International Snow Science Workshop Proceedings - Montana State University Library

Item: Statistical analysis and operational avalanche forecasting on the roads of northern Gaspésie, Québec, Canada

• • Title: Statistical analysis and operational avalanche forecasting on the roads of northern Gaspésie, Québec, Canada

    Proceedings: International Snow Science Workshop Proceedings 2018, Innsbruck, Austria

    Authors:

    • Francis Gauthier  [ Centre d’Études Nordiques (CEN), Laboratoire de géomorphologie et de gestion des risques en montagne (LGGRM), Université du Québec à Rimouski (UQAR) ]
    • Frédéric Banville-Côté  [ Centre d’Études Nordiques (CEN), Laboratoire de géomorphologie et de gestion des risques en montagne (LGGRM), Université du Québec à Rimouski (UQAR) ]  [ Ministère des Transports, de la Mobilité durable et de l’Électrification des transports ]
    • Dominic Boucher  [ Avalanche Québec ]
    • Francis Meloche  [ Centre d’Études Nordiques (CEN), Laboratoire de géomorphologie et de gestion des risques en montagne (LGGRM), Université du Québec à Rimouski (UQAR) ]  [ Avalanche Québec ]

    Date: 2018-10-07

    Abstract: Snow avalanches are a major natural hazard for road users and infrastructure in northern Gaspésie (Eastern Canada). Over the past 15 years, the occurrence of 642 snow avalanches on the two major roads servicing the area was reported by the Ministère des Transports, de la Mobilité durable et de l’Électrification des transports (MTMDET). Since 2016, Avalanche Québec (AvQc) issue avalanche bulletins for the MTMDET based on snowpack and weather analysis. As academic contributors, our research focus on analyzing the weather patterns promoting snow avalanche initiation and developing statistical tools to forecast snow avalanches on a daily basis. Using logistic regression (LR) and classification tree (CT) we analyzed 15 years of weather and avalanche data on a regional and local (road sections along the coast and inland) scale. We then test the best LR and CT models over the last three seasons in an operational forecasting perspective: each day, the probability of occurrence (LR) and the prediction (CT) computed by the models were compared to the avalanche hazard issue by AvQc. We also document the model performance using different forecast verification methods. Since most of the avalanche hitting the road seems to be controlled by direct snow loading more than persistent avalanche problems, stochastic models remain a highly effective complementary tool for the forecasters. Finally, we discuss the effects of climate change on avalanche activities.

    Object ID: ISSW2018_O12.6.pdf

    Language of Article: English

    Presenter(s):

    Keywords: Snow avalanche forecast, logistic regression, classification tree

    Page Number(s): 1126-1130

  • Get Full Article

Back to Home page