Total suspended solids concentration retrieval by hyperspectral remote sensing in Liaodong Bay

With high radiation and spectral resolution, hyperspectral remote sensing can be used to detect and extract small signals of submerged water features. This study investigated the potential of the satellite hyperspectral data-Hyperion image for mapping total suspended solids (TSS) concentration of coastal water in Liaodong Bay, China. After processing and atmospheric correction, the reflectance of water extracted from Hyperion image can be used to express the spectral characteristics of different TSS concentration. The estimated algorithms of TSS concentration based on water reflective spectra data collected in situ. The results indicated derivative spectra had been shown to be an effective technique for separating the contribution of TSS of the composite spectrum of coastal water. The significantly correlative coefficient (R) between TSS concentration and the first derivative of reflectance at 529nm was -0.73. Compared with the single band and band ratio reflectance data, first derivative of reflectance used in polynomial regression method can get better accuracy and the determination coefficient (R-2) can be improved to 0.58. The estimated TSS concentrations were evaluated by field truth data.