Accurate estimation of Alpine Snow Water Equivalent (SWE) is important for water resource management and hydropower operations, particularly under changing climatic conditions. Traditional hydrological assessments rely primarily on point-scale ground networks, such as rain and snow gauges, which are affected by several limiting factors, including the valley-bottom sampling bias, wind-induced undercatch, and thermal sensor limitations can lead to an underestimation of high-elevation snow accumulation. This study investigates a multi-source monitoring approach for the north-western Alps by combining ground observations with Sentinel-1 C-band Synthetic Aperture Radar (SAR) data. Mass-balance analysis indicates that rain gauge measurements underestimate SWE by approximately 31% on average. Sentinel-1 SAR provides reliable mapping of snow-covered areas (AUC = 0.935), but its performance is constrained under specific conditions, including reduced sensitivity to snow depths below 5 cm, signal saturation for snowpacks deeper than 2.5 m, and pulse absorption associated with wet snow during the spring melt season. The methodology was validated in the Toce River catchments using historical naturalized discharge records provided by ENEL. Results show that SAR-based information improves SWE estimation and reduces biases associated with gauge-based observations. These findings support the integration of SAR observations into operational hydrological monitoring and highlight opportunities for future work, including the use of lower-frequency SAR systems for deep snow retrieval and the incorporation of thermo-radiative to represent summer glacier ablation.
La stima dell’Equivalente Idrico della Neve (SWE) in ambiente alpino è un fattore indispensabile per la gestione sostenibile delle risorse idriche e l’ottimizzazione della produzione idroelettrica, specialmente nello scenario attuale di cambiamento climatico. La modellazione idrologica si è affidata storicamente a reti di monitoraggio puntuali al suolo, costituite da pluviometri e nivometri, che però soffrono di limiti strutturali e fisici: la dislocazione prevalentemente di fondovalle, l'azione del vento (wind undercatch) e le inefficienze termiche dei sensori (capping) generano infatti marcate sottostime volumetriche degli accumuli in alta quota. Per superare tali criticità, il presente lavoro sviluppa e valida un paradigma di monitoraggio multi-sorgente per le Alpi nord-occidentali (Piemonte e Valle d’Aosta), integrando le misurazioni a terra con il telerilevamento satellitare Radar ad Apertura Sintetica (SAR) in Banda C della missione Sentinel-1 (progetto C-SNOW). L'indagine quantifica le vulnerabilità tecnologiche: il bilancio di massa globale certifica che i pluviometri sottostimano circa il 31% dell'equivalente idrico reale; il SAR, di contro, eccelle nella mappatura dell'estensione nivale (AUC = 0,935), ma mostra vincoli operativi: cecità per manti inferiori a 5 cm, saturazione del segnale oltre i 2,5 metri e assorbimento dell'impulso causato dalla neve bagnata (wet snow) primaverile. La validazione dell’architettura integrata a monte di impianti idroelettrici conferma l'efficacia del telerilevamento nel correggere le sovrastime pluviometriche. I risultati delineano una base modellistica e tracciano i futuri sviluppi: l'impiego di SAR a bassa frequenza per i grandi accumuli e l'integrazione di un modulo termo-radiativo per calcolare l'ablazione glaciale estiva, essenziale per un sistema predittivo completo.
Integrazione di dati radar Sentinel-1 e reti meteorologiche terrestri per la stima dell'equivalente idrico della neve: validazione sul bilancio idrologico di bacini idroelettrici alpini
Fumagalli, Matteo
2025/2026
Abstract
Accurate estimation of Alpine Snow Water Equivalent (SWE) is important for water resource management and hydropower operations, particularly under changing climatic conditions. Traditional hydrological assessments rely primarily on point-scale ground networks, such as rain and snow gauges, which are affected by several limiting factors, including the valley-bottom sampling bias, wind-induced undercatch, and thermal sensor limitations can lead to an underestimation of high-elevation snow accumulation. This study investigates a multi-source monitoring approach for the north-western Alps by combining ground observations with Sentinel-1 C-band Synthetic Aperture Radar (SAR) data. Mass-balance analysis indicates that rain gauge measurements underestimate SWE by approximately 31% on average. Sentinel-1 SAR provides reliable mapping of snow-covered areas (AUC = 0.935), but its performance is constrained under specific conditions, including reduced sensitivity to snow depths below 5 cm, signal saturation for snowpacks deeper than 2.5 m, and pulse absorption associated with wet snow during the spring melt season. The methodology was validated in the Toce River catchments using historical naturalized discharge records provided by ENEL. Results show that SAR-based information improves SWE estimation and reduces biases associated with gauge-based observations. These findings support the integration of SAR observations into operational hydrological monitoring and highlight opportunities for future work, including the use of lower-frequency SAR systems for deep snow retrieval and the incorporation of thermo-radiative to represent summer glacier ablation.| File | Dimensione | Formato | |
|---|---|---|---|
|
2026_07_Fumagalli.pdf
accessibile in internet per tutti
Descrizione: testo tesi
Dimensione
5.47 MB
Formato
Adobe PDF
|
5.47 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/260549