Combined Modes of the Northern Stratosphere, Tropical Oceans, and East Asian Spring Rainfall: A Novel Method to Improve Seasonal Forecasts of Precipitation.

Seasonal forecasting models have a harder time accurately predicting regional precipitation anomalies in East Asia than the large‐scale drivers of these precipitation anomalies. These large‐scale drivers include both sea surface temperature (SST) anomalies and conditions in the polar stratosphere. A maximum covariance analysis is used to isolate the SST and stratospheric patterns most associated with anomalous East Asian precipitation, which are employed to correct seasonal forecasts by projecting the forecasted stratospheric circulation and SST onto the observed combined modes. The nonuniform predictability of East Asian rainfall in seasonal forecast models is improved after error correction with the observed linkage between the stratosphere–tropical ocean modes and East Asian rainfall. This method should be helpful for improving rainfall forecasts in other regions as well. Plain Language Summary: Seasonal predictability of East Asian rainfall in summer is mainly associated with sea surface temperatures (SSTs) across the tropical Pacific and Indian Oceans. In contrast, East Asian rainfall in spring can be affected by both the state of tropical oceans and downward propagation of extratropical stratospheric variability. This study reports the most common combinations between the extratropical stratosphere and tropical SST in early spring, which in turn are associated with East Asian rainfall variations in spring. Rainfall in some regions like Japan and near the China‐Russia border is poorly forecasted in most seasonal models. After error correction based on the physical linkage between the most frequent stratosphere–tropical SST combinations and East Asian rainfall, an improvement especially in less predictable regions is identified in seasonal forecast models. Key Points: A novel method is proposed to improve (i.e., error correct) seasonal forecasts of East Asian spring precipitationMaximum covariance analysis is used to identify large‐scale drivers associated with rainfallThese drivers are better predicted by models than rainfall, and can be utilized to improve upon the nonuniform predictability of East Asian spring rainfall [ABSTRACT FROM AUTHOR]