Assessment of surface roughness and fractional vegetation coverage in the CoLM for modeling regional land surface temperature

Abstract

The application of Land surface models (LSMs) at regional scales still faces challenges and large uncertainties exist in simulating regional energy fluxes and land surface temperature (LST). This is possibly caused by their poor representations of surface parameter properties (e.g., fractional vegetation coverage (FVC) and surface roughness length (z0m)) to effectively capture vegetation dynamics. In this study, satellite-based FVC and z0m schemes were incorporated into the Common Land Model (CoLM) and their impacts on modeling energy fluxes and associated LST were further assessed over grassland and forest areas in China. Our results show that the annual mean LST was obviously overestimated by the original CoLM for both grassland and forest. The overestimation was reduced by the revised CoLM when compared with the Moderate Resolution Imaging Spectroradiometer (MODIS) LST products. Concerning the seasonal patterns, the revised model performed evidently better in simulating LST in summer than in winter. Additionally, the decreased simulation of LST was dominated by a reduction of the surface absorbed short-wave radiation when applying the revised FVC scheme over both grassland and forest, while the increased sensible/latent heat fluxes contributed most to the reduction of LST for the revised z0m scheme in grassland/forest, respectively, indicating that land surface energy partitioning is sensitive to the FVC and z0m parameterizations. Results from this study suggest that the use of satellite-based dynamics of FVC and z0m parameterizations in the CoLM provides a valuable improvement in regional applications.