Hyperspectral remote sensing of chlorophyll-a in the Chagan Lake, China

An approach for the determination of chlorophyll-a from field reflectance spectra was presented in the Chagan Lake, China. Reflectance spectra were measured from May to September in 2004 with ASD FieldSpec spectrometer. Concurrently, water samples were collected directly from the lake to be analyzed in laboratory. Using correlations between the ground-truth data and combinations of spectral bands from the field spectral data, spectral indices including single band, spectral ratio, and reflectance first-derivative were developed which could be used to estimate chlorophyll-a. The latter two (spectral ratio and reflectance first-derivative) give better results with higher determination coefficient r² = 0.71 and r² = 0.48 with significance level p < 0.01, and are applied to estimate chlorophyll-a concentration. There is a satisfactory correspondence between the two spectral indices and chlorophyll-a concentration, and these algorithms may prove to be useful tools to detect potential water quality problems of the Chagan Lake using remote sensing data.