Piz Cengalo
(Hero map: copyright swisstopo)
The catastrophic 2017 Piz Cengalo rock-ice avalanche revealed the complex challenges of rock–ice avalanche modeling and marked the first real-world application of the RAMMS software to simulate such an event. In the immediate aftermath, attention focused not only on the massive collapse itself, but on the devastating debris flows that followed, reaching and severely damaging the village of Bondo. The Cengalo disaster highlighted the need for models that could simulate flow regime transitions and the composition and wetness of avalanche deposits.
Video 1: RAMMS::RockIce simulation of the 2017 rock-ice avalanche at Piz Cengalo.
Figure 1: One minute after initiation, the Cengalo rock/ice avalanche descends rapidly, with the frontal mass – a mixture of rock and ice -traveling at approximately 60 m/s and reaching flow heights of 10 to 20 meters. A dense powder cloud envelops much of the steep valley slope as the avalanche turns. Deposition heights at the front reach up to 30 meters.
Initial simulation efforts employed RAMMS::Extended, which accounted for phase changes and water transport, assuming that meltwater remained bound to the solid phase. While this method successfully captured the overall behavior—including the generation of large meltwater volumes—it proved insufficient for understanding the flow’s true composition. The avalanche core was a heterogeneous mixture of rock, ice, and water, yet the Extended model could not distinguish between these components. It applied a single friction law to the entire flow, limiting insight into the internal flow dynamics and therefore the possibility of secondary debris flows.
This challenge prompted the development of a new module: RAMMS::RockIce. Unlike earlier models, it was designed to handle multiphase flows by explicitly modeling rock, ice, and water as separate components. Crucially, it introduced three independent heat energy equations, allowing the simulation to track the temperature evolution of rock, ice, and water individually. Not only was ice heated by internal shearing, but also by convective heat transfer with the rock and water components. This capability proved essential for explaining why large quantities of ice were still present in the final deposits, and how localized meltwater accumulation triggered secondary debris flow activity.
Figure 2: Calculated maximum velocities of the Piz Cengalo rock/ice avalanche. The red lines indicate three distinct phases of the event: (1) the lateral extent of the rock/ice avalanche prior to deposition, (2) the area covered by the avalanche deposits, and (3) the width of secondary debris flows triggered by water or meltwater accumulation within the deposits. A debris flow surge is beginning to escape the avalanche deposits
The goal of the RAMMS::RockIce simulations was therefore to assess whether the metastable avalanche deposits contained sufficient meltwater to initiate debris flows. In the model setup, a 2.9 million m³ rock mass (approx. 40 m in height) detached and impacted a 610,000 m³ glacier, which failed instantly upon impact. The falling rock was modeled as ice-free but with 5 percent water by volume, while the glacier ice mass contained a slightly higher 10 percent water content.
Simulation results indicate that the avalanche attained a peak velocity of 80 m/s, with flow heights of rock and ice ranging between 10 and 20 meters. Meltwater generation began immediately upon impact with the glacier surface, triggered by the intense friction and heat from the descending rock mass. However, the majority of meltwater was produced during the final stages of the event, as the accumulated thermal energy within the rock effectively melted entrained ice. In the deposition zone, water saturation levels reached as high as 50 percent, underscoring the significant role of thermal-ice interactions in the later phases of the flow.
Figure 3: Calculated water content. Over 2m of water accumulated in the avalanche deposits. The location of the water content revealed the starting zones of the secondary debris flows.
Video 2: Simulated Water Content in the Cengalo Avalanche Deposits.
A crucial component of rock-ice avalanche hazard analysis is the ability to accurately predict both the distribution and quantity of water within the avalanche deposits. This information is essential for evaluating the potential for secondary debris flows, which may be triggered by the remobilization of water-saturated materials. In the case of the Cengalo event, the simulation reveals a significant concentration of water near the front of the deposit (highlighted in red), indicating a high-risk zone for potential post-event flow activity.
Video 3: Time-Evolution of Meltwater Production in the Cengalo Avalanche.
This video illustrates the dynamic progression of meltwater generation during the Cengalo avalanche event. Meltwater is initially produced upon impact of the rock mass with the glacier; however, the majority is generated later in the event, particularly within the avalanche’s frontal lobe. This late-stage meltwater accumulation played a critical role in triggering secondary debris flows, underscoring the importance of accurately modeling melt processes in rock-ice avalanche hazard assessments.
