A Data-Driven Analysis of Hydrological Trends and Correlations in Aosta's Springs

This thesis investigates the long-term thermal and hydrological dynamics of five alpine springs-Cheserod, Entrebin, Gabiet, Promiod, and Promise-located in the Aosta Valley (Northwestern Italy) over a 14-year period (2010–2024). Using high resolution hourly data, the study analyses water temperature, water level, and conductivity to identify trends, seasonal behaviors, and interspring relationships in response to climatic drivers. The time series were first cleaned using Kalman filtering and interpolation techniques, followed by sinusoidal modeling with linear trends. Results reveal that most springs exhibit a subtle but consistent warming trend in water temperature, with varying rates of increase. In contrast, water level trends are more heterogeneous, with some springs showing slight declines. In particular, Gabiet and Cheserod, both used for municipal supply, exhibit distinct patterns of variability. Cross-correlation analyses were performed between all spring pairs to evaluate temporal dependencies and regional hydrological coherence. Strong atmospheric correlations (e.g., air temperature r > 0.95) were found across all sites, while hydrological linkage varied significantly. In particular, only some springs showed synchronous flow or conductivity responses, indicating that even under shared climatic conditions, their aquifer systems respond differently. These findings highlight the influence of climate change on alpine groundwater systems and underscore the importance of site-specific analysis for sustainable water resource management in mountainous regions.