thus population levels of Herons (Reynolds 1979) and Rooks (Sage and Vernon 1978) have been monitored by fairly simple counts and with useful, even if not totally accurate, results. Waders which are difficult to count have been studied both in breeding (Smith 1983) and non-breeding (Prater 1981) habitats. The ideal in such studies is to have a uniform effort in different areas. Failing this, the effort needs to be known, especially if some places are not visited at all (Box 1.1). Purpose and Design in Counting Birds 3 Population counts with varying effort. (a) 40 r (b) 40, 1 j Wi 1 » 30 [ * k P- 1 0 10 20 30 40 50 -· r- • : I ·· 1 10 20 30 40 A hypothetical county happens to be a perfect rectangle and the recorder is interested in describing the Magpie population. Three different approaches are tried in succession. (a) Map of all Magpie nest areas in the past 5 years: what does it show? A possibility is that 10 km squares 01 and 02 are good for Magpies which are generally rare in the county. Perhaps there are about 20 pairs in the whole county allowing for a few which have been missed? (b) Sample survey, five randomly selected 10 km squares were fully covered and the 12 nests marked were found. What does this mean? Square 02 now looks less special. Four nests were also found in square 30 where Magpies had not been recorded casually as shown above. The total population for the county can be estimated from this survey as 12 x 20/5 = 48 pairs. These squares could be re-visited in future years to assess population changes. (c) A full survey, this shows that there were actually 50 pairs. Assuming that the method was accurate, this figure is correct. The sample survey took only a quarter o fthe effort and produced quite a good answer for the total population but did not, of course, actually locate most of the pairs. The sample survey was good enough to get total numbers but not sufficient if location was also important. The data from the full survey could be analysed to explain variation of density in terms of habitat. Example (a) was totally wrong. This was because the ringing group based in the town marked with a circle was very keen on Magpies and submitted all the records. No one else thought Magpies worthy of note! 4 Bird Census Techniques Mapping census and habitat map. Yellowhammer territories lb) 1 ί 'f-^\ ^\^ l ^üt\ v ~^rk^^ *?*·>^ \ y >s T">-^V I \ \ %/? T l\ \ 1 >\ / ■ V3g j^>* 1 1 ■ Powerlinei \ yi?c\ *V. Unmarked field boundaries ^ΚΧ ^ "i"i Pott and wire fence v J*T£^<<£l^ θΞϋ- HedBe ^ V^ -—— Hedge remnant (a) Map of Yellowhammer territories on a mapped plot (Williamson 1968). It means very little on its own if you happen not to know the site where it was done. (b) It makes more sense when some key habitat features such as field boundaries have also been mapped in. Yellowhammer seem to prefer territories along hedges or other field boundaries, with few in the centres of fields.
The list of books you might like
