Evaluation and Intercomparison of Multiple Dust Emission Schemes within the NASA Goddard Earth Observing System Model

Accurate modeling of the dust cycle—critical for understanding dust-climate interactions—is often hampered by the complexities of dust emission processes. These challenges arise from non-linear and highly sensitive nature of the emission processes. As a result, various dust emission schemes have been proposed, differing in the complexity of their physical representations. This global modeling study evaluates and compares several dust emission schemes within the NASA’s Goddard Earth Observing System (GEOS) model. The schemes range from relatively simple, such as the Goddard Chemistry Aerosol Radiation and Transport (GOCART) default model and its modified version with prescribed threshold velocity, to more complex models like the Dust Entrainment and Deposition (DEAD) scheme and the Kok2014 scheme. These schemes (except the Kok2014) have recently been updated to account for changes in vegetation cover, soil temperature, and snow cover. Additionally, the dust source function has been updated, which differs from the previous one and allows dust emission from high latitude sources. The results show that the schemes differ regionally and seasonally in simulating the dust aerosol optical depth, with a common issue of underestimating dust over the Taklimakan Desert. The presentation will compare the performances of these schemes, explore reasons for discrepancies, and include sensitivity tests examining the influence of model resolution or data (e.g., high-resolution ERA5-Land versus low-resolution GEOS wind speeds) and sub-grid scale wind variability.