Experimental assessment of large mammal population estimates from airborne thermal videography
- McElhinny, Julia S.
- Larsen, Gregory D.
- Messinger, Maxwell
- Newbolt, Chad H.
- Whitworth, Andrew
- Ditchkoff, Stephen S.
- Silman, Miles R.
- Beaver, Jared T. [ Montana State University: Animal & Range Sciences ]
Wildlife resource management requires reliable, fast, and affordable methods of surveying wildlife populations to develop and adaptively adjust policies. Thermal video from drones can yield high rates of detection over large areas with relative speed and safety. In wild populations estimating detection rates and accuracy of abundance estimates is difficult, as both study design and properties of the system can influence counts from thermal surveys and true population densities are unknown. Here we used a designed experiment on a captive white-tailed deer Odocoileus virginianus population of known size at the 174-ha Auburn Deer Facility in Auburn, AL, USA, to investigate the effects of observers, time-of-day, and day-to-day variation on abundance estimates using drone-based airborne thermal videography. The experiment consisted of 20 full-census flights over four days in winter at five times-of-day ranging from before sunset through after sunrise and counted by three trained observers. Counts across all trials and observers yielded a mean of 77 (95% CI 71–83) deer representing 81–97% of the true population range. Flights near sunset had the highest accuracy, within or close to the true population range. Estimate variability was primarily influenced by climatic conditions and time-of-day, with minor observer effects, highlighting the importance of day-to-day variability in environmental conditions and diel processes like nocturnal environmental cooling and crepuscular peaks in deer activity. Experimental results and model estimates show that censuses conducted at optimal times-of-day and averaged across days give estimates within the known population range without correction, and that by averaging across flights, even including suboptimal conditions, thermal videography returned estimates within 4% of the known population range with substantial overlap. Results show thermal videography as highly accurate when flights are planned with respect to diel activity patterns and landscape thermal properties; or, in the absence of that information, when multiple surveys are averaged across times and days.