A growing number of scientists with diverse backgrounds are studying debris flows. The difficulties in measuring parameters related to their initiation and propagation have progressively prompted research into a wide variety of laboratory experiments and monitoring studies. However, there is a need of improving the quality of instrumental observations that would provide knowledge for more accurate modelling and hazard maps. Nowadays, the combination of distributed sensor networks and remote sensing techniques represents a unique opportunity to gather direct observations of debris flows to better constrain their physical properties. At the same time, computer-aided simulations of physical processes, hazard assessment, and mitigation design are undergoing a revolution due to the widespread adoption of AI and data-driven numerical models. Not only do these developments mark an exciting era for advancing our understanding of complex natural hazards, but they also require researchers from diverse disciplines to collaborate in order to unlock their full potential.
Scientists working in the field of debris flows are invited to present their recent advancements. In addition, contributions from practitioners and decision makers are also welcome. Topics of the session include field studies and documentation, mechanics of debris-flow initiation and propagation, laboratory experiments, modelling, monitoring, impacts of climate change on debris-flow activity, hazard and risk assessment and mapping, early warning, and alarm systems.
Debris flow is one of the most destructive natural hazards which is typically distinguished by high solid content and significant interactions between the particles and interstitial fluid. This study focuses on fundamental inter-particle interaction pattern and the underlying mechanism in typical debris flows over bumpy bed with varying bed inclinations. Internal dynamic parameters of the debris flows are obtained based on the refractive index matching (RIM) technique and non-invasive sensor networks. It is found that the granular interaction pattern is vertically stratified with the near-bottom particles intensely colliding with each other in a gas-like state, while the near-surface particles sliding collectively in a solid-like state. Based on the observed flow behavior and measured parameters, a multi-state constitutive model is proposed, which incorporates the kinetic theory for the collisional stress and a newly developed frictional stress model. This constitutive model improves the overall granular stress modelling accuracy for the debris flow with highly heterogeneous flow structures.
How to cite: Sun, Y., Fang, H., and Liu, Q.: Debris Flows over Bumpy Bed: Experiments and the Constitutive Modelling, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8848, https://doi.org/10.5194/egusphere-egu26-8848, 2026.
The interaction between dense particle-liquid flows and obstacles plays a central role in debris-flow impact processes and the performance of protective structures, yet the associated flow regimes and impact loading characteristics remain insufficiently resolved by laboratory experiments. In this study, inclined dense particle-liquid flow impacts on a cylindrical obstacle are investigated using a laboratory-scale experimental system that combines synchronized multi-view high-speed imaging with direct force measurements. The experimental setup enables simultaneous observation of flow kinematics, particle-fluid distribution patterns, and load time histories during flow–structure interaction. Experiments are conducted over a range of slope angles and solid volume fractions representative of dense debris-flow conditions. The multi-view imaging configuration allows identification of three-dimensional flow features, including upstream shock formation, particle circulation zones, flow expansion, and localized particle-depleted regions near the obstacle.
Results indicate that the interaction process exhibits distinct flow regimes primarily controlled by solid volume fraction and the spatial structure of the upstream shock. At higher solid volume fractions (φ = 55%), the incoming flow develops a compact, high-shear shock front characterized by intense particle collisions and rapid momentum dissipation. This flow configuration promotes the formation of a stable upstream accumulation, accompanied by pronounced particle clustering and particle-liquid separation, and supports a clear transition from short-duration dynamic impact to a sustained reflection wave regime. In contrast, at lower solid volume fractions (φ = 45%), the shock structure is more diffuse and is frequently disrupted by persistent vertical jets and fragmented particle impacts. In this case, particle–liquid separation is weak or short-lived, and the loading remains strongly non-stationary without the establishment of a stable reflection structure.
These experimental observations provide new insights into flow-regime-dependent impact loading mechanisms of dense particle-liquid flows and offer a physical basis for improving debris-flow impact modelling and the design of protective structures.
How to cite: Yu, W., Liu, Q., and Wang, X.: Flow regimes and impact loading characteristics of dense particle–liquid flows interacting with a cylindrical obstacle, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16385, https://doi.org/10.5194/egusphere-egu26-16385, 2026.
Debris flows are rapid, high-energy mixtures of water and sediments that pose a severe threat to mountainous regions, often occurring with little warning and causing substantial loss of life and infrastructure damage. The design of effective structural countermeasures is therefore essential to mitigate their destructive potential. Open-type SABO dams are widely adopted to reduce the impact of stony debris flows by intercepting coarse material, while drainage systems enhance energy dissipation by removing part of the water content from the flowing mixture. This study investigates the novel approach of combining open-type SABO dams with drainage systems to enhance debris flow mitigation.
The complex multiphase physics governing debris flows severely limit the accurate reproduction of such events in both numerical simulations and laboratory experiments (Iverson, 1997), complicating the assessment and optimization of countermeasure performance. Although scaling effects introduce unavoidable uncertainty when scaling laboratory results to real-world environments, physical modelling remains a valuable tool for systematically testing alternative design configurations and identifying governing mechanisms relevant to preliminary engineering applications.
A total of 145 small-scale laboratory tests have been conducted, varying triggering discharges, channel slopes, SABO dam configurations (number and spacing of steel trestles), and drainage conditions. Starting from the framework proposed at EGU 2019 by Salandin and Lanzoni, the present study investigated two triggering discharges and two channel slopes, by including a SABO dam of varying numbers of steel trestles with different spacings between them, and multiple drainage configurations, allowing controlled variation of the degree of dewatering of the debris flow body. The spatio-temporal evolution of the sediment–water mixture surface was monitored using four ultrasonic sensors, water level was measured by a submersible pressure transducer, and debris-flow mass was quantified using a load cell.
The SABO dam efficiency is assessed in terms of energy dissipation, inferred from temporal changes in debris deceleration over time and from accumulation height upstream of the combined system. Results demonstrate that adding a drainage system significantly enhances the SABO dam energy dissipation capacity. This integration allows for larger trestle spacing while maintaining effective debris flow control. Moreover, under both drained and undrained conditions, our findings suggest optimal trestle openings that differ from current literature recommendations, highlighting the potential of integrated SABO–drainage systems to improve debris flow mitigation strategies.
How to cite: Giaretta, P., Lanzoni, S., and Salandin, P.: Experimental Assessment of Combined SABO Dam and Drainage System for Debris Flow Mitigation, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11574, https://doi.org/10.5194/egusphere-egu26-11574, 2026.
References
- Iverson, R. M., Reid, M. E., & LaHusen, R. G. (1997). Debris-flow mobilization from landslides. Annual Review of Earth and Planetary Sciences, 25, 85-138.
- Steers, L. J., Beddoe, R. A., & Take, W. A. (2024). Propagation velocity of landslide-induced liquefaction and entrainment of overridden loose, saturated sediments. Engineering Geology, 334, 107523.
- Sakai, N., Ishizawa, T., & Danjo, T. (2025). Experimental Research on Rain-Induced Landslide Mechanism Using Large-Scale Rainfall Experimental Facility: Findings and Challenges. In B. Abolmasov et al. (Eds.), Progress in Landslide Research and Technology (Vol. 3, Issue 2). Springer.
How to cite: Sakai, N.: Mechanisms of Rapid Entrainment and Acceleration in Landslide-Initiated Debris Flows: The Role of Static Liquefaction, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10193, https://doi.org/10.5194/egusphere-egu26-10193, 2026.
Entrainment significantly modifies the dynamics and runout of debris flows, yet the combined influence of water content, bed properties, and particle-scale characteristics remains poorly constrained. Building on our previous flume-based framework, this study integrates mesoscale flow experiments with micromechanical analysis of debris materials including grain shape, roughness, and fragmentation using X-ray micro-CT imaging. A series of controlled flume experiments were performed using erodible sand beds (4 cm thick) prepared via mist pluviation to minimize segregation. Sixteen flow tests were conducted across a range of volumetric water contents (20–50%), capturing high-speed flow kinematics, entrainment depth, and deposit morphology. Complementary micro-CT imaging of fluvial and colluvial grains enabled quantification of particle shape (sphericity, aspect ratio, surface irregularity) and its potential role in erosion thresholds.
Results show distinct morphological transitions with increasing water content. At low w/c (20–24%), flows exhibited limited mobility and formed short, conical lobes with minimal scouring. Around intermediate w/c (~28%), reduced bed dilatancy and moderate pore pressure generated thicker but shorter deposits, indicating partial suppression of entrainment. At higher w/c (30–50%), enhanced lubrication and basal shear promoted deeper scouring, larger entrainment volumes, and substantially longer runouts with wide, flattened deposits. A parabolic relationship emerged between bed water content and entrainment rate, highlighting the nonlinear coupling between fluid fraction, granular collisions, and bed resistance. Deposits exhibited poor sorting and layered structures similar to natural debris flows, confirming dynamic similarity. Preliminary micro-CT analyses suggest that more angular and elongated grains exhibit larger contact stresses and higher resistance to dislodgement, whereas smoother grains mobilize earlier potentially explaining material-dependent variability in erosion observed across tests. Ongoing work aims to link shape descriptors directly with measured entrainment rates. This combined experimental–micromechanical approach advances our understanding of debris-flow erosion by bridging particle-scale processes and mesoscale dynamics. The results provide new insights for improving entrainment parameterization in debris-flow models and for developing more reliable runout predictions in geophysical hazard assessments.
How to cite: Pandey, N. K. and Satyam, N.: Micro-mechanical controls on entrainment and depositional patterns in wet granular debris flows: Insights from flume experiments and particle-scale characterization, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-1042, https://doi.org/10.5194/egusphere-egu26-1042, 2026.
Deflection barriers are a critical mitigation measure for redirecting debris flows away from high-risk zones. However, their design is complicated by the oblique shock dynamics that occur upon impact, leading to complex runup and loading patterns that are not fully characterized. To address this gap, this study develops and validates an analytical model for predicting the runup height and impact loads generated by oblique debris-flow shocks. The theoretical model, derived from momentum conservation principles, explicitly links the runup height and peak impact load to the incoming flow conditions, notably the Froude number. A series of scaled flume experiments were designed to test the model's validity. By systematically varying the channel slope and gate opening height, we generated a range of supercritical flows to quantify the influence of incoming kinetic energy on shock phenomena. Results demonstrate the theoretical predictions show excellent agreement with experimental measurements across all tested scenarios. Furthermore, analysis confirms that the normal shock condition serves as a conservative upper bound for oblique shock impacts, providing a valuable simplified criterion for preliminary design. Importantly, we identify a key limitation: the model's accuracy decreases in flows where pronounced dead zones form downstream of the barrier, as the assumed shock geometry no longer holds.
How to cite: Li, X., Song, D., Liu, J., and Liu, Y.: Oblique Runup and Impact Load of Debris Flow on Deflection Barriers, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8652, https://doi.org/10.5194/egusphere-egu26-8652, 2026.
Debris flows are among the most devastating natural hazards in mountainous areas, posing a significant threat to infrastructure, particularly bridges that are crucial for regional connectivity. Climate change-induced increases in intense rainfall events have amplified both the frequency and magnitude of these sediment-laden flows. Consequently, bridge structures face growing exposure to extreme loading conditions. Bridge piers situated within active riverbeds are especially vulnerable, as debris flows generate highly impulsive forces that can surpass those accounted for in traditional design methodologies.
