[Relationship Between Source-Sink Landscape Pattern and Antibiotics in Surface Water in Peri-urban Watershed]

Quantitively identifying the effect of land use patterns on antibiotics in surface water has significance in maintaining water quality and protecting residents' health in urban and rural regions. In this study, a typical peri-urban watershed, located in the Yangtze River Delta, was selected as the study area. Based on surface water sampling, laboratory analysis, and source-sink landscape model (SSLM) analysis, the component and distribution characteristics of antibiotics in surface water in different sub-watersheds were analyzed. The effects of source and sink landscape patterns on antibiotic concentrations in surface water were identified. The results of this study showed substantial differences in types and concentrations of antibiotics in surface water in different sub-watersheds. The total concentrations of antibiotics in surface water ranged from 1.12 ng·L^-1 to 53.74 ng·L^-1. From upstream to downstream, the area of "source" landscape increased, and the area of "sink" landscape decreased based on landscape pattern analysis. The results of non-metric multidimensional scaling (NMDS) showed that sub-watersheds with similar "source-sink" landscape patterns were detected as having similar antibiotics types and concentrations in surface water. Land use composition, distance, elevation, and slope degree had substantial impacts on antibiotic concentrations in surface water. The results of this study also found that location-weighted landscape index (LWLI) was positive correlated with antibiotics concentrations in surface water based on correlation analysis and redundancy analysis. The sub-watersheds with high LWLI values usually had relatively higher antibiotic concentrations in surface water. This study indicated that optimization of "source" and "sink" landscapes at the watershed scale can decrease antibiotic contamination in surface water. Furthermore, SSLM is an effective tool in landscape optimization at the watershed scale.