Correcting Weather and Climate Models by Machine Learning Nudged Historical Simulations

Abstract

Due to limited resolution and inaccurate physical parameterizations, weather and climate models consistently develop biases compared to the observed atmosphere. Using the FV3GFS model at coarse resolution, we propose a method of machine learning corrective tendencies from a hindcast simulation nudged toward observational analysis. We show that a random forest can predict the nudging tendencies from this hindcast simulation with moderate skill using only the model state as input. This random forest is then coupled to FV3GFS, adding corrective tendencies of temperature, specific humidity and horizontal winds at each timestep. The coupled model shows no signs of instability in year-long simulations and has significant reductions in short-term forecast error for 500 hPa height, surface pressure and near-surface temperature. Furthermore, the root mean square error of the annual-mean precipitation is reduced by about 20%. Biases of other variables remain similar or in some cases, like upper-atmospheric temperature, increase in the year-long simulations.