Abstract: Chlorophyll-a is an important indicator to estimate water quality, and there are no easy solutions to accurately assess the chlorophyll-a of clean water. In this study, the GF-1 images and neural network method were applied, and a simulating model of chlorophyll-a in Qiandao lake was established. The distribution and variations of chlorophyll-a was detected and the superiority of the method was analyzed. The results show that it's feasible to simulate the concentration of chlorophyll-a in clean water using the proposed neural network method since the high value of coefficient; The spatial analysis revealed that the overall concentration of chlorophyll-a in Qiandao lake is quite low and the high values distributed mainly in the southwest and northeast regions. Moreover, we also found that the concentration of chlorophyll-a was quite stable with average value of 1.70 to 1.75 μg/L during the study period.