Loading...
Publication
Geospatial modelling for sinkhole hazard in the coastal area of Safi, Morocco
| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 4.18 MB | Adobe PDF |
Advisor(s)
Abstract(s)
Collapse sinkholes represent a geological hazard that is particularly associated with karst regions, characterized by the dissolution of soluble rocks leading to the formation of voids. This phenomenon is significantly more prevalent in areas with evaporate terrains, wherein the rapid dissolution of soluble rocks facilitates the swift alteration of gap locations and sizes. The ensuing discussion explores the distinctive nature of collapse sinkholes, emphasizing their occurrence in karst and evaporate terrains, shedding light on the factors contributing to their formation and frequency. The coastal region of Safi is characterized by abundant karst features overlying evaporitic rocks, along with historical and recent sinkhole collapse events. Our current investigation aims to develop a susceptibility map for sinkhole collapse in the Safi region. Initially, a comprehensive inventory map of collapse occurrences was created through meticulous fieldwork and the integration of remote sensing data. Nine predisposing factors were identified and incorporated into a Geographic Information System (GIS), including elevation, lithology, curvature, slope, aspect, drainage density, lineament density, topographic wetness index, and land use. To assess the relationship between these factors and sinkhole collapse, statistical analyses were conducted using spatial analysis based on the Frequency Ratio (FR) and Analytical Hierarchy Process (AHP). The FR analysis revealed that high and very high susceptibility zones constitute 14.19% and 6.89% of the region, respectively, while moderate, low, and very low susceptibility zones occupy 23.07%, 26.12%, and 29.74%, respectively. Utilizing the AHP method, it was determined that high and very high susceptibility zones cover 20.32% and 12.01% of the region, respectively, with moderate, low, and very low susceptibility zones encompassing 19.38%, 27.55%, and 20.74%, respectively. To compare and evaluate the outcomes of the AHP and FR methods, the Area Under the Curve (AUC) approach was employed. The success rates, as indicated by AUC, demonstrated that FR (90.5%) exhibited higher accuracy than AHP (73.5%). The validation of the landslide susceptibility map confirmed the acceptability of the methodologies employed. These results can be effectively utilized for hazard mitigation and land use planning in the coastal area of Safi.