The Mediterranean area belongs to the regions most exposed to hydroclimatic changes, with a likely increase in frequency and duration of droughts in the last decades. However, many climate records like, e.g., North Italian precipitation and river discharge records, indicate that significant decadal variability is often superposed or even dominates long-term hydrological trends. The capability to accurately predict such decadal changes is, therefore, of utmost environmental and social importance. Here, we present a twofold decadal forecast of Po River (Northern Italy) discharge obtained with a statistical approach consisting of the separate application and cross-validation of autoregressive models and neural networks. Both methods are applied to each significant variability component extracted from the raw discharge time series using Singular Spectrum Analysis, and the final forecast is obtained by merging the predictions of the individual components. The obtained 25-year forecasts robustly indicate a prominent dry period in the late 2020s/early 2030s. Our prediction provides information of great value for hydrological management, and a target for current and future near-term numerical hydrological predictions.

Robust decadal hydroclimate predictions for northern Italy based on a twofold statistical approach

Rubinetti S.;Taricco C.;Alessio S.;Bizzarri I.;
2020-01-01

Abstract

The Mediterranean area belongs to the regions most exposed to hydroclimatic changes, with a likely increase in frequency and duration of droughts in the last decades. However, many climate records like, e.g., North Italian precipitation and river discharge records, indicate that significant decadal variability is often superposed or even dominates long-term hydrological trends. The capability to accurately predict such decadal changes is, therefore, of utmost environmental and social importance. Here, we present a twofold decadal forecast of Po River (Northern Italy) discharge obtained with a statistical approach consisting of the separate application and cross-validation of autoregressive models and neural networks. Both methods are applied to each significant variability component extracted from the raw discharge time series using Singular Spectrum Analysis, and the final forecast is obtained by merging the predictions of the individual components. The obtained 25-year forecasts robustly indicate a prominent dry period in the late 2020s/early 2030s. Our prediction provides information of great value for hydrological management, and a target for current and future near-term numerical hydrological predictions.
2020
11
6
671
688
Decadal climate predictions; Drought; Neural networks; Po river discharges; Runoff; Statistical methods
Rubinetti S.; Taricco C.; Alessio S.; Rubino A.; Bizzarri I.; Zanchettin D.
File in questo prodotto:
File Dimensione Formato  
atmosphere-2020.pdf

Accesso aperto

Descrizione: Articolo principale
Tipo di file: PDF EDITORIALE
Dimensione 1.51 MB
Formato Adobe PDF
1.51 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/2318/1815201
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 6
  • ???jsp.display-item.citation.isi??? 6
social impact