Evaluating a Parameterization for Sublimation of Blowing Snow with In-situ Observations in the Arctic

Surface albedo, sea ice growth and glacier mass balance in the Arctic are all heavily dependent on snow and thus also impacted by blowing snow through redistribution and increased sublimation. The sublimation of blowing snow is significantly higher than that of ground snow due to the larger surface area of the suspended snow crystals and the continuous entrainment of dry air. Thus, sublimation of blowing snow impacts the exchange of energy, moisture and particles between the snow and atmosphere in windy conditions.

Because of the difficulty of modelling such a small-scale process for large areas, parameterizations of sublimation of blowing snow are necessary for snow mass balance and aerosol production studies. The widely used Déry and Yau (2001) parameterization has only been evaluated with model data from the Canadian Prairie, but never for other surface types, where it is applied, or with in-situ observations. Therefore, the goal of this work is to evaluate the parameterization by Déry and Yau (2001) with observations from the MOSAiC expedition in the central Arctic and the Intensive Observation Period for Water (IOP4H20) field measurements in Ny-Ålesund, Svalbard.

Here we show observations of blowing snow events that were detected and characterized by a snow particle counter and the Video In-Situ Snowfall Sensor (VISSS). During these events, measurements of latent heat fluxes from eddy covariance systems are used to evaluate the parameterized sublimation rate. To address challenges with eddy covariance observations in snowy conditions and calculating column-integrated values the observations are complemented with the 1D-column PIEKTUK-D blowing snow model.

In this way, comparing the parameterization with observations brings insights into its uncertainty or possible limitations for two different surface types and thereby improves the estimation of the accuracy of snow mass balance and aerosol production studies that apply this parameterization.