This session seeks contributions presenting novel methods, emerging trends, and case studies in landslide and slope instability reconnaissance, monitoring, and early warning. We particularly encourage submissions showcasing the integration of geophysical, geotechnical, geological, and remote sensing data to build a landslide model able to characterize the landslide architecture and track its evolution.
We especially invite abstracts demonstrating:
• Multi-method approaches combining geophysical, geotechnical, and remote sensing techniques.
• Applications of machine learning to landslide hazard assessment and prediction.
• Time-lapse geophysical surveys for monitoring subsurface changes.
• Determination of geomechanical parameters through integrated geological (e.g., borehole data, geotechnical surveys) and geophysical studies.
• Effects of climatic global changes and land use on the susceptibility and hazards towards shallow landslides.
• Field hydrological monitoring for the assessment of main pore-pressure build-up areas and triggering conditions of shallow landslides.
Recognizing the cross-disciplinary nature of this challenge, we welcome contributions addressing a broad range of slope instability types, including avalanches, natural and engineered slopes, and climate-induced failures.
Orals: Thu, 7 May, 08:30–12:30 | Room N2
Rainfall-induced shallow landslides represent a critical natural hazard in mountainous regions, with their frequency controlled by hydrological processes. Climate change is expected to alter both precipitation patterns and soil moisture dynamics but quantifying these impacts on landslide susceptibility remains challenging.
In this study, we integrate physically-based stability thresholds with distributed hydrological modeling to assess future landslide hazard evolution under multiple climate scenarios. The study is conducted for a small basin (~28 km2) located in the north-eastern Friuli Venezia Giulia (Italy).
Spatially explicit Critical Soil Moisture (CSM) and Critical Wetness Index (CWI) thresholds at 50 m resolution were derived in a previous effort for multiple failure depths (0.75 to 2.00 m) by inverting the infinite slope stability analysis. The thresholds represent hydrological conditions at which slope failure may initiate through either unsaturated zone processes or groundwater table rise. These thresholds were coupled with a calibrated distributed and physically-based hydrological model, the Triangulated Irregular Network‐based real‐time integrated basin simulator (tRIBS), which simulates hourly soil moisture and groundwater dynamics, to assess the occurrence of failure over 100-year periods for three synthetically generated climate scenarios: current conditions, moderate emissions (RCP4.5, 2050), and high emissions (RCP8.5, 2050). The synthetic series of meteorological variables, and particularly precipitation, were generated by combining the AWE-GEN (Advanced WEather GENerator) model with a procedure to correct the distribution of extreme events.
We quantify exceedance frequencies, i.e., the proportion of time during which CSM and CWI thresholds are exceeded, as a measure of temporal exposure to landslide-conducive conditions. Results reveal that, under RCP4.5, exceedance frequencies decrease by up to 14.6% (CWI) and 10.9% (CSM), due to a reduction in annual precipitation despite an increase in mean intensity per event. In contrast, RCP8.5 shows bidirectional patterns, with maximum increases reaching 5.1% (CWI) and 3.6% (CSM), indicating that precipitation intensification begins to overcome the reduction in annual precipitation. Critically, climate impacts amplify with failure depth; the 2.00 m failure depth exhibits changes in magnitude up to three times greater than those at 0.75 m, suggesting that deeper failures become disproportionately more sensitive to climate change.
This research received funding from European Union NextGenerationEU – National Recovery and Resilience Plan (PNRR), Mission 4, Component 2, Investment 1.1 -PRIN 2022 – 2022ZC2522 - CUP G53D23001400006.
How to cite: Arnone, E., Thomas, J., Ciriminna, D., and Francipane, A.: Assessing Climate-Driven Changes in Rainfall-Induced Landslide Probability Using Distributed Hydrological Modeling, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17967, https://doi.org/10.5194/egusphere-egu26-17967, 2026.
Exploring the internal structure of large landslides is crucial for understanding their deformation mechanisms and conducting stability assessments. However, traditional exploration methods, such as drilling, provide only localized information and fail to reflect the spatial continuity of subsurface structures. Single geophysical methods also face challenges in accurately characterizing deep-seated structures due to inversion non-uniqueness and interpretative ambiguity. Multi-source geophysical data fusion is considered an important approach to reduce ambiguity and improve modeling reliability, but existing research largely focuses on shallow landslides, lacking effective methods for the three-dimensional reconstruction of large deep-seated rock landslides. Taking the Tizicao deep-seated toppling on the eastern edge of the Tibetan Plateau as an example, this study proposes a multi-source geophysical data fusion modeling method based on the Gaussian mixture model (GMM). This method comprehensively utilizes electrical resistivity tomography (ERT), multi-channel surface wave exploration (MASW), the horizontal and vertical spectral ratio method (HVSR) for ambient noise, and UAV photogrammetry to achieve the fusion and classification of multiple parameters such as resistivity, shear wave velocity, and structural depth. By automatically partitioning the geophysical feature space using GMM, a three-dimensional model of the Tizicao toppling is constructed. The three-dimensional model is highly consistent with the borehole results, verifying the reliability of the fusion modeling method. In addition, the deep-seated structure revealed by the three-dimensional model plays a key controlling role in the initiation of slope instability. Overall, the proposed GMM-based multi-source geophysical fusion method not only enables accurate reconstruction of the internal structure of large deep-seated rock landslides but also provides a new technical pathway for mechanism analysis and hazard prediction of large deep-seated landslides.
Keywords: Deep-seated toppling; Multi-source geophysical integration; Gaussian Mixture Model (GMM); 3D structural modeling; Deformation evolution.
How to cite: Wang, H., Pei, X., Quan, Z., Cui, S., Xing, S., and Wang, Y.: 3D structure and deformation evolution of a large deep-seated toppling revealed by GMM-based multi-source geophysical integration, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17460, https://doi.org/10.5194/egusphere-egu26-17460, 2026.
Landslide motion spans a continuum from slow, steady creep to rapid catastrophic failure. However, the mechanisms controlling the timing, rate, and nature of sliding, the sensitivity of motion to perturbations driven by precipitation or human activity, and potential transitions from creep to catastrophic failure all remain poorly understood. The response of landslide basal shear zones to rainfall-driven changes in pore pressure and thus effective stress can be interpreted using rate and state friction, a framework that describes the constitutive behavior and sliding stability of frictional shear zones, and is widely applied to earthquake mechanics. Laboratory experiments provide direct constraints on these frictional properties, and thus hold the potential to illuminate the material properties and conditions that control basal slip. We investigate the frictional behavior of Oak Ridge earthflow, a slow-moving landslide in the Coast Ranges of central California hosted within a clay-rich mélange. We conduct a suite of direct shear experiments to characterize its frictional rheology, including both (1) the velocity dependence of friction measured from velocity step tests; and (2) frictional healing, or time-dependent restrengthening between slip events, measured via slide-hold-slide tests. Experiments are conducted across a range of normal stresses approximating the in-situ conditions of the active shear plane (0.3 – 2 MPa) and at sliding velocities that span the range of observed landslide creep (0.001 – 30 𝜇m/s).
The shear plane material exhibits uniformly velocity strengthening behavior, characterized by a positive rate parameter (a-b), indicating that friction increases with increased slip rate, and is consistent with stable sliding. The values of (a-b) from laboratory experiments ranges from 0.001 – 0.015, in agreement with values inferred from coupled field observations of slide motion and pore pressure. Our results suggest that velocity strengthening friction, combined with modulation of effective stress through pore pressure, can generate slip transients, providing a direct mechanistic link between laboratory scale behavior and field observations of landslide motion.
