Integrating the Fokker-Planck Approach for Sub-Grid Variability of Snow Depth into the RUC Land-Surface Model: Preliminary Results from Off-LineESM-SnowMIP Site Simulations

The Rapid Update Cycle Land Surface Model (RUC LSM) has been used to provide the lower boundary conditions for the NOAA Rapid Refresh (RAP) and High-Resolution Rapid Refresh (HRRR) models. Since RAP and HRRR are operational weather models, it is critical to have the accurate lower boundary conditions. Sub-grid variability of snow depth is important for the LSM and evolution of the atmospheric boundary layer, since snow spatial distribution is generally very heterogeneous while its properties are completely different from bare soil. In this study, a snow model representing sub-grid variability of snow depth was coupled into the RUC LSM to replace its uniform snow depth parameterization scheme. This new parameterization scheme is a stochastic model and describes snow spatial distribution through an advection-diffusion equation, which is also called Fokker-Planck equation (FPE). With this new scheme, the percentages of snow-cover area within a grid cell can be simulated. ESM-SnowMIP, which is a successor of the Snow Models Intercomparison Project (SnowMIP), provides meteorological forcing for well-instrumented reference sites. The performance of this coupled RUC LSM was tested at these ESM-SnowMIP sites. The preliminary results shown that the new sub-grid parameterization scheme could improve the performance of the RUC LSM.