Dr. Jordan Balfour Aaron

Debris flows are one of the most dangerous landslide types in mountainous regions. Their destructiveness is strongly controlled by their high peak discharge, which can be orders of magnitude larger than for floods. In order to reduce the associated hazards, detailed field measurements of natural debris flows are required to better constrain discharge and volume of these events. In this study, we used a high-frequency 3D LiDAR (light detection and ranging) scanner in combination with video cameras to measure key properties of a debris flow that occurred in the Illgraben catchment (Switzerland). Based on the LiDAR and video data, we directly measured i) front velocity ii) surge velocity iii) surface velocity and iv) cross-sectional area at sub-second intervals. We then estimated discharge and volume using these direct measurements and considering different channel bed geometry scenarios. We compared our results to estimates based on conventional methods and found that these more established methods substantially underestimate the (peak) discharge and volume for this event. Our results will be assessed in the future by analyzing more events, but our LiDAR-based method has the potential to provide much more detailed information on hazard-related debris-flow parameters, which will have important implications for understanding debris-flow processes and ultimately reducing their risk.

Différents systèmes de surveillance sont utilisés dans le monde entier pour l'observation et la saisie de paramètres importants des laves torrentielles, mais ils ont tous une résolution limitée dans le temps et/ou dans l'espace. Dans cet article, nous présentons une nouvelle méthode de surveillance basée sur des scanners laser 3D (LiDAR), qui fournit des données à haute résolution temporelle et spatiale sur les laves torrentielles. Dans l'Illgraben, plusieurs scanners LiDAR de ce type ont été installés le long du chenal et ont déjà enregistré onze événements de laves torrentielles. Grâce à ces données 3D, nous avons pu mesurer avec une grande précision la vitesse d'objets individuels dans le mélange des laves torrentielles et déterminer les débits et les volumes. Nos analyses montrent que la technologie LiDAR peut apporter une contribution essentielle à une description plus détaillée de la dynamique des laves torrentielles, d'autant plus que les premiers résultats indiquent des différences significatives, par exemple dans la détermination du débit, entre cette méthode et les méthodes conventionnelles.

Zur Beobachtung und Erfassung wichtiger Kenngrössen von Murgängen kommen weltweit verschiedene Monitoring-Systeme zum Einsatz, die alle eine zeitlich und/oder räumlich beschränkte Auflösung haben. In diesem Artikel präsentieren wir eine neue Monitoring- Methode basierend auf 3D-Laserscannern (LiDAR), die zeitlich und räumlich hochaufgelöste Daten von Murgängen liefert. Im Illgraben wurden mehrere solche LiDAR-Scanner entlang des Gerinnes installiert und haben bereits elf Murgangereignisse erfasst. Anhand dieser 3D-Daten konnten wir Geschwindigkeiten einzelner Objekte in der Murgangmischung mit hoher Genauigkeit messen sowie Abflüsse und Volumina ermitteln. Unsere Analysen zeigen, dass die LiDAR-Technologie einen wesentlichen Beitrag zur detaillierteren Beschreibung der Murgangdynamik leisten kann, zumal erste Resultate auf massgebliche Unterschiede etwa in der Bestimmung des Abflusses zwischen dieser Methode und konventionellen Verfahren hindeuten.

Estimating flow velocities is key to assessing hazards associated with debris flows. One approach to post-event velocity estimation is the superelevation method, which uses debris-flow mudlines to measure the cross-channel surface inclination, or superelevation, produced by centripetal forces acting on the flow in a bend. Flow velocities are then calculated using a subjective parameterization of the forced vortex equation modified to include a debris-flow specific correction factor. Subjective parameterization of this equation leads to substantial variability and uncertainty in the resulting flow velocities. We present an analysis of the reliability of the superelevation method using a large UAV-based data set of 14 debris flows with front velocities of ∼0.8–6.5 m s⁻¹ and cross-channel surface inclinations of ∼0.6–8.5°, as well as a validation for a single debris flow measured using high-resolution, high-frequency 3D lidar data fused to video imagery. The validation event indicates that when the flow surface inclination can be measured directly, the forced vortex equation produces excellent results without needing a correction factor for Froude numbers ranging from 0.7 to 1.5. This finding indicates that the main challenge with the superelevation method lies in obtaining accurate measurements of superelevation from the mudlines, and that a correction factor may serve to compensate for measurement difficulties rather than variable flow properties. For very small and highly subcritical flows, the superelevation method may generate a large overestimation of flow velocities.

