Transferability of a calibrated numerical model of rock avalanche run-out : application to 20 rock avalanches on the Nuussuaq Peninsula, West Greenland.

• Published in: Earth Surface Processes and Landforms
• Main Authors: Benjamin, J., Rosser, N.J., Dunning, S.A., Hardy, R.J., Kelfoun, K., Szczuciński, W.
• Format: Article in Journal/Newspaper
• Language: unknown
• Published: John Wiley 2018
• Subjects: Greenland; Nuussuaq; Paatuut; Vaigat
• Online Access: http://dro.dur.ac.uk/25473/; http://dro.dur.ac.uk/25473/1/25473.pdf; https://doi.org/10.1002/esp.4469

Long run‐out rock avalanches are one of the most hazardous geomorphic processes, and risk assessments of the potential threat they pose are often reliant on numerical modelling of their potential run‐out distance. The development of such models requires a thorough understanding of past flow behaviour inferred from deposits emplaced by previous events. Despite this, few records exist of multiple rock avalanches that occurred in conditions sufficiently consistent to develop a set of more generalised, and hence transferrable, rules. We conduct field and imagery‐based mapping and use numerical modelling to investigate the emplacement of 20 adjacent rock avalanches on the southern flanks of the Nuussuaq peninsula, West Greenland. The rock avalanches run out towards the Vaigat Strait, and are sourced from a range of coastal mountains of relatively uniform geology. We calibrate a three‐dimensional continuum dynamic flow code, VolcFlow, with data from a modern, well‐constrained event that occurred at Paatuut (AD 2000). The best‐fit model assumes a constant retarding stress with a collisional stress coefficient, simulating run‐out to within ±0.3% of that observed. This calibration was then used to model the emplacement of deposits from five other neighbouring rock avalanches before simulating the general characteristics of a further 14 rock avalanche deposits on simplified topography. Our findings illustrate that a single calibration of VolcFlow can account for the observed deposit morphology of a uniquely large collection of rock avalanche deposits, emplaced by a series of events spanning a large volume range. Although the prevailing approach of tuning models to a specific case may be useful for detailed back‐analysis of that event, we show that more generally applied models, even using a single pair of rheological parameters, can be used to model potential rock avalanches of varied volumes in a region and, therefore, to assess the risks that they pose.