EURING Newsletter - Volume 2, December 1998

EURING MEETINGS

Undoubtedly among the most important EURING initiatives, the analytical meetings have been able to involve statisticians and ornithologists in the joint effort of making the best possible use of mark-recapture data in the study of bird population dynamics. The last of these meeting, EURING ’97, was organised and kindly hosted by the British Trust for Ornithology. Gavin Siriwardena reports here on this successful and important conference.

REPORT ON THE EURING '97 CONFERENCE
‘LARGE-SCALE STUDIES OF MARKED BIRDS’

By Gavin Siriwardena

BTO, NATIONAL CENTRE FOR ORNITHOLOGY, THE NUNNERY, THETFORD, NORFOLK, IP24 2PU, UK
(Email: gavin.siriwardena@bto.org)

Since the inception of national ringing schemes worldwide, numerous insights into species' natural histories have been gained from straightforward examinations of ring-recovery data. Ringing has told us, for example, that many British passerines migrate across the Sahara desert (a fact that is still almost unbelievable when one looks at, say, a Whitethroat) and that Fulmars can live for over thirty years.

Such qualitative facts can easily be extracted from ring recoveries, but more quantitative data are much harder to produce: we need to find the methods by which the (relatively) tiny numbers of birds ringed can most effectively be used to reflect reality at the population or species level. Finding better ways to answer questions of scientific or conservation interest such as "are tree sparrows likely to be less long lived now than they were ten years ago?" has occupied some of the world's top statisticians for over twenty years.

The regular Technical Meetings organised by the European Union for Bird Ringing (EURING) represent the major world forum for analysts of ringing data, and they promote the development and application of appropriate statistical methods by bringing together statisticians and ornithologists in mutually beneficial exchanges of ideas. The sixth EURING Technical Meeting took place between 7-12 April 1997 at the University of East Anglia, and was hosted by the BTO. EURING '97 attracted over 70 scientists from 19 countries (including the USA, New Zealand, Latvia and Spain), who were joined by several BTO research staff.

The talks and posters presented covered methodological issues, new computer software and applications of the methods to real ornithological problems. There was also a short course in which the statisticians led us ecologists (via some occasionally terrifying maths) to a better understanding of the methods currently favoured for the analysis of movements and survival.

Basic statistical methods for calculating annual survival rates from mark-recapture (such as CES) and ring-recovery data have existed for several decades, and have been adapted and improved upon over time to allow closer approximations to reality (including, for example, consideration of possible changes in the probability of ringed birds being found and reported after death). New analytical procedures presented at EURING '97 included a protocol for mark-recapture studies allowing for transitory behaviour, i.e. allowing the separation of birds not recaptured due to death and those not recaptured because they have emigrated (Mike Conroy, USA). A technique allowing the inclusion of recapture, resighting (such as of colour rings) and ring-recovery data in the same model, so that the data from all possible sources could be combined to estimate survival, was presented by Richard Barker (New Zealand), and Roger Pradel (France) showed how the models used for capture-recapture studies could be adapted to investigate the recruitment of independent young into breeding populations (this phase of the life cycle is in general very difficult to study).

Several excellent new or updated software packages were demonstrated by their authors, showing how more and more complex analyses of variations in survival (among other parameters) are becoming easier and easier, year-by-year, to do. In particular, the new MARK program (Gary White, USA) not only allows just about any type of recovery or recapture analysis to be conducted, but is also written specifically for the extra-user-friendly Windows '95 interface, and is free over the World Wide Web!

The BTO's role in the biological-statistical collaboration that EURING Technical Meetings represent is very much at the "business end", putting the models and software into use with the huge data sets amassed by ringers (and those who have reported ringed dead birds) over the years. David Thomson of the BTO told the conference how his analyses of British ring-recovery data have shown that there has been a fall in the first year survival of Song Thrushes since the 1970s, and that this factor alone may well have caused the population decline shown by the Common Bird Census. This kind of study is particularly significant given that an equivalent would be impossible in most other parts of the world: in North America, for example, the sheer size of the country and its low human population density mean that although many rings are put on, they are only very rarely recovered from non-quarry species.

