Premium
Predicting snow velocity in large chute flows under different environmental conditions
Author(s) -
Rougier Jonathan,
Kern Martin
Publication year - 2010
Publication title -
journal of the royal statistical society: series c (applied statistics)
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.205
H-Index - 72
eISSN - 1467-9876
pISSN - 0035-9254
DOI - 10.1111/j.1467-9876.2010.00717.x
Subject(s) - snow , inference , bayes' theorem , hazard , probabilistic logic , bayesian inference , computer science , econometrics , environmental science , statistics , bayesian probability , mathematics , meteorology , artificial intelligence , geography , chemistry , organic chemistry
Summary. Observations, model evaluations and expert judgements are combined to make predictions of snow velocity in large chute experiments. Different experimental variables, namely the environmental conditions snow density and snow surface temperature, affect all aspects of this inference. We show how the effect of these two variables can be incorporated in our judgements regarding the uncertain parameters of the physical model, the discrepancy between the physical model and reality and the observation error. We adopt a Bayes linear approach to avoid the necessity of fully probabilistic belief specifications and demonstrate visual tools for statistical validation. Our results represent an important first step in improving the specification of uncertainty in model‐based avalanche hazard mapping.