A reliable estimation of debris-flow-induced thrust on bridge piers is essential to improve existing design methodologies and to ensure resilience of infrastructure in debris-flow-prone environments.
To address this critical need, an innovative experimental apparatus has been developed to investigate the impact of stony debris flows under controlled laboratory conditions. This setup reproduces both the initiation and propagation phases of debris flows, enabling a more comprehensive analysis of their dynamics and impact forces.
Experiments were conducted in a tilting flume measuring 3 m in length and 0.3 m in width. The flume features an erodible granular bed, allowing debris flows to initiate and evolve through bed erosion, closely mimicking the mechanisms observed in natural settings. This design significantly enhances the realism of the experimental simulations.
Within this framework, particular attention is devoted to the investigation of debris flows propagating under subcritical flow conditions, a regime that has received comparatively limited attention in experimental studies but may be relevant for specific geomorphological and hydraulic contexts.
Debris flows are initiated by the controlled release of a predetermined water discharge, which induces sediment mobilization and subsequent flow development along the channel. The experimental setup is instrumented with pressure transducers, sonar sensors, and load cells to measure flow depth, velocity, and impact forces exerted on model bridge piers of varying geometries and dimensions.
A dimensionless analysis carried out to characterize the flow regime reproduced in the laboratory indicates that the experimental conditions successfully reproduce a stony debris flow in terms of flow composition and propagation dynamics.
Following the preliminary comparison between measured impact forces and those predicted by classical hydrostatic and hydrodynamic theoretical models presented at EGU 2025, an integrated hydraulic model that combines the two approaches is proposed. This model is used to interpret a set of experimental results that has now more than doubled in size. Model parameters are calibrated using an Orthogonal Distance Regression (ODR) procedure, which allows for the joint consideration of uncertainties in both experimental observations and theoretical predictions.
Overall, the findings provide novel experimental insights into debris-flow impact processes under subcritical conditions and demonstrate the capability of integrated modeling approaches in predicting debris-flow-induced forces on bridge piers. These results contribute to the validation and refinement of existing design models, while supporting the development of more reliable, physically based design criteria for bridges exposed to debris-flow hazards.
How to cite: Cao, A., Giaretta, P., and Salandin, P.: Stony Debris Flows and Impact Forces on Bridge Piers: Insights from small-scale Laboratory Experiments, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20011, https://doi.org/10.5194/egusphere-egu26-20011, 2026.
The impact of debris flows against a rigid obstacle defines the critical loading scenario for structures in the track of a debris flow. The total force imparted by the debris flow on these structures can be approximated using a linear momentum approach. However, this method cannot be used to define the vertical pressure distribution, which is necessary to capture the position of the resultant force. The height of the resultant force is essential in defining the expected failure mechanism (i.e. sliding or overturning) of a structure on impact. Predicting the correct failure mechanism is critical for hazard mapping and emergency response, where estimation of phenomena like structural translation via sliding is needed for appropriate resource deployment.
Although total impact force has been widely investigated, comparatively few studies have reported spatially resolved pressure distributions for debris flow-barrier impacts. Typical empirical methods for the prediction of pressure distributions apply a constant dynamic pressure summed with a linear static pressure. This approach assumes velocity conservation and a circular flow path. However, observations from laboratory studies indicate that this may not always be true. To explore this, laboratory tests were conducted using a dense array of pressure sensors installed in a rigid barrier, impacted by varied releases of water and water-sediment mixtures. These experiments offer pressure measurements at high spatial and temporal resolution, correlated with visual high-speed camera data used to define the flow path and velocity field within the control volume.
Flow paths with variable velocity and curvature were observed for a range of material compositions. Based on these observations, a novel analytical model is proposed to predict pressure distributions using generalized approximations of the rate-of-change of flow properties and path curvature. This approach provides equivalent total force predictions to traditional linear momentum models but allows for direct determination of the position of the resultant force.
How to cite: Hirsch, E., Take, A., and Mulligan, R. P.: Turning the corner: How does debris flow path curvature affect the pressure distribution during impact on a rigid barrier?, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14828, https://doi.org/10.5194/egusphere-egu26-14828, 2026.
Excessive rainfall in mountain catchments may trigger landslides or destabilize saturated streambeds. The resulting debris flow may propagate along the drainage network and reach urbanized areas, causing damage and loss of life. To ensure an efficient delimitation of such risk-prone areas, numerical models are often adopted to compute the time evolution of the flow. To this end, we apply a custom made monophasic Shallow Water based finite volume solver to the mass-flow like events occurred in Cervinara (Southern Italy) on 15-16 December 1999 which caused 6 fatalities and serious damage to buildings and structures. Several landslides were triggered that day and one in particular was able to propagate downstream, reaching the urbanized areas of Ioffredo and Cervinara. The event was comparatively simulated using two different rheological models, i.e. Voellmy and O’Brien, implemented inside the solver in order to assess which of them was able to better replicate the main characteristics of the flow. Validation was performed considering the extension of the inundated areas in the town and maximum flow velocity recorded on the field previously available. The adoption of an unstructured grid allowed both the representation of the urbanized areas by introducing the buildings as holes inside the mesh and the computation of the forces exerted on the buildings by the flow.
How to cite: Bonomelli, R. and Pilotti, M.: Debris flow numerical simulation using multiple kinds of rheological models: a case study in Cervinara (southern Italy), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10728, https://doi.org/10.5194/egusphere-egu26-10728, 2026.
Ongoing improvements in monitoring are increasingly documenting the presence of quasi-regular trains of surge waves in debris flows. These phenomena exacerbate hazards associated with these events, since they can grow to reach depths and discharges greater than anywhere else in the flow. Using data from Illgraben, Switzerland, we track the development of these surges from small undulations on the free surface to waves with amplitudes of a metre or more. From this, we argue that the waves arise from a flow instability analogous to the classical 'roll wave' instability that occurs in flows of turbulent water. A complementary theoretical model is presented, which uses a basal drag parametrisation informed by the observational data. When initiated with measured upstream fluxes, the model develops waves that mature from small perturbations to large waves that are in excellent agreement with the field data. The underlying mathematics that governs the instability can be used to explain why waves are observed in some flows, but not others. Contributing factors include the bulk flow discharge and the shape of the channel.
How to cite: Langham, J., Aaron, J., Spielmann, R., Hirschberg, J., McArdell, B., Boss, S., Johnson, C., and Gray, N.: Explaining the formation of debris flow surges, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-22083, https://doi.org/10.5194/egusphere-egu26-22083, 2026.
Erosion and entrainment are key processes that modulate debris-flow mobility. However, the conditions under which erosive debris flows accelerate and attain longer runout or decelerate and come to rest earlier remain insufficiently understood. The Pudasaini and Krautblatter (2021) landslide mobility model attributes these divergent behaviors to inertial contrasts between the moving mass and the erodible bed, suggesting that the incorporation of inertially weaker, neutral, or stronger material can enhance, maintain, or reduce the flow mobility, respectively. Here, we use flume experiments and surface-based measurements to investigate how bed inertia influences the velocity, erosion, and runout of dry, single-phase debris flows by systematically varying solid densities. A quartz slide of constant solid density is released over erodible beds with lower, equal, and higher densities representing inertially weak, neutral, and strong scenarios. Our results reveal consistent and repeatable patterns. Debris flows over low-density beds exhibit higher apparent mean erosion rates, increased flow-front velocities before deposition, and longer runout than in the inertially neutral scenario. In contrast, debris-flow evolution over equal- and high-density beds is nearly identical, characterized by lower frontal velocities, reduced erosion, and shorter, thicker deposits. These findings indicate that the entrainment of the low-density material enhances debris-flow mobility relative to the inertially neutral scenario, whereas the incorporation of high-density material does not lead to the expected mobility reduction. This asymmetric response suggests that solid density alone does not fully explain the observed mobility behavior under the experimental conditions considered here. Additional influences related to particle shape and internal friction are likely involved, too. The low-density bed combines more rounded particles with a low internal friction angle facilitating entrainment, whereas the equal- and high-density beds comprise more angular grains with similar and higher internal friction angles, which may lead to comparable resistance to erosion despite their contrasting densities. Ongoing work focuses on resolving processes at the flow-bed interface to capture grain-scale dynamics at depth and resolve temporal variations in erosion intensity, which may help to identify subtle differences between the inertial scenarios that are not detectable using surface-based measurements alone.
How to cite: Wetterauer, K., Müller, S., Pudasaini, S. P., Krautblatter, M., and Baselt, I.: Exploring the effects of bed inertia on debris-flow mobility, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14068, https://doi.org/10.5194/egusphere-egu26-14068, 2026.
Runoff significantly influences the propagation of debris flows by transferring mass and momentum. The hydrodynamics of runoff, which are closely linked to contributions from sub-basins, determine the extent of this influence. In this paper, a cascade model is utilized to quantitatively analyze the contribution of sub-basins to runoff and, subsequently, to the propagation of debris flows, using the 2020 Meilong debris flow event as a case study. First, the propagation of debris flows, characterized by their high mobility and sediment entrainment, is well reproduced. This analysis examines how each sub-basin’s generated runoff contributes to debris flow propagation, revealing that both the area and location of a sub-basin are significant factors. Additionally, a series of scenarios with variations in basin features and debris flow types are simulated. The results suggest that as the basin area and internal relief decrease, the contribution of sub-basins to runoff-and consequently to debris flow propagation-also diminishes, aligning with trends observed in field data. Furthermore, the propagation of debris flows with lower viscosity is more effectively facilitated by runoff from sub-basins due to enhanced mixture between runoff and debris flow. This study provides significant insights into the propagation of debris flows, thereby supporting the assessment of this debris flow type.
How to cite: Liu, W., Tan, Z., Jian, J., and He, S.: How sub-basins runoff contribute to debris flow propagation at a basin scale? A numerical study, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9788, https://doi.org/10.5194/egusphere-egu26-9788, 2026.
Water related rapid mass movement such as landslides, debris flows, and slush flows are expected to become more frequent as the climate continues to change. Hazard zoning in urban areas is necessary to save lives and to prevent damages to existing and future buildings. Previous events have shown that the runout of such events can be strongly influenced by the buildings and infrastructure in the runout path. Buildings can stop or reduce, but also redirect the movement of the flow. Therefore, one of the challenges for hazard zoning in urban areas is determining how to take existing buildings into consideration when assessing runout of rapid mass movements.
In this work, we have explored to which degree RAMMS::Debrisflow can be utilized to estimate the effect that existing buildings have on the runout of landslides, debris flows and slush flows. We aim at developing a general procedure that can be applied for hazard mapping at a large scale. We have studied nine previous rapid mass movement events where buildings have affected the runout. For each event, the runout has been back-calculated in RAMMS::Debrisflow 1) without taking buildings into consideration, 2) using increased friction for areas with buildings, and 3) using “obstacle/dam”-mode for areas with buildings.
