Triaxial accelerometry allows to determine chick feeding events in a marine bird

DEL CAÑO, MONSERRAT; QUINTANA, FLAVIO; DELL'OMO, GIACOMO; GOMEZ LAICH, AGUSTINA

Hobart

Conferencia; Third World Seabird Conference; 2021

Over recent years, accelerometers have proven tobe a powerful tool for the identification and quantification ofanimal behavior. Even though in seabird species this technique hasallowed to identify activities such as diving, floating and flying,until now it has never been used to identify and analyze parentalfood provisioning events. In the present study, we applied theSupport Vector Machine (SVM) algorithm to identify food provisioningbehaviors from triaxial accelerometer data from breeding Imperialshags (Leucocarbo atriceps).A total of 15 breeding females were equipped with an accelerometer onthe head which was programmed to record data at 50 Hz and left on thebird for less than 24 h. In order to label acceleration data, femaleswere filmed during chick feeding and non-feeding events by means of avideo camera. Fed chicks were categorized according to age in youngand old (less and more than 7 days, respectively). Accelerationfeatures from feeding and non-feeding periods were calculated using asliding window of 1 s with a 0.98 s overlap. Finally, one SVM was runfor each chick category, using a cost value of 1. SVMs correctlyclassified the feeding and non-feeding events in the 84 and 93% ofthe cases, for young and old chicks, respectively. However, thespecific classification of food provisioning events (recall score),was correct for 79% of young and 95% of old fed chicks. Of thecalculated features, those derived from the surge and heaveacceleration axes and the pitch angle were the most important fordistinguishing feeding from non-feeding behaviors. Our resultssuggest that accelerometry can be successfully employed to recognizefood provisioning events in the Imperial Shag and can be applied in awider spectrum of seabird species.