Author(s): Daniele Dalla Torre; Andrea Menapace; Maurizio Righetti

Linked Author(s): Andrea Menapace, Maurizio Righetti

Keywords: Big-data; Data-driven models; Hydrology; Hydraulics; Sustainability; Weather forecast

Abstract: The performance of data-driven models is strongly affected by the data quality used and by the methods selected. The primary objective of this contribution is to assess the efficacy of data-driven methodology as forecast tools in water related applications (e. g. streamflow or network leakages forecast) employing various meteorological data types, including ground stations and reanalysis data. The second aim is to examine the impact of bias correction applied to the meteorological datasets on models' performance, to elucidate the outcomes of different data-driven approaches to a specific problem utilizing the distinct inputs. Results unveil interesting outcomes in forecasts through the integration of bias correction techniques. The study underscores the nuanced contributions of ground stations and reanalysis datasets used as forecasting data in tackling water management challenges. By presenting the outcomes of different data-driven approaches, the research provides valuable insights into the strengths and weaknesses associated with each model, thereby guiding the selection of an optimal tailored approach to specific forecasting requirements.

DOI: https://doi.org/10.3850/iahr-hic2483430201-375

Year: 2024