Item: Operational Use of a Statistical Forecasting Tool on the Seward Highway, Alaska
-
-
Title: Operational Use of a Statistical Forecasting Tool on the Seward Highway, Alaska
Proceedings: International Snow Science Workshop 2014 Proceedings, Banff, Canada
Authors:
- Matt Murphy [ Alaska Department of Transportation and Public Facilities, Anchorage, AK, USA ]
- Jordy Hendrikx [ Snow and Avalanche Laboratory, Department of Earth Sciences, Montana State University, Bozeman, MT, USA. ]
Date: 2014-09-29
Abstract: A variety of statistical avalanche forecasting tools using meteorological and snowpack predictors have been developed in many different snow avalanche environments around the World. They have also been used in "quasi-operational" settings in many places with mixed success. Despite numerous attempts using a range of statistical approaches, in different snow and avalanche climates, these statistical models tend to reach a statistical prediction ceiling of approximately 80-90% for overall accuracy. While encouraging, and of utility for learning and the transfer of institutional memory, this level of accuracy is still not widely considered adequate for real time operational uses. This paper will present an operational use case for a dual classification tree model which has been imbedded into the operational forecasting program for the Seward Highway in Alaska for the last two winters. We will discuss how this model is used as a tool, as part of the daily routine, and how the limitations in the model are operationally addressed. Of particular interest is that this model was of more use to predict the end of avalanche activity, rather than the onset of activity, which was typically more obvious. We will present two seasons of results and a case study to illustrate the utility of the model. These results have particular relevance to avalanche programs that want to consider incorporating a statistical tool as part of their program.
Object ID: ISSW14_paper_P1.17.pdf
Language of Article: English
Presenter(s):
Keywords: Avalanche Forecasting, Statistical Forecasting, Operational Decisions
Page Number(s): 511-515
-