On the systematic reduction of data complexity in multimodel atmospheric dispersion ensemble modeling

Abstract

[1] The aim of this work is to explore the effectiveness of theoretical information approaches for the reduction of data complexity in multimodel ensemble systems. We first exploit a weak form of independence, i.e. uncorrelation, as a mechanism for detecting linear relationships. Then, stronger and more general forms of independence measure, such as mutual information, are used to investigate dependence structures for model selection. A distance matrix, measuring the interdependence between data, is derived for the investigated measures, with the scope of clustering correlated/dependent models together. Redundant information is discarded by selecting a few representative models from each cluster. We apply the clustering analysis in the context of atmospheric dispersion modeling, by using the ETEX-1 data set. We show how the selection of a small subset of models, according to uncorrelation or mutual information distance criteria, usually suffices to achieve a statistical performance comparable to, or even better than, that achieved from the whole ensemble data set, thus providing a simpler description of ensemble results without sacrificing accuracy.Citation: Riccio A., A. Ciaramella, G. Giunta, S. Galmarini, E. Solazzo, and S. Potempski (2012), On the systematic reduction of data complexity in multimodel atmospheric dispersion ensemble modeling,

Two-scale multi-model ensemble: is a hybrid ensemble of opportunity telling us more?

“…This group is the one that best reduces the redundancies and optimizes the complementarity of the model results (Kioutsioukis and Galmarini, 2014). Other methods have been devised to determine the optimal number of models (Bretherton et al, 1999;Riccio et al, 2012) that are equally effective as the one used here, though they do not allow identifying the members of the subset. Beyond the use of the mmeS for the current analysis, given the diversity in the number of models comprising the two ensembles we have calculated the effective numbers of models (Bretherton et al, 1999) for the regional and global sets in the attempt to verify whether the effective numbers were close for the two sets.…”

“…In this study, we would also like to build upon the research performed in other multi-model ensembles over the years; rather than calculating only the classical model average or median ensemble (mme), we shall also calculate three ensembles based on the findings of Potempski and Galmarini (2009), Riccio et al (2012), Solazzo et al (2012aSolazzo et al ( , b, 2013, Galmarini et al (2013), and Kioutsioukis and Galmarini (2014). We shall therefore refer to the ensemble made by the optimal subset of models that produce the minimum RMSE as mmeS (Solazzo et al, 2012a, b); the ensemble produced by filtering measurements and all model results using the Kolmogorov-Zurbenko decomposition presented earlier and recombining the four components that best compare with the observed components into a new model set as kzFO ; and the optimally weighted combination as mmeW (Potempski and Galmarini, 2009;Kioutsioukis and Galmarini, 2014;Kioutsioukis et al, 2016).…”

<i>E pluribus unum</i>*: ensemble air quality predictions

“…The practice has been used in a wide range of applications in atmospheric and climate sciences (Galmarini et al, 2001;delle Monache et al, 2006;McKeen et al, 2005;Van Loon et al, 2007;Mallet and Sportisse, 2006;Solazzo et al, 2012a;Riccio et al, 2012;Potempski et al, 2008;Knutti et al, 2010;Tebaldi and Knutti, 2007) as well as in a range of other contexts. Over the years a large number of different approaches (Potempski and Galmarini, 2009) have been proposed from the very popular simple averaging of the result, to the construction of the median model to the application of weights derived from past skill scores or Bayesian model averaging theory (e.g., Delle Monache et al, 2006;Galmarini et al, 2004;Potempski et al, 2010;Riccio et al, 2007).…”

“…Recent findings point toward a deeper and more thorough analysis of the model results in an attempt to identify those that, within the ensemble, represent real original contributions to the improvement of the ensemble result. Toward this end, analyses aiming at promoting true model diversity such those used by Riccio et al (2012), Solazzo et al (2012a), Masson and Knutti (2011) seem to go in that right direction.…”

Two-scale multi-model ensemble: Is a hybrid ensemble of opportunity telling us more?

“…It has been widely demonstrated (e.g Potempsky and Galmarini, 2009) that when 81 multiple model results are distilled to retain only original and independent 82 contributions (Solazzo et al 2012) and thereafter statistically combined in what is 83 usually called an ensemble, one obtains results that are systematically superior to 84 the performance of the individual models and therefore can provide more accurate 85 and robust assessments or predictions. 86…”