Item: Sensitivity Analysis of the RAMMS Avalanche Dynamics Model in a Canadian Transitional Snow Climate
-
-
Title: Sensitivity Analysis of the RAMMS Avalanche Dynamics Model in a Canadian Transitional Snow Climate
Proceedings: International Snow Science Workshop Proceedings 2018, Innsbruck, Austria
Authors:
- Ryan Buhler [ Dynamic Avalanche Consulting Ltd., Revelstoke, BC, Canada ]
- Chris Argue [ Dynamic Avalanche Consulting Ltd., Revelstoke, BC, Canada ]
- Bruce Jamieson [ Snowline Associates Ltd., Calgary, AB, Canada ]
- Alan Jones [ Dynamic Avalanche Consulting Ltd., Revelstoke, BC, Canada ]
Date: 2018-10-07
Abstract: The Swiss avalanche dynamics model RAMMS is widely used to simulate avalanche flow for engineering, research, and hazard assessment. Since 2015, RAMMS modelling has been used as part of the Glacier National Park Avalanche Mitigation Project in British Columbia, Canada. Using 51 years of historic occurrence data from Parks Canada and a high resolution digital elevation model, modelled avalanches are calibrated to their predicted runout position. The sensitivity of the release volume in RAMMS is tested by varying the initially calibrated release volume for four paths and measuring changes to the runout distance and avalanche velocity in the runout zone. For each 10% increase in release volume, the runout distance increased by an average of 9 m and the velocity in the area of the runout zone with a 10-15° slope angle increased by an average of 1.0 m/s. The sliding and turbulent friction values used in RAMMS are functions of the return period and the avalanche volume classifications. Each of these classifications is tested for their effect on the runout distance and avalanche velocity in the runout zone. Results show that changes in avalanche volume classification have a larger effect on both runout distance and avalanche velocity compared to the return period classification. For each single step increase in volume classification, the runout distance and velocity increased by an average of 49 m and 4.4 m/s, respectively. Comparatively, for each single step increase in return period, the runout distance and velocity increased by an average of 12 m and 0.8 m/s, respectively. These results highlight the importance of choosing appropriate input parameters and testing the sensitivity of the model results to these inputs, especially in North America where limited calibration of the RAMMS model has been completed compared to Europe.
Object ID: ISSW2018_P08.13.pdf
Language of Article: English
Presenter(s):
Keywords: Dynamic modelling, RAMMS, sensitivity analysis, avalanche runout.
Page Number(s): 767-771
-