his research proposes a method of obtaining underwater terrain data of rivers based on Google Earth remote sensing images and using Kriging interpolation to generate terrain data for the simulation of hydrodynamic water quality of rivers in areas without measured topographical data. A two-dimensional hydrodynamic water quality model is used to analyze the effect of sewage outlets on the water quality of the Jiujiang section of the Yangtze River. The method proposed in this paper is compared with the calculation results of the Jiujiang Hydrology Bureau's 2017 1∶1 000 measured underwater terrain: the maximum error of the water level from 0.1 m to 0.2 m, the maximum relative error of flow rate increases from 10% to 20%, and the maximum relative error of water quality simulation increases from 17% to nearly 30%. The image data method has a large error and is not suitable for detailed environmental quality analysis. However, it can still be promoted as a technical method in the rapid analysis of the lack of measured terrain data and the need to understand the evolution trend of water quality.