Spatial and temporal evolution of the “source–sink” risk pattern of NPS pollution in the upper reaches of Erhai Lake Basin under land use changes in 2005–2020

Non-point source (NPS) pollution is a major contributor to water quality degradation of Erhai Lake, and identifying the key “source–sink” regions of NPS pollution risks is important for ecological protection of the lake. In the present work, an evaluation system of resistance base surfaces was built based on data of Erhai Lake Basin in 2005, 2010, 2015, and 2020; resistance surfaces were established, and risk levels were classified based on a minimum cumulative resistance model to analyze the temporal and spatial changes in NPS pollution risk levels over the 16 years of study and identify the driving factors of risk-level changes in the study area. The major findings were as follows: first, the average values of the resistance surface increased first and then declined, showing an overall increase of 833.94 in the 16 years. The distribution of resistance base surfaces and resistance surfaces was spatially heterogeneous: with an increased elevation, the resistance value was lower in the middle part but higher in surrounding areas of the basin, and the role of the “source” landscape was gradually replaced by the “sink” landscape. Second, extremely high-risk zones, high-risk zones, and medium-risk zones accounted for more than half of the total area, and the overall risk level of NPS pollution was high. The distribution of risk levels of NPS pollution in the study area was consistent with the local topography: risk levels in the middle and southern parts were higher than in the southern part of the study area, with Yousuo Town, Dengchuan Town, and Shangguan Town marking the highest risk levels among all townships. Third, governmental policies issued to protect the environment of Erhai have accelerated the risk level transfer, and the junctions of the “source” and “sink” landscapes witnessed the most drastic risk level transfer. From 2005 to 2020, the transferred areas of extremely high-risk zones, high-risk zones, medium-risk zones, low-risk zones, and extremely low-risk zones were 50.68 km2, 49.72 km2, 44.34 km2, 37.71 km2, and 8.36 km2, respectively. Forestland, grassland, and waters were the dominant landscape types of “sink” landscapes with higher impact intensities than the “source” landscapes. Though the area of cultivated land as a “source” landscape was diminishing, its role as a “source” was still stronger than its role as a “sink”. Finally, the risk level of NPS pollution had a strong correlation with the distribution of cultivated land, forestland, and grassland, but not with soil erosion. The intersections of three towns (Cibi Lake, Fengyu, and Yousuo) and Niujie Town marked the main regions of soil erosion, where the average soil erosion modulus has decreased by 8.29t/(hm2·a) over the 16 years. As governmental measures like “grain-to-green” initiatives, ecological zoning of cultivation, and industry restructuring took effect in the study area, the “source–sink” landscape began to play a more positive role in the control of NPS pollution, which in turn affected the spatial and temporal distribution of risk levels. The research findings are expected to provide some reference for decision-making on water environment management in the Erhai Lake Basin.