Reducing the Cold Bias of the WRF Model Over the Tibetan Plateau by Implementing a Snow Coverage-Topography Relationship and a Fresh Snow Albedo Scheme

Most climate models show systematic cold biases during snow-covered period over the Tibetan Plateau (TP), which is associated with snow and surface albedo overestimations. In this work, a snow cover fraction (SCF) scheme and a recently developed albedo scheme for shallow snow are implemented in the Noah-MP land surface model coupled with the Weather Research and Forecasting (WRF) model. The SCF scheme introduces subgrid orographic variability to reduce the SCF, and the shallow-snow albedo scheme parameterizes the fresh-snow albedo as a function of the snow depth (SD). Evaluations by remote sensing data show that both schemes can effectively alleviate the overestimation of the simulated surface albedo, SCF, snow water equivalent, and SD over the TP. The reductions in the modeled SCF and snow albedo directly lead to lower surface albedo values and thus more surface solar radiation absorption, which accelerates snow melting and causes surface warming effects. Further comparisons with Moderate Resolution Imaging Spectroradiometer data and station observations show that both schemes can significantly reduce the cold biases in the surface skin temperature (from -4.39 degrees C to 0.19 degrees C for the TP mean) and 2-m air temperature (from -4.48 degrees C to -1.05 degrees C for the station mean) during the cold season (October to May of next year) in the study region. This work provides guidance for advancing the snow-related physics in climate models and the improved WRF model could facilitate weather forecasting and climate prediction for the plateau region.