Key points are not available for this paper at this time.
Abstract Long‐range precipitation forecasting is crucial for flooding control and water resources management. However, making precise forecasting is rather difficult due to the complex climatic factors and large uncertainties arising from long lead times. Sea surface temperature anomaly (SSTA) is one of the strongest signals that influence regional precipitation, often used for the development of precipitation forecasts. Traditional models using SSTA for precipitation forecasting usually screen SSTA over fixed oceanic zones and neglect its preceding temporal fluctuation information. In this study, we introduce a multipole SSTA index and the preceding fluctuation modes of SSTA to develop a monthly precipitation forecasting model, which is applied to the upper Yangtze River basin in China where monthly precipitation during May–October for the period of 1961–2020 are forecasted. Results show that more significant SSTA poles correlated with precipitation are found for September than for the other months. The new approach is able to forecast monthly precipitation for May–October in the basin, particularly for September. It outperforms traditional statistical and dynamical models and has much more skill in forecasting precipitation for June–September when heavy precipitation is more likely to occur than for May or October. Our approach enriches the knowledge of the relationship between precipitation and SSTA, which is conducive to the improvement of long‐range precipitation forecasting.
Wu et al. (Fri,) studied this question.