Abstract

not-yet-known not-yet-known not-yet-known unknown One of the most difficult challenges for wildlife managers is reliably estimating wildlife populations. Camera traps combined with spatial capture recapture (SCR) models are a popular tool for population estimation. They have limitations, however, including long data processing times. Drones with thermal imagery are an emerging tool for estimating wildlife populations, but how they compare to other methods remain poorly studied. We compared the use of camera traps and SCR models to drone surveys for estimating population densities of Sitka black-tailed deer (Odocoileus hemionus sitkensis) on Afognak Island, Alaska. We deployed 26 camera traps from 1 September until 6 October 2022 and individually identified males using antler characteristics, for the SCR model. At the same site we conducted three drone surveys between October and December 2022, obtained deer counts, and identified sex composition. The estimated density from the SCR model was 3.7 males ± 0.8 (SE) /km2, and 14.1 ± 3.1 adults/km2 of clear-cut forest. Results from the drone survey produced similar estimates with 2.1 ± 0.9 males/km2 and 13.4 ± 1.6 adults / km2. The similarity in estimates suggests that both methods converged on an accurate representation of the population in this habitat, but these methods diverge in levels of sampling effort, duration, and financial cost. Camera traps offer further insights on behavior and home-range size but require longer data processing times, can be subject to malfunctions, and are difficult to deploy and maintain in remote areas. Drones are subject to legal restrictions, have difficulty in closed canopy habitat and can be initially costly, but they provide results faster and require less data analysis. Camera traps and drones are useful for determining population dynamics but are subject to their limitations. Wildlife managers should make survey decisions based on their specific goals, habitat type, focal species ecology and financial limitations.