Our study shows that modelling that takes buildings into account more accurately represent the runout of the different historic rapid mass movements than modelling without taking the effect of buildings into account. We therefore recommend using such an approach when assessing the hazard of rapid mass movements in urban areas.
We propose to classify the robustness of buildings to account for the varying effect that different types of buildings have on runout. This can be accomplished on a larger scale by using public datasets that include attributes for buildings, such as the Norwegian FKB-dataset. For example, a large, robust concrete building might fully stop runout and is best represented in RAMMS::Debrisflow as an obstacle. Wooden residential houses and other buildings with moderate robustness might retard, but not fully stop runout and are best represented using areas of higher friction. Small and fragile buildings such as sheds or small garages are expected to have negligible effect on runout, and we suggest to not take these into consideration when modelling runout.
Predictably, the effect that buildings have on runout is depending on the intensity of the flow and construction method of the building. There will therefore still be a need for expert judgement when assessing resistance of buildings to the mass flow and in interpretation of results on a detailed scale. Our proposed method can be viewed as a first step towards such an assessment. By utilizing the large building datasets (such as the FKB-dataset), the practitioner can make a quick and practical substitute for a tedious structural assessment of each building, thus increasing efficiency for the hazard engineer.
How to cite: Eidsvåg, E., Nordbrøden, H., Breien, H., and Kronholm, K.: Modelling the effect of buildings on water related rapid mass movements, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4709, https://doi.org/10.5194/egusphere-egu26-4709, 2026.
Debris flows are predominantly rainfall-induced phenomena in which progressive
degradation of soil shear strength plays a critical role in flow initiation and mobility. This
study presents an experimental investigation into the moisture-dependent reduction of shear
strength parameters of debris material collected from an active debris-flow site in the Sidhra
region of Jammu, India. The site lies in the Outer Himalayan belt and is underlain by
colluvial debris, weakly lithified Siwalik sandstones, siltstones, and mudstones, which readily
disintegrate into fine-grained, clay-rich soils during intense rainfall.
Systematic laboratory testing was performed on samples obtained from both the source and
deposition zones, with direct shear tests conducted at moisture contents of 0%, 20%, 30%,
and 50% to simulate dry to highly saturated field conditions. The results reveal a pronounced
reduction in both cohesion and internal friction angle with increasing moisture content,
indicating a clear transition from frictional-cohesive behaviour under dry conditions to
predominantly frictional and flow-like behaviour at high degrees of saturation. Beyond a
critical moisture threshold, cohesion becomes negligible, leading to a drastic reduction in
shear resistance and a strong increase in susceptibility to rainfall-triggered debris-flow
initiation.
The experimental results are further integrated into numerical simulations using the Rapid
Mass Movement Software (RAMMS). The influence of the Voellmy-Salm friction
parameters: the dry Coulomb friction coefficient (μ) and the turbulent friction coefficient (ξ),
is examined. These parameters, calibrated using laboratory-derived shear strength values,
significantly control simulated flow height, velocity, runout distance, and flow path. The
study highlights the importance of incorporating moisture-dependent shear-strength
degradation into debris-flow hazard assessments and demonstrates that realistic calibration of
RAMMS friction parameters is essential for reliable prediction of debris-flow dynamics.
How to cite: Jha, P. and Bhowmik, R.: Moisture-Induced Shear Strength Degradation of Debris Materials and Its Implications for Debris Flow Behaviour in a Himalayan Catchment, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7680, https://doi.org/10.5194/egusphere-egu26-7680, 2026.
Dense granular flow on rough slopes provides a simplified yet powerful analogue for gravity-driven natural hazards such as landslides and debris flows, in which particle shape is known to strongly influence the flow mobility and runout but remains difficult to parameterize. Using discrete element method simulations, we systematically investigate the effect of particle flatness (and elongation) on dense granular flows over rough inclined planes. Particles with identical volume but increasing flatness, from spherical to strongly flattened, are first considered. We follow the framework of Pouliquen’s flow rule [1] to identify the critical stopping conditions and then perform an analysis of steady uniform flows, which allows us to extract the velocity scaling with flow thickness and slope angle. We find that the velocity scaling for each particle shape collapses, but the corresponding mobility parameter exhibits a nontrivial dependence on the particle flatness. This shape dependence is characterized by an initial weak sensitivity near the spherical limit, a rapid mobility reduction at intermediate flatness, and a saturation regime for highly flattened particles. Microstructural analyses reveal that this behavior originates in shape-induced constraints on the particle kinematics, including suppressed particle rotation and the emergence of strong orientational ordering, with flat particles preferentially aligning parallel to the shear plane. Furthermore, comparing with recent results of elongated particles [2], we show that flat grains exhibit a characteristic bimodal distribution of preferred orientations, reflecting a distinct alignment mechanism under shear, which nonetheless leads to a comparable macroscopic reduction in mobility. Comparison with elongated particles also indicates that the effects of flatness and elongation may be unified by considering how the particle shape becomes different from perfect sphere. Indeed, when characterized by sphericity, the flow mobility data for both particle types collapse onto a unified trend. Future work will confirm whether this finding applies to other particle shapes. Our work provides a physically grounded route for incorporating particle shape effects into predictive models of landslides and debris flows.
References
[1] Pouliquen O. Scaling laws in granular flows down rough inclined planes. Physics of Fluids, 1999, 11(3): 542-548.
[2] Liu J, Jing L, Pähtz T, et al. Effects of particle elongation on dense granular flows down a rough inclined plane. Physical Review E, 2024, 110(4): 044902.
How to cite: Zhang, M. and Jing, L.: Non-spherical granular flow down a rough incline: understanding the role of particle flatness and elongation, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8797, https://doi.org/10.5194/egusphere-egu26-8797, 2026.
Regional-scale runout modelling is a critical component of landslide hazard assessment. The spatial prediction of debris-flow hazards over large regions requires the integration of source-area susceptibility with robust runout simulation. While various empirical and process-based models exist, there remains a need for flexible, cross-platform tools that integrate seamlessly with modern statistical and machine-learning workflows. We present runoutSIM, an open-source R package designed to facilitate data-driven regional runout modelling and source-area connectivity analysis.
By leveraging the R environment, commonly used for geoscientific computing and visualization, runoutSIM streamlines the transition from susceptibility mapping to runout distribution. The package implements a random-walk spreading algorithm to simulate potential runout paths, offering a statistical–physical framework to assess debris-flow spatial extent, velocity, and connectivity. Key features include the ability to estimate connectivity probability – the likelihood that a specific source area will impact downslope features of interest – and to adjust runout spatial probabilities and connectivity using spatial likelihoods from statistical and machine-learning predictions of source areas. This ensures that runout spatial and connectivity probabilities reflect the inherent variability in source-area initiation.
We demonstrate the application of runoutSIM through a case study in the central Andes of Chile, a region characterized by high-frequency debris-flow activity. The example couples machine-learning source-area prediction with optimization approaches, such as random grid search, to calibrate the runout model. The model is used to identify river-channel exposure and potential risks to water quality, highlighting the package’s utility for both spatial planning and local hazard mitigation. Overall, as a tool for applied landslide research, method development and teaching, runoutSIM aims to lower the barrier to accessing process-based models, enabling more comprehensive, source-to-impact hazard assessments. We anticipate that this open-source framework will support advances in quantitative geomorphic modelling and contribute to more reliable regional-scale debris flow risk management.
How to cite: Goetz, J. and Huang, J.: runoutSIM – An R package for regional debris-flow runout simulation and source-area connectivity, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8125, https://doi.org/10.5194/egusphere-egu26-8125, 2026.
Landslide runout and debris-flow inundation are crucial, yet often neglected, threats that can impact areas far beyond initial landslide sources. Understanding where mobile landslides initiate and how far they travel is essential to hazard and risk reduction worldwide, as mitigation strategies vary with landslide mobility. Moreover, debris flows can grow volumetrically as they travel, resulting in larger, faster flows with greater inundation. However, most landslide susceptibility maps focus on steep slopes and fail to address runout and inundation onto flatter ground, which typically encompasses more inhabitants and infrastructure. Given the widespread importance of runout and inundation, it is vital to have simple-to-use methods that rapidly map the effects of these mobile processes over large regions, especially in locations with limited geotechnical data.
We present an empirical approach for mapping areas susceptible to landslides and debris flows from their initiation to deposition. Using the publicly available USGS software package, Grfin Tools, we delineate landslide source areas, landslide runout, and debris-flow inundation zones within a DEM. Grfin is an acronym of Growth + flow + inundation and this computationally fast software uses simple, well-tested, and fully documented empirical models. Potential landslide runout is determined using angles of reach. Potential debris-flow inundation from volumetric growth is delineated using volume-area scaling relations based on worldwide observations, combined with a novel use of spatially integrated growth factors. These models require minimal input parameters and place an emphasis on regional geomorphic and topographic controls rather than specific material properties.
Using Grfin Tools, we illustrate our approach by mapping a spectrum of mass-movement mobility zones on three island states of the Federated States of Micronesia where landslide runout and debris-flow inundation onto flatter ground have resulted in fatalities. Future mobile events pose a deadly threat, yet previous landslide information is incomplete. To estimate the empirical model parameters needed to portray multiple mass-movement zones, we use satellite-derived landslide inventories combined with topographic thresholds obtained from 10-m resolution DEMs. Based on field observations of debris-flow deposits from 138 stream locations, our debris-flow inundation model incorporating spatially integrated growth has a prediction success of greater than 85%. Our methods using Grfin Tools can rapidly create preliminary regional assessments, provide multiple scenario assessments, or act as a screening tool to identify critical areas for further detailed studies across a wide variety of landscapes.
How to cite: Reid, M., Cerovski-Darriau, C., Brien, D., Leb, I., and Cyr, A.: Rapidly portraying landside runout and debris-flow inundation using simple, empirical methods, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8263, https://doi.org/10.5194/egusphere-egu26-8263, 2026.
Debris flows are one of the most common geomorphic processes in mountainous areas, and can form a great threat to local communities and infrastructure. Traditionally, mitigation efforts have focused on engineering solutions such as check dams or debris basins. Recently, focus has started to shift towards more nature-based solutions such as forest buffer zones, which require an understanding of interactions between debris flows and trees for their design. Some research into debris flow-forest interactions has been done using field data, aerial imagery or simplified physical experiments, however, quantitative knowledge of tree removal by debris flows is still lacking. This study aims to assess tree survival and removal by debris flows, and to identify controlling debris flow and vegetation properties.
We use multi-temporal, high resolution airborne laser scanning (ALS) data covering multiple debris-flow events over four different forested debris-flow fans in the Squamish-Lillooet region in British Columbia, Canada, to track sediment deposition and tree removal. Tree survival patterns are linked to tree and debris-flow characteristics (tree size, location and proximity to the next tree, and deposition and erosion depth, respectively) to gain insight into the interaction between debris flows and forests.
Preliminary results show that smaller trees have a higher chance of being removed by a debris flow, and that the chance of tree survival increases with distance from the fan apex and with higher tree density. Next steps include numerical simulations of debris-flow velocities to quantify the relationship between debris-flow impact forces and tree removal or survival. The results of this study will help identify optimal characteristics for resilient debris-flow forest buffer zones.
