Abstract: Sea surface temperature (SST) is the most important factor in ocean-atmosphere interaction, playing a crucial role in constraining the moist heat exchange between the air and the ocean. This study, based on linear regression, empirical orthogonal function (EOF) decomposition, wavelet analysis, and multifractal detrended fluctuation analysis (MFDFA), investigates the spatiotemporal variations of sea surface temperature in the South China Sea (SCS) from 1923 to 2022. The results indicate an overall increasing trend in SST over the past century, but with varying responses to the warming at different locations. EOF analysis reveals the first mode showing a basin-wide warming trend. The second and third modes indicate a northeast-southwest and northwest-southeast reversal in SST in the SCS. Wavelet analysis identifies periodicities of 2 years and 3－4 years in the SST changes, validating the trend characteristics and seasonal patterns. MFDFA suggests weak long-term autocorrelation in the SST changes, with different multifractal properties observed in different regions.