Due to limited resolution and inaccurate physical parameterizations, weather and climate models consistently develop biases compared to the observed atmosphere. These biases are problematic for forecasting on timescales from medium-range weather to centennial-scale climate. Using the FV3GFS model at coarse resolution, we propose a method of machine learning corrective tendencies from a hindcast simulation nudged towards an observational analysis. We show that a random forest can predict the nudging tendencies from this hindcast simulation 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 500hPa height, surface pressure and near-surface temperature. Furthermore, the root mean square error of the annual-mean precipitation is reduced by about 20%.

In Feb. 2017, a five-day sequence of atmospheric river storms in California, USA, resulted in extreme inflows to Lake Oroville, the state’s second-largest reservoir. Damage to the reservoir’s spillway infrastructure necessitated evacuation of 188,000 people; subsequent infrastructure repairs cost $1 billion. We assess the atmospheric conditions, snowmelt, and runoff against major historical events. The event generated exceptional runoff volumes (second-largest in a 30 year record) partially at odds with the event precipitation totals (ninth-largest). We explain the discrepancy with observed record melt of deep antecedent snowpack, heavy rainfall extending to unusually high elevations, and high water vapor transport during the atmospheric river storms. An analysis of distributed snow water equivalent indicates that snowmelt increased water available for runoff watershed-wide by 37% (25-52% at 90% confidence). The results highlight an acute flood risk to public safety and infrastructure projected to increase in severity in a warmer and more variable climate.