Exploring uncertainty analysis in GIS-based Landslide susceptibility mapping models using machine learning in the Darjeeling Himalayas DOI
Sumon Dey, Swarup Das, Abhik Saha

et al.

Earth Science Informatics, Journal Year: 2024, Volume and Issue: 18(1)

Published: Dec. 14, 2024

Language: Английский

Swarm optimization based heterogeneous machine learning techniques for enhanced landslide susceptibility assessment with comprehensive uncertainty quantification DOI
Sumon Dey, Swarup Das

Earth Science Informatics, Journal Year: 2025, Volume and Issue: 18(1)

Published: Jan. 1, 2025

Language: Английский

Citations

1
Evaluation of geological hazard susceptibility based on the multi-kernel density information method DOI Creative Commons
Yang Li, Yutian Lei,

Bo Chen

et al.

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: March 6, 2025

The increasing occurrence of geological hazards along roadway infrastructures presents a significant concern. Evaluating hazard susceptibility roads is critical aspect disaster emergency response and rescue efforts. Accurate evaluation outcomes are essential as they play crucial role in mitigating potential financial losses. However, previous studies on treated all samples independent entities, overlooking their spatial interactions. This study introduces novel assessment model termed the multi-kernel density information (MKDI) method. MKDI method integrates value with kernel estimation, effectively capturing dependencies among samples. Furthermore, distinct bandwidths prescribed for various scales disasters to facilitate estimation hazards. integration enables development comprehensive map, complexities distribution. To validate effectiveness proposed method, area selected investigation was G219 National Highway within Zayu County. Various factors were considered mapping, including slope, aspect, profile plan curvature, river road linear densities, peak ground acceleration, seismic spectrum characteristics, lithology, elevation, rainfall, landform. results show that outperformed methods, achieving an AUC 0.99. derived map expected offer scientific basis urban planning, construction, risk management area.

Language: Английский

Citations

1
A landslide susceptibility assessment method using SBAS-InSAR to optimize Bayesian network DOI Creative Commons
Xinyu Gao, Bo Wang, Wen Dai

et al.

Frontiers in Environmental Science, Journal Year: 2025, Volume and Issue: 13

Published: Feb. 27, 2025

Landslide susceptibility assessment is crucial to mitigate the severe impacts of landslides. Although Bayesian network (BN) has been widely used in landslide assessment, no study compared accuracy different BN structure construction methods for this purpose. SBAS-InSAR technology plays a vital role research, but its advantages combined with further improve prediction still need be studied. This paper takes Hanyuan County as area. First, 20 traditional impact factors were extracted from data such topography and meteorology. A new method GDSP was designed fuse GeoDetector SHAP dominant factor screening. Then, 8 learning using AUC value ROC curve, among which Tabu&K2 showed highest accuracy. The deformation calculated by then incorporated into model. optimized (OPT-BN) outperformed unoptimized version (ORI-BN) accuracy, mapping more reasonable. reverse inference highlighted that areas lower elevation, plow land, impervious cover, higher rainfall are prone provides valuable insights hazard prevention control future research.

Language: Английский

Citations

0
Multi-scenario landslide probabilistic hazard analysis based on a single rainfall event: A case of the Zhuzhou-Guangzhou section of Beijing-Guangzhou railway in China DOI Creative Commons
Zhiwen Xue, Chong Xu,

Jiale Jin

et al.

Research Square (Research Square), Journal Year: 2025, Volume and Issue: unknown

Published: March 10, 2025

Abstract This study calculates the absolute probability of landslides under varying rainfall scenarios along Beijing-Guangzhou Railway from Zhuzhou to Guangzhou, aiming enhance railway transportation safety. Using a Bayesian sampling strategy, Logistic Regression (LR) model was developed for landslide hazard assessment based on geological conditions and data railway. The demonstrated strong predictive performance with an AUC value 0.86 both training testing sets, showing no overfitting. Results indicated that when is less than 150 mm, over 70% area has below 0.1%. However, exceeding hazards increase significantly, rapid rise in areas where ranges 0.1–1%. When reaches 500 about 60% region exhibits 1%. Under real (e.g., cumulative during 10 days before June 7, 2020), probabilities greater 1% are mainly concentrated Fogang County, northeast eastern Zhuzhou, aligning heavy distributions. relationship between occurrence highly non-linear, increasing exponentially as rises. These results provide effective tool offer valuable support disaster warning prevention measures.

Language: Английский

Citations

0
Exploring uncertainty analysis in GIS-based Landslide susceptibility mapping models using machine learning in the Darjeeling Himalayas DOI
Sumon Dey, Swarup Das, Abhik Saha

et al.

Earth Science Informatics, Journal Year: 2024, Volume and Issue: 18(1)

Published: Dec. 14, 2024

Language: Английский

Citations

1