The effects of landscape pattern evolution on runoff and sediment based on SWAT model

Abstract

To reveal the hydrological effects on changes in land use and landscape pattern of the Wuhua River Basin (hereinafter referred to as the River Basin), this study investigated the upstream of Hezikou Hydrological Station on the Wuhua River; simulated the monthly runoff and sediment yield of the River Basin in a 25-year period (1986–2010) based on the widely used soil and water assessment tool (SWAT) model under the meteorological conditions from 1957 to 2012 and in the context of the landscape change; calculated the change rates in runoff and sediment yield throughout the year and examined the relativity between the rate of change in the runoff and sediment yield and that of land use type composition and landscape pattern using Spearman correlation coefficient. Results revealed the following results. First, the simulation accuracy of runoff and sediment yield based on the SWAT model reached the standard, so the SWAT model was applicable to mesoscale watersheds in hot and humid mountainous areas of southern China. Second, the River Basin was dominated by forest, accounting for around 75% of the whole basin area, followed by cropland, whereas the percentages of built-up land and water body are much lower. With the urbanization and agricultural expansion, a large area of deforestation occurred especially from 1995 to 2005, which potentially resulted in increases of runoff and sediment yield. Third, six typical landscape pattern indices of the River Basin were screened out through the factor analysis. The landscapes of the River Basin under the impact of human activities became fragmented as the spatial heterogeneity increased. Fourth, grassland had the largest soil erosion modulus followed by cropland and built-up land, cropland and forests contributing most to runoff and sedimentation. The changes in runoff and sediment yield mainly occurred near the watershed outlet and upstream in the southern of the river basin, runoff and sedimentation increased significantly in 2005, which are highly likely due to the heavy rain events and deforestation at the time. Fifth, the runoff was significantly affected by the area changes in the cropland and woodland, while the weaker and more insignificant relationship between the changes in landscape pattern indices and that of runoff and sediment yield were observed. Thus, we concluded that the changes in land use have higher and more significant impacts on the runoff and sediment yield in Wuhua River Basin, compared with landscape pattern indices. This work provides useful information for environmental managers to target efforts to reduce soil erosion and related environmental issues effectively. Further research is recommended to integrate climate change and human activities’ impacts on runoff generations and sediment yield processes.