Spatiotemporal characteristics of spring rainfall over Southwest China and their relationships with sea surface temperature during 1961–2017

The spatiotemporal variability of southwest China (SWC) spring (March–April–May) rainfall (SWCSR) during 1961–2017 is investigated using rainfall data from 75 meteorological stations over SWC in this study. It is found that (1) SWC can be divided into four subregions using the rotated empirical orthogonal function (REOF) method: southern Yunnan province (SY), northwestern Yunnan province (NWY), eastern Guizhou province (EG), and northeastern Sichuan province (NES). (2) All subregions show significant 2–8-year periodicity on the interannual timescale. However, only SY rainfall shows significant interdecadal periodicity. (3) The spring La Nina has a significant impact on the simultaneous rainfall over SY. The sea surface temperature anomalies (SSTAs) in the central North Pacific during late winter can be an important oceanic signal for the prediction of spring rainfall variability over NWY. The southern Indian Ocean dipole (SIOD)–like SSTA pattern during late autumn (early winter) can be an important predictor of rainfall variability over EG (NES) during the following spring. Moreover, the impact of the late autumn SIOD-like pattern on the following spring rainfall over EG is stronger than the impact of the early winter SIOD-like pattern on the NES rainfall during the following spring.