Abstract
During the ski season, professional avalanche forecasters working for the Utah Department of Transportation (UDOT) monitor one of the most dangerous highways in the world. These forecasters continually evaluate the risk of avalanche activity and make road closure decisions.
Previous research on this topic was funded, in part, by a grant from the National Science Foundation (SES-9212017).
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
See Bowles and Sandahl (1988).
- 2.
See Abromeit (2004).
- 3.
See Perla (1991).
- 4.
The word redundant more generally than correlation is used in this paper. This indicates when several variables are designed to measure the same thing or may be functions of each other.
- 5.
In computing SZAVLAG the measure which we use is the American size measure, which is perhaps less appropriate than the Canadian size measure. However, a similar adjustment might be relevant.
- 6.
Chipman et al. (2009).
- 7.
- 8.
This partition scores poorly but is only used to illustrate the concepts.
- 9.
The default value for a, 0.95, is selected. This implies a high likelihood of a split at the root node with a decreasing probability as the depth of the tree, d, increases. The default value of b is 2. However, b = 0.5 is used to obtain bushier trees (trees with more terminal nodes). The story used in the text had forecasters using less than 10 variables, but at least 3.
- 10.
Details are in Blattenberger and Fowles (1994, 1995). UDOT data indicate that, on average, there are 2.6 persons per vehicle, 2.5 of which are skiers. Of these skiers, 40 % are residents who spend an average of $19 per day at the ski resorts (1991 dollars). Sixty percent tended to be nonresidents who spent an average of $152 per day (1991). A road closure results in a revenue loss in 2005 of over $2.25 million per day based on average traffic volume of 5,710 cars during the ski season.
- 11.
The road must be closed while artificial explosives are used.
References
Abromeit D (2004) United States military for avalanche control program: a short history in time. In: Proceedings of the international symposium on snow monitoring and avalanches
Blattenberger G, Fowles R (1994) Road closure: combining data and expert opinion. In: Gatsonis et al. (eds.) Case studies in bayesian statistics, Springer, New York
Blattenberger G, Fowles R (1995) Road closure to mitigate avalanche danger: a case study for little Cottonwood canyon. Int J Forecast 11:159–174
Bowles D, Sandahl B (1988) Avalanche hazard index for highway 210—little Cottonwood Canyon mile 5.4–13.1. mimeographed
Chipman H, George EI, McCulloch RE (1998) Bayesian CART model search. J Am Stat Assoc 93(443):935–948
Chipman H, George EI, McCulloch RE (2009) BART: bayesian additive regression trees, http://CRAN.R-project.org/package=BayesTree
Chipman H, George EI, McCulloch RE (2010a) BART: bayesian additive regression trees. Ann Appl Stat 4:266–298
Chipman H, George EI, Lemp J, McCulloch RE (2010b) Bayesian flexible modeling of trip durations. Transp Res Part B 44:686–698
LaChapelle ER (1980) Fundamental processes in conventional avalanche forecasting. J Glaciol 26:75–84
Perla R (1970) On contributory in avalanche hazard evaluation. Can Geotech J 7:414–419
Perla R (1991) Five problems in avalanche research. In: CSSA symposium
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer Science+Business Media New York
About this chapter
Cite this chapter
Blattenberger, G., Fowles, R. (2014). Avalanche Forecasting: Using Bayesian Additive Regression Trees (BART). In: Alleman, J., Ní-Shúilleabháin, Á., Rappoport, P. (eds) Demand for Communications Services – Insights and Perspectives. The Economics of Information, Communication, and Entertainment. Springer, Boston, MA. https://doi.org/10.1007/978-1-4614-7993-2_11
Download citation
DOI: https://doi.org/10.1007/978-1-4614-7993-2_11
Published:
Publisher Name: Springer, Boston, MA
Print ISBN: 978-1-4614-7992-5
Online ISBN: 978-1-4614-7993-2
eBook Packages: Business and EconomicsEconomics and Finance (R0)