摘要

Dust emissions in climate and earth system models are associated with large uncertainties. These models often use the source erodibility (S) to constrain dust emissions and also lack explicit representations of the impact of surface roughness elements (SREs) on the threshold friction velocity (u(*t)). This study presents a process-oriented evaluation of dust emission parameterizations in the Community Earth System Model (CESM) by applying the model to simulate a severe dust storm during 19-22 March 2010 in East Asia. Through numerical experiments, we assess the applicability of S and investigate the impact of SREs on dust emissions by implementing the roughness correction factor (f) to u(*t). Simulation results are compared against the surface synoptic observations and station observations of dust concentrations. We found that the model can capture the main dust emission regions and reproduce the temporal-spatial evolution of surface dust concentrations in Mongolia and northern China. With a geomorphic S (S-g), the model tends to produce excessive dust emissions over the low-lying basins. Moreover, the high-resolution S-g performs worse with point sources of strong dust emissions than the low-resolution one. With the inclusion of f, total dust emissions are reduced by 24-34%, and the model reduces the overestimation of surface dust concentrations and improves their temporal variations over the vegetated regions. These results suggest that S-g may not be necessary when meteorology and land surface state are well simulated by the model and that f provides an important constraint on dust emissions through SREs.