Prediction of summer extreme precipitation over the middle and lower reaches of the Yangtze River basin

Abstract

Based on high-resolution observed daily precipitation data and atmospheric circulation data, a physics-based empirical model (PEM) is built for the prediction of summer extreme precipitation (SEP) over the middle and lower reaches of the Yangtze River basin (MLYRB; 27°–33°N, 108°–120°E). Two preceding predictors with significant physical influences (i.e., spring sea surface temperatures [SSTs] across the northern Indian Ocean [NIO; 20°S–20°N, 50°–95°E] and sea level pressure [SLP], which was defined as spring mean SLP anomalies averaged over the western North Pacific (WNP; 30°S–30°N,120°E–150°W) minus spring mean SLP anomalies averaged over the Aleutian Islands (AIs; 50°–70°N,160°E–160°W)) are selected. Analyses of these physical mechanisms show that when spring SSTs over the NIO are higher, the South Asian High (SAH) extends to the east and the western Pacific subtropical high (WPSH) extends to the west; therefore, the generated secondary circulation induces anomalous upward motions and increases water vapour transport to the MLYRB, which results in increased SEP. Meanwhile, the increase in SEP over the MLYRB is related to control by conditions via anticyclonic anomalies over the Philippines, which are maintained through the interaction between abnormally high WNP and dipole SST anomalies. The regression model is built over the period 1961–1999 with a correlation coefficient skill of 0.56 (p < 0.01), and the independent forecast of the PEM over the validation period 2000–2014 shows a skillful SEP prediction, with a significant correlation coefficient of 0.52 (p < 0.05).