Abstract:

Chlorophyll-a concentration in water is one of the most basic indices for water quality. In order to monitor chlorophyll-a concentration ration in Keqin Lake in Zhalong wetland, reflectance spectra were measured with FieldSpec spectrometer. Remote sensing models of Chlorophyll-a were developed by some methods, including normalized reflectance, first-derivative and spectral ratio, and aimed to monitor precisely chlorophyll-a concentration in Keqin Lake. The results indicated that the linearity algorithm between chlorophyll-a concentration and first-derivative of the reflectance at 697 nm was not high (R2=0.703). The accuracy of the power function algorithm using the normalized reflectance could be improved (R2=0.794). The relationship between chlorophyll-a concentration ration and the band-ratio of R703/R677 was the best, and the accuracy was the highest (R2=0.879). The present study showed that cholorophyll-a concentration was estimated with reflectance band-ratios model in Keqin Lake, and provided research foundation to estimate chlorophyll-a concentration using hyper-spectral remote sensing data.