How to cite: van Etten, J., Mitchell, A., McDougall, S., Eichel, J., and de Haas, T.: Quantifying debris-flow – forest interactions using high-resolution LiDAR data in the Squamish – Lillooet region, British Columbia, Canada, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7167, https://doi.org/10.5194/egusphere-egu26-7167, 2026.
Debris flows are among the most destructive mass movement processes affecting the mountainous regions of the Indian subcontinent, particularly the Indian Himalayas and the Western Ghats. These terrains are highly susceptible to mass movements, with debris flows posing significant hazards. In recent years, India has experienced several catastrophic debris flow events, including the Dharali (5 August 2025), Chasoti (14 August 2025), and Ramban (19-20 April 2025) debris flows in the states of Uttarakhand, Himachal Pradesh, and Jammu & Kashmir. These events have underscored the growing severity of debris flow hazards in the Indian Himalayan Region (IHR).
Recent events indicate that intense rainfall is the primary triggering factor for debris flows; however, extensive entrainment along the transport zone significantly amplifies their destructive potential. Although debris flows generally follow pre-existing channels, the magnitude of damage is largely governed by the presence of vulnerable elements within the deposition zones. Long runout mass movement processes can travel considerable distances, during which entrainment, bulking, and phase transitions occur.
Given the increasing frequency and impact of debris flow events in the Indian subcontinent, comprehensive hazard assessment studies are urgently required. These should include the identification of initiation zones, estimation of source volumes, characterization of entrainment zones and materials, runout modelling, and integrated hazard assessment. While numerical simulation models are effective tools for back-analysis and future hazard prediction, their reliability depends on the accurate estimation of input parameters. The escalating debris flow activity across India highlights the need for focused research, systematic monitoring, and improved mitigation strategies to reduce future risks.
How to cite: Dash, R. K. and Kanungo, D. P.: Destructive Debris Flows in the Indian Himalayan Region: Insights from Recent Events, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15648, https://doi.org/10.5194/egusphere-egu26-15648, 2026.
Debris-flow fans form by repeated deposition of debris-flow sediments. Catchment lithology affects debris-flow grain-size distribution, and thereby rheology, erosive potential, and depositional morphology. We can therefore expect that lithology also influences debris-flow fan characteristics. Here, we determine how catchment lithology affects the surface morphology and sedimentology of debris-flow fans, and by extension their spatio-temporal evolution. We study nine fans along the eastern margin of northern Owens Valley, California, USA, originating from catchments with contrasting lithologies, and similar climate, tectonics, and geological history.
Results show that debris flows originating from catchments comprising magmatic rocks are rich in cobble- to boulder-sized grains. The coarse sediment along the flow fronts and margins minimizes lateral spreading of debris-flow lobes, forming distinct levees and thick depositional mounds. In contrast, debris flows originating from catchments dominated by sedimentary rocks are rich in relatively fine gravel. Their fine-grained levees and lobes lack strongly frictional margins, spread more easily, and form distinctly thinner and wider deposits. Debris flows originating from catchments with metamorphic lithologies show intermediate grain-size and depositional morphology.
These contrasts in debris-flow characteristics guide the morphology and spatio-temporal development of debris-flow fans. Fine-grained debris flows spread laterally and tend to fill topographical lows, whereas lateral spreading of coarser-grained flows is hampered, instigating a low tendency to fill topographic lows. The more efficient topographic compensation on fans formed by fine-grained debris flows causes smaller elevation differences across a less rugged surface, and likely to higher avulsion frequencies. The limited mobility and spreading of coarse-grained debris flows promote frequent deposition on top of and directly adjacent to channel margins, forming well-defined channels bordered by thick composite levees, and raised fan sectors. These results illustrate how catchment lithology can affect the morphology, sedimentology, and evolution of debris-flow fans, providing guidelines for reading their depositional archives and avulsion hazard assessment.
How to cite: de Haas, T., Ventra, D., Densmore, A., and Binnie, S.: Influence of catchment lithology on debris-flow fan morphology, sedimentology and evolution – Field evidence from the White Mountains, southern California, USA, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3978, https://doi.org/10.5194/egusphere-egu26-3978, 2026.
Debris flows pose major hazards in the semi-arid Andes of Central Chile. Both, their regional spatial distribution and controlling factors, however, remain poorly understood. Our contribution addresses this gap in the upper Maipo River basin – a critical basin for Santiago’s water supply and recreational activities – which has experienced recent catastrophic events in 2017, 2021, and 2023 that resulted, among others, in the complete flooding of several villages.
Using multi-temporal imagery, we mapped 312 debris flows that occurred between 2007 and 2017, and modeled their occurrence through Bayesian logistic regression. We assessed the slope, contributing area, elevation, and lithology as potential controls, while testing the efficacy of slope–area relationships for susceptible terrain identification.
Our results demonstrate that slope and contributing area are primary predictors, exhibiting a credible positive interaction. Conversely, elevation showed a negative correlation, and lithology offered only negligible predictive power. Most strikingly, slope–area plots revealed that high-probability source areas cluster within a distinct morphometric domain, thus offering a simple, yet reliable, approach for delineating hazardous terrain from topographic data.
Despite our short observational window and restriction to debris flow events below 3700 m asl, our findings may help establishing a framework for regional susceptibility assessments in high-priority basins of the Central Andes and underscore the utility of simple models and open-access imagery for hazard mapping in data-scarce mountain regions and, potentially, providing a first step towards early warning.
How to cite: Mohr, C. H., Parra, E., Goetz, J., Brenning, A., Henriquez, C., Araneda, M. B., Bustos, M., and Korup, O.: Mapping and Predicting Debris Flows in the Central Chilean Andes, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5643, https://doi.org/10.5194/egusphere-egu26-5643, 2026.
The dynamics of landslides are strongly governed by their material composition and boundary conditions. When soil sediments mix with water, the fine fraction can markedly alter permeability—both within the flowing mass and in the underlying bed material—thereby influencing the generation and persistence of excess pore pressures and, consequently, shear strength. While new two-phase continuum models are increasingly capable of capturing these coupled hydro-geomechanical processes (e.g., Vicari et al., 2025b), a key challenge remains: can we reasonably measure the field material properties required to parameterize such models? Field sites are often steep, heterogeneous, and difficult to access, complicating in-situ characterization.
To address this challenge, we conducted a systematic geotechnical investigation of ten debris flow channels across Switzerland (Vicari et al., 2025a). Soil samples were collected to determine grain size distributions, revealing significant variability in fine content among sites. Higher fine contents were found to reduce sediment permeability, quantified using in-situ dual-head infiltrometer tests. Complementary Unmanned Aerial System (UAS) surveys provided high-resolution erosion and deposition patterns, allowing us to relate observed geomorphic changes to both channel and catchment morphology and sediment properties. Simple correlations suggest that higher fine contents correspond to enhanced erosion and more frequent debris flow activity, though these relationships are strongly modulated by channel geometry and sediment availability. Combining geotechnical and geomorphological parameters enabled us to classify the investigated channels into four distinct behavioral groups, ranging from small, coarse gullies through intermediate coarse- and fine-rich channels to large, fine-rich systems.
The methods developed and trained through this study proved invaluable for the investigation of the 28 May 2025 Blatten landslide. Modeling results indicate that a substantial frictional reduction was required to explain the extreme mobility of this event, implicating transient excess pore pressures as a likely mechanism. Geotechnical analyses of the landslide material revealed low permeability and high fine content, suggesting that excess pore pressure dissipation times may have greatly exceeded the event duration if even a 1 m flow layer became liquefied.
Our results highlight the importance of integrating geotechnical measurements with remote sensing to constrain key parameters for next generation two-phase numerical models.
References
Vicari, H., Bründl, F., Frieß, P., Ringenbach, A., Stoffel, A., Bühler, Y., Aaron, J., Mcardell, B., Walter, F., Graf, C., Herzog, R., Bebi, P., Gaume, J., 2025a. Linking debris flow erosion to channel-bed parameters: Geotechnical and remote sensing investigation of ten channels in Switzerland. ESS Open Archive. https://doi.org/10.22541/essoar.176126762.20405430/v1
Vicari, H., Tran, Q.-A., Metzsch Juel, M., Gaume, J., 2025b. The role of dilatancy and permeability of erodible wet bed sediments in affecting erosion and runout of a granular flow: Two-phase MPM–CFD simulations. Computers and Geotechnics 185, 107307. https://doi.org/10.1016/j.compgeo.2025.107307
How to cite: Vicari, H., Bründl, F., Friess, P., Bühler, Y., Stoffel, A., Herzog, R., Farinotti, D., Kang, J., Walter, F., Aaron, J., McArdell, B., and Gaume, J.: Measuring the unmeasurable? Geotechnical and UAS-based investigations of landslides, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4521, https://doi.org/10.5194/egusphere-egu26-4521, 2026.
The Réal Torrent is a very active debris-flow catchment of the Southern French Alps, monitored since 2011. A debris-flow intensity-duration threshold was established in 2017 (Bel et al., 2017), based on rainfall data collected with rain gauges during the 2011-2014 period, where 33 debris-flow events were observed. Our study aimed to update this threshold with rain gauge data collected during the 2014-2023 period, where 51 additional events were observed, to evaluate the stability of the threshold. Secondly, we tested the influence of rainfall data source through the comparison of the rain-gauge-based threshold with a radar-based threshold, established by using the Météo-France ANTILOPE product, which combines radar estimates and rain-gauge observations of precipitation. Finally, we tested the transposability of the latter threshold by comparing it to triggering rainfall events of various regional debris flows recorded on 82 catchments, spread across the Southern French Alps during the 2011-2024 period. The detailed dataset of debris-flow events in the study region was obtained from the ONF-RTM natural hazard database (French National Forest Office service dedicated to natural hazards in mountain regions).
The update of the threshold with 9 additional years of debris-flow monitoring allowed us to conclude that an amount of roughly 30 debris-flow events is sufficient to establish a stable intensity-duration threshold on a single torrent. We also observed that the radar-based threshold is much lower than the rain-gauge-based threshold. Therefore, we showed that the source of rainfall data has a strong influence on the threshold equation. Finally, the analysis of the intensity and duration of the regional debris-flow triggering rainfall events relative to the Réal radar-based intensity-duration threshold led us to conclude that using a threshold on only one very active catchment is not transposable at the regional scale due to the high proportion of false positives induced.
How to cite: Welfringer, T., Liébault, F., Laigle, D., and Fontaine, F.: Rainfall intensity-duration threshold of debris flows in the Réal Torrent from rain gauges and radar data: comparison and transposability, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11227, https://doi.org/10.5194/egusphere-egu26-11227, 2026.
