Mathieu Vrac

Followers

Following

Co-authors

Public Views

Michael Fischer

University of Kent

Aswani Kumar Cherukuri

VIT University

John Uebersax

California Polytechnic State University at San Luis Obispo

Cesar Merino

San Martin de Porres University

John Eustaquio

University of the Philippines Diliman

Ferda Esin Gülel

Pamukkale University

Sy Tien Do

Hochiminh University of Technology

José Francisco Pessanha

UERJ - Universidade do Estado do Rio de Janeiro / Rio de Janeiro State University

Dalson Figueiredo

Universidade Federal de Pernambuco

Adebayo D Agunbiade

Olabisi Onabanjo University, Ago-Iwoye, Nigeria

Interests

Uploads

Papers by Mathieu Vrac

Mélange de lois de lois par copules

EGC eBooks, 2002

Symbolic Class Descriptions

Studies in classification, data analysis, and knowledge organization, 2002

Our aim is to describe a partition of a class by a conjunction of characteristic properties. We u... more Our aim is to describe a partition of a class by a conjunction of characteristic properties. We use a stepwise top-down binary tree method. At each step we select the best variable and its optimal splitting to optimize simultaneously a discrimination criterion given by a prior partition and a homogeneity criterion. Moreover, this method deals with a data table in which each cell contains a histogram of nominal categories but the method can be extended or reduced to other types of data. The method is illustrated on both simulated data and real data.

M�lange de lois de lois par copules

F Egc, 2002

Copula analysis of mixture models

Computational Statistics, 2011

Contemporary computers collect databases that can be too large for classical methods to handle. T... more Contemporary computers collect databases that can be too large for classical methods to handle. The present work takes data whose observations are distribution functions (rather than the single numerical point value of classical data) and presents the computational statistical approach of a new methodology to group the distributions into classes. The clustering method links the searched partition to the decomposition of mixture densities, through the notions of a function of distributions and of multidimensional copulas. The new clustering technique is illustrated by ascertaining distinct temperature and humidity regions for a global climate dataset and shows that the results compare favorably with those obtained from the standard EM algorithm method.

Download

Décomposition de mélange de distributions et application à des données climatiques

Revue de statistique appliquée, 2004

Décomposition de mélange de distributions et application à des données climatiques Revue de stati... more

Download

Stochastic Model Output Statistics for Bias Correcting and Downscaling Precipitation Including Extremes

Http Dx Doi Org 10 1175 Jcli D 13 00604 1, Sep 10, 2014

ABSTRACT Precipitation is highly variable in space and time; hence, rain gauge time series genera... more ABSTRACT Precipitation is highly variable in space and time; hence, rain gauge time series generally exhibit additional random small-scale variability compared to area averages. Therefore, differences between daily precipitation statistics simulated by climate models and gauge observations are generally not only caused by model biases, but also by the corresponding scale gap. Classical bias correction methods, in general, cannot bridge this gap; they do not account for small-scale random variability and may produce artifacts. Here, stochastic model output statistics is proposed as a bias correction framework to explicitly account for random small-scale variability. Daily precipitation simulated by a regional climate model (RCM) is employed to predict the probability distribution of local precipitation. The pairwise correspondence between predictor and predictand required for calibration is ensured by driving the RCM with perfect boundary conditions. Wet day probabilities are described by a logistic regression, and precipitation intensities are described by a mixture model consisting of a gamma distribution for moderate precipitation and a generalized Pareto distribution for extremes. The dependence of the model parameters on simulated precipitation is modeled by a vector generalized linear model. The proposed model effectively corrects systematic biases and correctly represents local-scale random variability for most gauges. Additionally, a simplified model is considered that disregards the separate tail model. This computationally efficient model proves to be a feasible alternative for precipitation up to moderately extreme intensities. The approach sets a new framework for bias correction that combines the advantages of weather generators and RCMs.

