Modeling the mineral dust aerosol cycle in the climate system

Soil dust aerosol is an important factor of the climatic system. In order to evaluate the different aspects of the climatic effects of dust, estimates of its highly variable atmospheric distribution need to be computed by transport models. Such models also provide important means of evaluating the processes that govern changes in dustiness during different climatic periods. While models of the modern dust cycle are currently capable of simulating first-order patterns of its global distribution, the parameterization of dust emission in these models is still crude, since input information about soil properties and wind events cannot be resolved at a global scale. Regional models could be useful for evaluating emission parameterizations, as well as dust transport and depositional processes close to source regions. No single existing data set fully describes all aspects of the dust cycle. Validation of modeled dust distributions must therefore include comparisons with different types of observational data. While the compilation of such observational data sets is crucial for model development, model results can, in turn, provide guidance for new measurements of dust properties, which will be useful for future investigation of the dust cycle and its climatic effects. (C) 2003 Elsevier Ltd. All rights reserved.