We also find that the clay rich materials entrained along the base of the slide exhibit little to no healing (𝛽 ≈ 0). Near zero healing implies that the slide does not restrengthen during extended periods of low water pressure during the dry California summer. In the absence of healing, slip velocity responds directly and immediately to changes in pore pressure, independent of the duration of dry periods. Taken together, velocity strengthening friction and little to no healing are consistent with the persistent creep observed in the field, where the slip rate is governed by the stress state, pore pressure, and rate dependence of friction. Notably, Oak Ridge earthflow has been active since at least the 1930’s (the date of first air photos). The laboratory derived frictional rheology provides a quantitative framework to explain the observed landslide slip response to changes in pore pressure and suggests that friction laws can be used not only to interpret past slide behavior, but potentially to predict landslide responses to future climate-driven hydrologic forcing or other external perturbations.
How to cite: Schaible, K., Saffer, D., and Finnegan, N.: Experimental constraints on the slip response of a slow-moving landslide to rainfall driven pore pressure changes, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14669, https://doi.org/10.5194/egusphere-egu26-14669, 2026.
Relative seismic velocity changes (dv/v) derived from ambient noise interferometry serve as a proxy for the internal rigidity or structural health of landslide materials. Strong ground motion often induces coseismic velocity drops, indicating damage within the shallow crust or the landslide body. This study focuses on the deep-seated, slow-moving Wuhe landslide in eastern Taiwan, which exhibits stable creeping with daily displacement rates ranging from 4 mm to 25 mm(Weng et al., 2025), to investigate its response to the September 2022 earthquake sequence, specifically the ML 6.6 Guanshan and ML 6.8 Chihshang earthquakes.To monitor temporal variations in the landslide's internal state, we applied the single-station cross-component (SC) technique to the Wuhe landslide using continuous ambient noise records. The seismic monitoring network comprises one geophone installed directly on the sliding mass and three reference stations located on stable bedrock outside the landslide area. This configuration aims to differentiate between landslide-specific structural changes and regional reference variations. The preliminary results showed that a clear seismic velocity reduction was found spatially within the landslide area. Through dv/v measurements with in-situ real-time kinematic (RTK) GPS data and strong-motion records, the coseismic velocity drops are in response to the accelerating surface displacement and strong ground shaking, and the spatial relationships between dv/v, surface movement and peak-ground acceleration (PGA) are systematically compared . In fact, the earthquake did not trigger catastrophic landsliding at the Wuhe site, Thus, we further investigate the recovery of landslide material properties following strong ground shaking. The post-seismic recovery duration captured by dv/v observations can help us to better understanding recovery mechanism of landslide material after earthquakes.
How to cite: Weng, H.-K. and Chao, W.-A.: Coseismic Seismic Velocity Variations of a Deep-Seated Landslide Caused by Two M6.5+ Earthquakes in Eastern Taiwan, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15810, https://doi.org/10.5194/egusphere-egu26-15810, 2026.
Earthflows are flow-like landslides involving fine-grained, clay-rich materials that exhibit complex kinematics, long-term activity, and alternating phases of slow movement and sudden acceleration. Although their flow-like behaviour is commonly attributed to distributed internal deformation and plastic rheology, the mechanisms governing the transition from solid-like sliding to fluid-like flowing remain poorly understood, particularly with respect to boundary conditions and material properties. This transition is critical, as it may lead to surging events associated with high mobility and significant hazard.
This study investigates the role of mineralogical, geotechnical, rheological, and geomorphological factors in controlling earthflow mobility and material fluidization. A set of representative earthflows located in the southern Apennines was selected, covering a wide range of geological settings and morphological characteristics. Laboratory analyses were conducted on samples collected from different sectors of the landslides, including grain size distribution, Atterberg limits, mechanical behaviour, quantitative mineralogical composition. Moreover, rheometrical analysis of the fine fractions under controlled shear conditions were also performed. These data were integrated with long-term geomorphological analyses based on satellite imagery and morphometric reconstructions of landslide geometry.
Earthflow behaviour was analysed using a one-dimensional framework based on a Herschel–Bulkley viscoplastic rheological model, aimed at reproducing internal kinematic compartmentalisation in relation to variable water content.
The influence of water content variations, as a function of rainfall-induced infiltration conditions, on rheological parameters and mechanical response was investigated. The results highlight strong correlations between plasticity, occurrence of expandable clay minerals, rheology, and mobility, emphasizing the key role of fine-grained materials in promoting solid–fluid transitions.
By integrating multi disciplinary datasets, this work advances the understanding and prediction of earthflow fluidization and mobility-processes for which current forecasting capabilities remain notably limited.
How to cite: Annibali Corona, M., Calcaterra, D., Di Spirito, N. A., Izzo, F., Langella, A., Mercurio, M., Pasquino, R., Russo, G., Vitale, E., and Guerriero, L.: From Sliding to Flowing: Integrating Geotechnical, Mineralogical, and Rheological Controls on Earthflow Mobility, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14514, https://doi.org/10.5194/egusphere-egu26-14514, 2026.
Shallow landslides represent a major hazard in western Rwanda, where steep slopes, deeply weathered materials and intense precipitation frequently interact. This study, carried out in the framework of the WALL project (Grant ID: GCRW-CL001, https://www.wallatrwanda.org/), focuses on a landslide-prone area within the Karongi District and presents a proof-of-concept analysis aimed at investigating the sensitivity of slope stability to key geotechnical and pore pressure–related parameters.
Slope stability is analysed using Scoops 3D (Reid et al., 2015), which implements three-dimensional limit-equilibrium methods (LEM) and evaluates slope stability by testing a large number of potential spherical trial failure surfaces. This approach allows for a systematic exploration of potential instability mechanisms while maintaining a computationally efficient framework suitable for regional-scale and data-scarce applications. Due to the limited availability of site-specific geotechnical data, model parameters are defined within plausible ranges derived from published literature and regional information.
Under these conditions, a global sensitivity analysis based on Sobol indices (Saltelli and Sobol, 1995) represents a suitable and robust strategy to investigate model behaviour and uncertainty. The Sobol analysis is applied to investigate the influence of key geotechnical parameters, including cohesion, internal friction angle and unit weight, and additional pore pressure accounting for hydrological conditions on slope stability results. Both first-order effects and higher-order interaction terms are analysed, providing insights into the combined mechanical and hydraulic controls on slope stability.
The proposed workflow identifies the dominant sources of variability on the output and offers a structured basis for prioritizing the quantification of geotechnical parameters in future data acquisition and model refinement, also in connection with specific triggering factors relevant for the studied area, such as rainfall.
REFERENCES
Reid, M. E., Christian, S. B., Brien, D. L., & Henderson, S. T. (2015). Scoops3D: software to analyze 3D slope stability throughout a digital landscape (No. 14-A1). US Geological Survey.
Saltelli, A., Sobol’, I. M. (1995). Sensitivity analysis for nonlinear mathematical models: numerical experience. Matematicheskoe Modelirovanie, 7(11), 16–28.
How to cite: Zanetti, M., Armigliato, A., Angeli, C., Zaniboni, F., Barayagwiza, S., and Meriaux, C.: Exploring the stability of shallow landslides through global sensitivity analysis: a proof of concept from western Rwanda, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11274, https://doi.org/10.5194/egusphere-egu26-11274, 2026.
