Tropical depressions in Atlantic often develop in the presence of a large concentration of Sahara dust. Recently, effects of atmospheric aerosol on the precipitation was observed by Rosenfeld (2000) and simulated by Khain et al (2000; 2001) using the spectral micro-physics Hebrew University cloud model (HUCM). It was shown in these studies that an increase in concentration of aerosol leads to narrowing droplet spectra, resulting in significant decrease in rainfall formed by both warm rain and ice phase processes. It is anticipated that atmospheric aerosols may also be an important factor in TC development, although this has never been investigated in a numerical model.

In this study we investigate effects of atmospheric aerosol (in our case, Sahara dust) on the micro-physical structure of TC clouds and spatial distribution of latent heat release. In experiments, an axisymmetric cloud resolvable version of GFDL tropical cyclone model is applied. The TC model uses a spectral (bin) micro-physics scheme based on solving equation system for size distribution functions for liquid water, plate-, columnar,- and branch-type ice crystals, aggregates, graupel, hail/frozen drops. The scheme is designed in such a way to describe effect of atmospheric aerosol on formation of droplet and ice size spectra and precipitation. The model is specially designed to take into account the effect of atmospheric aerosols on the cloud development and precipitation formation.

Comparison of results of TC evolutions in clear and dusty atmosphere will be presented. According to results of our simulations of single deep convective cloud, as well as cloud ensembles, we expect that Sahara dust will retard raindrop formation and rainfall in the central zones of tropical cyclone. Small cloud droplets will continue growing up to higher levels increasing convective heating in the central zone of TC. At the same time, comparatively small ice particles formed as a result of droplet freezing will be transported toward TC periphery in the out flow layer, so that that sedimentation and melting of ice will take place further from the TC center than in case of clear air. We expect, therefore that increase in atmospheric aerosol concentration (increase in Sahara dust) will lead to a spatial separation of condensation heating within central TC zone and cooling by evaporation and melting of ice at significant distances from the TC center. An increase in the horizontal gradient of temperature in dusty air should foster deepening of TC disturbances. This hypothesis will be carefully tested in a set of experiments with different concentrations and size distributions of aerosol particles.