In-situ measurements of debris-flow properties are crucial for understanding their movement mechanisms and quantifying their impact. Here we present the first results of a field monitoring campaign, at Illgraben, Switzerland, to measure debris-flow parameters using high temporal (10 Hz) and spatial resolution LiDAR sensors at several locations along the channel. The point cloud data is projected onto video images to enhance visualization and aid in the interpretation of the measurements. We process the data using machine vision and deep learning based algorithms, and show that this system can accurately measure front and flow surface velocity, flow depth and bed elevation change, as well as the size, style of motion (e.g. rotating or floating without rotation) and trajectories of individual particles. This system thus provides a promising new method for inferring the internal dynamics of debris flows.

Surging debris flows are among the most destructive natural hazards, and elucidating the interaction between coarse-grained fronts and the trailing liquefied slurry is key to understanding these flows. Here, we describe the application of high-resolution and high-frequency 3D LiDAR data to explore the dynamics of a debris flow at Illgraben, Switzerland. The LiDAR measurements facilitate automated detection of features on the flow surface, and construction of the 3D flow depth and velocity fields through time. Measured surface velocities (2-3 m s^-1) are faster than front velocities (0.8-2 m s^-1), illustrating the mechanism whereby the flow front is maintained along the channel. Further, we interpret the relative velocity of different particles to infer that the vertical velocity profile varies between plug flow and one that features internal shear. Our measurements provide unique insights into debris-flow motion, and provide the foundation for a more detailed understanding of these hazardous events.

Extremely-rapid flowlike landslide are a major hazard in many parts of the world, and managing their risk requires an understanding of the mechanisms that drive motion, as well as reliable predictions of their potential destructiveness. Numerical runout models are one tool that can be used for both these applications, however their utility is presently limited by their computational runtime. In the present work, a new depth-averaged, smooth particle hydrodynamics based model is implemented to run on a graphical processing unit. The new implementation provides a speedup of over two orders of magnitude, compared to a commonly used CPU based runout model. The new model has been validated, and then is used to back-analyse the Johnsons Landing Landslide. It is shown that increasing model resolution results in an accurate simulation of complex topographic interactions between the flowing landslide and the surface topography. The new model runs on commercially available GPU’s, and should therefore be useful for both researchers and practitioners seeking to understand and quantify landslide hazard and risk.

The volume of a debris flow is a critical parameter in hazard analysis, yet accurate estimates of volume are often unavailable due to mixing with larger rivers at the downstream end of alluvial fans. We describe a method to calculate the volume of debris flows using flow depth data collected at a check dam, using a Manning friction relation to describe the velocity of the debris flow as a function of flow depth, and the geometry of the channel cross section. The method is evaluated using a published data set from the USGS debris-flow flume where event volume and stage information have been accurately measured, and results in volume estimates either somewhat smaller than or up to 50% larger than observed volumes. We further demonstrate the method to single-surge and a multiple-surge debris flows observed at Illgraben.

The Tamins rock avalanche lies adjacent to the Flims rock avalanche, the largest in the Alps. Its deposit forms a ridge across the Rhine Valley just downstream of the confluence of the Vorderrhein and Hinterrhein rivers. The deposit is dominated by a 1.6-km-long longitudinal ridge, Ils Aults, and two roughly 600-m-long transverse ridges. Several extensional scarps bear witness to spreading of the deposit. A breach through the deposit, where the Rhine River presently flows, reveals a carapace and intense fragmentation. Exposure dating using cosmogenic ³⁶Cl yields an age of 9420 ± 880 years. This suggests that the Tamins event occurred in a time frame similar to the Flims event but was slightly earlier than the Flims rock avalanche, as also required by stratigraphic relationships. 3D volume modeling reveals bulking of only 14%. The motion of the rock avalanche seems to have occurred first as a flexible block, which underwent fragmentation and simple shearing where the top moved faster than the bottom. The ensuing spreading led to the formation of extensional scarps. There is no identified weak layer along the sliding surface; nevertheless, modeling suggests a friction angle of 10°.

