Abstract Landslides in hilly regions can have devastating results in terms of life as well as financial losses. This research aims to create a map of the likelihood of landslides using the analytical hierarchy process and logistic regression techniques. The probable landslide locations were determined utilizing landslide causative factors in the study area. The susceptibility map was distributed into five classes: very low, low, medium, high, and very high. To evaluate the effectiveness of the model, the area under the receiver operating characteristic curve (AUC) was used. The outcomes showed that the AUC value for the logistic regression model (LR) was higher than the AUC value for the analytical hierarchy process model (AHP). Based on the criticality of the landslide susceptible zones, two critical slope sections are selected at the Bhabanagar and Tapri sites. The slope stability assessment has been conducted using a preliminary method of circular failure chart (CFC) and the results are compared with PLAXIS 2D software. To strengthen the slopes and increase their stability, mitigation measures are required to prevent slope failures that cause landslides. The findings of the current study provide a baseline for authorities to suggest sufficient landslide mitigation in the vicinity. Research Highlights <UnorderedList Mark="Bullet"> <ItemContent> Kinnaur district has been divided into five susceptibility classes using the integrated RSGIS LSZ. </ItemContent> <ItemContent> For data-driven multivariate modelling, logistic regression has been found to be better than AHP (AUC = 0.63 vs. 0.61), demonstrating superior predictive ability. </ItemContent> <ItemContent> Site checks reveal that the Tapri landslide zone is marginal (FOS 1.021 dry) but weakens with moisture, while the Bhabanagar site is critically unstable (FOS 0.828 dry; 0.706 moist). </ItemContent> <ItemContent> CFC trends were verified by PLAXIS 2D; both sites require stabilization; it is advised to combine regional mapping with focused site remediation and land-use controls. </ItemContent> <ItemContent> The study shows that combining targeted, site-specific geotechnical investigations with regional susceptibility mapping produces useful information for infrastructure management and land-use planning. </ItemContent> </UnorderedList>

Geospatial based prediction of landslide prone areas and stability assessment of Bhabanagar and Tapri landslides in district Kinnaur, Himachal Pradesh, India

期刊论文
发表日期：

2025年11月8日

查看全文

Areeba Qazi,
Kanwarpreet Singh,
Abhishek Sharma,
Aditya Kumar Tiwary,
Dinesh Kumar Vishwakarma,
Krishna Kumar Yadav,
Ghadah Shukri Albakri,
Maha Awjan Alreshidi

摘要

Abstract

Landslides in hilly regions can have devastating results in terms of life as well as financial losses. This research aims to create a map of the likelihood of landslides using the analytical hierarchy process and logistic regression techniques. The probable landslide locations were determined utilizing landslide causative factors in the study area. The susceptibility map was distributed into five classes: very low, low, medium, high, and very high. To evaluate the effectiveness of the model, the area under the receiver operating characteristic curve (AUC) was used. The outcomes showed that the AUC value for the logistic regression model (LR) was higher than the AUC value for the analytical hierarchy process model (AHP). Based on the criticality of the landslide susceptible zones, two critical slope sections are selected at the Bhabanagar and Tapri sites. The slope stability assessment has been conducted using a preliminary method of circular failure chart (CFC) and the results are compared with PLAXIS 2D software. To strengthen the slopes and increase their stability, mitigation measures are required to prevent slope failures that cause landslides. The findings of the current study provide a baseline for authorities to suggest sufficient landslide mitigation in the vicinity.

Research Highlights

Kinnaur district has been divided into five susceptibility classes using the integrated RSGIS LSZ.

For data-driven multivariate modelling, logistic regression has been found to be better than AHP (AUC = 0.63 vs. 0.61), demonstrating superior predictive ability.

Site checks reveal that the Tapri landslide zone is marginal (FOS 1.021 dry) but weakens with moisture, while the Bhabanagar site is critically unstable (FOS 0.828 dry; 0.706 moist).

CFC trends were verified by PLAXIS 2D; both sites require stabilization; it is advised to combine regional mapping with focused site remediation and land-use controls.

The study shows that combining targeted, site-specific geotechnical investigations with regional susceptibility mapping produces useful information for infrastructure management and land-use planning.