Strickfaden, K.M., A. Marshall, L. Svancara, D.E. Ausband, and T. Link. 2023. Identifying snow refugia in complex forested terrain using camera data. Environmental Research Letters. 18:044014. https://doi.org/10.1088/1748-9326/acbb90.
Abstract
Knowledge of snow cover properties on fine scales is imperative both for estimation of hydrologic processes and for habitat management for wildlife species that rely on snow cover. Identification of snow refugia, or places with relatively late snow disappearance dates compared to surrounding areas, are especially important as climate change continues to alter snow cover timing and duration. However, many snow data products are either too coarse-scale to capture variations in snow cover or are too expensive or logistically challenging to collect over broad spatial extents. The purpose of this study was to use remote cameras to collect snow data at fine spatial and temporal scales in a complex forested terrain for the identification of snow refugia. We built generalized linear models to relate the snow disappearance dates (SDDs) at the camera sites to their topographic and vegetation characteristics. We built a model to describe SDDs of camera sites which contained elevation, aspect, and an interaction between canopy cover and cold-air pooling potential. This model could predict SDDs to within 2 weeks and to within 1 week of true SDD for 93% and 71% of the camera sites, respectively. This model contains only data which are readily available as spatially distributed datasets, which allowed for mapping of SDDs across the entire study site. This model and map can be used to guide forest management for the retention of snow, increase the accuracy of hydrologic models, and inform habitat management for snow-dependent wildlife species.