Remote sensing inversion of lake water quality and its response to human activities in multi-scale buffer zones

ABSTRACT With rapid urbanization and industrialization in Hubei, assessing water quality and identifying key influencing factors are crucial for lake conservation. This study utilized remote sensing and machine learning to analyze water quality parameters, including the permanganate index, total phosphorus (TP), and turbidity, in Liangzi Lake (Hubei's second-largest lake) from 2019 to 2022. A spatial quantification model and statistical analysis were employed to assess human activity intensity at different buffer scales (1,000–4,000 m) and identify key influencing factors. Results showed significant seasonal and annual variations in water quality, with the highest pollution in autumn and the lowest in winter. Pollution levels decreased from 2019 to 2020 but gradually increased from 2020 to 2022, possibly related to changes in human activities during the COVID-19 pandemic. Spatially, the Gaotang sub-lake had the highest permanganate index and TP pollution, while the Manjiang sub-lake had the highest turbidity. Precipitation and livestock density were the primary factors, accounting for 26.7–30.8% and 12.9–17.6% of water quality variation, respectively. At the 1,000 m and 2,000–4,000 m buffer zones, industrial output and population density were the dominant human activity factors influencing water quality. This study provides practical insights for targeted lake management strategies and environmental protection efforts.