A dynamic snow accumulation simulation approach for forecasting snow distribution over regional‐scale terrain

Economic losses due to snow‐related disasters in pastoral areas of China have become severe in recent decades. Terrain and wind play prominent roles in determining snow distribution patterns. As complex topography and dynamic wind conditions are strong nonlinear multivariables with many uncertain factors, it is difficult to predict elevation‐ and wind‐affected snow depth distribution. The complex characteristics of the snow accumulation process call for highly reliable comprehensive simulation for disaster management. The accuracy of these simulations mainly depends on the integration of multiple models for multi‐stage snow accumulation processes. Focusing on coupling wind‐affected snowdrift and snowmelt processes, we put forward a virtual geographical environment snow accumulation simulation approach. The approach consists of three main steps: (1) computational fluid dynamics prediction of 3D dynamic wind‐flow fields; (2) wind‐affected snowdrift simulation; and (3) snowmelt simulation based on degree‐day model. To check its reliability and efficiency, this approach has been validated by reconstructing a historical snow event at Tacheng City, Xinjiang Province, China. The results have characterized the elevation‐ and wind‐induced snowdrift patterns and their specific levels of risk under various conditions. The comprehensive simulation method described here can be applied to help the government and the public examine future events, and stimulate consideration of appropriate policies and techniques for snow risk mitigation.