Evaluation of the Performance of the Global Weather Dataset AgERA5 for Sustainable Water Management in Agriculture: A Focus on Reference Evapotranspiration

Assessing water withdrawal for crop irrigation at large scale is a decisive step for arranging a sustainable development in the agriculture sector toward an optimal allocation and management of water resources. Since direct measurements of irrigation volumes are very often not available, indirect estimates are needed. These require the use of consistent time-series of gridded weather data for implementing crop-water balance models. So far, advances in meteorological numerical modelling have been encouraging the use of their outputs, including reanalysis data, as gridded weather data sources for similar purposes. As interest in meteorological weather reanalysis data increases, the need to evaluate their performance and compare the suitability of different databases for various sites becomes central. This study evaluates the performance of the weather dataset AgERA5, which was lately derived from ERA5 reanalysis with a finer resolution, for agricultural water management applications. The AgERA5 is compared with the ERA5-Land dataset, as well as with ground-based weather interpolated observations, which are the primary alternative weather database. The study focuses on the weather variables needed for computing the FAO Penman-Monteith reference evapotranspiration, ET0, such as wind velocity, surface shortwave radiation, temperature and relative humidity of air: key variables for irrigation volume estimates. The target area for the analyses is the Campania Region, a mediterranean-climate region in the South of Italy. The performances of the databases are evaluated from April to September, when irrigation occurs, respect to the years 2008–2024. Results show that AgERA5 is a reliable dataset for assessing agricultural indicators in regional studies by simply applying a local correction of the bias and it outperforms the weather dataset alternatives for assessing ET0.