Establishment of chlorophyll-a concentration distribution model in Dahuofang reservoir based on HJ-1 satellite

Abstract

With the latitude and longitude coordinates and same place, 26 field monitoring data of Dahuofang was associated with HJ-1 satellite multispectral CCD data. This paper analyses the correlation between chlorophyll-a concentration and four bands, band combination of multispectral data. The Pearson's correlation coefficient was calculated by MATLAB software to find band combination T1, T27, T32 which have higher Pearson coefficient and then establish the linear model with these independent variables. Due to the complex optical characteristics of the water body, the relationship between chlorophyll-a concentration and the spectrum can't simply be described with a linear model. So in the paper we use BP neural network for modeling and prediction. The results show that BP neural network is better than the linear model and it can be used for the prediction of chlorophyll-a concentration. © Springer-Verlag Berlin Heidelberg 2013.