Comparing results from a physical model with satellite and in situ observations to determine whether biomass burning aerosols over the Amazon brighten or burn off clouds

Biomass burning (BB) aerosol particles affect clouds through competing microphysical and radiative (semi-direct and cloud absorption) effects, each of which dominates at different degrees of aerosol loading. Here, we analyze the influence of competing aerosol effects on mixed-phase clouds, precipitation, and radiative fields over the Amazon with a climate-air pollution-weather forecast model that treats aerosol-cloud-radiative interactions physically. Extensive comparisons with remotely sensed observations and in situ measurements are performed. Both observations and model results suggest an increase in cloud optical depth (COD) with increasing aerosol optical depth (AOD) at low AODs, and a decrease in COD with increasing AOD at higher AODs in accord with previous observational and modeling studies. The increase is attributed to a combination of microphysical and dynamical effects, whereas the decrease is attributed to a dominance of radiative effects that thin and darken clouds. An analogous relationship is shown for other modeled cloud variables as well. The similarity between the remotely sensed observations and model results suggests that these correlations are physically based and are not dominated by satellite retrieval artifacts. Cloud brightening due to BB is found to dominate in the early morning, whereas cloud inhibition is found to dominate in the afternoon and at night. BB decreased the net top of the atmosphere solar+IR irradiance modestly, but with large diurnal variation. We conclude that models that exclude treatment of aerosol radiative effects are likely to over-predict the microphysical effects of aerosols and underestimate the warming due to aerosols containing black and brown carbon.