The Cold Biases in the Soil and Surface Air Temperature Simulations of RegCM4.7 Model Over the Tibetan Plateau in Cold Seasons Reduced by Adopting an Improved Snow Cover Fraction Scheme

Abstract

In cold seasons, global and regional climate models exhibit consistent cold biases in the soil and surface air temperature simulations on the Tibetan Plateau (TP), while the overestimated snow cover fraction (SCF) is treated as one of crucial factors leading to the cold biases. To partially solve this issue, this study adopts an improved SCF scheme that adequately consider the impact of sub-grid terrain relief on snow cover into the Regional Climate Model version 4.7 (RegCM4.7) coupled with the CLM4.5 land surface model to enhance the model skill. Results show that adopting the improved SCF scheme in RegCM4.7 model can significantly reduce the SCF overestimation on the TP produced by the original model, leading to obvious improvements in the surface albedo, soil and surface air temperature simulations in cold seasons. Mechanism analysis indicates that the sub-grid terrain relief suppresses the rapid increase of SCF with snowfall and makes the ground more difficult to be fully covered by snow, resulting in the overestimation of SCF obviously reduced and thereafter the surface albedo decreased. It further causes the ground surface to absorb more solar radiation and then more heat released to the overlying atmosphere and deep soils. Thereafter, the cold biases in the soil and surface air temperature simulations on the TP in cold seasons can be clearly reduced. This study highlights that considering the impact of the sub-grid topography relief on snow cover in numerical models is one of the effective ways for reducing the cold biases on the TP.