Glacier retreat and snow melting promote periglacial debris-flow occurrence in the Tibetan Plateau and surrounding mountains. We collect data of 32 historical events in the Zelunglung, Xueka, Tianmo catchments of the southeastern Tibet by retrospective analysis and on-site investigations. It is found that sedimentation on the Zelunglung debris flow fan reduces to pre-earthquake level about 40 years after the 1950 Assam Earthquake. In recent decades, debris flow occurrence lags behind the average annual temperature/summer temperature peaks by 2 to 3 years, indicating that the debris flows have shifted from being earthquake-driven to climate-warming-driven. 11 historical runoff-generated debris flow events were identified from 1940 to the present using dendrochronological analysis at the Xueka catchment, indicating the positive feedback between debris flow and climate warming. Large-scale debris flows transformed from ice avalanches or glacier collapse often result in dammed lakes and subsequent outburst floods that impose long-term impacts on downstream infrastructures and landscape evolution.
How to cite: Hu, K., Li, H., and Liu, S.: Recent periglacial debris flows driven by climatic warming in the southeastern Tibet, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2891, https://doi.org/10.5194/egusphere-egu26-2891, 2026.
Debris flows are extremely rapid landslides whose complex dynamics remain only partially constrained, largely due to the challenges associated with acquiring direct measurements in the field. Modern monitoring stations typically include cameras that, despite their relatively low cost, can provide highly valuable information for characterising recorded events. Recent studies have shown that Particle Image Velocimetry (PIV) algorithms, when paired with suitable orthorectification techniques to correct non-zenithal acquisition geometries, can serve as effective methods for reconstructing the surface velocity field of flow-like landslides, including debris flows.
In the present work, a PIV-based workflow is employed to analyse a debris-flow event that occurred on 22 October 2022 in the Blè Stream catchment, an active basin in the Camonica Valley (Lombardia, Italian Alps) within the municipality of Ono San Pietro. The 3.5 km² catchment reaches a maximum elevation of 2527 m a.s.l. and features a 2.9 km-long main channel, instrumented with several monitoring stations, each equipped with cameras and flow-depth radar sensors that documented the event.
The sequential application of two open-source MATLAB tools, PIVlab (Thielicke & Stamhuis 2014) and RIVeR (Patalano et al. 2017), yielded frame-by-frame, orthorectified surface velocity fields at each station. These velocity fields were integrated with cross-sectional areas derived from high-resolution pre- and post-event LiDAR and drone surveys, along with measured flow levels, to compute instantaneous discharge at key reference sections. By consistently applying this frame-by-frame procedure along the channel, while carefully accounting for the main sources of uncertainty associated with the continuously changing section geometry and the tendency of surface velocity to overestimate the actual depth-averaged velocity, depending on flow rheology, a range of plausible hydrographs was obtained at each monitoring station. These hydrographs, which provide estimates of the volume of material that passed through each section during the event, enabled a quantitative assessment of the relationship between the triggering water volume in the upstream reach and the fully-developed debris flow volume observed downstream, as well as estimates of entrainment rates along different sectors of the channel.
References:
Patalano A, García CM, Rodríguez A (2017) Rectification of image velocity results (RIVeR): a simple and user-friendly toolbox for large-scale water surface particle image velocimetry (PIV) and particle tracking velocimetry (PTV). Comput Geosci 109:323–330. https://doi.org/10.1016/j.cageo.2017.07.009.
Thielicke W, Stamhuis EJ, 2014. PIVlab – towards user-friendly, affordable and accurate digital Particle Image Velocimetry in MATLAB. J. Open Res. Softw. 2 http://dx.doi.org/10.5334/jors.bl.
How to cite: Zuccarini, A., Ioriatti, E., Albertelli, L., Beretta, L., Redaelli, M., Reguzzoni, M., Reguzzoni, E., Schimmel, A., and Berti, M.: Estimating debris flow entrainment from along-channel hydrographs reconstructed using low-cost field cameras and Particle Image Velocimetry, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11969, https://doi.org/10.5194/egusphere-egu26-11969, 2026.
The Simplon pass culminates at 2006 m over sea level and is one of the principal alpine roads crossing the Alps in the North-South direction. It is extremely important for intra-European goods transportation, as it is open over the winter and is only protected by avalanche galleries and does not cross a tunnel, so that dangerous goods like chemicals can be transported safely all year round on this road. On the 29th of June 2025 at ca. 4.30 PM, following a few days of high intensity precipitation, several massive debris flows originating from the region of the Hübschhorn rock glacier entered the Engi gallery that is normally protecting the road from avalanches in winter and deposited a layer of more than one meter debris over several tens of meters, causing the closure of the road. The Hübschhorn rock glacier had been melting substantially over the last decade so that a combination of high temperature and important cumulated precipitation could mobilize the debris. The road had to be reopened as rapidly as possible and the debris flow channel had to be raised to avoid new damages to the road in case of events, but this meant performing construction works in a region with frequent rockfall and high debris flow risk. The Federal Roads Office mandated Geoprevent to rapidly install both a monitoring system to provide some information about the state of the rock glacier and a multi-component alarm system to detect debris flows and rockfall, close the road and trigger a local alarm on the construction site in case of detections. This works introduces the different components of these complex monitoring and alarming systems and presents some insights about the challenges faced during their installation and operation.
How to cite: Stitelmann, Ó., St Pierre Ostrander, T., Wetter, J., Von Wartburg, J., Carrel, M., and Vincent, S.: Alarming and monitoring systems at the Simplon Pass, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2990, https://doi.org/10.5194/egusphere-egu26-2990, 2026.
Accurate identification of debris-flow events from seismic records is essential for developing high-resolution monitoring and early-warning systems. Here we develop an optimized Random Forest (RF) classifier designed to improve detection accuracy and, critically, to generalize across diverse geographic and environmental settings. We compile a global dataset of historical debris-flow events from 12 representative regions and construct an RF-based workflow that combines interpretable feature selection and automated model tuning. The Boruta algorithm is used to identify five informative predictors, improving interpretability while reducing redundancy in the feature set. In parallel, Bayesian optimization is employed to tune RF hyperparameters and enhance out-of-sample performance. We conduct three comparative experiments to quantify the contribution of each component. Results show that the combined Boruta–Bayesian RF consistently outperforms conventional RF approaches, achieving an accuracy of 96.25%, an F1 score of 0.9714, and an AUC of 0.9819. To further assess transferability, we apply the trained model to independent seismic data collected at Tianmo Gully in southeastern Tibet, China. The model successfully distinguishes debris-flow signals from background noise across the study period, demonstrating stable performance beyond the training regions. Overall, the proposed optimized RF framework offers an efficient, interpretable, and transferable solution for debris-flow detection using seismic signals, providing practical methodological support for the development of operational debris-flow early-warning systems.
How to cite: Qiao, Z., Wang, D., Yan, S., and Chen, H.: An Optimized Random Forest Model for Debris-Flow Event Detection from Seismic Signals, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11959, https://doi.org/10.5194/egusphere-egu26-11959, 2026.
Debris flows and landslides are frequently triggered by intense rainfall and are characterized by sudden onset and short warning lead times. Conventional early warning approaches that rely solely on rainfall thresholds are prone to false alarms or missed warnings due to spatial variability in rainfall and differences in actual slope conditions. To improve warning accuracy and operational applicability, this study proposes a novel early warning operational framework for debris flows that integrates rainfall thresholds, seismic monitoring, and near-real-time source classification into a multi-level, dynamic warning system. The proposed framework is implemented and evaluated in the Putunpunas River in Kaohsiung City, southern Taiwan, where a total of 46 documented debris-flow events were compiled and analyzed. Debris-flow occurrences were identified and confirmed through the combined use of riverine seismic signals and time-lapse camera observations, enabling reliable event detection and temporal validation. Based on reconstructed rainfall events, an empirical rainfall threshold was established using event duration (D) and effective cumulative rainfall (E),expressed as:
𝐸𝐷𝐹 = (14.1 ± 3.0)𝐷0.55±0.1
To assess whether a warning model trained on historical experience can successfully predict future debris-flow occurrences, this study further adopts a machine learning–based decision tree approach using the C5.0 algorithm to train the event classification model. This strategy allows objective evaluation of the predictive capability and generalization performance of the proposed integrated early warning framework under unseen event conditions, thereby enhancing its reliability and practical applicability for real-time debris-flow early warning operations.
The proposed system first evaluates rainfall conditions using real-time precipitation data and applies three warning levels—alert, management, and action—corresponding to exceedance probabilities of 5%, 10%, and 20%, respectively, as an initial risk screening mechanism. When rainfall conditions exceed the defined thresholds, modules of seismic source detection and landslide monitoring (GeoLoc scheme) are simultaneously activated to detect potential landslides in real time. Furthermore, artificial intelligence (AI) based debris flow classifier is adopted to identify whether debris flow events have actually occurred. Compared to conventional rainfall threshold–based debris-flow early warning systems, our proposed approach enables real-time monitoring of upstream sediment supply associated with landslide occurrence and provides a secondary verification using riverine seismic signals.
This operational early warning framework enables to real-time assess rainfall threshold, landslide detection, and classify debris flow source, thereby enhancing the reliability and practical value of debris flow early warning and serving as a core component for future smart disaster prevention and real-time risk management systems. The framework was evaluated during Typhoon Fung-wong in November 2025. A warning was issued once rainfall exceeded the alert threshold based on real-time precipitation data, followed by activation of landslide monitoring and debris-flow detection modules. Using microtremor seismic signal analysis and AI-based event classification, the system verified event occurrence. During the event, only one out of six rainfall stations in the Putunpunas River exceeded the rainfall threshold, highlighting strong spatial variability in rainfall-induced hazard potential; nevertheless,the system was able to reflect actual hazard conditions in near real time through postevent verification and status updating,demonstrating its operational reliability.
How to cite: Chu, C.-H. and Chao, W.-A.: An operational Early Warning Decision Framework For Debris FlowIntegrating Rainfall Thresholds and Seismic Signal Classification, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15826, https://doi.org/10.5194/egusphere-egu26-15826, 2026.
Basal sliding along the channel bed may play a significant role in debris flow propagation, however a lack of field measurements has limited our ability to understand the conditions that may occur in in-situ debris flows. Laboratory experiments have demonstrated that such sliding can occur under both fixed-bed and erodible conditions, driven by interactions between the heterogeneous debris flow material and the basal boundary. We introduce a novel monitoring setup designed to directly quantify basal slip velocities using paired conductivity sensors and report preliminary results from two natural debris-flow events recorded in the Lattenbach catchment (Tyrol, Austria) in June 2025.
The preliminary analysis indicates that basal slip was present in both events and consistently lower than independently measured surface velocities. Sixty-second binned median slip velocities were mostly below 2 m s⁻¹; fronts exhibited the highest values, followed by stabilization around 0.5–1 m s⁻¹. Event-scale ratios of daveraged approximately 0.2 for both events, with instantaneous values ranging from 0.1 to 0.5 for the 15 June event and from 0 to 1 for the 30 June event. The latter comprised three surge-like phases, including a middle surge that briefly matched surface velocity. We note that the effective detection depth of the sensor pairs remains uncertain and likely varies with mixture conductivity and fluid content; if substantial, measured velocities may reflect the motion of lowermost flow layers rather than true bed slip.
These observations suggest that no-slip boundary conditions on non-erodible channel sections may not adequately represent debris-flow mechanics. Future work will improve temporal resolution, constrain detection depth, analyse additional events, and conduct cross-catchment comparisons.
