Quantifying the Driving Forces of Water Conservation Using Geodetector with Optimized Parameters: A Case Study of the Yiluo River Basin

Accurately identifying the impact of different factors on water conservation is influenced by the spatial grid scale. However, existing studies on water conservation often overlook the Modifiable Areal Unit Problem (MAUP). MAUP is one of the key factors contributing to the uncertainty in spatial analysis results. The Qinling Mountains are a critical water conservation area, with the Yiluo River Basin (YLRB) as a key sub-basin. This study uses the Optimized Parameter GeoDetector (OPGD) model to analyze water conservation changes and influencing factors in the YLRB from 1990 to 2020. By optimizing spatial scale (2 km grid) and driving factor discretization, the OPGD model addresses spatial heterogeneity and the MAUP, enhancing analysis accuracy. Results show a fluctuating upward trend in water conservation depth, averaging 0.94 mm yearly, with a spatial decline from southwest to northeast. High–high and low–low clusters dominate the region, with some areas consistently showing high or low values. Key conservation zones expanded by 2748 km², reflecting significant enhancement. Natural factors, particularly precipitation, predominantly influence water conservation, outweighing human activities. The interaction between precipitation and temperature notably affects dynamic changes, while human impacts, such as land use, play a secondary role. The findings suggest water management should prioritize climatic factors and integrate land-use policies to enhance conservation. The OPGD model’s application improves factor identification and supports targeted ecological and water management strategies.