Improving the parameterization of dust emission threshold in the Community Earth System Model (CESM)

Importance of different parameterization changes for the updated dust cycle modelling in the Community Atmosphere Model (version 6.1)

“…Still, there exists large uncertainty in modeling the global and regional dust cycle in comparison with observations. Further development focusing on the following which the current model does not well represent or omits may be helpful for advancing the dust model on simulating the dust cycle in CESM: 1) the threshold friction velocity calculated in both BRIFT and DEAD does not account for the spatiotemporal variability of the soil properties (i.e., soil grain size distribution and aggregate state; Leung et al, 2021; mainly limited by the sparse information; Kok et al, 2014b) in addition to the soil moisture. The current dust module in CAM6.1 also does not consider the roughness effect due to the presence of non-erodible elements (i.e., rocks, pebbles, and vegetation) on the threshold velocity calculation (Marticorena and Bergametti, 1995); 2) crusted surface layer present at the erodible surface can greatly reduce the wind erodibility by increasing the particle cohesion, and, thus, the dust emission rate, compared to the surface that does not consist of consolidated aggregates (Rice and McEwan, 2001);…”