Revisiting spatiotemporal evolution process and mechanism of a giant reservoir landslide during weather extremes

Engineering Geology, Год журнала: 2024, Номер 332, С. 107480 - 107480

Опубликована: Март 21, 2024

Язык: Английский

---

Advanced integration of ensemble learning and MT-InSAR for enhanced slow-moving landslide susceptibility zoning

Engineering Geology, Год журнала: 2024, Номер 331, С. 107436 - 107436

Опубликована: Фев. 9, 2024

Язык: Английский

---

An interpretable attention-based deep learning method for landslide prediction based on multi-temporal InSAR time series: A case study of Xinpu landslide in the TGRA

Remote Sensing of Environment, Год журнала: 2025, Номер 318, С. 114580 - 114580

Опубликована: Янв. 2, 2025

Язык: Английский

---

Assessing the imperative of conditioning factor grading in machine learning-based landslide susceptibility modeling: A critical inquiry

CATENA, Год журнала: 2023, Номер 236, С. 107732 - 107732

Опубликована: Дек. 7, 2023

Язык: Английский

---

Enhanced Kinematic Inversion of 3‐D Displacements, Geometry, and Hydraulic Properties of a North‐South Slow‐Moving Landslide in Three Gorges Reservoir

Journal of Geophysical Research Solid Earth, Год журнала: 2023, Номер 128(6)

Опубликована: Июнь 1, 2023

Abstract Complete three‐dimensional (3‐D) movements of slow‐moving landslides are critical to enhancing the understanding landslide mechanism. Multi‐source synthetic aperture radar (SAR) observations provide an opportunity derive 3‐D movements. However, deriving complete faces potential challenges incoherent phases and ill‐posed inverse problem, which may result in incomplete inaccurate results, especially for slopes facing north/south. Here, we propose a topography‐constrained strain model, exploits spatial relationship deformations between neighboring points as well assumption surface parallel flow landslide, Both real datasets over north‐south Xinpu complex utilized, assess if implemented method can overcome condition retrieve movement field. With multi‐source SAR datasets, performance various NISAR time series assessed. Based on derived long‐term InSAR measurements, metrics, including elementary parameters geometry, spatial‐temporal patterns movement, thickness, hydraulic diffusivity, reveal that (a) thickest mass concentrates toe (b) effects precipitation more significant than those water level fluctuation complex, Three Gorges Reservoir areas.

Язык: Английский

---

Data-augmented landslide displacement prediction using generative adversarial network

Journal of Rock Mechanics and Geotechnical Engineering, Год журнала: 2024, Номер 16(10), С. 4017 - 4033

Опубликована: Март 8, 2024

Landslides are destructive natural disasters that cause catastrophic damage and loss of life worldwide. Accurately predicting landslide displacement enables effective early warning risk management. However, the limited availability on-site measurement data has been a substantial obstacle in developing data-driven models, such as state-of-the-art machine learning (ML) models. To address these challenges, this study proposes augmentation framework uses generative adversarial networks (GANs), recent advance artificial intelligence (AI), to improve accuracy prediction. The provides enhance datasets. A recurrent GAN model, RGAN-LS, is proposed, specifically designed generate realistic synthetic multivariate time series mimics characteristics real data. customized moment-matching incorporated addition during training RGAN-LS capture temporal dynamics correlations Then, generated by used long short-term memory (LSTM) particle swarm optimization-support vector (PSO-SVM) models for prediction tasks. Results on two landslides Three Gorges Reservoir (TGR) region show significant improvement LSTM model performance when trained augmented For instance, case Baishuihe landslide, average root mean square error (RMSE) increases 16.11%, absolute (MAE) 17.59%. More importantly, model's responsiveness mutational stages enhanced purposes. results have shown static PSO-SVM only sees marginal gains compared LSTM. Further analysis indicates an optimal synthetic-to-real ratio (50% illustration cases) maximizes improvements. This also demonstrates robustness effectiveness supplementing dynamic obtain better results. By using powerful AI approach, can high-fidelity critical improving advanced ML displacement, particularly there Additionally, approach potential expand use geohazard management other research areas.

Язык: Английский

---

Temporal stacking of sub-pixel offset tracking for monitoring slow-moving landslides in vegetated terrain

Landslides, Год журнала: 2024, Номер 21(6), С. 1255 - 1271

Опубликована: Фев. 28, 2024

Язык: Английский

---

Towards hydrometeorological thresholds of reservoir-induced landslide from subsurface strain observations

Science China Technological Sciences, Год журнала: 2024, Номер 67(6), С. 1907 - 1922

Опубликована: Май 29, 2024

Язык: Английский

---

Probing multi-physical process and deformation mechanism of a large-scale landslide using integrated dual-source monitoring

Geoscience Frontiers, Год журнала: 2023, Номер 15(2), С. 101773 - 101773

Опубликована: Дек. 20, 2023

The implementation of isolated heterologous monitoring systems for spatially distant borehole deployments often comes with substantial equipment costs, which can limit the effectiveness geohazard mitigation and georisk management efforts. To address this, we have developed a novel system that integrates fiber Bragg grating (FBG) microelectromechanical (MEMS) techniques to capture soil moisture, temperature, sliding resistance, strain, surface tilt, deep-seated inclination. This enables real-time, simultaneous data acquisition cross-validation analyses, offering cost-effective solution critical parameters in geohazard-prone areas. We successfully applied this integrated Xinpu landslide, an active super-large landslide located Three Gorges Reservoir Area (TGRA) China. resulting strain profile confirmed presence two shallow secondary surfaces at depths approximately 7 m 12 m, respectively, addition depth ∼28 m. lower was activated by extreme precipitation, while upper one primarily driven significant changes reservoir water levels secondarily triggered concentrated rainfalls. Anti-slide piles remarkably reinforced moving masses but failed control ones. gap between pile heads amplified rainwater erosion effect, creating preferential channel infiltration. Multi-physical measurements revealed mixture seepage-driven buoyancy-driven behaviors within landslide. study offers dual-source multi-physical paradigm collaborative multiple crucial boreholes on large-scale contributes evaluation improvement engineering measures similar geological settings.

Язык: Английский

---

Early warning system for landslide of gentle Piedmont slope based on displacement velocity, factor of safety, and effective rainfall threshold

International Journal of Disaster Risk Reduction, Год журнала: 2025, Номер unknown, С. 105232 - 105232

Опубликована: Янв. 1, 2025

Язык: Английский

---