In this study we use the National Centers for Environmental Prediction (NCEP) monthly sea surface temperature (SST) fields (1982-1994) to investigate the temporal and spatial variabilities of the South China Sea(SCS) warm/cool anomalies. Three steps of analysis were performed on the data set: ensemble mean ((T) double over bar), composite analysis to obtain the monthly mean anomaly relative to the ensemble mean ((T) over tilde), and empirical orthogonal function (EOF) analysis on the residue data relative to ((T) double over bar) + ((T) over tilde). The ensemble mean SST field ((T) double over bar) has a rather weak, horizontal gradient: 29 degrees C near the Borneo coast to 25 degrees-26 degrees C near the southeast China coast. Two areas of evident SST anomalies were found in the monthly (T) over tilde variation: west of Borneo-Palawan Islands (WBP) and southeast of the southern Vietnam coast (SVC). Four patterns, monsoon and transition each with two out-of-phase structures, were found. During the spring-to-summer transition (March to May) the warm anomaly is formed in the northern SCS with (T) over tilde > 1.8 degrees C located at 112 degrees-119 degrees 30'E, 15 degrees-19 degrees 30'N. During the fall-to-winter transition (October to November) the northern SCS (north of 12 degrees N) cool anomaly is formed in November with (T) over tilde < -0.6 degrees C located at 108 degrees-115 degrees E, 13 degrees-20 degrees N. We performed an EOF analysis on the residue data relative to (T) double over bar + (T) over tilde in order to obtain transient and interannual variations of the SST fields. EOF1 accounts for 47% of the variance and represents the northern SCS warm/cool anomaly pattern. EOF2 accounts for 14% of the variance and represents the southern SCS dipole pattern. Strong northern SCS warm anomaly (1 degrees C warmer) appears during October-November 1987 and January-February 1988, and strong northern SCS cool anomaly (1 degrees C cooler) occurs during March 1986 and November 1992. Furthermore, a strong cross correlation between wind stress curl and SST anomalies, computed from the European Centre for Medium-Range Weather Forecast analyzed wind stress data and the NCEP SST data for different lags, shows the existence of an air-sea feedback mechanism in the SCS deep basin.
