Favorable burrow site selection is critical to the survival of hibernating mammals, influencing thermoregulation, metabolic suppression, and overall fitness. Yellow-bellied marmots ( Marmota flaviventer ) hibernate for 7–8 months annually in high-altitude environments, exhibiting strong site fidelity, yet the selection of sites may be shifting in response to environmental change. Effective methods for monitoring burrows in subalpine environments remain limited. We evaluate drone and very-high-resolution satellite imagery for localizing yellow-bellied marmot burrows, including known hibernacula. Non-expert annotators correctly localized burrows in 63.3% of drone-image annotations. Localization accuracy declined modestly with flight altitude, from 68.7% at 10 m to 57.6% at 120 m, but was not related to site elevation, hibernaculum status, occupancy history or mean marmot abundance suggesting that fine-scale site characteristics may be driving localization potential. In contrast, no known burrows could be identified in 0.33 m satellite imagery, even when annotators were provided with precise coordinates. Drone imagery therefore offers a non-invasive tool for monitoring known marmot burrows in subalpine environments, whereas the spatial resolution of current optical satellite imagery is insufficient for this purpose. Extensions of this method using computer vision approaches to automate detection of burrows in drone imagery could provide a scalable method to better understand shifts in site selection of hibernating marmots in response to a changing climate.
Items connected by shared entities, co-authorship, citations, or semantic similarity.