Abstract: For river shoals exhibiting significant intra-annual area variations in response to water level fluctuations, we first extract time-series bank boundary lines using ESA/Sentinel-2 satellite remote sensing images, and incorporate water level observation data from the nearest hydrological stations upstream and downstream of the river, then establish an elevation information model of the shoal water level line or each characteristic point. Further by integrating the elevation information of the shoal boundary extracted from the satellite remote sensing images and the shoal boundary line, we construct a digital terrain model of the river shoal above the dry water level through TIN spatial interpolation. Then we take Nanyang shoal in Yueyang section of Changjiang River as an example, and use the 1 ∶10 000 measured topographic data of Nanyang shoal collected in 2016 to evaluate the accuracy of the digital terrain model generated by the algorithm of this paper. The results show that: the terrain data of Nanyang shoal above the withered water level reconstructed by the algorithm in this paper meets the requirements of 1 ∶25 000 topographic map specification, and the correlation coefficient R>90%, the average absolute error MAE=0.87 m, the root mean squared error RMSE=1.01 m. The combination of satellite remote sensing data and water level monitoring data to reconstruct the topography above the dry water level of shoals is of great significance for the protection and utilization of the river shoal′s shoreline and the evolution research of shoals.