CAM6 simulation of mean and extreme precipitation over Asia: sensitivity to upgraded physical parameterizations and higher horizontal resolution
The Community Atmosphere Model version 6 (CAM6), released in 2018 as part of the Community Earth System Model version 2 (CESM2), is a major upgrade over the previous CAM5 that has been used in numerous global and regional climate studies. Since CESM2–CAM6 will participate in the upcoming Coupled Model Intercomparison Project phase 6 (CMIP6) and is likely to be adopted in many future studies, its simulation fidelity needs to be thoroughly examined. Here we evaluate the performance of a developmental version of the Community Atmosphere Model with parameterizations that will be used in version 6 (CAM6α), with a default 1∘ horizontal resolution (0.9∘×1.25∘, CAM6α-1∘) and a high-resolution configuration (approximately 0.25∘, CAM6α-0.25∘), against various observational and reanalysis datasets of precipitation over Asia. CAM6α performance is compared with CAM5 at default 1∘ horizontal resolution (CAM5-1∘) and a high-resolution configuration at 0.25∘ (CAM5-0.25∘). With the prognostic treatment of precipitation processes and the new microphysics module, CAM6α is able to better simulate climatological mean and extreme precipitation over Asia, better capture the heaviest precipitation events, better reproduce the diurnal cycle of precipitation rates over most of Asia, and better simulate the probability density distributions of daily precipitation over Tibet, Korea, Japan and northern China. Higher horizontal resolution in CAM6α improves the simulation of mean and extreme precipitation over northern China, but the performance degrades over the Maritime Continent. Moisture budget diagnosis suggests that the physical processes leading to model improvement are different over different regions. Both upgraded physical parameterizations and higher horizontal resolution affect the simulated precipitation response to the internal variability of the climate system (e.g., Asian monsoon variability, El Niño–Southern Oscillation – ENSO, Pacific Decadal Oscillation – PDO), but the effects vary across different regions. For example, higher horizontal resolution degrades the model performance in simulating precipitation variability over southern China associated with the East Asian summer monsoon. In contrast, precipitation variability associated with ENSO improves with upgraded physical parameterizations and higher horizontal resolution. CAM6α-0.25∘ and CAM6α-1∘ show an opposite response to the PDO over southern China. Basically, the response to increases in horizontal resolution is dependent on the CAM version.