Influences of hydrometeorological factors on river water δ2H variability in the Xiangjiang River Basin

Climate change intensifies extreme hydrological events, river water stable isotopes serve as sensitive recorders of hydrometeorological changes, yet their controlling factors across scales remain unclear. This study characterized long-term isotope variability and assessed hydrometeorological influences at both site and basin scales in the Xiangjiang River Basin. Based on 13 hydrological years (2010–2022) at Changsha station, we integrated field observations (1668 precipitation events and 929 river water samples) with ERA5 reanalysis, isoGSM2 simulations, and GRACE water storage data. Standardized Runoff Index (SRI) classification and Partial Least Squares Path Modeling (PLS-PM) were applied to analyze 5-day interval variations in river water δ2H (Δδ2HR). Major findings include: (1) Basin-scale data exhibited significantly stronger correlations with Δδ2HR than site-scale data, demonstrating that river water integrated upstream basin information rather than merely reflecting local conditions; (2) Δδ2HR showed distinct non-linear responses to precipitation, with a pronounced “threshold effect” whereby rapid isotopic depletion occurred once basin-scale 5-day precipitation exceeded 70 mm; (3) Basin water storage variation integrated the competing effects of precipitation input and evaporative enrichment, explaining both significant isotopic depletion during water surplus periods and enrichment during deficit periods; (4) PLS-PM revealed a clear seasonal shift in controlling factors: precipitation dominated river water isotopic composition during drought-free periods (SRI > − 0.5), whereas evaporation became the primary controlling factor during drought periods (SRI ≤ − 0.5). These results emphasize the necessity of adopting basin-scale perspectives and incorporating non-linear relationships when interpreting river water isotope dynamics. The findings provide critical scientific insights for paleoclimate reconstruction and offer practical recommendations for improving isotope-enabled hydrological models and regional water resource management in monsoon-influenced river basins.