Assessing landslide susceptibility in northern Morocco: A geostatistical mapping approach in Al Hoceima-Ajdir

Abstract

Landslides are a significant global risk due to their impact on nature, society, and the economy. The Al Hoceima-Ajdir region is highly susceptible to geohazards such as earthquakes and ground instabilities caused by local tectonic movements and specific climatic and geomorphological conditions. This study aims to assess landslide susceptibility in Al Hoceima-Ajdir using a digital mapping approach that integrates the geostatistical Frequency Ratio tool, geomatic tools, and field observations. The frequency ratio method, a widely used technique in geospatial analysis, evaluates the relative frequency of landslides based on the spatial distribution of conditioning factors such as slope, lithology and land use – land cover. This method calculates the ratio of the frequency of landslides within specific classes of these factors to their respective frequencies in the entire study area. This research introduces a novel susceptibility map highlighting major threats to several natural and touristic sites crucial to the region's socio-economic activities. The susceptibility map was developed by analyzing 40 historical landslides along various predisposing factors, including topographic, geo-hydrological, and anthropogenic parameters. The results identify areas prone to landslides and classify their susceptibility into five levels ranging from very high to very low, with each level indicating a different degree of risk and requiring a specific set of preventive measures. The reliability of this map was evaluated using the Receiver Operating Characteristics (ROC) method, demonstrating a high accuracy of 89%, further validated by field visits. This high-precision susceptibility map is significant for pre- and post-risk management, enhancing prevention and mitigation strategies. It also protects society, the economy, and the environment by providing a reliable regional planning and hazard management tool. The findings of this study offer a significant advancement in the methodology for landslide susceptibility assessment and present a model that can be adapted to other regions facing similar geohazards. The research underscores the importance of integrating geostatistical tools with field data to produce accurate and actionable susceptibility maps, benefitting the global scientific community by improving hazard prediction and management strategies.