Assessing uncertainties in dust emission in the Aral Sea region caused by meteorological fields predicted with a mesoscale model

This study investigates how the choice of the planetary boundary layer (PBL) parameterization and dust emission scheme affects the prediction of dust entrainment simulated with a regional mesoscale model. For this analysis we consider a representative dust episode which occurred on April 2001 in the Aral Sea region. The meteorological fields were simulated using the PSU/NCAR MM5 modeling system considering two different boundary layer parameterizations. In each case, emitted dust fluxes were computed off-line by incorporating MM5 meteorological fields into the dust module DuMo. Several dust emission schemes with a prescribed erodible fraction and fixed threshold wind speed were the subject of our analysis. Implications to assessment of the anthropogenic fraction of dust emitted in the Aral region were investigated by conducting the full, half, and no lake modeling experiments. Our results show that the discrepancies in dust fluxes between the two different PBLs are much higher compared to the discrepancy associated with the use of considered dust production schemes. Furthermore, the choice of the PBL affects the timing and duration of a modeled dust event. We demonstrate that different combinations of the PBL parameterization and wind- or friction velocity-driven dust emission schemes can result in up to about a 50% difference in predicted dust mass caused by the Aral Sea desiccation. We found that the drying-up of the Aral cannot only affect the dust emission by expanding the source area, but also by affecting atmospheric characteristics, especially winds. These competitive factors add further complexity to quantification of the anthropogenic dust fraction in the region. (C) 2006 Elsevier B.V. All rights reserved.