Modeling the long-term impacts of land use change on surface runoff and non-point source pollution integrating remote sensing and GIS for the Liuxi River watershed

Abstract

In the context of integrated water resource management, water problems should be managed in a basin-wide context and considered in relation to land-based issues, among which land use change has been proven to have significant impacts on water resource. In this study, land use change and its long-term impacts on surface runoff and nonpoint source pollution are investigated for the Liuxi River watershed, which is the only one drinking-water-supply river basin located inside the administrative area of Guangzhou. The land use change analysis was conducted based on land use classification results generated from Landsat TM images. For the period 1990-2005, the Liuxi River watershed experienced a high-speed urbanization and appeared a decreasing trend in forest and agricultural land areas. The Long-Term Hydrologic Impact Assessment and Non Point Source Pollutant Model (L-THIA NPS), a GIS-based empirical model, was applied to assess the long-term impacts of land use change on surface runoff and nonpoint source pollution in terms of biological oxygen demand (BOD), total nitrogen (TN), total phosphorus (TP) and total suspended solids (TSS). The total runoff volume of the watershed was increased contributed by the increase of built-up areas. The loading of BOD has increased significantly as the concentration in runoff from urbanized areas is much higher than that from nonurban areas. Although urban runoff had relatively low contents of nutrients and suspended matters compared with nonurban runoff, the large increases in runoff volume from urban areas produced significant amounts of nutrients and suspended matters and so compensated the decrease of agricultural land, resulting in the overall increments of TN, TP and TSS. The spatial variation of land use change and consequent spatially varied impacts were also analyzed at sub-watershed scale. The results revealed that urbanization was intensified along the river, and the areas of agriculture land increased in the upstream and midstream parts of the watershed but decreased in the downstream part. Surface runoff depth was increased in almost all sub-watersheds, and the increases in urban areas provided main contribution. The increase in BOD load per unit area was obvious, and its spatial variation trend was in accordance with the spatial pattern of urban sprawl, presenting an increasing trend along the river and high risk of organic pollution at the river downstream. For the emissions of agriculture-dominated pollutants like TN, the increase of agricultural land undoubtedly has dominant effect, whereas the impacts of increased urban land may or may not exceed the decrease of agricultural land depending on their change extents.