RESUMO
River water quality is influenced by land use and landscape distribution patterns. Quantifying the relationship between land use, landscape pattern and water quality factor at different riparian buffer zone scales is of great significance for rational land use planning and water quality improvement. Based on water quality data from 91 sites in May 2021 in the Gaya River Basin, we analyzed the spatial characteristics of land use types and landscape patterns at the riparian buffer zone scales. With redundancy analysis (RDA) and generalized additive models (GAM), we examined the effects of land use and landscape patterns on river water quality. The results showed that water quality was primarily impacted by total nitrogen (TN). Farmland was the dominant land use type at riparian buffer zone of 50, 100 and 500 m. The sampling sites were classified into farmland dominant group and farmland other group. Forest was dominant at riparian buffer zone of 1000, 1500, 2000 m, and the sampling sites were classified into forest dominant group and forest other group. 100 m riparian buffer zone was the strongest scale in the Gaya River, and 1000 m was the second. Land use types in the forest dominant group were closely related with electrical conductivity, dissolved oxygen, phosphate, permanganate index and ammonium (NH4+-N) of water. NH4+-N was positively correlated with proportion of forest and farmland area. Phosphate was significantly affected by Shannon diversity index (SHDI). SHDI and largest patch index (LPI) was the key landscape indices affecting permanganate index. TN was significantly impacted by area proportion of forest, grassland and LPI in farmland dominant group, showing decreasing trend with the area proportion of forest increasing from 8% to 40%. Total suspended solids in farmland other group were significantly correlated with proportion of farmland area, while negatively correlated with proportion of forest area. Water quality in the Gaya River was mainly affected by proportion of forest area, followed by proportion of farmland area. The combined effects of LPI, SHDI and other land use types played an important role in affecting water quality.