Impacts of aerosols on deep convective clouds: clues from millions of observed cloud objects

Numerous studies have examined the response of storms to aerosol increases using numerical models as the primary tool. These experiments offer unique benefit of isolating aerosol impacts from other factors. The applicability and generality of conclusions from them are subject to uncertainties in model configuration of various physical and numerical schemes. Here we use extensive observations from long-term and large-scale satellite records to study this issue. Deep convective cloud objects are identified from observations and associated environmental conditions are characterized by reanalysis data sets. Two regimes of deep convective clouds are studied: tropical maritime and tropical rainforest in the polluted season. We analyzed the changes in precipitation PDF, cloud macro-physical properties and cloud micro-physical properties that can be attributed to aerosols after accounting for first-order environmental conditions. We found significant structural changes in cloud macro-physical properties, systematic changes in cloud micro-physical properties and rather limited precipitation PDF variations. Statistical results from our analysis of the large ensemble of cloud objects may prove useful in evaluating model behaviors. In addition, potential uncertainty and possible implications are discussed.