Search Results

You are looking at 1 - 2 of 2 items for

  • Author or Editor: Paul-Antoine Michelangeli x
  • Refine by Access: All Content x
Clear All Modify Search
Paul-Antoine Michelangeli, Robert Vautard, and Bernard Legras

Abstract

Two different definitions of midlatitude weather regimes are compared. The first seeks recurrent atmospheric patterns. The second seeks quasi-stationary patterns, whose average tendency vanishes. Recurrent patterns are identified by cluster analysis, and quasi-stationary patterns are identified by solving a nonlinear equilibration equation. Both methods are applied on the same dataset: the NMC final analyses of 700-hPa geopotential heights covering 44 winters. The analysis is performed separately over the Atlantic and Pacific sectors.

The two methods give the same number of weather regimes—four over the Atlantic sector and three over the Pacific sector. However, the patterns differ significantly. The investigation of the tendency, or drift, of the clusters shows that recurrent flows have a systematic slow evolution, explaining this difference. The patterns are in agreement with the ones obtained from previous studies, but their number differs.

The cluster analysis algorithm used here is a partitioning algorithm, which agglomerates data around randomly chosen seeds and iteratively finds the partition that minimizes the variance within clusters, given a prescribed number of clusters. The authors develop a classifiability index, based on the correlation between the cluster centroids obtained from different initial pullings. By comparing the classifiability index of observations with that obtained from a multivariate noise model, an objective definition of the number of clusters present in the data is given. Although the classifiability index is maximal by prescribing two clusters in both sectors, it only differs significantly from that obtained with the noise model using four Atlantic clusters and three Pacific clusters. The partitioning clustering method turns out to give more statistically stable clusters than hierarchical clustering schemes.

Full access
Malaak Kallache, Elena Maksimovich, Paul-Antoine Michelangeli, and Philippe Naveau

Abstract

The performance of general circulation models (GCMs) varies across regions and periods. When projecting into the future, it is therefore not obvious whether to reject or to prefer a certain GCM. Combining the outputs of several GCMs may enhance results. This paper presents a method to combine multimodel GCM projections by means of a Bayesian model combination (BMC). Here the influence of each GCM is weighted according to its performance in a training period, with regard to observations, as outcome BMC predictive distributions for yet unobserved observations are obtained. Technically, GCM outputs and observations are assumed to vary randomly around common means, which are interpreted as the actual target values under consideration. Posterior parameter distributions of the authors’ Bayesian hierarchical model are obtained by a Markov chain Monte Carlo (MCMC) method. Advantageously, all parameters—such as bias and precision of the GCM models—are estimated together. Potential time dependence is accounted for by integrating a Kalman filter. The significance of trend slopes of the common means is evaluated by analyzing the posterior distribution of the parameters. The method is applied to assess the evolution of ice accumulation over the oceanic Arctic region in cold seasons. The observed ice index is created out of NCEP reanalysis data. Outputs of seven GCMs are combined by using the training period 1962–99 and prediction periods 2046–65 and 2082–99 with Special Report on Emissions Scenarios (SRES) A2 and B1. A continuing decrease of ice accumulation is visible for the A2 scenario, whereas the index stabilizes for the B1 scenario in the second prediction period.

Full access