Use of geolocators for investigating breeding ecology of a rock crevice-nesting seabird: method validation and impact assessment

1: Investigating ecology of marine animals, imposes a continuous challenge due to their temporal and/or spatial unavailability. Light-based geolocators (GLS) are animal-borne devices that provide relatively cheap and efficient method to track seabird movement and are commonly used to study migration. Here we explore the potential of GLS data to establish individual behaviour during the breeding period in a rock crevice-nesting seabird, the Little Auk, Alle alle. 2: By deploying GLS on 12 breeding pairs, we developed a methodological workflow to extract birds' behaviour from GLS data (nest attendance, colony attendance and foraging activity), and validated its accuracy using behaviour extracted from a well-established method based on video recordings. We also compared breeding outcome, as well as behavioural patterns of logged individuals with a control group treated similarly in all aspects except for the deployment of a logger, to assess short-term logger effects on fitness and behaviour. 3: We found a high accuracy of GLS-established behavioural patterns, especially during the incubation and early chick rearing period (when birds spend relatively long time in the nest). We observed no apparent effect of logger deployment on breeding outcome of logged pairs, but recorded some behavioural changes in logged individuals (longer incubation bouts and shorter foraging trips). 4: Our study provides a useful framework for establishing behavioural patterns (nest attendance and foraging) of a crevice-nesting seabird from GLS data (light and conductivity), especially during incubation and early chick rearing period. Given that GLS deployment does not seem to affect the breeding outcome of logged individuals but does affect fine-scale behaviour, our framework is likely to be applicable to a variety of crevice/burrow nesting seabirds, even though precautions should be taken to reduce deployment effect. Finally, because each species may have its own behavioural and ecological specificity, we recommend performing a pilot study before implementing the method in a new study system.