Abstract: To address the limitations of existing deep learning methods in effectively capturing multi-scale evolution patterns of sea surface temperature (SST) and their insufficient capability in modeling SST nonlinear variations, this study first integrates Variational Mode Decomposition (VMD) with a 3D SwinLSTM network to develop a hybrid forecasting model. VMD is used to decompose spatiotemporal signals of SST across multiple scales, revealing the complex multi-scale variability characteristics of SST and enhancing the model's ability of capturing multi-scale information, thereby improving forecast accuracy. Moreover, by incorporating exogenous drivers such as heat flux and momentum flux, the model's prediction precision is further enhanced, and experimental validation for the model is conducted in the western pole area of the Indian Ocean Dipole. Compared to existing baseline models, the proposed hybrid model that integrates multi-scale variability features and exogenous drivers exhibits superior performance, which achieves a root mean square error (RMSE) of 0.70 ℃, suggesting the effectiveness of multi-scale periodicity information in SST forecasting. The introduction of exogenous drivers further reduces RMSE to 0.62 ℃, confirming their significant contribution to the improved prediction skill.