Doctoral Dissertations

Permanent URI for this collection

https://repository.univ-msila.dz/handle/123456789/26859

Browse

Browsing Doctoral Dissertations by Subject "Flooding"

Now showing 1 - 1 of 1
  • Thumbnail Image
    ItemOpen Access
    Etude de sensibilité de l’érosion hydrique et des zones à risques d’inondations aux pluies extrêmes dans les bassins versants (K’sob) Hodna / (Boussellam) Soummam
    (Université Mohamed Boudiaf - M’sila, 2025-06-29) MORAD BENAICHE; ENCA/Mokhtari Elhadj
    Water erosion and floods are among the most threatening natural hazards, particularly in ecologically fragile areas, due to the acceleration of climate change and the lack of effective preventive measures. This study aims to map the sensitivity of areas at risk by combining multi-criteria analysis techniques (AHP and FAHP) with erosion models (RUSLE and EPM) in a GIS environment, integrating various environmental and climatic factors to improve the accuracy of risk assessment. In the first phase, sensitivity maps for water erosion were produced based on different return periods (2, 5, 10, 100, and 1000 years), using AHP, FAHP, RUSLE, and EPM models. Several factors were analyzed, including slope, precipitation, soil characteristics, land use, and vegetation cover. The results revealed significant variations in sensitivity, with the most affected areas concentrated in the north and some parts of the west and south of the study area. ROC curves showed that the FAHP model performed best (AUC = 0.806), followed by AHP (0.783), while RUSLE (0.618) and EPM (0.638) showed lower predictive accuracy. The second part of the study focused on flood sensitivity mapping using the same AHP and FAHP methods. Several variables were considered, such as curvature, slope, elevation, rainfall, lithology, TWI, NDVI, MNDWI, land use, and proximity to watercourses. The results showed that high-risk flood zones are mainly concentrated in the Boussellam and K’sob sub-basins, particularly in low-altitude areas. According to the AHP model, high-risk zones covered between 2.19% and 25.55% of the study area, while under the FAHP model, they ranged from 1.53% to 24.51%. ROC analysis confirmed the strong performance of both models (AUC = 0.839 for AHP and 0.835 for FAHP). These findings highlight the importance of using spatial analysis tools and multi-criteria decision-making methods to improve natural hazard risk assessment and support sustainable adaptation strategies, enhancing infrastructure planning and community resilience to extreme weather events