Regional simulations of deep convection and biomass burning over South America: 1. Model evaluations using multiple satellite data sets

Multiple data sets, mostly from satellite observations, are used to evaluate the performance of the Weather Research and Forecasting model with Chemistry (WRF‐Chem) in simulating the distribution and evolution of aerosol, clouds, precipitation and chemistry during the dry season in South America. A 9‐day WRF‐Chem simulation with 36 km horizontal resolution is performed from 15 to 24 September 2006, during which frequent biomass burnings were observed. It is shown that the model reproduces the spatial distribution of aerosols produced by biomass burning and approximately captures convective transport of trace gases (e.g., CO and O3) into the upper troposphere. Surface precipitation is also in reasonable agreement with observation. The model simulations overestimate the magnitude of water vapor in the upper troposphere while the magnitude of cloud water content is lower than measurements from satellites, which may indicate problems in the cumulus and microphysical parameterizations. The model simulations capture temporal variations of outgoing longwave radiation at the top of atmosphere and downward shortwave radiation at the surface shown in the NASA GEWEX SRB data set. A sensitivity run at 4 km horizontal resolution shows similar results to the 36 km simulation, with a high bias of precipitation. The uncertainty and weakness in both satellite observations and model simulations are identified. This study demonstrates that satellite data are valuable to the evaluation of regional model simulations for climatologically important processes such as deep convection and biomass burning, especially in regions with little in situ observation.