Empirical estimation of total phosphorus concentration in the mainstream of the Qiantang River in China using Landsat TM data

Eutrophication is a serious environmental problem in Qiantang River, the largest river in the Zhejiang Province of southeast China. Increased phosphorus concentration is thought to be the major cause of water eutrophication. The objective of this study was to develop an empirical remote sensing model using Landsat Thematic Mapper (TM) data to estimate phosphorus concentration and characterize the spatial variability of the phosphorus concentration in the mainstream of Qiantang River. Field water quality data were collected across a spatial gradient along the river and geospatially overlaid with Landsat satellite images. Various statistical regression models were tested to correlate phosphorus concentration with a combination of other water quality indicators and remotely sensed spectral reflectance, including Secchi depth (SD) and chlorophyll-a (Chl-a) concentration. The optimal regression model was subsequently used to map and characterize the spatial variability of the total phosphorus (TP) concentration in the mainstream of Qiantang River. The results suggest that spectral reflectance from the Landsat satellite is spatially and implicitly correlated with phosphorus concentration (R-2 = 0.77). The approach proved to be effective and has the potential to be applied over large areas for water quality monitoring.Eutrophication is a serious environmental problem in Qiantang River, the largest river in the Zhejiang Province of southeast China. Increased phosphorus concentration is thought to be the major cause of water eutrophication. The objective of this study was to develop an empirical remote sensing model using Landsat Thematic Mapper (TM) data to estimate phosphorus concentration and characterize the spatial variability of the phosphorus concentration in the mainstream of Qiantang River. Field water quality data were collected across a spatial gradient along the river and geospatially overlaid with Landsat satellite images. Various statistical regression models were tested to correlate phosphorus concentration with a combination of other water quality indicators and remotely sensed spectral reflectance, including Secchi depth (SD) and chlorophyll-a (Chl-a) concentration. The optimal regression model was subsequently used to map and characterize the spatial variability of the total phosphorus (TP) concentration in the mainstream of Qiantang River. The results suggest that spectral reflectance from the Landsat satellite is spatially and implicitly correlated with phosphorus concentration (R(2) = 0.77). The approach proved to be effective and has the potential to be applied over large areas for water quality monitoring