Abstract:With the development of watershed urbanization, the impact of landscape pattern on river basin water quality has gradually increased. The types of pollution caused by urbanization are random, complex, and uncertain. Therefore, it is important to understand how a landscape pattern affects water quality in an urbanized watershed. The Yuqiao Reservoir is an important water source for the Luanhe-Tianjin Water Diversion Project. However, it is unclear how urbanization of the Yuqiao Reservoir watershed influences water quality. In this study, we investigated how the "source-sink" landscape pattern of the urbanized watershed affected nitrogen and phosphorus spatial variations in local rivers. This can potentially provide theoretical support for optimizing the landscape pattern in the Yuqiao Reservoir watershed. The main objective of this study was to identify the relationship between landscape pattern and water quality. The quantitative relationship between the landscape index and water quality was analyzed using landscape indices (including the location-weighted landscape contrast index (LWLI) combined with 33 water quality data for 2013-2015). The results showed that the amount of "sink" landscape, fell, but the amount of "source" landscape rose in the whole basin from upstream areas to downstream areas. The proportion of the land used for residential construction reached 34.6% on average in the middle and lower sub-basin, and the forest area accounted for 33.5% of the "sink" landscape. LWLI's Moran's I for the Yuqiao Reservoir watershed was 0.637 (P<0.01), and this tended to cluster at the spatial scale. LWLI and the spatial patterns for nitrogen and phosphorus were significantly correlated. The coefficient of determination (R2) between the LWLI and total nitrogen (TN) was 0.811 during the normal-water period, and R2 was 0.741 during the high-water period. The NH4+-N, TN, NO3--N, TP, and LWLI water quality parameters tested in the sub-basins were gathered together in the fourth quadrant. LWLI had a greater effection nitrogen and phosphorus in the river than did other landscape pattern indices. It had widely distributed and apparently aggregating characteristics in urban residential land. Urban residential land surrounding Zunhua City was an important source of water pollution in the watershed. Identifying the quantitative relationships between landscape pattern and water quality indices is of theoretical and practical significance when attempting to optimize the regional landscape pattern and improve water quality.