PNNL

Abstract

Large-scale dynamical and thermodynamical processes are common environmental drivers of extreme weather events. However, such large-scale environmental fields often display systematic biases in climate simulations, posing challenges to evaluating the changes in extreme weather events and associated risks in current and future climate. In this paper, a machine learning (ML) approach was employed to bias correct the large-scale wind, temperature and humidity simulated by the E3SM atmosphere model at $\sim 1^\circ$ resolution, in order to obtain a more realistic representation of the large-scale storm environment associated with extreme weather events. The usefulness of the proposed ML approach for extreme weather analysis in low-resolution climate models was demonstrated with a focus on three different types of extreme weather events, including tropical cyclones (TCs), extratropical cyclones (ETCs) and atmospheric rivers (ARs). We show that the ML model can effectively reduce the mean climate bias in large-scale wind, temperature, and humidity while preserving their responses to imposed climate change perturbations. Specifically, the bias correction is found to directly improve the water vapor transport associated with ARs. More realistic representations of ETC structure and ETC-induced changes in water vapor transport and thermodynamical flows are also obtained in the simulations with ML bias correction. When the ML bias-corrected large-scale winds are used to drive a synthetic TC track forecast model over the Atlantic basin, the resulting TC track density agrees better with that of the TC track model driven by observations. In addition, the ML bias correction does not significantly interfere with the mean climate change signals of large-scale wind, temperature and humidity as well as the occurrence and intensity of the three types of extreme weather events. The findings in this study suggest that the proposed machine learning bias correction can be a useful tool to improve the downscaling of extreme weather events in low-resolution climate models by providing more realistic large-scale storm environment information.