With intensifying precipitation events, landslides pose increasing environmental hazards. Unsaturated slopes are key monitoring targets due to their rapid, and sometimes severe, response to rainfall. This study investigates how hydrological changes in an unsaturated slope in Eidsvoll (Norway) influence seismic velocities through time and space using a physics-based modelling framework. Vertical effective stress and density fields from hydromechanical simulations in GeoStudio are used as inputs to the Biot-Gassmann relationship to estimate time-varying P- and S-wave velocities. These velocities are used to compute Rayleigh wave phase velocity dispersion curves and sensitivity kernels for selected days throughout a 250-day (September 2019-May 2020) simulation period. The results reveal a strong coupling between infiltration, effective stress, and seismic velocities, especially in the upper part of the unsaturated slope. Rayliegh wave sensitivity is highly frequency- and depth- dependent: high frequencies (above 60 Hz) are sensitive to near-surface changes, while lower frequencies probe deeper layers. A persistent blind zone in an intermediate high-velocity layer limits the surface waves sensitivity to certain depths, underscoring the importance of survey design and the usefulness of surface waves depending on the geologic scenario. This forward modelling approach enables identification of optimal frequency ranges and target depths, providing critical input for future field investigations. These findings contribute to the development of focused site-specific seismic monitoring strategies, including passive surveys using anthropogenic noise sources or active source MASW. By bridging hydromechanical modelling and the associated seismic response using slope-scale physical processes, this approach can support early warning systems and landslide hazard assessment under changing climate conditions.
How to cite: Mikesell, T. D., Lorentzen, E. B., Piciullo, L., and Sørensen, M. B.: Linking Hydrological Forcing to Seismic Sensitivity in an Unsaturated Slope Using Physics-Based Modelling, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16829, https://doi.org/10.5194/egusphere-egu26-16829, 2026.
Shallow landslides triggered by intense and prolonged precipitation represent a major geohazard in many soil-dominated landscapes. This study presents the development of an integrated monitoring and modelling framework for the early detection and characterisation of hydraulically induced shallow landslides. The approach is based on the selection of three representative pilot sites and the implementation of comprehensive field investigations (engineering-geological, pedological, geotechnical, hydrological) and laboratory testing to determine the chemical, physical, and mechanical properties of characteristic soil horizons. A real-time monitoring system has been established to continuously record soil volumetric water content and suction, together with precipitation, providing high-resolution hydro-meteorological and hydrological data. Geoelectrical measurements and field investigations were applied to characterise soil structure and depth, and to establish relationships between geophysical parameters and physico-mechanical soil properties. These analyses enable the development of a non-invasive monitoring approach capable of diagnosing landslide initiation, delineating landslide geometry, and estimating potentially unstable volumes. Based on the monitoring data obtained at pilot sites, hydro-meteorological thresholds and critical soil parameters controlling shallow landslide occurrence are derived for key soil types. Safety factors and probabilistic landslide occurrence models are developed to identify dominant triggering mechanisms. The results contribute to a national-scale framework for shallow landslide susceptibility mapping and provide a transferable methodology for operational landslide early-warning systems. This research is supported by the Slovenian Research and Innovation Agency through research projects: A holistic approach to Earth surface processes driven by extreme weather events (J7-60124) and Geospatial information technologies for a resilient and sustainable society (GC-0006).
How to cite: Jemec Auflič, M., Maček, M., Smolar, J., Kure, K., Peternel, T., Grčman, H., Turniški, R., Zupan, M., Zupanc, V., Žvokelj, L., and Pulko, B.: Framework for early detection and characterisation of hydraulically induced shallow landslides, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9556, https://doi.org/10.5194/egusphere-egu26-9556, 2026.
INTRODUCTION
Underground mining frequently leads to surface instability such as subsidence, sinkholes, and landslides. In the Bulqiza chrome mine in Albania, decades of extraction and the transition from cut-and-fill to sublevel stoping have increased rock-mass deformation, resulting in fissures, caving, and surface failures. This study focuses on Profile XIV, where both continuous subsidence and a sinkhole are present, in order to evaluate the accuracy of predictive methods used to assess mining-induced deformation.
AIM
This study aims to assess the surface impacts of underground mining in the Bulqiza district by applying both empirical subsidence modelling and numerical simulations using Finite Element Methods. The study compares predicted results with observed deformation, evaluates the influence of caved zones (goaf) and tectonic structures, and verifies the suitability of using a combined empirical and numerical approach for deformation assessment.
METHODS
Geological and mechanical properties were defined through field investigations and archived mine data. An empirical model with a subsidence coefficient of K = 0.9 was used to calculate the critical collapse depth (Hcal) and compare it with the effective mining depth (Hfac). Numerical simulations were then performed with the Rocscience FEM software for two surface-deformation profiles: one exhibiting continuous subsidence and the other featuring a surface sinkhole. Each profile was modelled under different conditions, including the presence or absence of goaf and the inclusion or exclusion of tectonic influence. Surface displacement was used as the main indicator for assessing deformation.
RESULTS
The empirical model indicated a low likelihood of funnel formation in the subsidence profile, where Hcal was smaller than Hfac, while in the sinkhole profile, Hcal exceeded Hfac, confirming a high probability of collapse consistent with field observations. Numerical modelling supported these findings. In the subsidence profile, vertical displacement remained small around 14 mm regardless of whether the goaf was included, and no funnel formation was predicted. In the sinkhole profile, displacement increased to 24.3 mm when the goaf was considered without tectonics. When tectonic effects were included, displacement increased substantially to values between 40.4 and 61 mm, closely reproducing the actual sinkhole conditions. These results show that tectonics strongly amplifies surface deformation.
CONCLUSIONS
This study demonstrates that both empirical and numerical methods effectively reproduce the types and magnitudes of surface deformation observed in the Bulqiza mine. Numerical modelling closely matched actual conditions, particularly when tectonic effects were incorporated. While goaf conditions had little effect in the subsidence zone, they significantly increased deformation in the sinkhole area. The findings confirm that tectonic structures are a major factor controlling surface collapse and that a combined empirical and numerical approach provides a reliable method for assessing mining-induced surface impacts in Bulqiza and comparable underground mining environments.
How to cite: Belba, P.: Surface Deformation Assessment in the Bulqiza Chrome Mine Using Empirical and Numerical Modelling, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-798, https://doi.org/10.5194/egusphere-egu26-798, 2026.
How to cite: Agustina, L., Arnhardt, C., Van Wyk De Vries, M., Hussain, E., Large, D., and Turnbull, B.: Understanding and Zoning Rainfall-Induced Landslide Hazards in Indonesia: Insights from Observation to Forecasting, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-22967, https://doi.org/10.5194/egusphere-egu26-22967, 2026.
Understanding landslide (LS) distribution in deglaciated mountains is key to landscape evolution and geohazard risk. We present an orogen-scale assessment of 1,691 Post-Little Ice Age (LIA) LSs (91% shallow) along the South Patagonian Icefield (SPI, 48–52°S) margins. Mapped via high-resolution multitemporal imagery (2010–2025) and multi-operator validated, kernel densities (10 km bandwidth) show clustering in western and southern SPI—central peak, northwest secondary—amid ~20% ice loss (since the end of the LIA) and uplift >40 mm/yr.
Environmental variables from LS/non-LS areas fed Bayesian horseshoe variable selection. Sparse Gaussian process regression (R2=0.96, SPAEF ≥0.85) identified precipitation, fault density, and uplift as dominant controls. Precipitation destabilizes slopes via pore pressures, triggering shallow LSs (positive correlation); fault density signals structural weakness/seismic facilitation; uplift shows complex negative LS correlation, as active deformation/steep slopes favor erosion over accumulation, reducing LS buildup. Lithology, permafrost, retreat rates exert weaker, context-dependent influences. LS versus non-LS distinctions underscore the value of integrating correlation-based and predictive approaches. Coupled climate-deglaciation-tectonics govern landslide distribution in the SPI.
