Issue |
ESAIM: Procs
Volume 60, 2017
Journées MAS 2016 de la SMAI – Phénomènes complexes et hétérogènes
|
|
---|---|---|
Page(s) | 104 - 113 | |
DOI | https://doi.org/10.1051/proc/201760104 | |
Published online | 14 December 2017 |
Is future climate predictable with statistics?
CNRM, Météo France and CNRS, 42 Avenue Gaspard Coriolis, 31057 France
The purpose of this note is to briefly introduce the statistical models and methods used in climate sciences to estimate, from observations, the sensitivity of the Earth's climate to Greenhouse Gases. First the context of climatology is described with an explanation of how statistics can interact with the use of climate models. A description of the main models used, which are original variants of Error-in-Variables models, follows. Then a few issues for which methodological progresses would be helpful are mentioned. This includes the inference of large covariance matrices and uncertainty quantification.
Résumé
Cette note a pour but d’introduire brièvement les modèles et outils statistiques utilisés en sciences du climat afin de quantifier, à partir d’observations, la sensibilité du climat à l’augmentation de l’effet de serre. Je commence par présenter le contexte actuel des sciences du climat, et la façon dont la statistique interagit avec l’utilisation des modèles de climat. Je décris ensuite les principaux modèles utilisés, qui sont des variantes originales des modèles à erreur. Enfin je mentionne quelquesuns des problèmes méthodologiques rencontrés, liés notamment à l’estimation de grandes matrices de covariance et à la quantification d’incertitudes.
© EDP Sciences, SMAI 2017
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.