资源环境科技发展态势分析平台

Finer grids in global climate models could lead to an improvement in the simulation of precipitation extremes. We assess the influence on model performance of increasing spatial resolution by evaluating pairs of high- and low-resolution forced atmospheric simulations from six global climate models (generally the latest CMIP6 version) on a common 1 degrees x 1 degrees grid. The differences in tuning between the lower and higher resolution versions are as limited as possible, which allows the influence of higher resolution to be assessed exclusively. We focus on the 1985-2014 climatology of annual extremes of daily precipitation over global land, and models are compared to observations from different sources (i.e., in situ-based and satellite-based) to enable consideration of observational uncertainty. Finally, we address regional features of model performance based on four indices characterizing different aspects of precipitation extremes. Our analysis highlights good agreement between models that precipitation extremes are more intense at higher resolution. We find that the spread among observations is substantial and can be as large as intermodel differences, which makes the quantitative evaluation of model performance difficult. However, consistently across the four precipitation extremes indices that we investigate, models often show lower skill at higher resolution compared to their corresponding lower resolution version. Our findings suggest that increasing spatial resolution alone is not sufficient to obtain a systematic improvement in the simulation of precipitation extremes, and other improvements (e.g., physics and tuning) may be required.