Key points are not available for this paper at this time.
Landslide susceptibility mapping using GIS-based logistic regression: The city Al Hoceima and its suburbs (Morocco) case studyPublic authorities in the Al Hoceima region have become concerned about landslides given the increasing demand for urban development in the region.Therefore, the identification of landslide susceptibility has become necessary to meet the needs of current and future habitats and to minimize possible landslide disasters.This paper is a demonstration of the application of the logistic regression method to produce the landslide susceptibility map for Al Hoceima.A landslide database with nine predictive factors was constructed by processing cartographic documents, Google Earth images, the restitution plan, the geological map, and the landslide inventory.The obtained susceptibility map shows that 22.1% of the study area is extremely susceptible to landslides, 7.6% of the area is moderately susceptible to landslides and 70.3% of the area is weakly susceptible to landslides.The validation of the map obtained by the logistic regression model used in this analysis gave good results.The superposition of the landslides reserved for validation with the obtained map allows for classifying the majority of the observed landslide pixels into high and very high hazard classes.The results of the ROC curve obtained for the approach used proved that the multivariate approach by logistic regressions is performing better (AUC= 0.894) for the prediction of the landslide hazard in the city of Al Hoceima and its periphery.The map obtained is a major contribution to various urban development plans and is the future orientation of urbanization.
Taoufik Byou (Fri,) studied this question.