Characterizing the signature of topographic controls of seasonal SWE on the microwave behavior of mountain snowpacks

Abstract:

The spatial distribution of snow depth and snow water equivalent is highly heterogeneous in mountainous regions because of complex topography, spatially-variable vegetation cover, and wind driven snow redistribution, roughning and crusting of the snowpack top layers. The distriubuted version of the Multilayer Snow Hydrology Model coupled to a snow microwave radiative transfer model was implemented over Senator Beck Basin (SBB), Colorado at high-resolution (100 m, 5 min). Atmospheric forcing obtained from the High-Resolution Rapid Refresh (HRRR) model was downscaled and adjusted to sub-grid scale topography. Multi-year simulations of snowpack properties (SWE, snow depth, snow density and temperature profiles) and snowmelt runoff were evaluated against existing long-term observations including stream gauge data at the SBB outlet, SnowEx observations, and snow cover imagery. The results show that the model captured the snow accumulation and melt processes over the basin reasonably as spatial-temporal variability in field measurements is generally consistent with the simulated patterns. The temporal evolution of the spatial fields of backscatter and brightness temperatures at L-, C-, X-, Ku- and Ka-bands is examined against the spatial and temporal evolution of snow depth and SWE including upscaling from 100 × 100 m² resolution to 1 km² toward synthesizing the signature of key topographic controls of snow hydrologic processes on microwave behavior. The model is subsequently applied to a much larger domain in the same region, and model results are analyzed using a deep learning algorithm validated against the SBB data toward identifying and predictinggeneral modes of microwave behavior organization by topography.