Four decades of observations from NGI's full-scale avalanche test site Ryggfonn—Summary of experimental results

Abstract

The Norwegian Geotechnical Institute (NGI) has run full-scale avalanche experiments at the Ryggfonn test-site in Western-Norway for close to 40years. The construction of an avalanche catching dam in 1981 laid the cornerstone of the so-called “Ryggfonn project”. Over these years of operation, various kinds of instrumentation and structures have been placed along the avalanche path to gain in-depth understanding of avalanche dynamics and the interaction of avalanches with structures. Avalanche measurements provide benchmarks for the development and calibration of numerical avalanche models. Increasingly, these models are used for hazard zoning to estimate runout distances or impact pressures with varying return periods for assessing endangered areas. However, these models are imperfect and require a high degree of expert judgment for specifying the required model parameters. At this point, it is valuable to have reference events to evaluate simulation results.In this paper, we summarize runout, velocity, and impact pressure observations from the Ryggfonn test site as well as some derived quantities such as the retarding acceleration or density estimates. We try to relate the measurements to ambient (in-situ) conditions during the events. For Ryggfonn, the runout observations suggest a 10 to 15% increase in runout distance comes along with a decrease of a factor 10in probability. The expected front velocity of an avalanche as it enters the runout area at Ryggfonn is about 27 ± 10ms−1, but may reach 50ms−1. The impact pressures at the beginning of the runout area are typically of the order of 100kPa, but may reach several hundred kPa. Observations from a power line assembly give some insight in the vertical pressure distribution and its dependency of the avalanche velocity. The combination of these measurements provides some implications for the avalanche density suggesting a range of flowing densities. Our observations affirm a dependency of dynamical parameters on the ambient conditions, which were made at other locations.