Abstract: Due to limited monitoring points and insufficient pollution source data,the water quality simulated by traditional mathematical models has low accuracy. In this study,the Xiangxi River in the Three Gorges,where water blooms have frequently occurred in recent years,was used as the research area. A method was proposed to simulate water quality,using water quality remote sensing results as the initial conditions of the Environmental Fluid Dynamics Code（EFDC）. The multiple linear regression model was selected for remote sensing inversion,and the multiple correlation coefficients of the simulated and measured values of chlorophyll a（Chl-a）,total nitrogen（TN）and total phosphorus（TP）reached 0.983,0.848 and 0.736,respectively. The results show that the proposed technology combining remote sensing with mathematical model can improve the effect of model simulation. Compared with using only limited in-situ data,this technology decreased the average error of the simulation of Chl-a from 3.96% to 3.84%,that of TN from 44.57% to 39.27%,and that of TP from 22.17% to 19.44%.The study shows that the technology coupling remote sensing with EFDC model has high practical value for the water quality simulations of rivers lacking observational data.