Assessing climate change impacts in the Tagus-Segura water transfer (Spain) through artificial intelligence

Being the most extensive water infrastructure in Spain, the future of the Tagus-Segura interbasin transfer is contested by climate change and increasing controversy regarding its use. It is a crucial infrastructure for regional socioeconomic development since it significantly contributes to the sustainability of one of Europe's most important agricultural areas. Diverse stakeholders' interests have generated conflicts impacting its decision-making processes. Several regulations have established maximum transferable volumes in the last decades depending on water availability and demands in the involved basins. However, the volumes transferred do not only account for those limits but also include expert criteria and ad-hoc considerations, making it complex to predict them. Artificial intelligence emerges as an effective solution to address this challenge and to reconcile all the factors affecting its current operating rules.To this end, this contribution combines artificial intelligence (fuzzy logic) with climate change scenarios and hydrological and water resource management models to predict future water transfers from the upper Tagus (donor basin) to the Segura (receiver basin). Climate change scenarios refer to five CMIP6 (Coupled Model Intercomparison Project Phase 6) climate models and four scenarios: historical (1979-2014), SSP126, SSP370, and SSP585 (2015-2100). Using their meteorological projections, the eco-hydrological model TETIS is used to obtain future time series of streamflows in response to them. The current operation of the water transfer is inferred through fuzzy logic systems that take into account the hydrological discharge of the upper Tagus estimated by TETIS, the storage level of the upper Tagus reservoirs (Entrepeñas and Buendia), the storage levels of the rest of the Tagus and the Segura basins, the regulatory limits, and the month of the year. The results show how foreseen streamflows in the upper Tagus would affect the transfer, providing valuable information for water planning in both basins, particularly in the Segura, for its adaptation to any decrease in water received from the Tagus basin.AcknowledgmentsThis study has received funding from the European Union's Horizon 2020 research and innovation programme under the GoNEXUS project (grant agreement No 101003722); and from the SOS-WATER project under the European Union's Horizon Europe research and innovation programme under grant agreement No. 101059264.