How to cite: Nagl, G., Ender, M., Klein, F., McArdell, B., Aaron, J., and Kaitna, R.: Slip happens: Field evidence of basal sliding in natural debris flows, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18863, https://doi.org/10.5194/egusphere-egu26-18863, 2026.
Debris flows are gravity-driven, channelized mass flows with a highly variable composition of solids and fluids. Due to the variability of grain size distribution and water content, flow resistance is expected to vary within single events as well as between different events. One approach to constrain the flow resistance of debris flows involves the measurement of vertical velocity distributions, i.e., average velocities and velocity fluctuations at different heights above the channel bed.
This study investigates vertical velocity distributions in natural debris flows observed at a monitoring station at the Gadria creek in South Tyrol, Italy. The first aim is to establish a robust methodology for estimating these distributions through a comprehensive parameter sensitivity analysis which forms the foundation of the present work. The second aim is to contrast velocity profiles during single debris-flow events and along different debris-flow events. For this we differentiate between relatively short, “quasi-steady” flow sections, characterized by no significant changes in bulk flow velocity, flow depth, or visually assessed composition of the passing debris, and unsteady flow periods, which are characterized by rapid and pronounced variations in velocity, flow depth, and mixture composition over short time scales, as typically occurring in debris-flow surges/waves or at granular debris-flow fronts.
In the current setting, we achieve a maximum temporal resolution for derivation of continuous vertical velocity profiles of 0.4 seconds. We observe substantial differences in the vertical velocity distributions of quasi-steady and unsteady flow regimes. Quasi-steady flow exhibits a constant velocity profile. For an initial analyzed quasi-steady section the profile follows a S-shape, which we interpret as indication of non-homogenous mixture composition along depth. For the unsteady flow section, represented by a sequence of waves/surges, we identify changing profile shapes, progressing from linear to S-shaped and finally to slightly concave.
In the future, we will analyze (quasi-)steady and unsteady flow sections of many debris-flow events and connect these with independent measurements of basal normal stresses and pore fluid pressure, as well as analyses of material samples and laboratory experiments. The outcomes of this study provide a basis for improved debris-flow model representation and validation.
How to cite: Ender, M., Klein, F., Nagl, G., Hübl, J., and Kaitna, R.: Vertical Velocity Profiles in Natural Debris Flows: Insights into Different Flow Regimes, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6488, https://doi.org/10.5194/egusphere-egu26-6488, 2026.
Debris flows are a significant geohazard in the semi-arid mountain belts of southwestern Türkiye, where short-duration, high-intensity rainfall frequently triggers rapid sediment mobilization, generating destructive debris-flow hazards that threaten settlements, transportation corridors, and agricultural land. The catastrophic 6-7 August 2018 debris-flow event, reported to have caused severe damage to agricultural fields, livestock, and road infrastructure (in both two villages, one day apart) together with more recent rainfall-triggered events in the area, highlights the vulnerability of the region; since these basins drain into the Elmalı polye, a critical agricultural hub, assessing debris-flow susceptibility is essential for future risk mitigation. This study presents a regional debris-flow susceptibility assessment for the Elmalı Basin (Western Taurus Mountains) in Antalya, Türkiye. Using a 5m resolution DEM, NDVI-based vegetation change analyses, topographic thresholds (slope, curvature, flow accumulation), and lithological data, potential source areas were identified, and runout paths were modeled with the empirical Flow-R approach. Model calibration was supported by geomorphic evidence of the 2018 event, and NDVI difference maps provided an effective tool for evaluating the accuracy of runout angle calculations. The results highlight several channels where steep, concave slopes coincide with high-susceptibility zones, indicating that certain settlements and agricultural fields lie within potential impact zones.
How to cite: Özkulluk, A. N., Görüm, T., Yılmaz, A., Karabacak, F., Avcıoğlu, A., Akbaş, A., Çömert, R., and Fidan, S.: Debris-Flow Susceptibility Assessment in a Semi-Arid Mountain Belt: Western Taurus, Turkiye, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-423, https://doi.org/10.5194/egusphere-egu26-423, 2026.
Debris flows, composed of water, soil, sand, rocks, and organic materials such as woody debris, are highly destructive phenomena commonly occurring in mountainous regions. The presence of woody debris can significantly modify flow mobility, depositional characteristics, and overall debris-flow dynamics. In this study, woody-debris suspensions composed of clay-silt, water, and woody debris were systematically prepared to investigate the effects of woody-debris proportion (Cvg) and woody-debris size (Lw) on rheological properties and flow behavior. Rheological parameters were measured using a Brookfield DV-III rheometer. The results show that increasing Cvg significantly increases yield stress (τB) and viscosity (μB), whereas increasing Lw leads to a reduction in both parameters. Inclined-channel tests were further conducted to examine flow dynamics. Higher Cvg results in lower entry speeds, shorter runout distances, and thicker, wider deposits. In contrast, larger Lw generates higher entry speeds, leading to longer runout distances with thinner and narrower deposits. A strong correlation is observed between rheological parameters and flow-test parameters, indicating that inclined-channel tests provide a practical alternative for estimating rheological properties of debris flows containing woody debris.
How to cite: Nguyen, L.-T. and Jan, C.-D.: Experimental study on the rheological and flow behavior of woody-debris suspensions, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4368, https://doi.org/10.5194/egusphere-egu26-4368, 2026.
On Earth, hillslope processes are typically driven by gravity and lubricated by liquid water. The slope angle, availability of water, and material composition ultimately determine the type of mass-movement, the flow dynamics, and the morphology of the resulting depositional landforms. Therefore, terrestrial hillslope landforms have served as our guide in the interpretation of hillslope landforms and their formation processes on other planetary bodies (e.g. the Moon, Mars). However, pioneering work has shown that gravity has a significant effect on the dynamic angle of repose (Kleinhans et al., 2011), the transition of bedload to suspended load in fluvial sediment transport (Braat et al. 2024), and the settling speed of fine sediment in water (Kuhn et al., 2015). This raises the questions if and how gravity affects the non-linear flow dynamics of hillslope mass movements and the morphology of their depositional landforms.
In this study, we experimentally explored the effects of gravity on the dynamics of dry mass movements and those lubricated by a liquid. We performed rotating drum experiments under varying gravity (from ~0.1g to 2g, with g=9.81ms-2). The lower and hyper-gravity conditions were created by flying, respectively, parabolic trajectories and steep turns with a Cessna Citation II aircraft (PH-LAB), in which the rotating drum set-up was installed. In the rotating drum (diameter=50 cm), we tested how dry and wet granular flows responded to different gravity by measuring flow depth, density, compaction and dilation, and internal grain dynamics. Reference experiments with varying drum-rotation speeds were performed under Earth gravity to determine the relative effects of centrifugal force versus gravity, and aircraft vibrations.
Preliminary analyses show that gravity changes the dynamics of both dry and wet granular flows in our drum, and that these effects are more pronounced for wet granular flows. Under higher gravities (>1g), the granular flows become more compacted, which pushes the water out of the mixture and decreases the water content of the granular flow itself. As a result, the interparticle friction increases and the centre of mass shifts upslope in the drum. At lower gravities (<0.7g), the granular flows dilate, increasing the pore space in the sediment-water mixture, resulting in an increase in air in the inter-particle pore space. This increases the relative importance of flow resisting forces relative to lubricating forces within the mixture, shifting the center of mass of the mixture upslope. The results under varying gravities seem to imply that, for a given ratio of sediment to water, an optimum gravity exist for peak water-lubricated granular flow mobility.
Comparison of the results under varying gravity with those of the reference experiments with varying drum rotation speeds under 1g confirm that gravity has a unique effect on the flow dynamics of granular flows. In particular, on the dilation of the flowing mixture and the interparticle behaviour. However, as changing drum-rotation speed also shifts the centre of mass of the flowing mixture, further analysis will focus on the combined effects of dilation, shifting centre of mass, and the steepening slope in the drum for all experiments.
How to cite: Roelofs, L., van Dam, B., van Eijk, A., Klaassen, M., den Toom, G., Mulder, H., de Vet, S., Kleinhans, M., ten Kate, I. L., van Westrenen, W., and de Haas, T.: Experimental debris flows and rock avalanches under different gravities – To the Moon and Mars in an airplane, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4902, https://doi.org/10.5194/egusphere-egu26-4902, 2026.
Debris-flow dynamics are governed by the internal deformation of sediment–water mixtures. Due to the destructive potential of natural debris flows, constitutive models predicting flow velocity, impact forces, erosion, and deposition are desired. While simplified sediment–water flows are well understood, existing debris-flow models typically rely on constitutive assumptions for internal friction, sediment concentration, and solid–fluid interaction. Systematic experiments exploring the effects of grain-size distribution and fine material content on internal deformation and flow resistance are essential to better constrain and improve these models.
In this work we introduce a novel methodological setup to measure vertical velocity profiles within steady shallow (~15 cm) flows of sediment–water mixtures in a ~2.5 m diameter rotating drum. Measurements are performed in the central, most uniform flow region using an array of paired-conductivity sensors of varying geometry. Velocities are obtained via established cross-correlation methods of adjacent signals. Spectral properties of the conductivity signals are also explored as a potential complementary source of velocity information.
A low-cost capacitance probe is being developed to enable non-intrusive estimation of sediment concentration during flow. Proof-of-concept tests in air demonstrate feasibility, and further testing in water is planned to realize its use in ongoing experiments. Upcoming work will systematically investigate how grain-size distribution, particularly the fine material content and the uniformity of the coarse fraction, influence internal deformation, pore-fluid pressure, and bulk flow resistance.
How to cite: Klein, F. M., Ender, M., Nagl, G., and Kaitna, R.: Towards systematic measurements of velocity profiles and sediment concentration in a wide range of laboratory debris-flow mixtures, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6540, https://doi.org/10.5194/egusphere-egu26-6540, 2026.
Debris flows are rapid mass movements that move down steep mountain creeks and are a major threat to human life, properties, and infrastructure. As the debris flow travels down the channel, the impact force of the debris on the channel bed generates ground vibrations that propagate to and can be recorded by the seismometer. The impact force is an important parameter in the design of debris-flow damage mitigation, such as check dams. Direct measurements of impact force from debris flows are limited by the high cost of instrumentation and the risk of instruments being destroyed in the process. Installing and maintaining a seismic network outside the debris flow channel keeps it protected from the hazard and can be a suitable alternative to direct measurements of the impact force. Connecting seismicity to the generating impact force is complex due to the complicated environment. Bridging this gap using deep learning could help estimate physical information to improve debris flow warning.
In this study, we train an extended-LSTM (xLSTM) model to invert impact force from seismic signals generated by debris flows in the Illgraben catchment, in Switzerland. We chose the xLSTM model ahead of others due to its ability to process long and complex sequences of data. We used seismic signals generated by debris flows as they pass through CD 27 and impact force signals recorded at CD 29 by an 8m2 force plate. The xLSTM model is compared to the LSTM model architecture as a baseline, and we show that the xLSTM model performs better at capturing the distribution of the impact force and producing lower mean error. Along with this, it inverts the peak impact force with an absolute error of less than 1kN to the measured impact force. Furthermore, we find a strong correlation between the volume and the cumulative impact (CIF) force for debris flows, showing that the xLSTM inverted impact force can be used to derive an initial constraint on the volume of a debris flow event. This method can support early warning systems for debris flow by allowing for quick impact force analysis and providing initial constraints on some physical characteristics, for example, debris-flow volume.
