The disaster in Blatten

Abstract/Description

The rockslide and glacier collapse in Blatten was the fourth largest rockslide ever observed in Switzerland and, due to the combination of several collapse processes, highly complex, if not unique.

However, the data collected during this event can certainly be described as unique. It is no exaggeration to say that no similar event has ever been observed and monitored so closely. This presents a unique opportunity to analyze the event, learn from it, and gain a better understanding of the relevant processes. The canton of Valais therefore wishes to make the data collected during the event available to the scientific community.

In addition to an overview of this data, we will present a brief chronology of the event. We will also focus on the practical questions that remain open to us, which we hope to answer together with the scientific community.

Abstract/Description

In this contribution we will present results obtained by performing a targeted back-analysis of synthetic aperture radar dataset over the Kleines Nesthorn. The results are key to understand potential and challenges of early detection and warning based on satellite radars.

Abstract/Description

We have analyzed a number of digital elevation models (DEMs) and orthophotos to reconstruct the pre-event behavior of Birchglacier, the volumes that were involved in the catastrophic collapse, as well as the dynamics of the runout of the rock-ice avalanche. The analysis of the pre-event DEMs shows the unusual behavior of the Birchglacier in the years prior to the event. Between 2011 and 2025, Birchglacier advanced by several tens of meters, thinning up-stream and thickening at the terminus. It is critical to understand how this behavior may have influenced the 2025 collapse.

We also compared pre- and post event DEMs to estimate the volumes, the deposited and eroded amount of material that was involved in the event. From orthophotos we mapped deposit structures that can help reconstruct the dynamics of the event, offering valuable information to calibrate numerical models.

Abstract/Description

Ongoing atmospheric temperature rise causes deep warming of perennially frozen mountain slopes. The thereby governing process is heat diffusion, in places markedly slowed down by latent heat exchange from local ice melt. This extremely slow combination of processes induces strong, deep and long-lasting delays in thermal disturbance at depth. Such pronounced paleoclimatic effects also affect the mechanical conditions and hydraulic characteristics of rock masses with ice-filled cracks and fissures, the strength, permeability and stability of which are known to be strongly temperature-dependent.

In order to better understand deep, warming-induced effects in the permafrost slope at Kleines Nesthorn, the destabilization of which initiated the catastrophic destruction by a large rock-ice avalanche of the village of Blatten, Valais Alps, Switzerland, a combined thermo-mechanical 2D model calculation is being carried out for the detachment zone before its failure. Present-day near-surface temperatures are first parameterized. A paleo-correction is then applied for an assumed steady-state temperature field at the end of the Little Ice Age around 1850. In a next step, this assumed steady state 2D temperature field is used for a transient calculation of present-day temperature distribution at depth. The resulting changes in the 2D-temperature field at depth are finally interpreted in view of stability changes following laboratory results reported in the literature and geotechnical slope analyses using various shear horizons.

First results indicate a characteristic asymmetric permafrost in a ridge with a warm and a cold side. Depths of permafrost on the cold destabilized slope reached up to more than 100 meters below surface. Since the Little Ice Age, warming and related stability decrease have reached down to roughly 50 – 100 meters below surface, hence affecting large frozen rock masses. At Kleines Nesthorn, permafrost warming and related mechanical weakening at centennial to decadal and annual time scales may most probably have ultimately triggered the failure of a rock mass, which had reached sub-critical topographic and geological stability conditions through much longer time scales. Such a situation and development in time is likely to occur in many places of the Alps and of cold mountains worldwide.

Abstract/Description

On May 28, 2025, a massive rock-ice avalanche struck near Blatten (Lötschental Valley, Switzerland), triggering seismic waves equivalent to a magnitude 3.1 earthquake. While the 300 residents were evacuated in time, one shepherd was killed and 90% of the village was buried. The event underscores rising alpine hazards linked to permafrost thaw and glacier destabilization, and the need for effective monitoring and early warning systems.

To better understand the dynamics of such events, we conduct research that combines seismic signal analysis, geomorphological observations, and numerical simulations. Since the pioneering work of Kanamori and Brodsky, it has been well established that landslides, including rock, debris, and snow avalanches, generate seismic waves that contain critical information about the source: flow duration, mobilized mass, granular friction, particle impacts on the ground, and interactions with water.

These seismic waves, recorded by dense sensor networks at distances of up to 1000 km depending on landslide volume and frequency content, exhibit a broad spectral range. High-frequency waves are primarily generated by grain-scale flow processes, while long-period waves are linked to the overall accelerations and decelerations of the flow. High-frequency signals are significantly distorted by topography and subsurface heterogeneities, making their interpretation challenging. In contrast, long-period waves are less sensitive to such effects and can be used to retrieve the force exerted by the landslide on the ground through seismic waveform inversion.

This force is obtained by deconvolving the low-frequency seismic signal with the Earth’s Green’s function and provides a means to interpret the mechanical characteristics of the flow. It can be analyzed using physical models ranging from single-block approximations to more advanced granular flow models. However, the inverted force is inherently limited by the bandwidth of the seismic instruments, which may distort its amplitude, duration, and even polarity.

To overcome these limitations, we incorporate additional constraints from independent observations, such as deposit geometry and volumes derived from differential digital elevation models, as well as from numerical simulations, which provide both physical constraints and a means to explore plausible scenarios of the flow dynamics. In this contribution, we introduce the general principles of our combined approach, using the Blatten avalanche as a case study and as an entry point to the more specialized analyses conducted by our research group.

Abstract/Description

Rock avalanches like in Blatten can not be mitigated with structural measures but only with organizational measures, for which the prediction of the expected runout is crucial. While the reconstruction of rock avalanches is rather easy with an appropriate model, their prediction is still challenging due to uncertainties in the parameterization. The runout of the rock avalanche in Blatten is reconstructed with a Savage-Hutter model, achieving an areal match about 60 % regarding the complete envelope and about 80 % excluding the distal part of the run up. Additionally, the reconstruction with a simple Fahrböschung model achieves an areal match of almost 100 % including the distal part of the run up but overestimating the lateral spread. The only parameter needed for these models is the kinetic friction angle, which is determined by the estimated volume of 9000000 m³ to 19° using the empirical formula of Scheidegger, which also fits well to the geometric value estimated from the envelope. This reconstruction shows a promising path for future predictions that could overcome parameterization issues. Machine learning could analyze the kinetic friction angle and various covariates from past rock avalanches in order to predict values with probabilities of exceedance for parameterizing and predicting future rock avalanches. Assigning a probability of exceedance to the kinetic friction angle would enable a transformation from a deterministic to a probabilistic approach. Furthermore, several models such as the Savage-Hutter model and the Fahrböschung model could be combined in an model ensemble. Such an approach would combine statistical analyses with physical models and could contribute to robust predictions for the runout of rock avalanches in preliminary assessments.