Desert Storms - Towards an Improved Representation of Meteorological Processes in Models of Mineral Dust Emission

Description of the project

This project is conducted in close collaboration with the University of Leeds (see http://www.see.leeds.ac.uk/research/icas/research-themes/atmosphere/groups/dust-storms/research-projects/desert-storms/). It aims at improving the way the emission of mineral dust from natural soils is treated in numerical models of the Earth system. Dust significantly affects weather and climate through its influences on radiation, cloud microphysics, atmospheric chemistry and the carbon cycle via the fertilization of ecosystems. To date, quantitative estimates of dust emission and deposition are highly uncertain. This is largely due to the strongly nonlinear dependence of emissions on peak winds, which are often underestimated in models and analysis data.<\p>

The core objective of this project is therefore to explore ways of better representing crucial meteorological processes such as daytime downward mixing of momentum from nocturnal low-level jets, convective cold pools and small-scale dust devils and plumes in models.<\p>

To achieve this, we are undertaking (A) a detailed analysis of observations including station data, measurements from recent field campaigns, analysis data and novel satellite products, (B) a comprehensive comparison between output from a wide range of global and regional dust models, and (C) extensive sensitivity studies with regional and large-eddy simulation models in realistic and idealized set-ups to explore effects of resolution and model physics. In contrast to previous studies, all evaluations are made on a process level concentrating on specific meteorological phenomena. Main deliverables are guidelines for optimal model configurations and novel parameterizations that link gridscale quantities with probabilities of winds exceeding a given threshold within the gridbox. The results have substantially advanced our quantitative understanding of the global dust cycle and will reduce uncertainties in predicting climate, weather and impacts on humans.<\p>