Biased Aerosol Wet Deposition CAM5 Simulations: A Result of Misrepresented Convective-Stratiform Precipitation Partitioning When Benchmarked Against SPCAM

Wet deposition is a major sink for atmospheric aerosols, but its representation in conventional global climate models (GCMs) remains highly uncertain, partly as a result of the partitioning between convective and stratiform precipitation. Using the Super-parameterized Community Atmosphere Model (SPCAM) as a benchmark, we evaluate the performance of the conventional CAM5 model in simulating precipitation and aerosol wet deposition. SPCAM explicitly resolves convection and provides a more physical representation of cloud and precipitation processes. Compared to SPCAM, CAM5 overestimates the frequency of light convective rainfall by up to 50% at rain rates from 1 to 20 mm day−1 and underestimates heavy convective precipitation, leading to a more than 90% contribution from convective precipitation to total rainfall in the tropics, far exceeding that in satellite observations. Accordingly, this bias results in an overestimation of aerosol wet removal by convective precipitation (74.2% in CAM5 versus 47.6% in SPCAM) and an underestimation by large-scale precipitation, as well as an overestimation of aerosol wet removal by light rain (84.0% in CAM5 versus 65.5% in SPCAM). As a result, CAM5 shows systematic biased wet deposition fluxes simulations across aerosol types and sizes compared to SPCAM, particularly in tropical regions. The misrepresentation of convective-stratiform rainfall partitioning in conventional GCMs like CAM5 significantly distorts aerosol lifetime and distribution. Improving convective parameterizations to better capture precipitation frequency distribution and partitioning is essential for credible aerosol-climate projections.