Critically, ~17% of LSs overlap glacial lake upslope areas (30 m buffer), preconditioning glacier lake outburst flood risks at, e.g. Torre Glacier's ~8 Mm³ failure—shallow dominance may temper severity, sea-proximal cases extend threats. Findings illuminate paraglacial responses to glacier retreat, offering predictive hazard frameworks for warming cryosphere.
How to cite: Seier, G., Slíva, M., Pánek, T., and Winocur, D.: Environmental Controls on Post-Little Ice Age Landslide Distribution Around the South Patagonian Icefield, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3757, https://doi.org/10.5194/egusphere-egu26-3757, 2026.
Road construction on hillslopes has increased explosively due to the rapid socioeconomic development in China’s mountainous areas. The exposure of steep and rapidly weathering slopes caused by road construction accelerates slope movements, especially roads building on residual soil. Residual soil slopes are prone to slow movements and may evolve to failure in response to infiltration of rainwater. Engineering works on residual soil (e.g., excavation, filling for buildings and roads) exacerbate these problems through altering the internal and external stress of slopes. Yet our understanding of the interactive effects of rainfall and road construction on slope dynamics or even failure in subtropical residual soils remains elusive. Here, we used three-decadal radar remote sensing data to quantify the time series deformation before a catastrophic slope failure, occurring at Meida Highway in China that caused 52 fatalities. Physics-based decomposition of the time series movements over the past 8 years reveals that there is a constant seasonal movement related to rainfall and a precursory accelerated movement triggered by slope reinforced measures before failure occurrence in May 2024. Emergency mitigations of reinforced measures modified the infiltrates and routes of surface and subsurface water, leading to an adverse impact of reducing slope failure risk. Analysis of numerical simulation indicates that rainfall-induced pore water pressure reduced the shear strength of granite residual soils, ultimately triggering slope failure. This improved understanding of the slope dynamics in response to different forces will be important to avoid economic and life loss, strengthen emergency planning and identify potential risks.
How to cite: Huang, X. and Ma, P.: Satellite images reveal progressive slope deformation triggered by mountainous road construction in subtropical South China, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4787, https://doi.org/10.5194/egusphere-egu26-4787, 2026.
Landslides are a global phenomenon occurring in several climatic and geomorphologic contexts, generating billions in economic losses and causing thousand of causalities each year. This phenomenon is often characterized as a local problem, but its effect and cost frequently cross local jurisdiction and may become a national problem [1]. Landslides, resulting from disturbance in slope equilibrium induced by the movement of a mass of rock, debris or earth down a slope and pose a significant threat to landscapes, infrastructure and human life [2]. Landslides can be labelled into different categories depending on the type of movement and the type of material involved. They may be triggered by several phenomena; the primary are seismic activities and heavy rainfall. More precisely, rainfall-induced landslides typically occur in regions prone to heavy precipitation, with steep slopes and poorly consolidated soil or rock [2]. In Italy, the most recent case study, is the Cormons (Gorizia, Italy) landslide occurred on November 17th 2025. Here, intense rainfalls caused a mud-flow inducing the collapse of several buildings and two casualties. In this area, landslides are the most frequent type of instability. These are mostly small and medium-sized landslides, located on flyschoid hills, affecting vineyards and only locally affecting roads and rural settlements [3]. Identifying these phenomena through satellite-based remote sensing techniques offers essential data and insight for landslide studies. Information regarding timing, location and spatial extent of detected landslides, along with changes in surface materials, plays a key role in risk and susceptibility assessments as well as in effective disaster management, monitoring and response activities. For the purpose of this work, optical satellite images provided by Sentinel-2, together with the Lidar provided by the Italian Civil Defense have been used with the aim to identifying the Cormons landslide and its characteristics in terms of dimensions, shape and amount of material moved during the event. The use of optical imagery from Sentinel-2 it’s been used to evaluate spectral indices like Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI) and Bare Soil Index (BSI). Instead Lidar and DEM have been used to define the ground changes in terms of elevation and also the amount of material involved in the event. From the GIS analysis, the results confirm the presence of a mudflow within a watershed located in the Cormons area. Additionally, from the Lidar other small collapse features have been highlighted in the surrounding area.
REFERENCES:
- Highland, L. M., & Bobrowsky, P. (2008). The landslide handbook-A guide to understanding landslides(No. 1325). US Geological Survey.
- Peters, S., Liu, J., Keppel, G., Wendleder, A., & Xu, P. (2024). Detecting coseismic landslides in GEE using machine learning algorithms on combined optical and radar imagery. Remote Sensing, 16(10), 1722.
- https://www.isprambiente.gov.it/files/pubblicazioni/rapporti/rapporto-frane-2007/Capitolo_11_Friuli_Venezia_Giulia.pdf
How to cite: scalabrini, A., fornasari, S. F., and costa, G.: Cormons landslide characterization using Lidar and remote sensed data., EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6955, https://doi.org/10.5194/egusphere-egu26-6955, 2026.
An increase in soil water content (SWC) from rainfall infiltration reduces the matric suction and shear strength; hence, rainfall is a primary trigger of shallow landslides. While accurate SWC monitoring is critical for predicting slope failure, traditional point-based sensors lack the spatial resolution required for effective field-scale assessment. This study aims to bridge this gap by integrating hyperspectral and multispectral imaging technologies with advanced machine learning (ML) models. Based on 114 in-situ soil samples collected from landslide-affected areas across South Korea, correlations between physical soil properties (e.g., void ratio, soil color) and hyperspectral data in the visible and near-infrared (Vis-NIR) regions were analyzed. Two ML algorithms, Random Forest (RF) and Multilayer Perceptron (MLP), were employed to develop predictive models for SWC. In this study, statistical evaluation indicated that the RF model demonstrated superior accuracy and robustness in handling high-dimensional spectral data compared to the MLP model. To validate the method's applicability for landslide monitoring, field tests were conducted in the mountainous region of Pyeongchang, South Korea, using a multispectral camera mounted on an unmanned aerial vehicle (UAV). The RF model successfully predicted the spatial distribution of SWC using spectral reflectance and geotechnical parameters. Although the model showed limitations in extrapolating beyond the training data range, it effectively captured critical variations in soil moisture relevant to slope stability. These results suggest that integrating UAV-based remote sensing with ML offers a promising, non-contact approach for high-resolution monitoring of shallow landslides, contributing to more proactive disaster prevention strategies.
How to cite: Lim, H.-H., Cheon, E., and Lee, S.-R.: UAV-Based Multispectral Assessment of Soil Water Content for Shallow Landslide Monitoring: A Machine Learning Approach, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2879, https://doi.org/10.5194/egusphere-egu26-2879, 2026.
Rapid and safe identification of slip-surface geometry is essential for efficient landslide investigation and mitigation. Conventional approaches to slip-surface determination rely primarily on borehole surveys and in situ instrumentation; however, these methods require long investigation periods and substantial labor.
In this study, we propose a new method that automates slip-surface reconstruction using high-density ground-surface displacement vectors derived from multi-temporal topographic data collected by a laser-equipped UAV at two landslides developed in Neogene formations in northeastern Japan. The analysis estimates two-dimensional slip-surface profiles along multiple cross sections (following Ogita et al., 2024), which are subsequently integrated to construct a quasi–three-dimensional slip-surface geometry. For validation, the landslide moving mass volumes estimated using the proposed method were compared with those identified from dense borehole data. The results show agreement rates of 87% and 96%, respectively. These findings demonstrate that the proposed method achieves sufficient accuracy for practical application in future landslide mitigation planning.
References：
OGITA, S., HAYASHI, K., ABE, S., TSOU, C.-Y. (2024): Estimation of slip surface geometry from vectors of ground surface displacement using airborne laser data : case studies of the Jimba and Tozawa landslides in Akita Prefecture, Journal of the Japan Landslide Society, 61(4) 123-129 （in Japanese with English abstract）.