More detailed field measurements are required for a better understanding of surging debris flows. In this work, we analyze a debris flow at the field-scale using timelapse point clouds from a high-resolution, high-frequency 3D LiDAR sensor, which has been installed over a check dam on the fan of the Illgraben catchment in Switzerland. In our investigations, we manually measured the front velocity and tracked individual features such as large boulders and woody debris over a 25 m long channel segment. We observed a change in the front velocity as well as a difference in the velocity of large boulders and woody debris (v_boulder ≈ 0.6 v_wood) during the second surge of the event. We also estimated the discharge for different closely spaced channel sections based on automated measurements of the cross-sectional area and the surface velocity, which enabled us to infer spatial variations in the bed geometry and the velocity profile. From the discharge estimates, we then derived the volume of this event. Over the course of the next year, the amount of field-scale LiDAR data from the Illgraben will increase substantially and allow for an even more detailed analysis of fundamental debris-flow processes.

Accurate debris-flow modelling depends on the ability to simulate surging and pulsing behaviour. However, our understanding of these phenomena is starved of observational constraints. Here we propose the use of seismic measurements, which resolve the arrival of coarse-grained roll wave fronts in debris flows at Illgraben, Switzerland. Roll waves likely play a key role in flow pulses but are typically only observed with point measurements like depth gauges. We compare in-torrent force plate measurements with near-torrent seismic records and discuss how these data can test existing roll wave theories.

Existing work on scene flow estimation focuses on autonomous driving and mobile robotics, while automated solutions are lacking for motion in nature, such as that exhibited by debris flows. We propose DeFlow, a model for 3D motion estimation of debris flows, together with a newly captured dataset. We adopt a novel multi-level sensor fusion architecture and self-supervision to incorporate the inductive biases of the scene. We further adopt a multi-frame temporal processing module to enable flow speed estimation over time. Our model achieves state-of-the-art optical flow and depth estimation on our dataset, and fully automates the motion estimation for debris flows. Source code and dataset are available at project page.

The forced vortex equation, based on the cross-stream inclination of a flow surface as it passes through a bend, is a common approach to estimating debris flow velocities. Here, we present the preliminary results of a study of superelevation and the correction factor k, used to adapt the forced vortex equation to debris flows, based on data from the Illgraben torrent in Switzerland. The definition of the radius of curvature, a factor in the calculation of superelevation velocities, is not found to exercise a large influence on the calculated velocities when using high resolution aerial images, with the choice of cross-section location and k-factor exercising a more significant influence. The k-factors found here fall within the range previously reported in the literature, ranging from approximately 1 to 7, and a previously suggested non-linear relationship with Froude numbers is evident in the dataset. Following the debris flow season of 2022, the study will be continued with additional debris flow events and the investigative methods will be extended to include high-resolution LiDAR sensors installed along the Illgraben torrent.

Rock avalanches can be a significant hazard to communities located in mountainous areas. Probabilistic predictions of the 3D impact area of these events are crucial for assessing rock avalanche risk. Semi-empirical, calibration-based numerical runout models are one tool that can be used to make these predictions. When doing so, uncertainties resulting from both noisy calibration data and uncertain governing movement mechanism(s) must be accounted for. In this paper, a back-analysis of a database of 31 rock avalanche case histories is used to assess both of these sources of uncertainty. It is found that forecasting results are dominated by uncertainties associated with the bulk basal resistance of the path material. A method to account for both calibration and mechanistic uncertainty is provided, and this method is evaluated using pseudo-forecasts of two case histories. These pseudo-forecasts show that inclusion of expert judgement when assessing the bulk basal resistance along the path can reduce mechanistic uncertainty and result in more precise predictions of rock avalanche runout.

