The research team, led by scientists at the University of Bern, combined detailed geomorphological field mapping, surface exposure dating with the cosmogenic nuclide chlorine-36, and numerical runout modeling to reconstruct the landslide from beginning to end. Their central finding is striking: the Sierre rock avalanche occurred about 10.7 thousand years ago, give or take 1.5 thousand years, at the transition from the Younger Dryas cold period to the Early Holocene warming. By that time, the Rhone valley had been ice-free for thousands of years, meaning the collapse happened on dry, exposed slopes rather than onto or against a glacier. This conclusion directly challenges hypotheses proposed since the late nineteenth century, which variously invoked the Rhone trunk glacier, tributary valley ice, or icebergs rafting debris across a proglacial lake to explain the strange geometry of the deposit.

The scale of the event is difficult to overstate. The release area, located on the northern valley slope between the municipalities of Salgesch and Varen at elevations between roughly 700 and 2000 meters above sea level, spans about 5.9 square kilometers. The researchers estimated that approximately 1.4 cubic kilometers of rock detached from the source zone, and after accounting for the roughly 20 percent volume increase that occurs when solid rock breaks into a loose, porous debris mass, the total deposit volume at the time of failure reached about 1.7 cubic kilometers. Today, the surviving deposit holds an estimated 0.9 cubic kilometers of material spread over 24.3 square kilometers, with thicknesses ranging from less than a meter to as much as 87 meters in places. The difference between the emplaced volume and what remains today points to substantial erosion and reworking by the Rhone River in the millennia since the collapse.

Dating such an ancient landslide required careful sampling. Because the bedrock involved is limestone, chlorine-36 is the only cosmogenic nuclide suitable for surface exposure dating in this setting. The team identified about 170 large boulders protruding from the deposit and applied strict selection criteria: each candidate had to measure at least 2.5 meters along its long axis, protrude at least half a meter above the ground, sit on stable terrain, and carry minimal vegetation that could shield the surface from cosmic rays. Thirteen boulders met these standards, and the researchers also collected two samples from the exposed detachment surface itself. All fifteen samples were processed at the Surface Exposure Dating Laboratory at the University of Bern and measured at the 6 MV TANDEM accelerator mass spectrometer at ETH Zurich, where chlorine-36 to chlorine ratios were determined using the isotope dilution method.

The resulting exposure ages ranged widely, from about 2.3 to 13.8 thousand years before present when calculated without erosion correction, and from 2.4 to 15.1 thousand years after applying a conservative denudation correction of 10 millimeters per thousand years. To make sense of this scatter, the team used a statistical clustering tool called P-CAAT, which identified four probability peaks: around 2.5, 6.1, 10.7, and 14.4 thousand years ago. The youngest two clusters were rejected as failure ages because wood fragments embedded in debris-flow deposits that cut into the landslide near Salgesch had previously been radiocarbon dated to more than 9 thousand years old, providing an independent minimum age that rules out anything younger. Those younger boulder ages instead reflect post-depositional disturbance, such as the gradual exhumation of boulders once buried under a thin debris drape or the loss of a few centimeters of surface rock through spalling and frost damage.

The oldest cluster, at 14.4 thousand years, was likewise set aside, but for a different reason: inheritance. Some boulders likely sat on the pre-landslide slope and absorbed cosmic radiation before the collapse, carrying a pre-existing chlorine-36 inventory onto the passive carapace of the moving mass. The bedrock sample from the failure surface showed a similar excess, explainable if rupture stripped away only 10 to 30 centimeters of previously exposed rock rather than the full cosmic-ray penetration depth. With the youngest and oldest modes explained away, the dominant cluster of six samples at 10.7 plus or minus 1.5 thousand years stands as the most plausible failure window, placing the Sierre collapse squarely at the dawn of the Holocene.

