Simulating Dust Impacts on a Severe Thunderstorms Outbreak during April 2003

Three-dimensional numerical simulations were conducted for a severe storms outbreak that occurred in the U.S. Southern Plains during April 2003. The convective line of interest was associated with a significant dust plume, which originated from the desert southwest. Potential dust microphysical and radiative impacts were simulated and put in context with other environmental factors. Increased solar absorption within the dust plume resulted in modest increases in pre-storm atmospheric stability at low levels. This yielded weaker convective updrafts and generally less widespread precipitation. Dust indirect microphysical impacts on convection were slight in comparison, due in part to the fact that the lofted dust concentrations were relatively few in number when compared to the non-dust background aerosol populations. Dust preferentially serving as cloud condensation nuclei (CCN) versus giant CCN (GCCN) had opposing influences on initial warm-rain production. However, both scenarios resulted in ample supercooled water and subsequent glaciation aloft, and thus larger graupel and hail mean sizes. The associated latent heating from condensation and freezing contributed little to overall updraft invigoration. With reduced rain production overall, the simulations that included dust-atmosphere-cloud interactions experienced slightly reduced grid-cumulative precipitation, as well as noticeably warmer and spatially smaller cold-pool regions. Dust serving as ice nucleating particles (INP) played no significant role in these simulations as related to overall storm severity or longevity. Additional sensitivity simulations are needed to isolate the magnitude of any INP influences on storm anvil characteristics as a function of INP parametrization scheme. The presence of dust generally resulted in fewer supercells produced. However, the mesocyclones of these supercells were consistently associated with higher values of relative vertical vorticity than those simulated in the non-dust experiments. Overall, dust microphysical and radiative affects were relatively small in magnitude when compared to those from changing the background pre-storm environments (i.e., convective available potential energy and vertical wind shear). It is difficult to generalize such findings from a single event, due to a number of case-specific environmental factors.