The mobility of debris flows is highly variable, which makes runout prediction challenging. Numerical runout models are often used to estimate inundation area, flow depth and velocity for debris-flow analyses. Runout models typically use initiation conditions that disregard the fact that debris flows commonly occur in multiple surges. In this work, we modified the Dan3D runout model to enable a more realistic hydrograph initiation condition. We analyzed a 2019 debris flow at Mount Currie in southwestern British Columbia to examine the effects of complex inflow hydrographs on model outputs. The impact area, flow depths and velocities were sensitive to the inflow hydrograph, particularly depths and velocities higher on the fan. The results from the case study demonstrated that, in addition to variations in material and rheological parameters, the simulated debris flow runout is sensitive to variations in the inflow hydrograph. The numerical model results are also interpreted using an empirical runout relationship to assess the plausibility of the numerical results.

Inversion of low-frequency regional seismic records to solve for a time series of bulk forces exerted on the earth by a landslide (a force-time function) is increasingly being used to infer volumes and dynamics of large, highly energetic landslides, such as rock avalanches and flowslides, and to provide calibration information on event dynamics and volumes for numerical landslide runout models. Much of the work to date using landslide runout modeling constrained by seismic data has focused on using single-phase models with frictional or velocity-weakening rheologies. Awareness of multistage landslide initiations is increasing, with discrete failures separated in time contributing to the final impact of an event. Our work utilizes a method for incorporating seismic data as a calibration constraint for landslide runout models, considering variable rheologies and different initiation conditions. This study presents a systematic examination of multiple rheologies and initiation conditions, and shows how these factors affect the force-time function derived from the landslide runout model. Our work confirms that, while rheology and fragmenting or initially coherent initiations affect the force-time function, multiple collapses separated by tens of seconds have the greatest impact on the shape and amplitude. We apply this method to the analysis of three real rock avalanches to better constrain plausible initiation conditions and rheology parameters using both seismic and field data. This study provides insights on how assumptions about the initiation dynamics of the source zone and the runout model definition can aid in the interpretation of seismic inversions for multistage rock avalanches.

Debris flows affect people and infrastructure around the world, and as a result, many numerical models and modelling approaches have been developed to simulate their impacts. Observations from instrumented debris-flow channels show that variability in inflow depth, velocity, and discharge in real debris flows is much higher than what is typically used in numerical simulations. However, the effect of this natural variability on numerical model outputs is not well known. In this study, we examine the effects of using complex inflow time series within a single-phase runout model utilizing a Voellmy flow-resistance model. The interactions between model topography and flow resistance were studied first using a simple triangular hydrograph, which showed that simulated discharges change because of local slopes and Voellmy parameters. Next, more complex inflows were tested using time series based on 24 real debris-flow hydrographs initiated from three locations. We described a simple method to scale inflow hydrographs by defining a target event volume and maximum allowable peak discharge. The results showed a large variation in simulated flow depths and velocities arising from the variable inflow. The effects of variable-inflow conditions were demonstrated in simulations of two case histories of real debris flows, where the variation in inflow leads to significant variations in the simulation outputs. The real debris-flow hydrographs were used to provide an indication of the range of impacts that may result from the natural variability in inflow conditions. These results demonstrate that variation in inflow conditions can lead to reasonable estimates of the potential variation in impacts.

Investigating surface displacements in high alpine environments is often subject to challenges due to the difficult accessibility or harsh climatic conditions. Measurement systems have improved greatly in recent years regarding accuracy, range, or energy consumption. Continuously receiving high-precision, real-time monitoring data from a remote location can still support a better understanding of slope dynamics and risk. We present the design, construction, operation, and performance of a complex surface displacement monitoring system installed in the surroundings of the Great Aletsch Glacier in the Swiss Alps, based on two robotic total stations to continuously measure 3D displacements with high accuracies. In addition, GNSS stations are also considered in order to pass from a local to a geographic reference system, as well as to improve the measurement accuracy. The monitoring network is aimed at studying several types of deformation processes, i.e., (i) gravitationally driven and irreversible rockslide movements around the tongue of the Great Aletsch Glacier, (ii) reversible rock slope deformations caused by annual cycles of groundwater recharge and depletion, and (iii) small irreversible deformations of stable rock slopes resulting from progressive rock damage driven by glacier retreat and cyclic hydraulic and thermal loading. We describe the technical details of the monitoring system, which has been in operation successfully for 6 years, and discuss the system performance in terms of its robustness and accuracy.