Abstract. Extreme drought events have frequently occurred in Southwest China (SWC) in this century. The rainfall of SWC could be related to several climate systems such as the East Asia monsoon, Indian monsoon and El Nino-southern Oscillation. Also it tightly depends on the variety of landforms and the complex terrain of SWC. Therefore the mechanism and prediction of rainfall in the area have became a difficult and central issue in climatology. Here we develop a novel multi-variable network method to delineate the relation between the global sea surface temperature anomalies (SSTA) and the rainfall of SWC. Our results show the important degree patterns in the Western, Eastern Pacific Ocean and Indian Ocean, which significantly influence the rainfall in SWC. Particularly the patterns will change with season and connect to some specific subareas within SWC. The strongest teleconnection is observed for the spring rainfall. The underlying mechanism of our observed teleconnection could be related to the large-scale ocean–atmosphere circulations. Moreover, we can identify the time-lag of the teleconnection links that can potentially improve the prediction of rainfall in SWC.