Chris Wernham talked about a more applied use of BTO ringing data, investigating the causes of the recent inland expansion of Cormorants, and the novel methods she has had to use to try to discover which traditional breeding areas have acted as the source of the influx (final results were still unknown when the talk was given). Other applied studies included one by Steven Piper (South Africa), who told of how colour ring resighting data had been used to find firstly that the threatened Cape Griffon Vulture tended to have very poor survival in the period immediately following fledging, and secondly that the provision of supplementary food at this time (in the form of carcasses) dramatically increased this survival rate. Michael Samuel (USA) then described how mark-resighting, radio-tracking and ring-recovery methods had been used to show that avian cholera regularly affected Lesser Snow Goose populations significantly via an effect on survival, and that vaccination against the disease improved the prospects of individual geese (although it was not 100% effective).

Several other speakers talked about studies with results directly relevant to ringing in practice. For example, David DeSante (USA) showed how data from the US equivalent of the CES scheme (known as MAPS: Monitoring Avian Productivity and Survival) could be used to produce good measures of annual productivity, which in turn could lead to strong predictions of future population trends at both the site-local and regional scales. Juan Carlos Senar (Spain) talked about an ingenious study in which Great Tits' trap-shyness towards baited funnel traps was tested for using both recaptures and video-taped resightings at a feeder adjacent to the trap. The results suggested that yearlings are more likely to be recaptured than adults, and adult males more likely than females, but that the different capture probabilities are due to the birds' dependence on supplementary food, and not to trap-shyness.

Finally, Rhys Green (RSPB) presented a valuable example of how useful information on the survival and dispersal of a very rare bird, the Corncrake, could be gleaned from ringing data despite the fact that (necessarily) very few Corncrakes are ringed, and only tiny numbers recovered. It is important that even when sample sizes would be considered too small by conventional wisdom, as is frequently the case for rare and declining species, data from a well-directed ringing programme can still answer significant conservation questions. EURING '97 was a great success, and the BTO organizers received many positive reactions from those who attended.

The formal and informal communication between the statisticians and biologists has led immediately, at least at the BTO, to several novel methods which will allow us to conduct new and better analyses of the data sets the Trust's members have generated. In this way the links forged at EURING '97 (and the past Technical Meetings) will continue to help improve the range and quality of the information we can provide for the science in conservation. Anyone interested in further details of the presentations at EURING '97 is directed to the conference proceedings which will be published early next year in Bird Study (spoken papers) and The Ring (poster papers).

Proceedings of EURING Technical Conferences

1. Wageningen, The Netherlands, 4-7 March 1986 North, P.M. (Ed.) 1987. Ringing recovery analytical methods. Acta Ornithologica 23, 1. Available from British Trust for Ornithology, The Nunnery, Thetford, Norfolk, IP24 2PU, UK, price £10.50.

2. Sempach, Switzerland, 12-14 April 1989 North, P.M. (Ed.) 1990. The statistical investigation of avian population dynamics using data from ringing recoveries and live recaptures of marked birds (ed. P.M. North). The Ring (1990) 13, 1 & 2. Business address: Przebendowo, 84-210 Choczewo, Poland.

3. Montpellier, France, April 1992 Marked individuals in the study of bird population (ed. J.-D. Lebreton & P.M. North) (1993) Birkhauser Verlag, Basel. ISBN 3-7643-2780-4 (Basel, Switzerland).

4. Patuxent, USA, 19-24 September 1994 Statistics and Ornithology (ed. P.M. North & J.D. Nichols). Journal of Applied Statistics (1995) 22, 5 & 6. ISSN 0266-4763. Carfax Publishing Company, PO Box 25, Abingdon, Oxfordshire, OX14 3UE, UK.

5. Norwich, UK, 7-12 April 1997 Proceedings to be published in 1998 as a special supplement to Bird Study.

Back to index of EURING Newsletter Volume 2