The InTheMED Decision Support System source code is freely as an official toolbox in MathWorks.
The DSS tool is demonstrated using surrogate groundwater models developed for five sites in the Mediterranean region under the scope of the European project InTheMED (https://inthemedprima.com/), promoted by the PRIMA program. The DSS tool (Varouchakis et al., 2023) works within a fuzzy logic framework and is available online (http://147.27.70.139:9988/webapps/home/).
The DSS operation employs data-driven techniques tailored based on the case study and data availability.
The Random Forest method (Godoy et al., 2022) is used for the Requena-Utiel area (Spain), Artificial Neural Networks (Todaro et al., 2023) for the Konya basin (Türkiye), while spatio-temporal geostatistical modelling is applied to the Tympaki site (Greece)(Lino Pereira et al., 2023). For the Grombalia (Tunisia) and Castro Verde (Portugal) sites, the surrogate models are developed using a statistical approach based on regression models (Secci et al., 2021).
The DSS tool is used to classify the vulnerability of the demo sites using a fuzzy clustering method. The clustering algorithm inputs the difference or absolute difference in groundwater levels between two scenarios the user selects. These scenarios are defined by changing parameters related to climate scenarios, groundwater pumping, and simulation periods. The output clusters groundwater vulnerability areas, reflecting variations in climate conditions and groundwater utilization across different time horizons. The DSS tool can classify the sites into six categories: very low, low, low to medium, medium to high, high, and very high vulnerability. Based on this information, groundwater managers can decide on remediation measures related to groundwater use and apply them to areas in the same cluster. The tool is freely accessible and readily transferred to other regions for policy and educational purposes.