Multi-scale summer rainfall variability over China and its long-term link to global sea surface temperature variability

Multi-scale summer (Jun-Aug) rainfall variability over China and its long-term link to global sea surface temperature (SST) variability are studied for the period of 1955-1997. First, the dominant spatial and temporal patterns of the observed rainfall anomaly are studied by empirical orthogonal function (EOF) analysis. By a wavelet transform, interannual and decadal-interdecadal variabilities as well as a trend are found, with different dominance, in the first two EOF modes. EOF1 shows a sudden shift in rainfall anomaly over China in the late 1970s, representing overall wetter conditions in central China and drier conditions in northern and southern China in the 1980s than the conditions in the 1960s. This sudden shift is associated with a quasi-in-phase reinforcement between bidecadal and quadridecadal variabilities. EOF2 represents an increasing trend in the rainfall anomaly in broad central and southern China, especially in the Yangtze River valley, without an apparent shift in the late 1970s. The lack of such a shift is associated with an out-of-phase partial cancellation between a bidecadal cycle and the trend around that time. Second, to understand the long-term rainfall variability that is linked to global SST variability, the singular value decomposition (SVD) analysis for the two fields is carried out. SVD1 links drought conditions in northern China and flood conditions in central China to an Fl Nino-like SST anomaly distribution. This mode shows both an apparent trend and a regime shift in the late 1970s, which do not coexist in the rainfall EOF modes. SVD2 links the rainfall anomaly in the area between the Yangtze River and the Yellow River and the opposite anomaly in southern China to a wave-like SST anomaly distribution in the eastern Pacific from tropics to extratropics. SVD3 links the rainfall anomaly in the Yangtze River valley to the SST anomaly in the western Pacific centered near 20 degrees N 140 degrees E. The rainfall variability in different areas of China that can be attributed to SST effects results from the interplay of the SVD modes. The most significant links found from SVD analysis are verified by cross-correlation functions. A scenario for a long-term link on the trend scale between the rainfall over China and global SST variabilities, through the associated large-scale circulation, is presented.