Evaluating the potential effects of capturing and handling on subsequent observations of a migratory passerine through individual acoustic monitoring

Many aspects of the ecology and behaviour of birds can only be studied and understood through the identification of marked individuals. Most commonly, individual recognition is achieved by fitting rings that can either be read during recapture or, in the case of colour rings, viewed in the field. The fitting of rings, together with the collection of important body measurements and samples used in DNA studies, requires the capture and handling of individuals, a process that might alter the behaviour of an individual or change its future prospects.

It is important that we understand the potential for such effects to bias our understanding of behaviour, movements, survival rates and other important aspects of bird ecology. The ideal approach would be to look for any possible effects by comparing a sample of birds that had been captured and handled with a sample that had never been caught or handled. The problem with this approach is how to identify those individuals who have never been caught, since you can only identify them as individuals by marking them in some way, and you can only mark them by catching and handling them.

A study led by Tereza Petrusková at Charles University in Prague, and based on a Czech dataset, has addressed this problem in a novel way. The research team, which included BTO’s Dr Niall Burton, used individual acoustic monitoring to compare the apparent survival and re-encounter probabilities of male Tree Pipits (Anthus trivialis) that were captured in a mist net and intensively handled (measured, blood-sampled and ringed) with those of a control group of unhandled birds tracked solely by their vocalisations. Previous work has highlighted the variability of Tree Pipit song repertoires, and that individuals can be reliably identified from recordings of approximately 5 mins (containing 20+ songs), and thus the potential for acoustic monitoring of movements and return rates of territorial males.

There were no differences between handled and unhandled birds in apparent survival rates, either in simple annual estimates, or in jointly estimated within- and between-breeding season survival estimates. Re-encounter probabilities in the latter analysis peaked at the start of the breeding season, when song activity would be expected to be greatest, but were greater for handled than unhandled birds in mid to late May. Apart from demonstrating that Tree Pipits were not affected in the long-term by capture and associated handling, the study also confirms that individual acoustic monitoring is a valuable non-intrusive method for long-term monitoring studies.