The Impact of Model Horizontal Resolution on Simulating Regional Climate Over East Asia Using Variable‐Resolution CESM2.

In this study, a variable‐resolution version of the Community Earth System Model (VR‐CESM) with mesh refinement (∼0.125°) over East Asia is used to simulate the regional climate in this region. For the evaluation of model performance and sensitivity to model resolution, the simulated near‐surface temperature and precipitation are compared with observations and simulation results from a globally quasi‐uniform (∼1°) CESM (UN‐CESM). Results show that VR‐CESM better simulates the spatial patterns and seasonal variations of mean temperature and precipitation than UN‐CESM over China. For extreme events, VR‐CESM improves the simulation of the occurrence frequency of wintertime daily minimum temperature and heavy precipitation. In regions with complex terrains, VR‐CESM better resolves the topographic forcing and captures the observed fine‐scale spatial patterns of temperature and precipitation, although precipitation is still overestimated. For East Asian summer monsoon precipitation, both UN‐CESM and VR‐CESM tend to overestimate (underestimate) the precipitation over northern (southern) China, which is associated with too strong meridional water vapor transport in the models and biases in the large‐scale circulation in the middle and upper troposphere. Different from previous studies with different physics parameterizations and refined domains, as the model resolution increases, simulated monsoon precipitation evolution is not obviously improved, and convective precipitation intensity decreases over eastern China. Despite this, our results indicate that VR‐CESM simulates regional climate, topographical forcing, and large‐scale circulations over East Asia reasonably well, and thus it can be applied for the future climate projection in the region. Plain Language Summary: Horizontal resolution is one of the key factors that affect the climate model's fidelity. Current global climate models (GCMs) are usually run at a coarse resolution due to the limitation of computational resources. To compromise, variable‐resolution GCM with higher resolution over domains of interest is developed to reduce the computational costs of high‐resolution modeling. This study evaluates the impact of model horizontal resolution on simulating regional climate in terms of temperature and precipitation using the model with a regionally‐refined high resolution over East Asia. Compared to observational data sets and the coarse‐resolution model, the regionally‐refined model better simulates spatial distributions and seasonal variations of climatological mean temperature and precipitation over China, as well as the occurrence frequency of extreme cold events and heavy precipitation. The regionally‐refined model with high‐resolution terrain improves precipitation simulation in regions with complex terrains. Our results demonstrate the regionally‐refined model's capability in simulating the regional climate over East Asia. Key Points: Variable‐resolution CESM2 (VR‐CESM2) performs better in simulating seasonal variation of precipitation over China but overestimates its magnitude than the quasi‐uniform 1° CESM2VR‐CESM2 reasonably simulates frequency of cold temperature extremes and heavy precipitationPrecipitation improvements at higher resolution depend on model physics parameterizations and refined domain size [ABSTRACT FROM AUTHOR]