Non-stationary probabilistic downscaling of extreme precipitation

General circulation model (GCM) output or reanalysis data typically provide information at a coar... more General circulation model (GCM) output or reanalysis data typically provide information at a coarse spatial resolution, which cannot directly be used for local impact studies. Downscaling methods are developed to overcome this problem and to obtain local-scale surface weather from regional-scale atmospheric variables. Here the derivation of local-scale extremes still is a challenging topic. We present a probabilistic downscaling approach where the cumulative distribution function (cdf) of large- and local-scale extremes is linked by means of a transform function. In this way the cdf of the local-scale extremes is obtained for a projection period and statistical characteristics like quantiles or return levels are inferred. The extreme values used for downscaling are assumed to be distributed according to a Generalized Pareto distribution (GPD). This allows to apply the approach to many different empirical data series. Out of the resulting cdf, realisations can be generated to provide, for example, uncertainty measures. In case the large-scale variable does not solely determine the evolution of the local-scale variable in the projection period, further variables may be included in the analysis in form of covariates of the GPD parameters. We apply our methodology to downscale NCEP reanalysis precipitation rate in winter to obtain daily precipitation at five stations in Southern France. The calibration period of 1951 to 1985 is used to project to the time period 1986 to 1999. The applicability of the approach is checked by a comparison study with verifying observations by means of, e.g., quantile-quantile plots or the continuous ranked probability score. It shows that covariates have to be chosen with care, otherwise they may worsen the results.

Download

Assessing the performance of dynamical and statistical downscaling techniques to simulate crop yield in West Africa

Agu Fall Meeting Abstracts, Dec 1, 2010

Global circulation models (GCM) are increasingly capable of making relevant predictions of season... more Global circulation models (GCM) are increasingly capable of making relevant predictions of seasonal and long-term climate variability, thus improving prospects of predicting impact on crop yields. This is particularly important for semi-arid West Africa where climate variability and drought threaten food security. Translating GCM outputs into attainable crop yields is difficult because GCM grid boxes are of larger scale than the processes governing yield, involving partitioning of rain among runoff, evaporation, transpiration, drainage and storage at plot scale. It therefore requires the use of downscaling methods. This study analyzes the performance of both dynamical and statistical downscaling techniques in simulating crop yield at local scale. A detailed case study is conducted using historical weather data for Senegal, applied to the crop model SARRAH for simulating several tropical cereals (sorghum, millet, maize) at local scale. This control simulation is used as a benchmark to evaluate a set of Regional Climate Models (RCM) simulations, forced by ERA-Interim, from the ENSEMBLES project and a statistical downscaling method, the CDF-Transform, used to correct biases in RCM outputs. We first evaluate each climate variable that drives the simulated yield in the control simulation (radiation, rainfall, temperatures). We then simulate crop yields with RCM outputs (with or without applying the CDG-Transform) and evaluate the performance of each RCM in regards to crop yield simulations.

Detecting spatial patterns with the cumulant function - Part 2: An application to El Ni�o

Nonlinear Process Geophys, 2008

The spatial coherence of a measured variable (e.g. temperature or pressure) is often studied to d... more The spatial coherence of a measured variable (e.g. temperature or pressure) is often studied to determine the regions of high variability or to find teleconnections, i.e. correlations between specific regions. While usual methods to find spatial patterns, such as Principal Components Analysis (PCA), are constrained by linear symmetries, the dependence of variables such as temperature or pressure at different locations is generally nonlinear. In particular, large deviations from the sample mean are expected to be strongly affected by such nonlinearities. Here we apply a newly developed nonlinear technique (Maxima of Cumulant Function, MCF) for detection of typical spatial patterns that largely deviate from the mean. In order to test the technique and to introduce the methodology, we focus on the El Niño/Southern Oscillation and its spatial patterns. We find nonsymmetric temperature patterns corresponding to El Niño and La Niña, and we compare the results of MCF with other techniques, such as the symmetric solutions of PCA, and the nonsymmetric solutions of Nonlinear PCA (NLPCA). We found that MCF solutions are more reliable than the NLPCA fits, and can capture mixtures of principal components. Finally, we apply Extreme Value Theory on the temporal variations extracted from our methodology. We find that the tails of the distribution of extreme temperatures during La Niña episodes is bounded, while the tail during El Niños is less likely to be bounded. This implies that the mean spatial patterns of the two phases are asymmetric, as well as the behaviour of their extremes.