To understand how the landslide moved, the team turned to DAN3D, a semi-empirical dynamic modeling program that simulates the flow of large debris masses over three-dimensional terrain. The first step was reconstructing the pre-landslide topography by adjusting modern elevation data with field observations, cross-sections, and borehole records, including a modeled lowering of the valley floor by 15 to 20 meters to account for post-failure sedimentation. The researchers then ran 21 simulation scenarios, varying the rheological parameters that govern how the sliding mass resists motion. The best match to the mapped deposit came from a multi-material scenario, labeled V20, which applied a friction coefficient of 0.05 in the source zone and an exceptionally low value of 0.0125 in the valley sediments, with a turbulence coefficient of 500 square meters per second squared held constant.

The winning simulation paints a vivid picture of the catastrophe. The collapse began at roughly 2000 meters elevation, and within the first 10 seconds the detached mass was still largely confined to the release area. By 50 seconds it had reached the valley floor and begun spreading laterally. At 60 seconds it slammed into the opposite valley wall, and by 100 seconds it was dispersing broadly across the flat terrain. The entire 12-kilometer journey down the valley was complete within about 10 minutes, with maximum modeled velocities reaching 138 meters per second, equivalent to nearly 500 kilometers per hour. The total transport distance from source to the farthest deposit was about 15 kilometers, corresponding to a height-to-length ratio of 0.1 and a travel angle of just under 10 degrees, figures that place Sierre among the longest-runout landslides ever documented in the Alps.

The exceptionally low friction coefficient required for the valley sediments is the key to this extraordinary runout. Modest changes in the basal friction coefficient shifted the modeled runout terminus substantially, whereas variations in the turbulence coefficient had only a second-order effect. Comparisons with other well-studied Alpine rock avalanches reinforce this picture: the Kander landslide required a segmented parameterization with friction values ranging from 0.04 to 0.4 across different path segments, while the Flims rockslide, the largest Alpine landslide at 11.3 cubic kilometers, achieved its tens-of-kilometers runout with an intermediate bedrock friction coefficient of 0.16, aided by sheer mass. The Sierre values, particularly the 0.05 friction coefficient in the source zone, are lower than typical bedrock shearing values, which the authors suggest may reflect underestimated slope angles in the distal runout area rather than truly anomalous material behavior.

Placed in the broader context of Alpine post-glacial landscape evolution, the Sierre landslide marks the onset of a remarkable episode of instability. It postdates a series of smaller Lateglacial failures in the Eastern Alps, such as Hohe Munde, Auernig, Almtal, and Pletzachkogel, and precedes the giant Early Holocene collapses at Flims, Köfels, and Tamins, all of which exceeded one cubic kilometer. Notably, most of the largest Alpine rock avalanches, including Sierre, Flims, and Tamins, occurred in massive Mesozoic limestone of the Swiss carbonate Alps, while failures in crystalline bedrock rarely exceed 0.4 cubic kilometers, with Köfels as the sole major exception. The timing of the Sierre collapse, coinciding with rapid Early Holocene warming and the debuttressing of slopes recently relieved of glacial ice, fits a paraglacial framework in which climate-driven environmental change destabilizes slopes over millennia. As modern climate change accelerates permafrost thaw and extreme precipitation in the Alps, and as urban development continues to expand into hazard-prone valleys, understanding the mechanics and timing of ancient failures like Sierre becomes not just an academic exercise but a practical foundation for assessing the risks that tomorrow’s slopes may pose.

Subject of Research: Timing, dynamics, and glacial history of the Sierre rock avalanche in the Swiss Rhone valley

Article Title: The Sierre landslide: chronology, dynamics, and insights into Alpine post-glacial instabilities

Article References: The Sierre landslide: chronology, dynamics, and insights into Alpine post-glacial instabilities. (n.d.). https://doi.org/10.3897/sjg.119.180346

Image Credits: AI Generated

DOI: 10.3897/sjg.119.180346

Keywords: Sierre landslide, rock avalanche, cosmogenic chlorine-36 dating, surface exposure dating, Rhone valley, Swiss Alps, DAN3D modeling, Younger Dryas, Early Holocene, paraglacial slope failure, landslide hazard, tuma hills