Landslides pose significant risks to human life and infrastructure, particularly in mountainous regions like Inje, South Korea. This study aims to develop detailed landslide susceptibility maps (LSMs) using statistical (i.e., Frequency Ratio (FR), Logistic Regression (LR)) models and a hybrid integrated approach. These models incorporated various factors influencing landslides, including aspect, elevation, rainfall, slope, soil depth, slope length, and landform, derived from comprehensive geospatial datasets. The FR method assesses the likelihood of landslides based on historical occurrences relative to specific factor classes, while the LR method predicts landslide susceptibility through the statistical modeling of multiple predictor variables. The results from the FR, LR, and hybrid methods showed that the cumulative area covered by high and very high landslide susceptibility zones was 13.8%, 13.0%, and 14.28%, respectively. The results were validated using Receiver Operating Characteristic (ROC) curves and the Area Under the Curve (AUC), revealing AUC values of 0.83 for FR, 0.86 for LR, and 0.864 for the hybrid method, indicating high predictive accuracy. Subsequently, we used K-mean clustering algorithms on the hybrid LSI to identify the higher LSI cluster of the region. Furthermore, sensitivity analysis based on landslide density confirmed that all methods accurately identified high-risk areas. The resulting LSMs provide critical insights for land-use planning, infrastructure development, and disaster risk management, enhancing predictive accuracy and aiding in the prevention of future landslide damage.

Landslides are significant geological hazards in mountainous regions, arising from both natural forces and human actions, presenting serious environmental challenges through their extensive damage to properties and infrastructure, often leading to casualties and alterations to the landscape. This study employed GIS-based techniques to evaluate and map the landslide susceptibility in the Bekhair structure located within the Zagros mountains of Kurdistan, northern Iraq. An inventory map containing 282 landslide occurrences was compiled through intensive field investigations, as well as the interpretation of remote sensing data and Google Earth images. Ten potential influencing factors, including elevation, rainfall, lithology, slope, curvature, aspect, LULC, NDVI, distance to roads and rivers, were selected to construct susceptibility maps by integrating the frequency ratio (FR) and analytical hierarchy process (AHP) approaches, with the goal of understanding how these factors relate to landslides occurrence. The Bekhair core area was divided into 5 hazard zones on the landslide susceptibility maps. The regions classified as very low and low hazard zones are mainly occur in flat or gently sloping plains that characterized by resistant rocks, dense vegetation, minimal rainfall, shallow valleys, and are distant from riverbanks and roads. The areas designated as high and very high hazard zones are found in steep slopes and rough terrain with bare soil, intense weathering, high rainfall, sparse vegetation, highly fractured rocks, deep valleys, and close proximity to construction projects. The moderate hazard zones are mainly located between the other 4 zones across the Bekhair anticline. Results of the susceptibility analysis indicate that the occurrence of landslides in Kurdistan mountains are primarily controlled by factors related to the tectonic structure, surface characteristics and environmental conditions, such as rock lithology (competency), terrain slope, rainfall intensity, and human impacts. The delineation of landslide hazard zones offers important guides for government decision-makers engaged in regional planning, infrastructure development, and the formulation of strategies to mitigate landslides and protect lives and properties in Kurdistan. The accuracy of susceptibility maps was evaluated using the R-index and the AUC-ROC curve. The landslide susceptibility index (LSI) values allocated to different susceptibility classes derived from both FR and AHP models are consistent with the values obtained from the R-index. Moreover, the FR model demonstrated superior performance compared to the AHP model, with a success rate of 85.3% and a predictive rate of 81.2%, in contrast to the AHP model's success rate of 75.2% and predictive rate of 72.4%.

The growing popularity of GIS technology in Ethiopia has encouraged multiple scholars to investigate landslide hazards using quantitative approaches, despite its limitations. The present review examined the approach used in the evaluation of landslide hazards by five prior studies that shared catchments. The review results reveal that the controlling factors assumed by the five researchers were inconsistent and resulted in highly divergent frequency ratio (FR) values, even for the same factors. This implies that the contribution of a single instability factor can be inferred sufficiently for landslide hazard assessment and mapping; otherwise, the results are highly subjective and disputable. Since the soil type in the region was alluvial-colluvial in the five studies, and a majority of the failures occurred shortly after rainfall, rainfall data and basic soil properties (classification and shear strength) should not be overlooked. In addition to the nonstandard use of morphometric parameters, the inherent limits of GIS methodologies, the omission of hydrogeotechnical properties, and the observed subjective outcomes make the GIS-based approach imprecise, error-prone, and doubtful. The total effect will result in ineffective early warning systems and unworthy mitigation measures, resulting in significant life costs and damage. As a result, it is recommended that GIS technology should be coupled with software (TRIGRS, Scoops3D, SINMAP, OpenLISEM, GLM, and SLIP) that considers hydrogeotechnical properties to provide more reliable conclusions. In addition to using instability factors consistently, regional statistical correlations of all morphometric parameters can be developed, allowing for less complex and realistic empirical models to be used.