How to cite: Kar, K., Tang, H., and Zhou, Q.: Deep Learning Reveals Debris Flow Impact Forces from Seismic Signals , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6967, https://doi.org/10.5194/egusphere-egu26-6967, 2026.
Debris flows increase in size by channel bed and bank erosion, enhancing their hazardous potential. While bed erosion by debris flows has been studied extensively through field measurements, laboratory experiments, and numerical modeling, our understanding of bank erosion remains limited. Therefore, we have no information on the spatial and temporal dynamics of debris flow bank erosion. Due to the infrequent occurrence of debris-flow events and the difficulty in accessing debris-flow channels, there have been no torrents for which there are (1) detailed measurements of debris-flow properties, (2) high-resolution topographic measurements of bank erosion before and after debris flow events, and (3) detailed measurements of bank composition and strength. We need the combination of these three processes to be able to physically explain the conditions that lead to high bank erosion rates. We use a comprehensive, long-term dataset from the Illgraben torrent, one of the most active debris-flow channels in the European Alps, to investigate bank erosion processes. This unique record includes field measurements of debris-flow characteristics and 51 high-resolution DEMs, spanning over 70 debris flow events between 2020 and 2025. By generating DEMs of Difference (DoD) to quantify bank erosion and integrating these with flow parameters derived from RAMMS modeling and field measurements, we investigate the controls on debris-flow bank erosion. Our preliminary results indicate that bank erosion often lags behind major debris-flow events. Large erosion episodes commonly occur after a high-magnitude flow. Smaller flows can gradually erode the bank toe during successive events, creating progressive undercutting that reduces stability until a sudden, larger bank failure occurs. A better understanding of debris-flow bank erosion processes and controls provides insights into the timing and magnitude of volume amplification, improving the accuracy of debris-flow models and fostering the development of strategies to reduce debris-flow erosion and mitigate its hazards.
How to cite: van den Broek, A., McArdell, B., Draebing, D., Zwarts, M., Huguenin, P., Nijland, W., and de Haas, T.: Deciphering Debris-Flow Bank Erosion: Insights from the Illgraben Torrent, Switzerland , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7067, https://doi.org/10.5194/egusphere-egu26-7067, 2026.
Debris flows are composed of solid grains and fluid, in which the grains span a range of size, and the interstitial fluid is viscous. Erosion and deposition processes have significant impacts on post-event morphology, and their mechanisms are closely related with the grains and the viscosity of the interstitial fluid within the flow body. In the present study, we present a two-phase erodible model, extended from Wong et al., (2024) in which mono-grain-size is assumed, for modeling heterogeneous grain-fluid mixtures composed of multiple grain sizes and a viscous interstitial fluid. That is, the solid phase within the flow body is supposed to consist of grains of various sizes. In this simplex approach, the effects of grain size are explicitly incorporated into the erosion-deposition processes. The erosion rate is proportional to the shear stress and follows the Shields parameter (Shields, 1936) for the threshold magnitude, while deposition is assumed to be induced by settling speed and to follow the regressed Hjulström-Sundborg diagram (Hjulström, 1935). Because both the Shields parameter and the settling speed depend on grain size and fluid viscosity, the resulting entrainment or deposit patterns vary with the grain-size composition of the flow body. For example, sediments of smaller size are entrained first and settle latter, whereas larger grains tend to deposit at earlier stage. The key features of this simplex approach will be demonstrated through numerical investigations on flows in chutes with simple geometry, as well as through an application to a back-calculation of a large-scale historical event.
How to cite: Tai, Y.-C., Feng, F.-W., Sarno, L., Pai, P.-H., and Kan, H.-C.: A Simplex Solid–Fluid Model for Debris Flows over Erodible Beds with Multi-Size Sediments, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9257, https://doi.org/10.5194/egusphere-egu26-9257, 2026.
The presence of pore fluid can great change landslide dynamics, significantly enhancing sliding mobility and resulting in high velocities and long runout distances. Our study presents a two-phase model with dilatancy/contraction for dense solid-fluid mixture based on the material point method. In the constitutive model, we consider the dilatancy/contraction effect on the two-phase system and the rate-dependent frictional law derived from granular flow rheology (the μ(K) and Φ(K) relationships). Numerical benchmarks including saturated granular column collapse and flume experiments were conducted to see the performance of the model. Furthermore, simulations of the 2014 Oso landslide in Washington State, USA, were performed to investigate the mechanisms governing its high mobility. The liquefaction of saturated sediment and the induced excess pore pressure at the base of the slide, which contributes to the high mobility of the landslide, were well captured in our simulations.
How to cite: Tang, X., Sun, Y., Hou, R., Zhu, L., and He, S.: Two-phase model with dilatancy/contraction for dense solid-fluid mixture in landslide mobility, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11194, https://doi.org/10.5194/egusphere-egu26-11194, 2026.
The study presents an example of application of the guidelines for the assessment of hydraulic hazard and risk assessment in small catchments, developed within the research project RETURN-PB (https://www.fondazionereturn.it/en/portfolio/nuovi-approcci-per-la-valutazione-della-pericolosita-idraulica-nei-piccoli-bacini-montani-return-pb/). The case study refers to the limestone reliefs of Campania (Italy), characterized by diffuse presence of loose pyroclastic soil deposits, originated by air-fall deposition from several eruptions of the volcanic complexes of the area (Somma-Vesuvius and Phlegrean Fields).
The soil deposits are a few meters thick and consist of ashes (loamy sands) and pumices (gravels with sand), characterized by very high porosity (up to 75% in the ashes) and saturated hydraulic conductivity (in the order of 10-4 m/s). These characteristics make the infiltration and retention capacity of the soil deposit so high that, even during the most intense rainfall, overland runoff is quite small, with runoff coefficient rarely exceeding 2%. The deposits are fairly cohesionless and present effective friction angles in the range of 36° to 38°. Nonethless, unsaturated soil deposits with thickness of 1 to 2 meters rest also on slopes with inclination higher than the friction angle, thanks to the apparent cohesive contribution given by soil suction. After intense and prolonged rainfall, the increase of soil moisture and the consequent reduction of suction can lead to the instability of the soil deposit and the triggering of shallow landslides featured as debris avalanches. Owing to the unstable loose soil fabric and the coarse particles, the deposits undergo volumetric contraction under shear deformation, which can lead to the establishment of positive pore water pressure, favoring soil liquefaction. This behavior is responsible for the frequent evolution of landslides in the form of debris flows. Thanks to the steepness of the slopes, the flows reach speed as high as tens of m/s, running out long distance from the original landslide scarp, often channelized through streams that reach nearby towns and villages, with huge damage.
As an example of how debris flow hazard can be assessed in the studied context, the debris flow occurred on 16 December 1999 in Cervinara is modelled. The debris flow was triggered after a rainfall with more than 300 mm in 48 hours, as recorded by a rain gauge less than 2 km from the failed slope. The failure involved a volume of around 30000 m3 of soil, that flew in the form of a liquefied mud hitting the village of Ioffredo, a hamlet of Cervinara, where several buildings were destroyed and five people were killed.
A modelling chain consisting of rainwater infiltration modelling, slope stability analysis, debris flow propagation and impact is applied. The effects of the uncertainty of slope and soil properties, as well as of debris flow behavior are discussed, with an ensemble modelling approach. Specifically, the propagation of the debris flow is simulated with different modelling approaches under different hypotheses (i.e., fixed or erodible bed; single- or double-phase fluid; various rheological formulations with dilution-dependent parameters). The results highlight how the application of ensemble modelling allows introducing the effects of uncertainty in the assessment of hydraulic hazard and risk.
How to cite: Greco, R., Bonomelli, R., Bouraour, O., Marino, P., Molica, S., Pai, P.-H., Roman Quintero, D. C., Aronica, G. T., Papa, M. N., Pilotti, M., Righetti, M., Santonastaso, G. F., Sarno, L., Tai, Y.-C., and Larcher, M.: Debris flow hazard assessment in small catchments with diffuse pyroclastic soil deposits: a case study in Cervinara (southern Italy), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11810, https://doi.org/10.5194/egusphere-egu26-11810, 2026.
The 13 June 2015 Vere River disaster, which caused multiple fatalities in Tbilisi, exposed the extreme vulnerability of Georgia’s capital to debris flow processes generated in the steep, landslide-prone headwaters of the Vere basin. Although damaging flash floods occur frequently in the Vere catchment, the recurrence of debris flow events prior to the start of instrumental observations has remained unknown. To determine whether the 2015 event was exceptional or part of a persistent natural regime, we conducted an integrated geomorphological, sedimentological and chronological analysis of the basin.
High-resolution UAV and satellite imagery, combined with field mapping, were used to identify paleo debris flow and landslide deposits along the main channel and its tributaries. Flow directions and sediment pathways were reconstructed from palaeocurrent indicators, including clast imbrication, allowing depositional units to be linked to specific source areas.
Radiocarbon dating of organic material from multiple stratigraphic sections within individual depositional complexes provides a chronology of major sediment-delivery episodes. The results reveal repeated debris flow events during the Late Holocene. It demonstrates that the 2015 event can be intrinsic to the long term behavior of the Vere basin rather than a rare anomaly. Because the Vere River drains directly into the densely urbanized centre of capital city of Tbilisi, this palaeohazard record has critical implications for hazard assessment and confirms that future catastrophic events are expected unless exposure is reduced.
How to cite: Sukhishvili, L., Gogoladze, S., Merebashvili, G., Javakhishvili, Z., and Kvlividze, K.: Late Holocene Sedimentary Records of Recurrent Debris Flow Hazards in Tbilisi, Georgia, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12614, https://doi.org/10.5194/egusphere-egu26-12614, 2026.
Debris flows are mountain hazard processes, that are among the most devastating natural disasters in the alpine region. Therefore, reliable simulation tools are indispensable for identifying areas affected by debris flows and for developing and evaluating mitigation measures. In engineering practice, the use of depth averaged single-phase models for debris-flow hazard assessment is challenged by the question of the optimal representation of flow resistance and erosion, the respective uncertainty of model parameters and unknown starting conditions. Open-source availability and documentation, including a database of case studies, pose further challenges. The DebrisFrame project (opennhm.org/about_debrisframe) is a collaborative, open-source, Python-based framework for depth-averaged single-phase debris-flow simulations. It offers a user-friendly, modular, and extensible architecture that allows for the flexible configuration of initial conditions, flow resistance, and erosion formulations. Here we present the first results of a sensitivity analysis that quantifies how different types of input hydrographs and friction models influence deposition behavior. To estimate uncertainties, stochastic approaches and scenario analyses are applied. First, simplified, synthetic topographies are used. Subsequently, real-world case studies from the Austrian Alps are employed, while accounting for variable input data and model parameters. Future studies will focus on the role of erosion and its interaction with initial conditions and flow resistance in controlling debris-flow dynamics. The results of our work will help practitioners to better understand how the choice of input data and parameters affects debris-flow runout simulation.
