Validating an Unmanned Aerial Vehicle (UAV) Approach to Survey Colonial Waterbirds

Abstract

Unmanned aerial vehicles (UAVs) can improve detection and count accuracy of avian species and provide an inexpensive alternative to surveying waterbirds compared to ground counts and other aerial methods. To validate this method for surveying colonial waterbirds in logistically challenging areas, we compared counts from images collected with a fixed-wing UAV to ground counts using a marked-subsample method or direct counts at three study sites: 1) in an open field using decoys that represented three colonial waterbird species; 2) constructed platforms in a swamp with colonial nesting Great Egrets (Ardea alba); and 3) an island with a large nesting colony of Brown Pelicans (Pelecanus occidentalis). Estimates or direct counts from UAV images were not significantly different from a known number of decoys or ground counts for Great Egrets but were almost twice as high compared to ground counts for juvenile Brown Pelicans. UAV flights did not cause either Great Egrets or Brown Pelicans to flush from nesting areas while ground surveys flushed the majority of adult Brown Pelicans from the survey area. UAV surveys captured images of eggs in nests that could not be observed by ground surveyors. Our results indicate that UAVs can provide a cost-effective and efficient method of surveying waterbirds with similar or improved results and little disturbance of colonies compared to ground counts.