A Distributed Snow‐Evolution Model for Sea‐Ice Applications (SnowModel)

Abstract

AbstractSnow‐depth distributions on sea ice have substantial impacts on winter ice growth and summer ice melt. There are two types of wind‐related snow distributions in this environment: snowdrifts that form around ice pressure ridges; and snow dunes and other snow bedforms that form on relatively level, undeformed ice. A snow‐evolution modeling system (SnowModel) was tested against winter snow observations collected during the Norwegian young sea ICE expedition (N‐ICE2015) north of Svalbard, with an emphasis on reproducing these two types of snow distributions. The SnowModel simulation spanned 1 year, summer 2014 through summer 2015, over a 1.5 km by 1.5 km domain, using a 1 m horizontal grid increment and 3 h time step. Existing SnowModel components, created originally for terrestrial applications, performed energy balance, snow property, snowdrift, and data assimilation calculations in response to prescribed meteorological forcings. A new SnowModel component, SnowDunes, was created to simulate snow‐bedform distributions on level, undeformed ice. Model results reproduced observed snowdrift profiles around pressure ridges and snow‐depth distributions over level, undeformed ice, resulting in physically realistic snow heterogeneity and property information. We conclude that the SnowModel toolkit contains the components required to simulate most processes driving the seasonal distribution of snow on sea ice. SnowModel can evolve snow on sea ice over local to pan‐Arctic areas, using grid increments ranging from 1 m to tens of km, and time increments of subhourly to daily.