Geospatial Assessment and Mapping Landslide Susceptibility for the Garo Hills Division, Meghalaya, India

Geological Journal, Journal Year: 2025, Volume and Issue: 60(5), P. 1184 - 1201

Published: Feb. 19, 2025

ABSTRACT Creating accurate and effective Landslide Susceptibility (LS) maps can aid disaster prevention mitigation efforts provide sufficient public safety. The primary aim of this study is to develop an LS map for the Garo Hills region in Meghalaya, India, using weight evidence (WoE), frequency ratio (FR), Shannon entropy (SE) methods. A comprehensive landslide inventory catalogued 98 events from 2000 2023 analysis, nine key geographical environmental parameters were prepared. Conducted multicollinearity correlation analysis identify mitigate collinearity issues between factors. model's performance was analysed through area under curve (AUC) value receiver operating characteristic (ROC) curves three recent landslides. results showed that FR method achieved highest accuracy, with successive rate (SRC) AUC predictive (PRC) values 0.860 0.940, respectively, classified susceptibility at sites as high, moderate, low. WoE effectively identified landslides site high very zones, achieving SRC PRC 0.844 0.915, respectively. SE robust predicting landslide‐prone areas, comparable other methods (0.913), though its (0.771) lower. Developed revealed zones account approximately 10% 3% area, predominantly near roads, steep slopes, higher elevations. information valuable civilians government authorities involved hazard monitoring management.

Language: Английский

---

Comprehensive review of remote sensing integration with deep learning in landslide forecasting and future directions

Natural Hazards, Journal Year: 2025, Volume and Issue: unknown

Published: March 10, 2025

Language: Английский

---

High-resolution earthquake-induced landslide hazard assessment in Southwest China through frequency ratio analysis and LightGBM

International Journal of Applied Earth Observation and Geoinformation, Journal Year: 2024, Volume and Issue: 131, P. 103947 - 103947

Published: June 5, 2024

Earthquake-induced landslides cause extensive harm, necessitating accurate predictions for effective risk management. To tackle the dual challenges posed by inadequate model accuracy and absence of frequency-derived landslide intensity as critical information, this paper presents a robust modelling method that generates high-resolution hazard map assessment. Firstly, Frequency Ratio (FR) is employed to quantitatively explore correlation between geo-environmental factors occurrences, enabling exclusion less significant factors. Subsequently, these FR values are integrated into Light Gradient Boosting Machine (LightGBM) model, resulting in composite FR-LightGBM. test proposed data from areas affected Luding earthquake based on seismic following 2022 Ms 6.8 Sichuan, China, examined. Model validation, using area under curve (AUC) receiver operating characteristic curve, highlighted superior predictive FR-LightGBM marking 3.5 % improvement AUC value compared both Convolutional Neural Network (CNN) Logistic Regression (LR) models. The results have identified high-risk along Dadu riverside Xianshuihe fault zone correspond well with actual distribution. Overall, practical applicability significantly enhanced its capacity prioritize lithology, faults, aspect crucial Additionally, an extended inventory established high spatial resolution (2 m) satellite imagery Chinese Gaofen series, approach delivered (12.5 across area, contributing relevant studies

Language: Английский

---

Algal blooms forecasting with hybrid deep learning models from satellite data in the Zhoushan fishery

Ecological Informatics, Journal Year: 2024, Volume and Issue: 82, P. 102664 - 102664

Published: June 6, 2024

Algal blooms are increasingly frequent in coastal areas, posing a significant threat to ecosystems. The Zhoushan fishery, one of the most affected regions along Chinese coast, faces severe challenges from algal blooms. In this study, Convolutional Neural Network (CNN), Long Short-term Memory (LSTM) and hybrid CNN-LSTM deep learning models were constructed forecast chlorophyll (Chl) concentrations satellite data. model outperformed individual models, achieving highest determination coefficient lowest root mean square error for Chl concentration forecasts. It also excelled predicting blooms, with probability detection Heidke skill score, effectively capturing trends bloom development. areas high concentration, parameter significantly influences forecasts, while meridional wind current main influence factors medium low concentration. powerful provided by offers valuable support efficient management sustainable development fishery.

Language: Английский

---

Landslide susceptibility assessment in Eastern Himalayas, India: a comprehensive exploration of four novel hybrid ensemble data driven techniques integrating explainable artificial intelligence approach

Environmental Earth Sciences, Journal Year: 2024, Volume and Issue: 83(22)

Published: Nov. 1, 2024

Language: Английский

---

Prediction of urban surface water quality scenarios using hybrid stacking ensembles machine learning model in Howrah Municipal Corporation, West Bengal

Journal of Environmental Management, Journal Year: 2024, Volume and Issue: 370, P. 122721 - 122721

Published: Oct. 13, 2024

Language: Английский

---

Unveiling soil coherence patterns along Etihad Rail using Sentinel-1 and Sentinel-2 data and machine learning in arid region

Remote Sensing Applications Society and Environment, Journal Year: 2024, Volume and Issue: 36, P. 101374 - 101374

Published: Oct. 7, 2024

Language: Английский

---

Assessment of Landslide Susceptibility using Geospatial Techniques: A Comparative Evaluation of Machine Learning and Statistical Models

Geological Journal, Journal Year: 2024, Volume and Issue: 60(5), P. 1129 - 1149

Published: Nov. 20, 2024

This study delineates landslide susceptibility zones in the Kalimpong district by integrating multi‐sensor datasets and assessing effectiveness of statistical machine learning models for precision mapping. The analysis utilises a comprehensive geospatial dataset, including remote sensing imagery, topographical, geological, climatic factors. Four were employed to generate maps (LSMs) using 16 influencing factors: two bivariate models, frequency ratio (FR) evidence belief function (EBF) random forest (RF) support vector (SVM). Out 1244 recorded events, 871 events (70%) used training 373 (30%) validation. distribution classes predicted RF SVM produced similar distributions, predicting 13.30% 14.30% area as highly susceptible, 2.42% 2.82% very respectively. In contrast, FR model estimated 20.98% susceptible 4.30% whereas EBF 17.42% 5.89% these categories, Model validation receiver operating characteristic (ROC) curves revealed that (RF SVM) had superior prediction accuracy with AUC values 95.90% 86.60%, respectively, compared (FR EBF), which achieved 74.30% 76.80%. findings indicate Kalimpong‐I is most vulnerable, 6.76% its categorised high 24.80% susceptibility. Conversely, Gorubathan block exhibited least 0.95% 6.48% classified susceptibility, research provides essential insights decision‐makers policy planners landslide‐prone regions can be instrumental developing early warning systems, are vital enhancing community safety through timely evacuations preparedness measures.

Language: Английский

---

Landslide susceptibility prediction and mapping in Taihang mountainous area based on optimized machine learning model with genetic algorithm

Earth Science Informatics, Journal Year: 2024, Volume and Issue: unknown

Published: Aug. 31, 2024

Language: Английский

---

Empowering NGOs with Remote Sensing and CNN-LSTM Models for Social and Environmental Transformation

Remote Sensing in Earth Systems Sciences, Journal Year: 2024, Volume and Issue: 7(4), P. 554 - 568

Published: Oct. 7, 2024

Language: Английский

---