Causality among landscape characteristics, seasonality and stream water quality in the Paraopeba river basin

Abstract

Anthropogenic pressures on the environment are increasingly evident, characterized by uncontrolled changes in land use that adversely affect water quality. This study aims to assess how land use and land cover contribute to water quality and to evaluate the influence of spatial landscape metrics on water quality variability in eight tributary sub-basins of the Paraopeba River. The analysis considers two seasonal periods reflective of the region's tropical climate. The dataset includes spatial data on land use and land cover, digital elevation models, soil types, geology, geomorphology, spatial-temporal data, and landscape fragmentation metrics. First, spatial differences in water quality data collected at each sampling site were tested, and the significance of seasonal variations was assessed. Correlation analyses were then conducted to determine the relationships between landscape metrics and water quality parameters across the eight sub-basins, considering both seasonal periods. Key findings include the identification of mixed pollution sources, such as pasture, urban areas, and mining, which significantly affect water quality, particularly during the rainy period. Conversely, forest plantations were found to be the land use category that most positively contributed to the preservation of water quality. The relationships between landscape patterns and water quality, analyzed using redundancy analysis, revealed that the influence of landscape metrics on the variation of water quality parameters was significantly more pronounced during the dry period, explaining 75 % of the variation, compared to 49 % during the rainy period.