Irina Sokolik, School of Earth and Atmospheric Sciences, Georgia Institute of Technology

Abstract: Mineral aerosol (or dust) has been long known as an important agent of the Earth system because its influence on the radiative balance, and the hydrological and biogeochemical cycles. Yet, most recently, there has been a growing recognition of the complexity associated with quantitative assessments of the dust impacts that necessitate consideration of dust interactions and coupling with the Earth system components, especially land and atmosphere. Given the heterogeneity in dust sources and distinct regional changes in land cover and land use (LCLU) and climate, an improved understanding of region-specific systems would be highly desirable.

Focusing on drylands in Central and East Asia, this talk will address the major processes controlling variability of Asian dust based on an integrative analysis of satellite aerosol and land products, ground-based observations and a fully-coupled regional model WRF-Chem-DuMo. The model, a modified version of the public NCAR Weather Research and Forecasting (WRF) Chem, includes advanced treatments of size- and composition-resolved dust aerosol, which is coupled with cloud/precipitation and radiation processes. WRF-Chem-DuMo also offers different dust emission schemes, ranging from “simple” schemes to physically-based schemes, which are coupled with land surface model and meteorology.

Results will be presented on Asian dust emission, burden, and radiative forcing from the 1950s to present. Building on the synergy between observational data and modeling results, the interactions among dust, LCLU, and climate are investigated across the broad range of spatial and temporal scales to gain an improved process-level understanding of the underlying mechanisms within the Earth system framework. The talk will also address some critical issues related to the integration of satellite data and regional models, including the role of spatial and temporal resolution and treatments of dust aerosols.