How to cite: Ogita, S., Sanuki, S., Hayashi, K., Itou, K., Abe, S., Nguyen, D. D. N., and Tsou, C.-Y.: Automated quasi-3D reconstruction of landslide slip surfaces using UAV-derived surface displacement, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7148, https://doi.org/10.5194/egusphere-egu26-7148, 2026.
Slope creep refers to the imperceptibly slow and gradual downslope movement of soil and rock driven by gravity. It is mainly driven by moisture-induced expansion of clay-rich materials and the resulting decrease in shear strength. Although subsurface conditions can influence slope creep vulnerability, identifying their effects remains challenging. In recent years, electrical resistivity and seismic surveys have been widely used to characterize the spatial and temporal variability of subsurface soil properties. These geophysical methods provide a non-destructive means of investigating subsurface physical characteristics. In this study, electrical resistivity and seismic surveys were conducted to assess slope creep vulnerability associated with subsurface conditions. Geophysical survey data were obtained from 124 slope sites, and their slope creep vulnerability was classified into two groups (low and high) based on field investigations. Cross-plot analysis was applied to integrate electrical resistivity and seismic velocity, and the resulting data points were classified into four quadrants according to threshold values of seismic velocity and electrical resistivity. The threshold values were statistically determined using a t-test. The composition ratios of the four quadrants were used as input variables for deep learning training, and the bedrock proportion based on seismic velocity included as an additional input. As a result, a total of five input variables were used, and deep learning training was performed by classifying slope creep vulnerability into two groups. As a result, a total of five input variables were used to train a deep learning model for classification of slope creep vulnerability into two groups. Due to the limited dataset size, five-fold cross-validation was applied for model evaluation. As a result, the deep learning model achieved an accuracy of 81.5% and a recall of 83.0% in classifying slope creep vulnerability, indicating its effectiveness in identifying slope creep–prone areas.
Acknowledgments: This study was carried out with the support of ´R&D Program for Forest Science Technology (RS-2025-02213490)´ provided by Korea Forest Service (Korea Forestry Promotion Institute).
How to cite: Bong, T., Jeon, J., Jeong, E., Lee, S., Heo, J., and Seo, J.: Deep Learning-Based Assessment of Slope Creep Vulnerability Using Geophysical Survey Data, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10608, https://doi.org/10.5194/egusphere-egu26-10608, 2026.
Posters on site: Thu, 7 May, 16:15–18:00 | Hall X3
The interplay between internal structure, deformation mechanisms, and subsurface hydrogeological processes controls the long-term stability of large landslides. A key unresolved issue is whether infiltrating groundwater is confined to the landslide body or can migrate into the underlying bedrock along deep-seated structural discontinuities. This problem is particularly relevant in areas underlain by steeply dipping flysch formations, where structural anisotropy may promote vertical groundwater connectivity and influence landslide reactivation. This study focuses on the Cisiec landslide in the Żywiec district of southern Poland, aiming to identify groundwater percolation pathways and their relationship to slope deformation. The landslide affects a ski slope located in a forest–meadow transition zone and moves predominantly east–northeast, with an elevation difference of approximately 100 m. Previous monitoring indicated complex kinematics but did not resolve the depth extent of groundwater infiltration or its coupling with deep geological structures.
We apply an integrated electromagnetic approach explicitly designed to resolve processes across complementary depth ranges. Shallow groundwater dynamics were monitored using time-lapse Frequency Domain Electromagnetics (FDEM), which is sensitive to depths of approximately 0–3 m and was repeated over a three-year interval. FDEM conductivity variations were used to map spatial and temporal patterns of near-surface water percolation within the landslide body. In addition, the in-phase component of the FDEM signal was exploited to detect positional changes of buried infrastructure on the ski slope. When combined with high-precision Differential GPS (DGPS) measurements, these data provided quantitative constraints on surface displacement and landslide activity. To resolve the intermediate-depth range and provide robust constraints for deep imaging, Electrical Resistivity Tomography (ERT) was conducted along five profiles across the landslide. The resulting resistivity sections, which image the subsurface to approximately 30 m depth, were incorporated as a priori resistivity constraints and starting models for the inversion of Audio-Magnetotelluric (AMT) data. This constrained inversion strategy significantly reduced ambiguity in the AMT results and ensured consistency between shallow, intermediate, and deep resistivity structures.
AMT imaging extended the investigation below 30 m depth and enabled the construction of a three-dimensional resistivity anomaly model of the landslide and its geological basement. The model reveals pronounced, near-vertical resistivity structures associated with the Carpathian flysch beneath the landslide, interpreted as preferential pathways for deep groundwater migration. The integrated interpretation of FDEM, ERT, and AMT data indicates that infiltrating groundwater is not restricted to the landslide mass but can penetrate into the bedrock along steeply oriented discontinuities. This hydrogeological connectivity between shallow infiltration zones and deep structural features provides a plausible mechanism for delayed landslide reactivation and long-term slope instability. The study highlights the importance of multi-scale, constraint-driven electromagnetic imaging for improving hazard-relevant conceptual models of complex landslide systems.
How to cite: Oryński, S., Marciniak, A., Kowalczyk, S., Flores-Orozco, A., and Majdański, M.: Hydrological links between shallow and deep zones in a flysch landslide revealed by repeated FDEM surveys and 3D AMT imaging, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13476, https://doi.org/10.5194/egusphere-egu26-13476, 2026.
Cuolm da Vi (CdV) is a deep-seated gravitational slope deformation in central Switzerland with an estimated unstable volume of around 150 million m3. In the central part, surface displacement rates are on the order of 10 to 20 cm/yr. The ongoing south-westward deformation, which is dominated by toppling, is expressed by scarps, graben-like structures, tension cracks, and local instabilities. These landforms suggest gravitational movement guided by inherited tectonic structures. Despite detailed geomorphological mapping, geological-geotechnical investigations, and more than two decades of surface-displacement monitoring, fundamental uncertainties remain regarding, for example, the maximum depth of the unstable mass and the internal deformation processes.
Here, we integrate multiple geophysical and geological constraints into a 3-D structural model of the instability. To establish the model, we combined a 3-D P-wave velocity volume from first-arrival travel-time tomography, microseismicity detected during five months of continuous distributed acoustic sensing (DAS) monitoring, and distributed strain sensing (DSS) observations from around two years of periodic measurements, together with detailed mapping of tectonic features and available geotechnical information. We feed the geophysical and geological data into a 3-D structural and probabilistic geological modelling framework to establish a complex model of the structural features of CdV. The model covers about 1 km² at the surface and extends to a few hundred meters depth.
Low P-wave velocities (Vp < 2000 m/s) spatially coincide with mapped unstable terrain, indicating that velocity variations can help delineating comparatively intact versus more fractured/damaged rock volumes. Based on the geometry of the low-velocity domain, the maximum depth of the unstable mass in the central part is estimated at about 180-200 m. Microseismicity is concentrated within low-velocity regions and clusters near mapped tectonic features, consistent with deformation localized on key planar discontinuities. Key tectonic features are also associated with distinct DSS strain events. The resulting 3-D “static” model provides a quantitative framework for future analyses of temporal changes in microseismicity, with direct relevance for process understanding and the continued development of early-warning strategies at CdV.
How to cite: Schmelzbach, C., Kiers, T., Chudalla, N., Amann, F., and Bonanomi, Y.: The 3-D anatomy of the Cuolm da Vi slope instability, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14001, https://doi.org/10.5194/egusphere-egu26-14001, 2026.
