Enhancing landslide disaster prediction by evaluating non landslide area sampling in machine learning models for Spiti Valley India

Abstract Landslides are well-known natural or quasi-natural hazards which can harm human lives, infrastructure, and the environment, particularly in mountainous regions like Spiti Valley, Himachal Pradesh, India. The Spiti Valley in Himachal Pradesh, India, is prone to landslides due to its fragile...

Full description

Saved in:
Bibliographic Details
Main Authors: Devraj Dhakal, Kanwarpreet Singh, Kennedy C. Onyelowe, Steven Alejandro Salazar Cazco, Abhishek Sharma, Nassir Alarifi, Fakhrul Islam, Randeep, Krishna Prakash Arunachalam, Youssef M. Youssef
Format: Article
Language:English
Published: Nature Portfolio 2025-04-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-95087-7
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849726538751148032
author Devraj Dhakal
Kanwarpreet Singh
Kennedy C. Onyelowe
Steven Alejandro Salazar Cazco
Abhishek Sharma
Nassir Alarifi
Fakhrul Islam
Randeep
Krishna Prakash Arunachalam
Youssef M. Youssef
author_facet Devraj Dhakal
Kanwarpreet Singh
Kennedy C. Onyelowe
Steven Alejandro Salazar Cazco
Abhishek Sharma
Nassir Alarifi
Fakhrul Islam
Randeep
Krishna Prakash Arunachalam
Youssef M. Youssef
author_sort Devraj Dhakal
collection DOAJ
description Abstract Landslides are well-known natural or quasi-natural hazards which can harm human lives, infrastructure, and the environment, particularly in mountainous regions like Spiti Valley, Himachal Pradesh, India. The Spiti Valley in Himachal Pradesh, India, is prone to landslides due to its fragile geology, steep slopes, and unpredictable climate. These dangers threaten life, infrastructure, and the environment, requiring comprehensive landslide susceptibility mapping (LSM) for mitigation and planning. Advanced machine learning and new sampling procedures may improve LSM accuracy and reliability in high-risk zones. This investigation aims to enhance LSM through a systematic evaluation of innovative sampling methodologies for non-landslide areas. This study focuses on assessing the effectiveness of two novel sampling methods: Buffer Zone Safe Points (BZSP) and Slope Buffer Safe Points (SBSP). Detailed susceptibility zonation maps were created employing advanced statistical techniques, specifically Extreme Gradient Boosting (XGBoost), Random Forest (RF), and K-Nearest Neighbors (KNN), allowing for an in-depth comparison of their predictive performance. The findings clearly indicate the advantages of the SBSP technique, showcasing notable improvements in performance across all metrics. In the analysis of Category-II, XGBoost showed a significant rise in the Area Under Curve (AUC) from 0.91 to 0.97, RF increased from 0.89 to 0.97, and KNN improved from 0.87 to 0.94, with corresponding enhancements in accuracy, sensitivity, kappa values, and F1-scores. These advancements highlight the capability of the SBSP method to enhance susceptibility predictions and reduce overestimation in areas of high vulnerability. The Landslide Density Index (LDI) supports these findings, as Category-II sampling provides more dependable estimates across all susceptibility classes, minimizing variability and improving interpretive confidence. This study emphasizes the essential importance of sophisticated sampling techniques in enhancing the dependability of LSM and establishes a fundamental framework for upcoming research focused on reducing landslide hazards in intricate landscapes. The results highlight the importance of integrating various conditioning factors and flexible approaches to enhance regional hazard evaluations and strengthen disaster readiness.
format Article
id doaj-art-50a9c4e23c4442ee834946c17d5afc2b
institution DOAJ
issn 2045-2322
language English
publishDate 2025-04-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-50a9c4e23c4442ee834946c17d5afc2b2025-08-20T03:10:09ZengNature PortfolioScientific Reports2045-23222025-04-0115112510.1038/s41598-025-95087-7Enhancing landslide disaster prediction by evaluating non landslide area sampling in machine learning models for Spiti Valley IndiaDevraj Dhakal0Kanwarpreet Singh1Kennedy C. Onyelowe2Steven Alejandro Salazar Cazco3Abhishek Sharma4Nassir Alarifi5Fakhrul Islam6Randeep7Krishna Prakash Arunachalam8Youssef M. Youssef9Department of Civil Engineering, Chandigarh UniversityDepartment of Civil Engineering, Chandigarh UniversityDepartment of Civil Engineering, School of Engineering and Applied Sciences, Kampala International UniversityFacultad de Informática y Electrónica, Escuela Superior Politécnica de Chimborazo (ESPOCH)Department of Civil Engineering, Chandigarh UniversityDepartment of Geology & Geophysics, College of Science, King Saud UniversityUniversity of Chinese Academy of SciencesDepartment of Civil Engineering, Dayalbagh Educational Institute (Deemed to be University)Departamento de Ciencias de la Construcción, Facultad de Ciencias de la Construcción Ordenamiento Territorial, Universidad Tecnológica MetropolitanaGeological and Geophysical Engineering Department, Faculty of Petroleum and Mining Engineering, Suez UniversityAbstract Landslides are well-known natural or quasi-natural hazards which can harm human lives, infrastructure, and the environment, particularly