Applying unmanned aerial systems (UAS) and thermal infrared technology for the detection and surveying of wild ungulates

Abstract

The use of unmanned aerial systems (UAS) for wildlife surveying and research has widely expanded in the past decade, but with varying levels of success. Applying UAS paired with Forward Looking Infrared (FLIR) technology to survey forest-dwelling species has been particularly challenging because of unreliable animal detection. There is also little understood about how the novel technology of UAS compares to conventional methodologies for surveying wildlife populations. The goal of this thesis project was to evaluate the efficacy of FLIR-equipped UAS for surveying wild animal populations. We focused on moose (Alces alces) and white-tailed deer (Odocoileus virginianus) as study species for examining the efficacy of UAS for surveying large-bodied, forest-dwelling species. My first chapter presents a background on wildlife population surveying, with a literature review of several methodologies including; fecal pellet counts, remote camera surveys, aerial surveying, thermal infrared sensing, and the relatively new approach of UAS surveys for wildlife. These topics are discussed because they are all widely applied for surveying wildlife and estimating population parameters. My second chapter describes our work using a quadcopter-style UAS and FLIR sensor to detect wild GPS-collared moose and conduct calf counts. We collected environmental variables to model moose detection success in order to improve UAS moose detection rates. We found that UAS thermal detection increased with greater cloud cover, and was hindered by increased forest canopy, and increased vegetative greenness. Overall, we report that FLIR-equipped UAS shows potential for monitoring the reproductive success and survival of wildlife species in densely forested regions. My third chapter reports our application of a FLIR-equipped fixed-wing UAS to estimate the population density of wild white-tailed deer at the Cedar Creek Ecosystem Science Reserve. We compared density estimates from UAS methodology to density estimates generated from fecal pellet counts to describe how novel technology compares to conventional approaches of population surveying. We modeled deer counts and detection probabilities, and calculated both point estimates and bootstrapped prediction intervals for deer density from UAS and pellet-group count data. We found that there was overlap for density and abundance estimate values between methodologies; however, UAS surveys were more efficient and required less field effort than conventional pellet count surveys.