Abstract: Seasonal monitoring of surface water resources provides crucial insights for understanding the dynamic changes in water environments and rationally allocating water and land resources. However, water resource monitoring in Africa faces challenges such as insufficient spatial resolution and inadequate periodic fluctuations. This study investigated the typical Congo and Zambezi river basins in Africa by integrating Sentinel-1, Sentinel-2, and elevation multi-source data. Specifically, the synthetic aperture Radar (SAR) images, optical images, topographic features, and spectral end-member abundance information were integrated. Based on the coverage of optical and SAR data, a zoned modeling strategy was employed to divide the study areas into optical-SAR-covered and SAR-covered zones. Accordingly, differentiated parameter sets were constructed for these zones. These parameter sets, coupled with local climate characteristics and annual water accumulation frequency, enabled fine-scale monitoring of permanent and seasonal water bodies. The results indicate that the extraction accuracies of surface water bodies in 12 months all exceeded 90%. The boundaries of extracted water bodies align highly with the high-spatial-resolution images. The water bodies exhibit uneven spatial distributions within the river basins, with distinct spatiotemporal variations observed across different water body types. Permanent water bodies are primarily distributed in the eastern and central parts of the Congo River and the eastern part of the Zambezi River. In contrast, seasonal water bodies are concentrated in the western part of the Congo River and the central and northwestern parts of the Zambezi River. Overall, this study enables the extraction of surface water bodies and analysis of seasonal variation patterns under varying image coverage conditions, thereby supporting the dynamic monitoring and management of water resources within river basins.