Spatial scale effects of the variable relationships between landscape pattern and water quality: Example from an agricultural karst river basin, Southwestern China

• Published in: Agriculture, ecosystems & environment ecosystems & environment, 2020-09, Vol.300, p.106999, Article 106999
• Main Authors: Xu, Sen, Li, Si-Liang, Zhong, Jun, Li, Cai
• Format: Article
• Language: English
• Subjects: Agricultural karst basin; Geographical characteristic; Landscape pattern; Spatial scale; Water quality
• ISSN: 0167-8809; 1873-2305
• DOI: 10.1016/j.agee.2020.106999

•The relationships between landscape patterns and water quality are scale-dependent.•Water quality variation is better explained by landscape pattern at catchment scale.•The influence of landscape pattern on water quality is stronger in the dry season.•Geographical characteristics influence water quality by interacting with land use. Understanding the spatially varying relationships between landscape pattern and water quality in the agricultural karst river system is helpful for landscape planning and management to protect water quality. In this study, empirical models were established to examine the impact of landscape pattern on the seasonal water quality at multi-spatial scales in a typical karst river basin in Southwestern China. Water samples from the Chishui River basin were collected and analysed during both the wet and dry seasons. Kruskal–Wallis tests revealed significant seasonal variations in water physic-chemical parameter variables, the values of which were all generally higher in the wet period than those in the dry period except for pH and SO42−. Moreover, the nutrient indicators including NO3−-N, PO43−-P, DOC and DSi showed obvious spatial variations, some of which were related to the spatial distribution of cropland that reflected the agricultural activities of the basin. Redundancy analysis results indicated that the landscape pattern metrics can better explain the variation of water physic-chemical parameters in the dry period than in the wet period. The results revealed that the overall variations of water physic-chemical parameters were better explained by landscape pattern metrics at catchment scales. Further, the results of multiple linear regression model showed that the scale effects of the relationship between landscape pattern and water quality varied across different metrics and parameters. The geographical characteristics of catchment had significant relationships with some water physic-chemical parameters, suggesting that it may have impacts on water quality through combining with land use. This study highlights the importance of spatial scale in reflecting land-water interactions.