The impacts of different land surface parameterization schemes on Northeast China snowfall simulation

Abstract

In this paper, hindcast experiments were carried out on the snowfall that occurred on 12–13 April 2010 in Northeast China using the WRF model. Three land surface parameterization schemes (Noah, Noah-MP, and SSiB) and three microphysics schemes (WSM5, Lin, and CAM5.1) were chosen for nine ensemble members. The results indicate that spatial distribution of snowfall is sensitive to the choice of land surface schemes. The Noah-MP land surface scheme yields the smallest spatial distribution among the three land surface schemes. The simulations with the Noah and Noah-MP land surface schemes are better than those with the SSiB land surface scheme in terms of spatial correlation coefficient. Microphysics schemes have an influence on snowfall amount in the nine simulation experiments. The Lin and CAM5.1 microphysics schemes lead to much less snowfall than the WSM5 microphysics scheme. Overall, the ensemble means of simulations with three different land surface schemes can capture the features of snowfall distribution, especially when combined with the CAM5.1 microphysics scheme.