Coupling global climate drivers to dust emission dynamics at Etosha Pan, Namibia

Ephemeral lake beds are globally significant sources of atmospheric mineral dust aerosols, but emissions from these topographic features are characterised by considerable spatial and temporal variability. Resolving the complex relationships between climatic and surface drivers that control aeolian emissivity has proven extremely challenging, leading to substantial uncertainty in model predictions of atmospheric dust loading and its impacts on other global-scale Earth system processes. This study uses a range of satellite-derived measurements (e.g., MSG-SEVIRI, MODIS) of mineral dust aerosols, lake inundation, catchment precipitation, and climate data (from meteorological stations and model reanalysis data sources) from 2000 to 2022 to assess the influence of regional climate and basin-scale hydrology on the emission of mineral dust from Etosha Pan in Namibia, a major Southern Hemisphere dust source. Significant associations were identified between interannual and seasonal variations in catchment rainfall, ephemeral pan surface inundation, and the frequency and magnitude of dust emissions over the two-decade analysis period. Results indicate that dust production is modulated by seasonal precipitation and ephemeral flooding events, which suppress dust emissions during wet periods and also lead to the delayed production of dust emissions on surface drying. During dry periods, wind speed variability was identified as the primary driver of dust emissions. These regional climate and hydrological factors were found to be closely linked to El Niño-Southern Oscillation (ENSO) phases and shifts in the subtropical Indian Ocean Dipole (SIOD), where seasonal increases (reductions) in dust emissions were associated with positive (negative) ENSO and negative (positive) SIOD events. The identification of robust links between dust emissions and global climate drivers in this study makes a significant contribution to the improved integration of ephemeral lake systems in dust emission models and opens the potential for forecasting future dust trends based on large-scale climate cycles.