Download

Description symbolique de classes

F Egc, 2001

... Mathieu Vrac* — Edwin Diday* ... Parmi les nombreuses méthodes monothétiques supervisées, act... more ... Mathieu Vrac* — Edwin Diday* ... Parmi les nombreuses méthodes monothétiques supervisées, actuellement, les plus mentionnées sont celles développées par Breiman et al [BRE 84] dans l&amp;#x27;algo-rithme CART (Classification And Regression Tree), par Quinlan [QUI 86] dans l&amp;#x27;al ...

Dynamical and Statistical downscaling of Mediterranean climate: comparison and uncertainty assessment in the MED-CORDEX and HYMEX context. EGU conference, Vienna, april 2011. (poster)

ABSTRACT

MOS correction of GCM- and RCM-simulated daily precipitation

Understanding long-term changes in daily precipitation characteristics, particularly those associ... more Understanding long-term changes in daily precipitation characteristics, particularly those associated with extreme events, is an important component of climate change science and impact assessment. Estimates of such changes are required at local scales where impacts are most keenly felt. However, the limited spatial resolution of General Circulation Models (GCMs) makes direct estimates of future daily precipitation unrealistic. A popular downscaling approach is to use GCMs to drive high-resolution Regional Climate Models (RCMs). Whilst able to simulate precipitation characteristics at smaller scales, RCMs do not represent local variables and remain limited by systematic errors and biases. It is possible to apply statistical corrections, known as Model Output Statistics (MOS), to RCM-simulated precipitation. The simplest form of MOS (including bias correction) follows a 'distribution-wise' approach in which the statistical link is derived between long-term distributions of si...

A complete hydro-climate model chain to investigate the influence of sea surface temperature on recent hydroclimatic variability in subtropical South America (Laguna Mar Chiquita, Argentina)

Climate Dynamics, 2015

as input to simulate the lake behavior. The results indicate that the long-term lake level trend ... more as input to simulate the lake behavior. The results indicate that the long-term lake level trend is fairly well depicted by the LMDz-SD-cpHM simulations. The 1970s level rise and high-level conditions are generally well captured in timing and in magnitude when SST-forced AGCM-SD variables are used to drive the cpHM. As the LMDz simulations are forced solely with the observed sea surface conditions, the global SST seems to have an influence on the lake level variations of Laguna Mar Chiquita. As well, this study shows that the AGCM-SD-cpHM model chain is a useful approach for evaluating long-term lake level fluctuations in response to the projected climate changes.

Download

Intercomparison of statistical and dynamical downscaling models under the EURO- and MED-CORDEX initiative framework: present climate evaluations

Climate Dynamics, 2015

ABSTRACT

Climate change impact on renewable energy sources during the 21st century over France

The impact of climate change on three types of green electricity production over France, wind, hy... more The impact of climate change on three types of green electricity production over France, wind, hydro and solar energies, is studied through the evolution of related climate variables during the 21st century. Climate projections of these variables, obtained from the IPCC model database, are downscaled to higher resolution grids (25 km for solar radiation and 50 km for precipitation) and

Assessing climate change impacts on European wind energy from ENSEMBLES high-resolution climate projections

