Collapse dolines susceptibility mapping using frequency ratio method and GIS in Sahel-Doukkala, Morocco

Abstract

This study aims firstly, to build the collapse dolines susceptibility maps in the Sahel-Doukkala area of central Morocco, using the modified frequency ratio (FR) method. Secondly, to determine the relationships between factors responsible for collapse doline occurrence and their spatial distribution. To accomplish these objectives, a total of 58 collapses were identified in the field and integrated into the Geographic Information System (GIS) database. A collapse inventory map was prepared from the locations of the identified collapses. This inventory was subdivided randomly in two equal groups, one for FR model development and one for model validation. In this study, six parameters that control the development of karstic collapse dolines were selected. These factors are slope, appearance, elevation, geological nature of the terrain, distance from the underground drainage axis and, distance from faults. The combination of the chosen factors allowed the creation of Collapse Dolines susceptibility map of the study area. The validity of the obtained model was verified using the Area Under the Curve (AUC) highlighting a high prediction degree. The FR value obtained for Collapse dolines susceptibility were comprised between 2.5 and 25.8, these levels have been divided into five susceptibility classes: very low, low, moderate, high and very high. Overall, the obtained results show that the study area was dominated by very low and low or moderate susceptibility classes. However, the Sahel and the NW part of the study area was characterised by a concentration of high and very high classes. Moreover, the AUC value of 90% was obtained, which indicates the reliability of the susceptibility map for the karst collapse. This study’s findings can help decision-makers in land use planning and preventing areas of potential collapse.