Towards a Classification of Tunisian Dams for Enhanced Water Scarcity Governance: Parametric or Non-Parametric Approaches?

Classifying dams is important to ensure proper management, safety, and maintenance based on their size, purpose, and risk level. This helps in planning for emergency responses, structural inspections, and efficient water resource utilization. This study used the analysis of variance (ANOVA) technique to categorize the main Tunisian dams according to their precipitation to potential evapotranspiration (P/PET) ratio. The data were obtained from the NASA POWER platform, with potential evapotranspiration estimated using the Oudin model. Despite the violation of the normality assumption, the robustness of the ANOVA test for classification purposes remained unaffected. A comparison between Duncan’s test (parametric) and the Kruskal–Wallis test (non-parametric) revealed similar class structures, although Duncan’s test provided greater precision. The analysis identified four primary dam classes, reflecting regional differences in water availability and evaporative demand, and included dams in north-west Tunisia, considered the ‘water tower’ of the country, and those in semi-arid and arid regions.