How to cite: Lahrssen, J., Spannring, P., Oesterle, F., Fischer, J.-T., Hagen, K., Moser, M., Puschmann, L., Kammerlander, J., Scheidl, C., and Kaitna, R.: Interplay between input hydrograph and flow resistance within the open-source debris-flow framework DebrisFrame, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12647, https://doi.org/10.5194/egusphere-egu26-12647, 2026.
Debris flows are highly destructive landslide processes involving water, air, and sediments mobilizing by gravity. In 2024, the state of Rio Grande do Sul, southern Brazil, experienced the most extensive disaster in its history, with widespread mass movements and flooding, particularly in the Taquari-Antas Basin, one of the most important basins in the state. During this event, Santa Tereza recorded some of the largest debris flows.
The present study aimed to simulate a huge debris flow occurrence in Santa Tereza, integrating computational modeling and field survey observation. The debris flow was simulated using Morpho2DH (v. 2.1), a solver of iRIC software (v. 4.1) for unsteady horizontal two-dimensional bed deformation analysis.
Santa Tereza is characterized by a humid subtropical climate and steep terrain, which increases its susceptibility to landslides. During the 2024 extreme rainfall event, 411 properties were affected by landslides in the municipality. The studied debris flow traveled 1.5 km resulting in the destruction of two houses. Deposition occurred on an alluvial fan along the Marrecão Stream, a tributary of the Taquari River. The event was mapped by aerophotogrammetry by the Latitude/UFRGS research group, producing a high-resolution orthophoto.
Landslide initiation areas were defined based on orthophoto as two rupture polygons that converged into the channel and developed the debris flow. The digital elevation model used was from ALOS-PALSAR. Field observations indicated a maximum erosion depth of 2 m. Mean grain diameter of 0,001 m was obtained from granulometric analysis of eight in-situ samples. Vegetation parameters were set based on field data, assuming a density of 1 tree km-2 and a mean vegetation height of 6.2 m. Post-event vegetation erosion depth was set to zero, reflecting the complete removal of vegetation cover observed in most of the affected area. The time step of 0.001 s was adopted. Remaining input data followed default model settings. Simulation tests indicated a total event duration of approximately 280 s, indicating high flow velocity and consistent with eyewitness accounts.
Model calibration was performed by comparing the simulated affected area and the flow route with orthophoto interpretations. The simulation estimated an affected area approximately twice as large as the visible scar mapped in the orthophoto, excluding the stretch above the Stream, which could not be calibrated. Despite the overestimation of the affected area, the model accurately reproduced the flow route. These results demonstrate that Morpho2DH can capture debris flow dynamics in Santa Tereza, and the conservative area estimates may be advantageous for disaster risk management applications.
Acknowledgements: This study was supported by FAPERGS under Grant Agreement No. 24/2551-0002124-8 (Call FAPERGS 06/2024).
How to cite: Lahiguera Cesa, L., Andrades Paixão, M., Becker Bobsin, A., and Rosa de Almeida, A. J.: Integration of numerical simulation (Morpho2DH) and fieldwork for a 2024 debris flow event in Santa Tereza, Rio Grande do Sul, Brazil, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14685, https://doi.org/10.5194/egusphere-egu26-14685, 2026.
Regarding whether submarine landslides' mobility decreases linearly or varies non-monotonically with increasing clay content in current studies is still under debate. To address this issue and further investigate the long-runout distance mechanism of submarine landslides, we conducted experiments with clay content ranging from 5% to 30% in a flume with an inclination angle of 10°. Through analysis of the rheological properties of the sediment slurry, the pore pressure and the total stress at the bed bottom along the channel, and the flow velocities, the dynamics of the submarine landslide were characterized. The experiments show that as the clay content increases, the flow transits from liquid-like to solid-like behavior. The peak values of both the pore pressure and the total stress, and the pressure loading rate at the bed bottom monotonically increase as the clay content increases. The velocity analysis supports the conclusion of a non-monotonic variation of mobility, which refers to the flow velocity exhibiting an initial increase followed by a subsequent decrease with the increase of clay content. The critical clay content, at which the maximum flow velocity occurs, is within the range of 10~15%. The mechanism analysis shows that the submarine landslide with the critical clay content has both lower apparent viscosity and higher pore pressure that is sufficient to generate hydroplaning, resulting in the highest mobility. The dimensional analysis shows that the dimensionless yield stress positively correlates to the clay content. It is also found that within the range of approximately three orders of magnitude from 5×10-3 to 3, the dimensionless yield stress and the densimetric Froude number (Frd) exhibit a non-monotonic relationship, which also supports a non-monotonic behavior of the mobility. In summary, this study enhances our understanding of submarine landslide processes and further contributes to better disaster prediction.
How to cite: Zhou, S., Bi, Y., Tan, X., Guo, Z., Zhu, C., Sun, J., and Huang, Y.: The effects of clay content on the dynamics of submarine landslides: New insights from flume experiments, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15700, https://doi.org/10.5194/egusphere-egu26-15700, 2026.
Climate change is expected to increase the likelihood of hydro-geomorphic hazards in active tectonic areas, particularly debris flows. Early warning systems are considered one of the most effective and economical methods for mitigating debris-flow risk. However, current approaches still face challenges in providing accurate quantitative predictions and are subject to considerable uncertainty due to limited observational data. In this study, we develop a new framework for predicting rainfall thresholds for debris-flow initiation by combining numerical simulations with machine learning methods. A small catchment in the Italian Dolomites was selected as a test site to evaluate the efficiency of the framework in areas with limited historical records. Preliminary results suggest that the rainfall threshold can be represented by a piecewise function with an inflection point rather than by the commonly used power-law relationship. Our results suggest that, in the Dimai catchment, rainfall intensity is the dominant factor controlling debris flow initiation for the most rainfall events lasting longer than one hour. While sensitivity analyses indicate that infiltration capacity acts as a key control by regulating the partitioning between infiltration and runoff, thereby influencing the rainfall intensity required to trigger debris flow initiation. These findings provide insight into the hydrological processes governing debris flow initiation and demonstrate the potential of the proposed framework for improving threshold-based early warning systems under limited data conditions.
How to cite: Cheng, W. and Tang, H.: Towards more reliable Debris Flow Rainfall ID Thresholds under Changing Climate Scenarios, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18650, https://doi.org/10.5194/egusphere-egu26-18650, 2026.
High mobility of granular flows is commonly attributed to basal lubrication and fluid–solid interactions, yet the role of internal shear and velocity fluctuations in promoting flow runout remains insufficiently quantified. Here we present a series of controlled flume experiments designed to isolate the effects of internal deformation on granular‐flow mobility. Using synchronized measurements of surface velocity fields, basal forces, and high‐frequency velocity fluctuations, we quantify the spatial and temporal evolution of shear localization, fluctuation intensity, and basal stress transmission.
Results show that intense internal shear zones generate pronounced velocity fluctuations, which propagate downward through the flow depth and modulate basal stresses. The amplitude of basal stress fluctuations increases systematically with both shear rate and fluctuation intensity, indicating an efficient transfer of internal agitation toward the base. This process weakens effective basal resistance and enhances slip, leading to significantly increased runout and mean flow velocity under otherwise identical conditions.
By integrating kinematic measurements with stress analysis, we identify a scaling relationship that links basal friction, flow thickness, inertial number, and normalized fluctuation stress through a power‐law form. This law provides a quantitative bridge between internal dynamics and macroscopic mobility. Our findings demonstrate that internal shear and velocity fluctuations are not merely byproducts of granular motion, but key drivers of enhanced mobility, offering new insights into the mechanics of long‐runout granular flows such as landslides, debris avalanches, and dry granular surges.
How to cite: Yu, X. and He, S.: Internal Shear and Velocity Fluctuations Promote Granular Flow Mobility: Insights from Flume Experiments, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20951, https://doi.org/10.5194/egusphere-egu26-20951, 2026.
Debris-flow hazard assessment relies on the accurate estimation of (peak) discharge and volume. However, traditional methods for inferring these hazard-related parameters often encounter significant limitations, especially in natural, dynamic channels without mitigation measures. Furthermore, existing assessment methods often overlook the characteristic surging behavior of debris flows and the influence of material composition. While laboratory experiments have demonstrated that mixtures with larger grain sizes produce more pronounced levees and extended runout distances, field evidence remains largely qualitative and anecdotal. Consequently, the combination of measurement uncertainties and the omission of flow composition continues to result in substantial uncertainties in debris-flow hazard assessments. Therefore, high-resolution and accurate measurements are needed to better understand debris-flow hazards.
Here, we present high-resolution measurements of over 130 debris flow surges, which occurred in the natural debris-flow channel of the Oeschibach in Kandersteg,Switzerland,in 2024. The sediment source is a rock slope instability in permafrost (Spitze Stei), which is closely monitored. Its recent acceleration has also led to increased debris-flow activity downstream. In 2024, we installed a high-resolution camera and 3D LiDAR sensor, which recorded several debris flows at 10 fps. Using a set of processing algorithms including Particle Image Velocimetry (PIV) on hillshade images, point cloud differencing, and a deep-learning based boulder detection model on camera images, we derived spatially distributed flow velocity, depth, discharge, and material properties (grain count and grain size).
Our key findings include that coarser surges tend to be faster, deeper, levee-forming and erosive. These findings are in line with laboratory experiments, whereas the levee-formation likely also causes surges to be more confined and therefore faster and deeper. Furthermore, we observed that while all events consisted of a series of surges, the bigger the first surge, the more surges were to follow. As traditional monitoring techniques cannot capture these dynamics in sufficient detail, we provide a comprehensive and novel data set in a natural channel, which helps bridging the gap between laboratory experiments and field evidence to reduce uncertainties in debris-flow hazard assessment.
How to cite: Hirschberg, J., Lehmann, R., Spielmann, R., and Aaron, J.: High-resolution measurements of debris-flow surges in a natural channel, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7638, https://doi.org/10.5194/egusphere-egu26-7638, 2026.
Debris flows in steep mountain channels are commonly triggered due to heavy precipitation events mobilising the sediment in the channel bed and from the banks. The magnitude of these events is heavily influenced by rainstorm intensity as well as the sediment availability in those channels. After a debris flow has occurred, the system recharges with material from the upstream catchment until the next event. This poses the question of how large these sediment recharge rates are and how they are connected to rainfall intensities.
Here, we present a 7-year monitoring campaign of a debris flow channel in the Northern Calcareous Alps between 2015 and 2022. Biannual measurements resulted in ten terrestrial laser scans and five UAV surveys to observe the sediment deposition and erosion magnitudes. Additionally, the local precipitation was measured in the vicinity of the channel from the second year of the campaign. We analysed how sediment recharge rates change after a debris flow event and how they are influenced by season and precipitation.
How to cite: Stammberger, V. and Krautblatter, M.: Sediment recharge of a debris flow channel: Insights from a 7-year monitoring campaign in the Northern Calcareous Alps, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18320, https://doi.org/10.5194/egusphere-egu26-18320, 2026.