Estimating the volume of potentially unstable rock masses is a critical yet challenging task in landslide characterization. Traditional methods often struggle to accurately define the height and actual separation of rock blocks because of the hidden nature of fracture persistence. In engineering geology and geophysics, natural frequency (f0) refers to the fundamental modes of vibration of materials, rock masses, soil layers, entire slopes, as well as different man-made structures. A variety of studies have explored the natural frequency and resonance phenomena across contexts using both experimental and numerical approaches.
This work is based on the principle that specific peaks in the Horizontal to Vertical Spectral ratio (H/V) curves of rock blocks are linked to their eigenfrequencies rather than stratigraphic resonance proposes. These frequencies are characterized by strong polarization and linearity normal to the fracture network. Thus, the frequency (fHV) estimated from H/V measurements, is considered a good approximation/estimator of f0 (the block eigenfrequency) and an innovative approach to estimate block volumes from an abacus is proposed. The eigenfrequency-volume abacus was build using Finite Element Method (FEM) simulations. Rock blocks were modelled as rectangular cuboids with fixed boundary conditions at the base, similar to an Euler–Bernoulli cantilever. The simulations integrated site-specific mechanical parameters (Young’s modulus, density, and Poisson’s ratio) consistent with a S-wave velocity of approximately 850 m/s.
The procedure was validated using seismic noise datasets from two test sites on Malta Island (Anchor Bay and Il-Qarraba), where independent volume data from UAV-Digital Photogrammetry and satellite imagery were available. The proposed six-step workflow - ranging from data acquisition to the integration into the abacus of fHV with independent surface area (A) measurements - provides a reliable approximation of the volume's order of magnitude, even with errors in frequency selection.
A key advantage of this method is the ability to use easily obtainable seismic noise data to infer structural properties. Furthermore, discrepancies between abacus-derived volumes (Vest) and field-calculated volumes (Vcalc) can serve as indicators of fracture persistence: Vest < Vcalc suggests fractures are less persistent than they appear, while Vest > Vcalc indicates higher isolation from the rock mass. While the current abacus is site-specific, the methodology is adaptable to different geological backgrounds. This tool represents a significant step forward for rapid, non-invasive rockfall hazard assessment and the characterization of block-release susceptibility.
How to cite: Pazzi, V., Fornasari, S. F., Devoto, S., Costa, G., and Forte, E.: Rapid estimation of block volumes from seismic noise measurements and an eigenfrequency abacus , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17252, https://doi.org/10.5194/egusphere-egu26-17252, 2026.
Shallow landslides frequently occur on natural slopes and cause flow-like disasters. The authors have previously developed 3-Dimensional Translational Slide (3DTS), a physically-based 3D shallow landslide susceptibility model accounting for side resistance and vegetation effects, to efficiently evaluate the slope stability in terms of the factor of safety (FS) over a regional scale. Traditionally, a deterministic slope stability analysis was performed by assigning representative values to rainfall history, soil layers, and soil properties; however, new design standards demand reliability-based analyses that account for the uncertainty and variation in precipitation, subsurface conditions, soil hydro-geotechnical properties, and vegetation root reinforcement. Therefore, this research proposes extending the developed model into a 3-Dimensional Translational Slide-Probabilistic (3DTSP) model to enable reliability-based landslide susceptibility assessment. The developed 3DTSP model combines the generalized Green-Ampt infiltration model and the 3D Janbu simplified slope stability model. The 3D slope stability analysis accounts for additional soil frictional resistance at the side regions in translational slides and additional reinforcements from tree roots. The 3DTSP model uses a Monte Carlo simulation with a random-field approach to determine the FS statistical distribution from variations in the following input parameters: soil thickness, hydraulic properties, Mohr-Coulomb criterion-based shear strength properties, unsaturated soil strength properties, and vegetation resistance properties. Based on the statistical distribution and characteristic length, the model generates a random field of input parameters that accounts for spatial variation in the horizontal direction. For each Monte Carlo simulation iteration, a new random input field is generated to compute FS. The performance and applicability of the developed 3DTSP for probabilistic assessment of landslide susceptibility over regional scales were demonstrated by analyzing landslide case studies. A sensitivity study was conducted to assess the sensitivity of FS to variations in soil thickness, soil properties, and vegetation properties.
How to cite: Cheon, E., Gotaas, M., Pettersen, S., Oguz, E. A., DiBiagio, A., and Piciullo, L.: Sensitivity Analysis of Physically-based 3D Landslide Susceptibility Model from Variation of Input Parameters, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19964, https://doi.org/10.5194/egusphere-egu26-19964, 2026.
Global warming has accelerated glacier retreat and permafrost degradation in high-elevation regions, significantly increasing the frequency and magnitude of glacier-related debris flows. This study focuses on Tianmogou, a debris-flow-prone catchment on the Tibetan Plateau, where three broadband seismometers were deployed for continuous monitoring during the active period. Using ambient noise interferometry, relative seismic velocity changes (dv/v) and the effective decorrelation coefficient (dCe) were calculated to achieve high-resolution characterization of the temporal evolution of subsurface mechanical properties.
The results show that dv/v exhibits pronounced seasonal variations and is significantly negatively correlated with soil temperature, while short-term hydrological processes, such as intense rainfall and snowmelt, lead to rapid dv/v decreases accompanied by marked dCe increases. Notably, several hours prior to multiple debris-flow events, persistent dv/v reductions and rapid dCe increases were consistently observed as precursory signals, with rainfall-triggered events (e.g., 10 July 2020) displaying particularly prominent precursory characteristics. By jointly analyzing seismic velocity changes, precipitation, and soil moisture, this study reveals the progressive degradation of subsurface media during debris-flow initiation and demonstrates the potential of seismic methods for long-term hazard monitoring in glacial and periglacial environments.
How to cite: Lyu, A. and He, S.: Seismic Precursory Velocity Changes Associated with Debris Flows in Tianmogou Inferred from Ambient Noise Interferometry, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17858, https://doi.org/10.5194/egusphere-egu26-17858, 2026.
In recent years, extreme climate events characterized by heavy rainfall and seismic activity have significantly intensified the risks of slope disasters in Taiwan's mountainous regions. This study focuses on Zhongxing Village, Liugui District, Kaohsiung City, Taiwan, an area marked by steep topography and a recurrent history of severe landslides and debris flows. The primary objective is to evaluate slope stability under diverse environmental scenarios using numerical simulation. The methodology utilizes the STEDwin slope stability analysis software, specifically employing the Bishop method, which is based on limit equilibrium theory. A representative geographic profile near Shanping Villa was established, with soil parameters calibrated from 16 localized borehole records obtained from engineering geological databases. The analysis examines three critical conditions: normal, heavy rain, and earthquake. The findings indicate that under normal conditions, the factor of safety (FS) is 1.30, which falls short of the official standard threshold of 1.5 for permanent slope structures. Under the heavy rain scenario (with groundwater at the surface), the FS drops drastically to 0.66, representing a critical 49.23% reduction in stability. In the earthquake scenario, incorporating parameters from the 2016 Meinong earthquake, the FS reached 1.01. These results align closely with historical records from Typhoons Morakot and Kaemi, highlighting significant risks to Shanping Villa, Shanping Forest Road, and Highway 27. In conclusion, the drastic rise in the groundwater level is the primary driver of slope failure in this region. The study recommends the prioritized implementation of deep drainage systems, such as drainage galleries, to enhance soil effective stress. Furthermore, establishing a real-time monitoring and early warning system is essential to facilitate mandatory evacuations during extreme rainfall, thereby ensuring public safety and infrastructure resilience.
