Relationships of groundwater quality and associated health risks with land use/land cover patterns: A case study in a loess area, Northwest China

This study investigated the relationships of groundwater quality and associated health risks with land use/land cover (LULC) patterns. Twenty-nine groundwater samples were collected from a loess area in northwest China, and analyzed for twelve water quality parameters. Water quality was assessed using entropy water quality index (EWQI) and the non-carcinogenic risks caused by NO3− and Cr6+, and the carcinogenic risks caused by Cr6+ through drinking water exposure pathway were considered. The LULC information was extracted from remote sensing image data and classified using a neural net classification method. A curved streamline searchlight shaped model (CS-SLM) was proposed and applied to determine the domain around a well where LULC influences well water quality. Kendall’s tau (τ) was calculated to determine the relationship of water quality and associated health risks with LULC. As suggested by EWQI, 51.72% of the groundwater samples were unacceptable for drinking proposes. The main LULC types in the study area are loess, forest, grassland, agricultural land and urban land, and the groundwater quality is influenced by forest, grassland and loess. Non-carcinogenic risks caused by NO3− and Cr6+ in the groundwater are related to grassland, loess and urban land, and carcinogenic risks caused by Cr6+ are associated with loess and urban land.