High-Resolution Simulations with the Community Earth System Model (CESM): An Update

As impacts of climate change are being felt by the society through sea level rise, increased intensity and occurrences of heat waves, droughts, extreme rainfall events and / or tropical cyclones (TCs), just to list a few, decision makers and stakeholders need reliable weather and climate information at increasingly finer spatial and temporal scales. Beyond such actionable aspects, there are numerous science questions regarding representation of and changes in importance of various processes with increased model resolution as well as their interactions with each other such as how TCs and oceanic mesoscale eddies interact with each other and with large-scale circulations. It is generally anticipated that with less reliance on uncertain parameterizations and their parameter choices, high-resolution models will represent various processes and coupled interactions of the Earth system with increased fidelity. To address these needs and challenges, we have made significant advances in high-resolution global climate modeling and predictions. Specifically, we have performed an unprecedented set of simulations at a TC-permitting and ocean-eddy-rich horizontal resolution using the Community Earth System Model (CESM 1.3), with additional modifications and improvements (hereafter referred to as CESM-HR). CESM-HR uses a 0.25° grid in the atmosphere and land and a 0.1° grid in the ocean and sea-ice components. These simulations include: a 500-year pre-industrial control simulation; 150-year 1%CO₂ per year increase and 4xCO₂ simulations; a 10-member ensemble of historical simulations; 10-member ensembles each of RCP8.5 and RCP6.0 future scenario simulations; 1 member each of RCP4.5 and RCP2.6 future scenario simulations; all HighResMIP coupled and AMIP simulations; and 10-member ensembles of 5-year decadal prediction simulations for the 1980-2023 period with May and November start dates for each year. The presentation will introduce these simulations and provide a few highlights from our extensive analysis. In general, high‐resolution simulations show significant improvements in representing global-mean surface temperature, oceanic heat uptake, sea level changes, extreme events such as TCs and winter-time extreme precipitation, and recent cooling and expanding sea-ice trends in the Southern Ocean. There are also improvements in prediction skill for several fields of interest. Our analysis shows that the projected increase in daily extreme precipitation over global land by the end of this century under the business-as-usual scenario is nearly double in the high-resolution simulations compared to its low-resolution counterpart, suggesting that current low-resolution models may significantly underestimate the future threat. Moreover, high-resolution simulations suggest that future precipitation intensifications arise from both moisture and circulation changes. This finding is in contrast with low-resolution simulations which primarily attribute such changes to increased moisture with warming. While not a panacea to address all the biases, these high-resolution simulations certainly offer promising potential to reduce model biases and uncertainties in comparison with their low-resolution counterparts and to improve our understanding of processes. Datasets from many of these simulations are now available to the broader community.