How to cite: Hsu, H.-H., Deng, X.-X., Chen, Y.-H., Chang, Y.-C., and Chen, Y.-H.: Slope Stability Analysis and Hazard Potential Assessment in Zhongxing Village, Kaohsiung City: Numerical Simulation under Extreme Rainfall and Earthquake Scenarios Using STEDwin, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2417, https://doi.org/10.5194/egusphere-egu26-2417, 2026.
Shallow landslides represent a frequent geomorphological process in the study region, located in northeastern Romania. The area is characterized by gently undulating interfluves, fragmented slopes, and deeply incised valleys, developed predominantly on clayey substrates. These predominantly shallow slope failures have significant impacts on intensive agriculture, rural infrastructure, and slope stability. Recent climatic variability and anthropogenic modifications of land use amplify the vulnerability of this geomorphological unit. This study presents a detailed assessment of shallow landslide susceptibility through the integration of an extensive landslide inventory with conditioning factors derived from high-resolution geospatial data. The landslide inventory was developed predominantly using digital elevation models generated from LiDAR data (1–2 m resolution), complemented by current orthophotos, drone aerial imagery, slope maps, and selective field validation. The use of LiDAR data substantially improves the precision of delineating shallow unstable features and reduces propagation errors associated with conventional lower-resolution DEMs. This methodology enabled the precise delineation of hundreds of active and relict shallow landslide features, surpassing the limitations of traditional inventories based on photogrammetry or global DEMs.
Relevant conditioning factors for slope dynamics in this region included slope angle, aspect, plan and profile curvature, lithological units (predominantly Miocene-Pliocene clayey deposits), land use, and distance to the drainage network. The dataset was divided into 70% for calibration and 30% for independent validation. The Presence Only Model performance was evaluated through ROC curves and AUC metrics, with values consistently demonstrating excellent predictive performance of the hybrid approach employed.
Results highlight zones of high and very high susceptibility to shallow landslides concentrated along major valleys and their tributaries, and on slopes exceeding 12–15°, where favourable lithological conditions overlap with intensive agricultural land uses or reduced vegetation cover. Methodologically, this study aligns with established international approaches for landslide susceptibility assessment but distinguishes itself through the use of high-resolution LiDAR data (1–2 m), specifically adapted to the morphological context of the region—an area with gently rolling relief and deeply incised valleys. This choice enables substantial reduction of topographic uncertainties inherent in models based on medium or low-resolution DEMs, thereby improving the precision of shallow instability feature delineation and the robustness of local predictions. The result is a susceptibility model with high transferability potential to other similar geomorphological units in plain-to-hill transition zones affected by shallow landsliding.
How to cite: Irimia, R., Sandric, I., and Ilinca, V.: Mapping Landslide Susceptibility in the Moldavian Plain, Romania, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21563, https://doi.org/10.5194/egusphere-egu26-21563, 2026.
Landslides are commonly investigated in mountainous regions characterized by steep slopes. In contrast, the low-plateau region of the French Vexin (Paris Basin) is shaped by slopes resulting from ancient low-energy mass movements. The objective of this study is to describe the geometry and outcrops of an ancient landslide in order to obtain data to geologically characterize its dynamics and processes. In the French Vexin area, valleys are incised into a limestone plateau whose multilayered stratigraphy - comprising coarse limestone, fine sand and clay - controls the water table position. This water table can induce seepage erosion within the sand layers beneath the limestone layers and can be considered as a predisposing factor. This leads to their fragmentation (rotational blocks) and/or their progressive dipping (i.e. cambering) towards the valley bottoms to adapt to the topography subjected to gravitational constraints.
Recent studies conducted in a similar geological setting in the Champagne vineyards in France have improved our understanding of the links between these mass movements, substrate properties and hydrogeological conditions. However, the French Vexin region exhibits distinctive characteristics: the upper limestone layer is particularly thick and densely fractured, resulting in slope shapes that have never been studied before.
A representative site in Blamécourt (Magny-en-Vexin, Val d’Oise) was investigated to develop methodology for characterizing slope processes and their geological context. The area includes three disused quarries, multiple outcrops and a complex morphology. Field observations, high-resolution LiDAR, GIS mapping and electrical geophysical data were combined to analyse this complex landslide. Detailed morphological studies and characterization of geological structures in quarries beneath the plateau have revealed the state of the rock without the influence of the valley. The limestone blocks are fractured in two directions of tectonic origin, corresponding to the regional structural directions. From the plateau edge, a third structural trend aligned with the valley orientation is observed. These three structural directions persist downslope to the base of the slope, as confirmed by field observations and structural analysis. The limestone blocks covering the slope have therefore been affected by gravitational movements, whose structural boundaries result from the combined influence of inherited faults and newly formed structures.
How to cite: Vandecapelle, G., Robion, P., Antoine, R., Souloumiac, P., Finco, C., Lacquement, F., Leturmy, P., Betard, F., and Frizon-de-Lamotte, D.: Multi-methodology characterisation of low energy landslide : Example of Blamécourt (Vexin region, France), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13251, https://doi.org/10.5194/egusphere-egu26-13251, 2026.
Landslides are among the most destructive natural hazards in Türkiye, where high susceptibility is linked to active tectonic structures, steep topography, and complex climatic conditions. For instance, the Eastern Anatolian Region is recognized as a high-risk zone regarding seismicity and vast mass movements; therefore, reliable predictive tools for hazard mitigation are needed. Although several studies have already applied Machine Learning (ML) methodology for Landslide Susceptibility Mapping (LSM) problems in Türkiye, no systematic comparative evaluation of different modelling hierarchies has been performed so far for this particular area in a tectonically complex environment.
This study attempts to fill this gap by developing and rigorously comparing three disparate modeling methods: a statistical baseline, Logistic Regression (LR); an ensemble, Random Forest (RF); and a state-of-the-art deep learning method, Convolutional Neural Networks (CNN). The study was conducted using a landslide inventory and twelve landslide conditioning factor layers, including topographic data: DEM, Slope, Curvature, TWI; geological data: Lithology and Distance to Fault; environmental data: NDVI and Land Cover.
The core methodology embraced a systematic optimization of dataset splitting, whereby model performance was compared across different test/train ratios in order to identify the most stable and accurate data partition. Results are presented using key statistical metrics, including Accuracy and the Area Under the Receiver Operating Characteristic Curve (AUC-ROC), for LR, RF, and CNN. The best-performing model and its corresponding optimal test/train ratio were used to generate the final high-resolution LSM map for the Muş-Bingöl area. This forms a scientifically validated tool that can be used for regional land-use planning and risk management.
How to cite: Erdoğan, N. and Akgün, H.: Landslide susceptibility mapping of the Muş-Bingöl region: a comparative analysis and optimization of machine learning models, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-498, https://doi.org/10.5194/egusphere-egu26-498, 2026.
How fluid impact frictional slip is a central question in various geological settings, from tectonic faults to friction at the base of glaciers. In this work, we study the impact of fluid infiltration on the creep dynamics of the shear zone located at the base of a densely monitored landslide in Western Norway. In Åknes, approximately 50 million cubic meter of rock mass continuously creeps over a shear zone made of rock fragments, with seasonal accelerations that strongly correlate with rainfall. In this natural laboratory for fluid-induced frictional creep, unprecedented monitoring equipment reveals low fluid pressure across the shear zone, thereby challenging the conventional theory of fluid-driven instability in landslides. Here, we show that a generic micromechanical model can disentangle the effects of fluid flow from those of fluid pressure, and demonstrate that seepage forces applied by channelized flow along the shear zone are the main driver of creep accelerations. We conclude by discussing the significance of seepage forces, the implications for hazard mitigation and the broader applicability of our model to various geological contexts governed by friction across saturated shear zones.
