Abstract
Over the past years, drones have become increasingly more common in wildlife research and monitoring, although their effectiveness has varied. One area of difficulty has been the use of drones combined with thermal infrared cameras (TIR) to observe species living in forest environments, primarily due to inconsistent detection of animals.
In our study we explore the use of drones equipped with TIR to estimate moose densities in southeastern Norway, comparing these results with pellet count estimates. We conducted 33 drone flights over GPS-collared moose to assess detectability under varying conditions, followed by 34 grid flights to estimate local moose density.
Our results indicate that moose detectability decreases under sunny conditions, potentially due to thermal cluttering effects. Our analysis found a trend towards decreased detectability under sunny conditions, and increased detectability with higher tree cover density (TCD) in drone-captured images. Estimated moose densities based on drone observations ranged from 1.85 to 2.11 moose km², slightly higher than pellet count estimates. While drones show promise for improving density estimates and wildlife management practices, challenges such as limited flight time and line-of-sight requirements need to be addressed for this method to be applied across larger areas.