in mountainous regions like Spiti Valley, Himachal Pradesh, India. The Spiti Valley in Himachal Pradesh, India, is prone to landslides due to its fragile geology, steep slopes, and unpredictable climate. These dangers threaten life, infrastructure, and the environment, requiring comprehensive landslide susceptibility mapping (LSM) for mitigation and planning. Advanced machine learning and new sampling procedures may improve LSM accuracy and reliability in high-risk zones. This investigation aims to enhance LSM through a systematic evaluation of innovative sampling methodologies for non-landslide areas. This study focuses on assessing the effectiveness of two novel sampling methods: Buffer Zone Safe Points (BZSP) and Slope Buffer Safe Points (SBSP). Detailed susceptibility zonation maps were created employing advanced statistical techniques, specifically Extreme Gradient Boosting (XGBoost), Random Forest (RF), and K-Nearest Neighbors (KNN), allowing for an in-depth comparison of their predictive performance. The findings clearly indicate the advantages of the SBSP technique, showcasing notable improvements in performance across all metrics. In the analysis of Category-II, XGBoost showed a significant rise in the Area Under Curve (AUC) from 0.91 to 0.97, RF increased from 0.89 to 0.97, and KNN improved from 0.87 to 0.94, with corresponding enhancements in accuracy, sensitivity, kappa values, and F1-scores. These advancements highlight the capability of the SBSP method to enhance susceptibility predictions and reduce overestimation in areas of high vulnerability. The Landslide Density Index (LDI) supports these findings, as Category-II sampling provides more dependable estimates across all susceptibility classes, minimizing variability and improving interpretive confidence. This study emphasizes the essential importance of sophisticated sampling techniques in enhancing the dependability of LSM and establishes a fundamental framework for upcoming research focused on reducing landslide hazards in intricate landscapes. The results highlight the importance of integrating various conditioning factors and flexible approaches to enhance regional hazard evaluations and strengthen disaster readiness.https://doi.org/10.1038/s41598-025-95087-7Landslide susceptibility mapping (LSM)Landslide predictionNon-Landslide samplingsMachine learning modelsSpiti ValleyGeoinformatics
spellingShingle Devraj Dhakal
Kanwarpreet Singh
Kennedy C. Onyelowe
Steven Alejandro Salazar Cazco
Abhishek Sharma
Nassir Alarifi
Fakhrul Islam
Randeep
Krishna Prakash Arunachalam
Youssef M. Youssef
Enhancing landslide disaster prediction by evaluating non landslide area sampling in machine learning models for Spiti Valley India
Scientific Reports
Landslide susceptibility mapping (LSM)
Landslide prediction
Non-Landslide samplings
Machine learning models
Spiti Valley
Geoinformatics
title Enhancing landslide disaster prediction by evaluating non landslide area sampling in machine learning models for Spiti Valley India
title_full Enhancing landslide disaster prediction by evaluating non landslide area sampling in machine learning models for Spiti Valley India
title_fullStr Enhancing landslide disaster prediction by evaluating non landslide area sampling in machine learning models for Spiti Valley India
title_full_unstemmed Enhancing landslide disaster prediction by evaluating non landslide area sampling in machine learning models for Spiti Valley India
title_short Enhancing landslide disaster prediction by evaluating non landslide area sampling in machine learning models for Spiti Valley India
title_sort enhancing landslide disaster prediction by evaluating non landslide area sampling in machine learning models for spiti valley india
topic Landslide susceptibility mapping (LSM)
Landslide prediction
Non-Landslide samplings
Machine learning models
Spiti Valley
Geoinformatics
url https://doi.org/10.1038/s41598-025-95087-7
work_keys_str_mv AT devrajdhakal enhancinglandslidedisasterpredictionbyevaluatingnonlandslideareasamplinginmachinelearningmodelsforspitivalleyindia
AT kanwarpreetsingh enhancinglandslidedisasterpredictionbyevaluatingnonlandslideareasamplinginmachinelearningmodelsforspitivalleyindia
AT kennedyconyelowe enhancinglandslidedisasterpredictionbyevaluatingnonlandslideareasamplinginmachinelearningmodelsforspitivalleyindia
AT stevenalejandrosalazarcazco enhancinglandslidedisasterpredictionbyevaluatingnonlandslideareasamplinginmachinelearningmodelsforspitivalleyindia
AT abhisheksharma enhancinglandslidedisasterpredictionbyevaluatingnonlandslideareasamplinginmachinelearningmodelsforspitivalleyindia
AT nassiralarifi enhancinglandslidedisasterpredictionbyevaluatingnonlandslideareasamplinginmachinelearningmodelsforspitivalleyindia
AT fakhrulislam enhancinglandslidedisasterpredictionbyevaluatingnonlandslideareasamplinginmachinelearningmodelsforspitivalleyindia
AT randeep enhancinglandslidedisasterpredictionbyevaluatingnonlandslideareasamplinginmachinelearningmodelsforspitivalleyindia
AT krishnaprakasharunachalam enhancinglandslidedisasterpredictionbyevaluatingnonlandslideareasamplinginmachinelearningmodelsforspitivalleyindia
AT youssefmyoussef enhancinglandslidedisasterpredictionbyevaluatingnonlandslideareasamplinginmachinelearningmodelsforspitivalleyindia