Impacts of land-use on surface waters at the watershed scale in southeastern China: Insight from fluorescence excitation-emission matrix and PARAFAC

Sci Total Environ. 2018 Jun 15:627:647-657. doi: 10.1016/j.scitotenv.2018.01.279. Epub 2018 Feb 2.

Abstract

In recent years, the Chinese government has strengthened its efforts in surface water protection and restoration through strict policies and heavy investments. A clear understanding of the impacts of land use on water quality is necessary in order to ensure an effective and efficient implementation of the ongoing surface water restoration program in China. To this end, four small watersheds (less than 5000 ha) in southeastern China, which have clear gradients in the intensities of agriculture (17.0-45.4%), forest (35.2-73.6%) and built-up area (3.3-8.5%), were investigated regarding the impacts of land use on water quality. In addition to the general water quality indices, characteristic components derived by fluorescence excitation-emission matrices (FEEMs) coupled with parallel factor analysis (PARAFAC) were employed to explore a more accurate association between land use and water quality. The results show that agricultural intensity has significant effects by elevating the concentrations of dissolved organic carbon (DOC, an approximate six-fold increase) and total phosphorous (TP, an approximate four-fold increase) in the surface waters. A total of five PARAFAC components representing terrestrial (three components) and protein-like (two components) substances were identified. The PARAFAC results indicate that land-use patterns affected the dissolved organic matter (DOM) in the aspects of both amount and composition. The intensity (R.U.) of the terrestrial components showed a strong correlation (r2 = 0.95, p = 0.01) with agricultural land percentage. Moreover, although the proportion of built-up area varies with a relatively small range, a protein-like component could predict its impact with excellent sensitivity (r2 = 0.94, p = 0.02), whereas the general water quality indices were incapable of predicting the impact due to their multiple sources. The results of this study demonstrate that the FEEMs-PARAFAC technique provides an inexpensive and effective tool for policy makers to overcome the insensitivity of general water quality indices, particularly for the restoration of watersheds with complex land-use patterns.

Keywords: Agriculture; DOM; Ningbo; Sewage; Water restoration.