Login
A Framework for the Cross‐Validation of Categorical Geostatistical Simulations
Résumé The mapping of subsurface parameters and the quantification of spatial uncertainty requires selecting adequate models and their parameters. Cross‐validation techniques have been widely used for geostatistical model selection for continuous variables, but the situation is different for categorical variables. In these cases, cross‐validation is seldom applied, and there is no clear consensus on which method to employ. Therefore, this paper proposes a systematic framework for the cross‐validation of geostatistical simulations of categorical variables such as geological facies. The method is based on K‐fold cross‐validation combined with a proper scoring rule. It can be applied whenever an observation data set is available. At each cross‐validation iteration, the training set becomes conditioning data for the tested geostatistical model, and the ensemble of simulations is compared to true values. The proposed framework is generic. Its application is illustrated with two examples using multiple‐point statistics simulations. In the first test case, the aim is to identify a training image from a given data set. In the second test case, the aim is to identify the parameters in a situation including nonstationarity for a coastal alluvial aquifer in the south of France. Cross‐validation scores are used as metrics of model performance and quadratic scoring rule, zero‐one score, and balanced linear score are compared. The study shows that the proposed fivefold stratified cross‐validation with the quadratic scoring rule allows ranking the geostatistical models and helps to identify the proper parameters.
   
Mots-clés Cross‐validation framework is developed for testing simulations of
categorical variables; The methodology is generic, and competing models are based on a single score; It requires a set of observation points and can be used with any spatial simulation method
   
Citation Juda, P., Renard, P., & Straubhaar, J. (2020). A Framework for the Cross‐Validation of Categorical Geostatistical Simulations. Earth and Space Science, 2020, 1152-1169.
   
Type Article de périodique (Anglais)
Date de publication 6-2020
Nom du périodique Earth and Space Science
Volume 2020
Pages 1152-1169
URL https://doi.org/10.1029/2020EA001152