Assessing the Impact of Climate Change on Water Scarcity in the Tormes ‎Catchment, Spain: A Human-Water System Modeling Approach

Climate change presents a pressing challenge to global water availability resulting in ‎increased variability in precipitation and increased temperatures, imposing more stress ‎on existing water resources and the economic activities that depend on them. The ‎Tormes catchment, located in a semi-arid region, is facing increasingly severe water ‎shortages, which may be further aggravated under climate change. This catchment is ‎extensively employed for agricultural purposes, and a potential reduction in the ‎availability of water for irrigation emerges as a significant concern.‎
This study evaluates the impact of climate change on water availability, and the ‎responses implemented by irrigators to adapt to growing scarcity, in the Tormes ‎catchment. To this end, we develop a human-water system model that couples the Soil ‎and Water Assessment Tool (SWAT) model and a Positive Multi-Attribute Utility ‎Programming (PMAUP) model using a dynamic and modular approach. The coupled ‎model runs the water (SWAT) and human (PMAUP) system models iteratively and ‎over time using inputs from six different bias-corrected Global Climate ‎Models(GCMs) under CMIP6 scenarios, as follows: i) CMIP6 climate change scenario ‎simulations are fed to the SWAT model to estimate relevant hydrological data ‎including water availability in March (beginning of the irrigation campaign); ii) ‎information on water availability is fed to the PMAUP model to simulate the adaptive ‎responses of irrigators in terms of water and land allocation; iii) land and water use ‎choices by irrigators are fed into the SWAT model, which reproduces the ‎consequences of human decisions on the water system; iv) when the hydrological year ‎is over, a new iteration starts where CMIP6 climate change scenario simulations for ‎the following year are fed into the SWAT model and the process is repeated again. The ‎non-linearity and modular approach in both the hydrological and economic models ‎imply complex and interconnected interactions within these systems, with behaviors ‎that may or may not follow linear patterns.‎
Six bias-corrected GCMs under CMIP6 scenarios were employed for the climate ‎change scenario simulations. The dataset covered precipitation, maximum and ‎minimum temperatures for the historical period (1981–2010) and projections for ‎SSP245 and SSP585. Future data was analyzed for three periods: 2020–2039, 2040–‎‎2059, and 2060–2100. A multi-model ensemble approach was applied, averaging ‎outputs from the six models. Precipitation and temperature data were integrated into ‎the SWAT model.‎
The hydrological analysis revealed a downward trend in projected precipitation, with ‎reductions of 0.7% (2020s), 0.3% (2040s), and up to 5.3% (2060s) under SSP245. ‎SSP585 showed declines of 6.4% (2020s), 6.6% (2040s), and 16.1% (2060s). ‎Maximum and minimum temperatures exhibited an upward trend under both ‎scenarios. Simulated mean annual runoff under SSP245 experienced a drastic ‎reduction of 48.1% in the 2020s, followed by 43.8% (2040s) and 53% (2060s). ‎Similarly, under SSP585, mean annual runoff decreased by 47.2% over the entire ‎projection period. While the hydrological analysis reveals concerning trends in ‎precipitation, temperature, and mean annual runoff under different scenarios, the ‎economic results, reflecting the effects of these hydrological changes on human ‎activities, are still being investigated and are not yet finalized.‎