Monsoon-ocean coupled modes in the South China Sea (SCS) were investigated by a combined singular value decomposition (CSVD) analysis based on sea surface temperature (SST) and sea surface wind stress (SWS) fields from SODA (Simple Ocean Data Assimilation) data spanning the period of 1950 1999. The coupled fields achieved the maximum correlation when the SST lagged SWS by one month, indicating that the SCS coupled system mainly reflected the response of the SST to monsoon forcing.Three significant coupled modes were found in the SCS, accounting for more than 80% of the cumulative squared covariance fraction. The first three SST spatial patterns from CSVD were: (Ⅰ) the monopole pattern along the isobaths in the SCS central basin; (Ⅱ)the north-south dipole pattern; and (Ⅲ) the west-east seesaw pattern. The expansion coefficient of the SST leading mode showed interdecadal and interannual variability and correlation with the Indo-Pacific warm pool (IPWP), suggesting that the SCS belongs to part of the IPWP at interannual and interdecadal time scales. The second mode had a lower correlation coefficient with the warm pool index because its main period was at intra-annual time scales instead of the interannual and interdecadal scales with the warm pools.The third mode had similar periods to those of the leading mode, but lagged the eastern Indian Ocean warm pool (EIWP) and western Pacific warm pool (WPWP) by five months and one year respectively, implying that the SCS response to the warm pool variation occurred from the western Pacific to the eastern Indian Ocean, which might have been related to the variation of Indonesian throughflow. All three modes in the SCS had more significant correlations with the EIWP, which means the SCS SST varied much more coherently with the EIWP than the WPWP, suggesting that the SCS belongs mostly to part of the EIWP. The expansion coefficients of the SCS SST modes all had negative correlations with the Ni(n)o3 index, which they lag by several months, indicating a remote response of SCS SST variability to the El Ni(n)o events.