This study investigates the intricate effects of lithology, temperature, discharge, and land use changes on headwater stream chemistry by analysing two decades of hydrochemical data from twenty karst headwater catchments in the Garonne River basin, France. Focused on the Pyrenees and the lowland (LM) and upland (UM) regions of the Massif Central, this study identified significant regional variations and commonalities in water chemistry. The headwater streams were clustered based on their hydrological and hydrochemical profiles, revealing strong similarities between upland sites, i.e. UM and Pyrenees, despite their geographical distance. The findings revealed a predominance of water driven by calcite dissolution, with specific influences from minor lithologies. Seasonal variations in water chemistry were primarily driven by hydrological conditions. Trend analyses highlighted increased pCO2 concentration in both the Pyrenees and UM, linked to higher forest density and agricultural activities, respectively. In contrast, LM exhibited increasing Ca2+ and HCO3- concentrations alongside decreasing trends in pCO2 and discharge, and increased nitrate concentration. While overall water temperatures increase, only a few sites exhibited significant warming trends, consistent with similar studies in the region and worldwide. These findings underscore the complex interplay between land use changes and hydrochemical dynamics in karst headwaters. They reveal that rising pCO2 concentration trends in upland regions are driven by reforestation and agricultural practices, which have significant implications in CO2 emissions, and consequently for regional and global carbon budgets and carbon-related policies. In lowland areas, declining water resources and increasing ion concentrations highlight potential challenges for water management, particularly in sensitive karst catchments. This study provides a baseline for understanding how karst headwaters respond to environmental changes. Expanding this research to other karst systems worldwide, under different climates, would help validate and model these findings, and improve our understanding of the global carbon cycle.
Read full abstract