Climatic Change, 2014

Climate change may alter the geographical pattern and intensity of near-surface winds which are t... more Climate change may alter the geographical pattern and intensity of near-surface winds which are the "fuel" for wind turbines. In a context of fast current and planned development of wind power worldwide, investigating the impacts of climate change on wind power generation is necessary. This study aims at assessing future changes in the potential for wind power generation over the whole Europe and in the effective wind power production from national wind farms operating at the end of 2012 and planned by 2020. For this purpose, a simplified wind power generation model is applied to an ensemble of 15 regional climate projections achieved from 10 Regional Climate Models downscaling six Global Climate Models under the SRES A1B emission scenario from the ENSEMBLES project. The use of a relatively large multi-model ensemble allows the identification of robust changes and the estimation of a range of uncertainties associated with projected changes. We show with a high level of confidence that, under the A1B scenario, over most of Europe, changes in wind power potential will remain within ±15 and ±20 % by mid and late century respectively. Overall, we find a tendency toward a decrease of the wind power potential over Mediterranean areas and Climatic Change (2015) 128:99-112 an increase over Northern Europe. Changes in multi-year power production will not exceed 5 and 15 % in magnitude at the European and national scale respectively for both wind farms in operation at the end of 2012 and planned by 2020. Therefore, climate change should neither undermine nor favor wind energy development in Europe. However, accounting for climate change effects in particular regions may help optimize the wind power development and energy mix plans.

Download

Comparison of GCM- and RCM-simulated precipitation following stochastic postprocessing

Journal of Geophysical Research: Atmospheres, 2014

ABSTRACT In order to assess to what extent regional climate models (RCMs) yield better representa... more ABSTRACT In order to assess to what extent regional climate models (RCMs) yield better representations of climatic states than general circulation models (GCMs) the output of each is usually directly compared with observations. RCM output is often bias-corrected and in some cases correction methods can also be applied to GCMs. This leads to the question of whether bias-corrected RCMs perform better than bias-corrected GCMs. Here, the first results from such a comparison are presented, followed by discussion of thevalue added by RCMs in this setup. Stochastic postprocessing, based on Model Output Statistics (MOS), is used to estimate daily precipitation at 465 stations across the United Kingdom between 1961–2000 using simulated precipitation from two RCMs (RACMO2and CCLM) and, for the first time, a GCM (ECHAM5) as predictors. The large-scale weather states in each simulation are forced toward observations. The MOS method uses logistic regression to model precipitation occurrence and a Gamma distribution for the wet-day distribution, and is cross-validated based on Brier and quantile skill scores. A major outcome of the study is that the corrected GCM-simulated precipitation yields consistently higher validation scores than the corrected RCM-simulated precipitation. This seems to suggest that, in a setup with postprocessing, there is no clear added value by RCMs with respect to downscaling individual weather states. However, due to the different ways of controlling the atmospheric circulation in the RCM and the GCM simulations, such a strong conclusion cannot be drawn. Yet, the study demonstrates how challenging it is to demonstrate the value added by RCMs in this setup.

Stochastic Model Output Statistics for Bias Correcting and Downscaling Precipitation Including Extremes

Journal of Climate, 2014

Multivariate—Intervariable, Spatial, and Temporal—Bias Correction*

by Mathieu Vrac and Petra Friederichs

Journal of Climate, 2015

Mélange de lois de lois par copules

Extraction et Gestion des Connaissances, 2002

Mélange de lois de lois par copules

EGC eBooks, 2002

Symbolic Class Descriptions

Studies in classification, data analysis, and knowledge organization, 2002

M�lange de lois de lois par copules

F Egc, 2002

Copula analysis of mixture models

Computational Statistics, 2011

Download

Décomposition de mélange de distributions et application à des données climatiques

Revue de statistique appliquée, 2004

Décomposition de mélange de distributions et application à des données climatiques Revue de stati... more

Download

Stochastic Model Output Statistics for Bias Correcting and Downscaling Precipitation Including Extremes

Http Dx Doi Org 10 1175 Jcli D 13 00604 1, Sep 10, 2014

Non-stationary probabilistic downscaling of extreme precipitation

Download

Assessing the performance of dynamical and statistical downscaling techniques to simulate crop yield in West Africa

Agu Fall Meeting Abstracts, Dec 1, 2010

Detecting spatial patterns with the cumulant function - Part 2: An application to El Ni�o

Nonlinear Process Geophys, 2008

Download

Description symbolique de classes

F Egc, 2001

Dynamical and Statistical downscaling of Mediterranean climate: comparison and uncertainty assessment in the MED-CORDEX and HYMEX context. EGU conference, Vienna, april 2011. (poster)

ABSTRACT

MOS correction of GCM- and RCM-simulated daily precipitation