Long-term vegetation landscape pattern with non-point source nutrient pollution in upper stream of Yellow River basin

Abstract

Grassland, forest, and farmland are the dominant land covers in upper catchments of the Yellow River and their landscape status has direct connection with dynamics of non-point source (NPS) pollution. Understanding the correlations between landscape variables and different formats of NPS nutrients pollutants is a priority in order to assess pollutants loading and predicting the impact on surface water quality. The regional vegetative cover in 1977, 1996, 2000 and 2006 was determined by classifying historical multi-temporal Landsat imagery and clipping data from the National Landcover Database. The landscape pattern is expressed means of metrics such as patch density, edge density, fractal distribution index, all of which were calculated by FRAGSTATS. The soil and water assessment tool (SWAT) was used to analyze and visualize the fate of NPS nitrogen and phosphorus loads in diverse formats from different land cover types in different years. Statistical analysis indicated that the grassland landscapes played a major role in NPS nutrient pollution dynamics and grassland patch edges benefited pollution control. However, the presence of more forest and farmland lead to more NPS nitrogen emissions. It was found that grassland areas reduced nitrogen loss and had a multi-function role in the nutrient pollution process. Farmland was the direct source of organic nutrients, but did not have great impacts on sediment P and soluble N loadings. Forest areas contributed NPS nutrients pollution loading. The statistical models derived in this study can be used to estimate watershed NPS nutrient pollution losses. These equations can help identify pollution sources and suggest appropriate and effective solutions for planing basin management practices.