Abstract: Monitoring and assessing eutrophication of lake water are basic work, which is important for water resource management and water environment protection. Based on GF-1 WFV images,82 measured data and the method of trophic level index, we built the multiple linear regression model and RBF neural network model, and we got the trophic level index of lakes in Wuhan and its surrounding areas. The inversion result shows that most lakes are in a state of light eutropher and mesotropher, and the partial in middle eutropher. The results show that GF-1 WFV multispectral images are feasible in monitoring the water quality of numerous lakes; we can retrieve the trophic level index and dynamically monitor water quality with functions established by measured data and remote sensing information; the RBF neural network model has a R2 of 0.742 3 and a root mean square error of 3.72. The inversion accuracy of RBF is higher, which is more suitable for monitoring the change of water quality in inland lakes.
Key words：Wuhan; the inversion of lake water quality; trophic level index; GF-1; RBF neural network