How to cite: Barras, F., Aspaas, A., Aharonov, E., and Renard, F.: Unveiling the role of seepage forces in the acceleration of frictional creep in fluid-saturated shear zone, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17167, https://doi.org/10.5194/egusphere-egu26-17167, 2026.
Himalayan slopes are highly exposed to landslides, primarily triggered by earthquakes and monsoon precipitation. Satellite methods offer unrivalled spatial coverage of surface displacements on a weekly scale. However, they do not directly provide details of deformation at depth, nor do they offer sufficient temporal resolution to elucidate the continuity or intermittent nature of the landslide deformation during phases of heavy rainfall, strong rise in the water table or during intermediate seismic shaking. To address these issues in the context of the ANR/FNR project "SLIDE", we have recently deployed in late October 2025 a geophysical network at the level of one active, km-scale cultivated landslide in Nepal consisting in 16 co-located seismic and GNSS stations and one metereological station.
In this presentation, we will present the practical aspects of deploying and maintaining these instruments in remote Himalayan terrain. Each system required specific installation techniques and careful site selection to ensure stable measurements and long-term performance. Field operations were challenged by difficult access, variable road conditions, limited power availability, and unpredictable weather. Beyond technical challenges, community engagement is essential and close collaboration with local residents guided several site choices. We will also show the preliminary analysis of seismic, GNSS and meteorological data over the first 6 months of operation, which will be applied in the next three years to derive temporal and spatial changes of the landslide properties.
How to cite: Jaspard, M., Lave, J., Sitaula, B., Barrière, J., Gajurel, A., and Paudel, T. and the Team Slide: Multi-instrument geophysical monitoring of a km-scale slow-moving landslide in Nepal: Technical insights and preliminary results, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10241, https://doi.org/10.5194/egusphere-egu26-10241, 2026.
Rainfall-induced landslides are among the most widespread and destructive natural hazards, yet their physical reconstruction has rarely been explored beyond local or regional scales. We present a simplified slope-stability framework driven entirely by globally available rainfall, soil, and topographic datasets, and demonstrate its ability to reproduce thousands of rainfall-triggered landslides documented in the Global Landslide Catalog (GLC).By avoiding computationally intensive hydrological simulations while retaining physical interpretability, the proposed approach enables large-scale reconstruction of rainfall-induced slope failures across diverse environmental settings. Sensitivity analyses indicate that slope geometry and rainfall forcing primarily control proximity to failure and its timing, whereas soil bulk density exerts a disproportionate influence on model uncertainty due to its structural role in both mechanical resistance and hydrological response.Model performance is strongest in tropical and temperate regions, while reduced skill is observed in arid and cold climates, where failures tend to be conservatively predicted, favouring early-warning applications. Under scenarios characterised by intensified extreme rainfall, the framework suggests an overall increase in global slope instability. These results demonstrate the feasibility of reconstructing rainfall-induced landslides at the global scale using simplified physical representations, and highlight key directions for further improvement, including vegetation effects, subsurface heterogeneity, and hydrological process representation.
How to cite: Xia, Y. and Zhang, K.: Reconstructing rainfall-induced landslides at the global scale, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19557, https://doi.org/10.5194/egusphere-egu26-19557, 2026.
An increasing population in mountainous regions, where gentle and stable topography is scarce, drives residents to settle on steep slopes. These slopes are particularly prone to shallow landslides, which involve the displacement of the upper soil layers and are more easily triggered by rainfall. Therefore, accurate landslide hazard models are needed to safeguard populations.
These models typically include spatial data, such as soil thickness and rainfall. However, the lack of detailed inputs often means that models operate at coarse scales, which can mask local variability and potentially underestimate hazard levels. To address this gap, our research question is whether simulations of shallow landslides can be improved by enhancing the spatial resolution of two critical variables derived from coarse satellite data: (i) soil thickness determining the volume of material available for sliding, and (ii) rainfall controlling soil saturation and pore-water pressure dynamics.
To demonstrate the scalability and applicability of the method to other regions prone to landslides, we tested this approach in La Estrella, Colombia, a municipality with a long history of landslides and rapid population growth on steep slopes. For soil thickness, we applied a geomorphological model that relates soil depth to slope angle and distance to the drainage network. We validated the estimates against borehole measurements, finding strong agreement at three of five test sites. For rainfall, we integrated CHIRPS with local rain-gauge data, using spatial interpolation and regression-based downscaling to produce high-resolution rainfall fields. The downscaling model was then evaluated using statistical metrics, including the Pearson correlation coefficient (r), bias, and Nash–Sutcliffe efficiency (NSE).
In the next step, we will feed these two outputs into a Landlab shallow landslide probability model that couples hydrological response with soil mechanical stability. This will allow us to quantify the influence of input resolution on predicted landslide probability patterns.
How to cite: Cortes, P., Vega, J., Reinecke, R., and Ozturk, U.: Beyond Coarse Data: Soil Thickness and Rainfall forLandslide Hazard Modelling, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2976, https://doi.org/10.5194/egusphere-egu26-2976, 2026.
Under the combined effects of Tibetan Plateau uplift and global climate warming, the transition zone between the northeastern Tibetan Plateau and the Loess Plateau has become one of the most landslide-prone regions worldwide. Intense tectonics, abundant material supply, and densely developed faults produce landslides with large volumes, multi-stage evolution, and complex failure mechanisms, posing severe threats to infrastructure and human safety. However, progressive deformation processes and multi-scale controls remain poorly understood.
This study investigates the Lade–Lijiaxia landslide using an integrated “space–air–ground–subsurface” framework. Field investigations, systematic mapping of cracks and rupture surfaces, high-resolution remote sensing, SBAS-InSAR monitoring (140 SAR images), XRD mineralogical analysis, and SEM observations are combined to elucidate the landslide’s structural features, time-dependent deformation, and material basis.
Results indicate: (1) The landslide’s spatial distribution, boundaries, and internal structure are strongly controlled by regional tectonics. It develops along tectonically weakened zones, with the main sliding direction aligned with dominant lineaments. The landslide comprises a distinct sliding block and a creeping block (~1.5 × 10⁸ m³), representing a tectonically controlled progressive failure mode; (2) Crack and rupture surface analysis shows dominant crack orientations of ~30° and 125°, and rupture dip directions of 130°, 310°, and 20°, reflecting rear scarp tension, internal creep, and sliding surface geometry; (3) SBAS-InSAR indicates slow deformation, with the creeping block reaching ~170 mm/yr, accelerating seasonally during summer–autumn and warm spring due to rainfall and freeze–thaw cycles; (4) XRD reveals vertical heterogeneity: clay content is ~22% in the upper Quaternary deposits and ~38% in underlying Miocene mudstone, dominated by illite. SEM shows localized clay enrichment, fragmented microstructures, and well-developed pores, providing microstructural evidence for long-term creep and strength reduction.
Overall, long-term deformation is primarily controlled by deep-seated tectonics and lithology, while shallow deformation is triggered by seasonal hydrothermal processes. These results improve understanding of progressive failure and creep evolution of large landslides at the northeastern Tibetan Plateau margin and provide insights for hazard assessment and long-term monitoring in the plateau–loess transition zone.
Map of Location Study Area
Geological Map of Study Area
How to cite: Jingqi, Z. and Genhou, W.: Deformation Characteristics and Mechanisms of a Large Landslide at the Northeastern Margin of the Tibetan Plateau Based on Multi-source Data Integration: A Case Study of the Lade–Lijiaxia Landslide, Qinghai Province, China, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11055, https://doi.org/10.5194/egusphere-egu26-11055, 2026.
