Spatial patterns of snow distribution in Arctic permafrost landscapes estimated from ground- and UAV-based observations

Abstract:

Permafrost landscapes of the Arctic are undergoing massive changes, driven in part by the changing spatial distribution of snow. Snow plays a vital role in Arctic climate, hydrology, and ecology due to its fundamental influence on the water balance, thermal regimes, vegetation, and carbon flux. To capture this influence and define key hydro-ecological drivers, we carried out intensive field studies over multiple years for two small (2017-2019) subarctic study sites located on the Seward Peninsula of Alaska. Using the ground-based observations of snow (~19,000 data points), we developed statistical models of snow water equivalent (SWE) distribution for individual years and basins using factors such as macro- and micro-topography, greenness indexes, and vegetation characteristics. SWE distribution was also estimated from spatially continuous measurements of snow surface elevation and inferred snow depth collected using an Unmanned Aerial Vehicle (UAV) for one of the study sites. The most successful statistical models were the non-linear and randomForest models, which illustrate the complexity and variability of snow characteristics across the sites. Approximately 75% of the SWE distribution could be accounted for on average at the study sites. Factors that impacted year-to-year snow distribution included exposure, elevation, NDVI, microtopography, and curvature, while slope and vegetation type were less important. Each of the most successful models was then used to predict SWE for the study areas. The characterization of SWE spatial distribution patterns and the statistical relationships developed between SWE and its impacting factors will be used for the improvement of snow distribution modeling in the Department of Energy’s Earth System Model, and to improve our understanding of hydrology, topography, and vegetation dynamics in the Arctic.