Drone-conducted counts as a tool for the rapid assessment of productivity of Sandwich Terns (Thalasseus sandvicensis)

Abstract

Drones are revolutionizing the methodological approaches to the study of bird population ecology, and pioneering studies are providing evidence that drones could effectively be used to study breeding success. This study compared the accuracy and precision of drone-conducted chick counts to traditional ground counts in a population of Sandwich Terns Thalasseus sandvicensis, and also assessed the safety and degree of disturbance of these methods for both adults and chicks. Results from ground and drone counts were compared to a third count which consisted of a combined count, accounting for uncounted individuals in each method. Fledging success, determined by ground count, drone count, and their combination, was, respectively, 0.54, 0.55, and 0.56 fledglings/nesting pair in 2018, and 0.62, 0.68, 0.69 fledglings/nesting pair in 2019. Mean creche size of Sandwich Terns was similar for ground counts and counts made using drone-derived imagery, but these counts were lower than the mean creche size obtained from combined counts. Drone counts were more accurate and precise than ground counts, and the distance walked from the nest site by creches was far shorter for drone surveys compared with ground counts; likewise, the time spent by creches far from the nesting site was far longer for the ground method compared with drone surveys. Our drone-based approach to repeated chick counts maximizes effectiveness and safety for birds without increasing resource absorption. The use of drones, in combination with the traditional ground-based approach, is a promising approach for the assessment of seabird productivity that can contribute to our understanding of the breeding biology of seabirds.