scispace - formally typeset
Search or ask a question

A Model for Avalanche Forecasting on the Bonaigua Pass, Spain, Using Classification Trees

TL;DR: In this paper, the authors used a classification tree method to determine periods of significant avalanche activity in terms of the predefined avalanche day concept, which is performed for the entire road in a combined analysis and also for three individual sub-areas within the Bonaigua Pass.
Abstract: The highway C-28 is located in the Central Pyrenees and links the Aran valley with Catalonia along 20 km over the Bonaigua Pass. It constitutes a key access route for winter visitors. Most of the slopes affecting the road face to the south, with heights varying between 1600 and 2300 meters. We started from 12 years of meteorological and avalanche data collected by the local avalanche warning service of Aran Valley. Weather data were obtained from two automatic weather stations and a flowcapt, whereas avalanche activity was manually recorded in a GIS. We selected several weather parameters including snow drift, elapsed time, trend and categorical parameters. Using a classification tree method, we have developed a model to determine periods of significant avalanche activity in terms of the predefined avalanche day concept. The model is performed for the entire road in a combined analysis and also for three individual sub-areas within the Pass. Results showed that conventional factors describing snow depth were more significant than temperature and precipitation factors. Derived snow drift parameters from snow depth and water precipitacion showed more importance than drift data from the flowcapt. Radiation and wind direction variables had low importance in all the tests. The detailed analysis by subareas has not achieved the objective due to the reduction of the database. However, it has allowed to confirm the differences between one of the sub-areas and the dynamics of the rest of the highway.

Content maybe subject to copyright    Report

Citations
More filters
Journal Article
TL;DR: In this article, a CART inductive model based on Breiman's algorithm was proposed to estimate the probability of intrahospital death from acute myocardial infarction (AMI).

2 citations

References
More filters
Journal ArticleDOI
TL;DR: This article gives an introduction to the subject of classification and regression trees by reviewing some widely available algorithms and comparing their capabilities, strengths, and weakness in two examples.
Abstract: Classification and regression trees are machine-learning methods for constructing prediction models from data. The models are obtained by recursively partitioning the data space and fitting a simple prediction model within each partition. As a result, the partitioning can be represented graphically as a decision tree. Classification trees are designed for dependent variables that take a finite number of unordered values, with prediction error measured in terms of misclassification cost. Regression trees are for dependent variables that take continuous or ordered discrete values, with prediction error typically measured by the squared difference between the observed and predicted values. This article gives an introduction to the subject by reviewing some widely available algorithms and comparing their capabilities, strengths, and weakness in two examples. © 2011 John Wiley & Sons, Inc. WIREs Data Mining Knowl Discov 2011 1 14-23 DOI: 10.1002/widm.8 This article is categorized under: Technologies > Classification Technologies > Machine Learning Technologies > Prediction Technologies > Statistical Fundamentals

16,974 citations


"A Model for Avalanche Forecasting o..." refers methods in this paper

  • ...This is based on the algorithm developed in publishing Breiman (1984). CART is a non - parametric binary segmentation method which aims to generate homogeneous populations....

    [...]

  • ...The main objective of this work is to use this database to determine the most significant meteorological variables and potentially, develop a model to predict avalanche days on this stretch of road C 28 using classification trees based on the algorithm of Breiman (1984). Knowledge of the dynamics of avalanches in different sectors of the route has allowed to distinguish three sub-areas (Fig....

    [...]

  • ...The statistical tool chosen to develop the model is a classification tree developed in the work “Classification and regression trees” (CART) de Breiman (1984)....

    [...]

  • ...This is based on the algorithm developed in publishing Breiman (1984)....

    [...]

  • ...The main objective of this work is to use this database to determine the most significant meteorological variables and potentially, develop a model to predict avalanche days on this stretch of road C - 28 using classification trees based on the algorithm of Breiman (1984)....

    [...]

Journal ArticleDOI
TL;DR: In this article, the authors evaluated 31 factors in terms of their importance to explaining avalanche activity indices at two ski areas: Alta, UT and Mammoth Mountain, CA using classification and regression tree models.

64 citations


"A Model for Avalanche Forecasting o..." refers methods in this paper

  • ...Also classification and regression trees have been used in many occasions (Davis et al., 1992), (Davis and Elder, 1994), (Davis et al., 1999), (Rosenthal et al., 2002)....

    [...]

Journal ArticleDOI
TL;DR: In this article, a 10-year data set of meteorological parameters and over 1800 individual avalanche occurrences from the Transit New Zealand Milford Road Avalanche Programme were used to determine periods of significant avalanche activity in terms of an avalanche day.

59 citations


"A Model for Avalanche Forecasting o..." refers methods in this paper

  • ...More recently, Hendrikx et al. (2005) and Murphy and Hendrikx (2014) have complemented this method using cross-validation for test sample....

    [...]

Journal ArticleDOI
TL;DR: A CART model intended to estimate the probability of intrahospital death from acute myocardial infarction (AMI) was developed and found that the C ART model was much easier to use and interpret, because the decision rules generated could be applied without the need for mathematical calculations.

28 citations