A computationally efficient tool for capturing heterogeneous wind-affected snow accumulations from commonly available data

In non-forested or sparsely forested mountain regions, wind strongly affects snow accumulation, scouring snow from some areas and forming large drifts in others to create a highly heterogeneous snow cover. A mixed vegetation structure of shrubs and sparse forest stands adds additional complexities to snow distributions. The resulting heterogeneous snow distribution affects runoff, soil moisture, vegetation, and effective habitat. However, the complexities of the controlling wind fields and lack of snow data have hindered explicit representations of these important heterogeneities. A computationally efficient tool for simulating these complex snow accumulation patterns from typically collected data is presented. Forcing data consisted of undercatch-adjusted point measurements of precipitation from wind-sheltered locations similar to NRCS SNOTEL sites. Distributed wind speeds (Winstral et al., 2009) were employed in conjunction with terrain analysis to modulate the measured data and formulate hourly effective precipitation inputs to Isnobal, a distributed mass and energy balance snow model. Snow simulations were conducted across a range of basin sizes. Simulated snow distributions were compared to field measurements and remotely sensed snow patterns. Basin-integrated surface water inputs were compared to stream discharge. In all cases, the distributed snow accumulation algorithm capably captured the hydrologically-significant aspects of snow heterogeneity and greatly improved runoff simulations.