Investigations on the impact of explicitly predicted snow metamorphism on the microclimate simulated by a meso-β/γ-scale non-hydrostatic model

A snow model is developed, coupled to and tested within the framework of the meso-beta/gamma-scale non-hydrostatic model, Geesthacht's simulation model of the atmosphere (GESIMA). An evaluation of the snow model is conducted both in a stand-alone version and within GESIMA. In the stand-alone mode, it is evaluated at local scales using data routinely observed at Brandis (51.32degreesN, 12.62degreesE, 133 m NN, Saxony) between 1993 and 1997. The snow model reproduces reasonably the temporal evolution of the snow depth; however, it slightly underestimates snow depth, on average. In the coupled mode, simulations are performed with and without the snow model for a winter-storm snow event and a melt period in East Germany to examine the influence of explicitly modeled snow metamorphism on the simulated microclimate. The snow model reasonably predicts the effects typically associated with snow cover. Accuracy of predicted snow depth and extension depends on the lateral boundary conditions and snow prediction by the host model. Evaluation of the simulated air temperatures as well as humidity shows that the inclusion of the snow model improves the model performance as compared to the simulations without snow model. The results show that changing only the values of albedo and emissivity to those typical for snow, as often done in meso-beta/gamma-scale modeling of snow events, can even lead to opposite effects in simulated latent heat fluxes, ground heat fluxes, soil- and near-surface air temperatures than those typically associated with a snow cover. A rigorous evaluation of the snow simulations in coupled meso-beta/gamma-scale non-hydrostatic models requires datasets of snow properties (e.g., albedo and emissivity, snow cover extent, snow depth, snow water equivalent, snow temperature) in a high quality and resolution for the region under study. The available datasets are not yet ready to fulfil this objective. (C) 2002 Elsevier Science B.V. All rights reserved.