Climate models help us understand the complexity of Earth’s climate, forecast the next seasons and predict the influence of anthropogenic forcings. It is therefore important to evaluate the performance of these models relative to observational datasets, to build confidence and to improve them.

Recently multiple ‘initial-condition large ensembles (LEs)’ have been computed to isolate the role of internal variability (natural variability arising from processes intrinsic to the coupled ocean–atmosphere–land–biosphere–cryosphere system). These LEs are particularly useful for the evaluation of modes of climate variability.<br>In this presentation I will discuss the art of climate model evaluation, using the example of El Niño–Southern Oscillation (ENSO). I will illustrate how internal variability affects ENSO evaluation and how LEs allow us to estimate the precision of the evaluation, which is necessary to decide on a new model version among multiple candidates. To deepen the discussion on the art of model evaluation, I will then illustrate the importance of choosing the right observational datasets for the right applications when evaluating and developing climate models.

Date & Time: Tue Jul 18 2023 11:00:00 GMT+0200 (Central European Summer Time)
Room: salle du LOCEAN, couloir 45-55, 4ème étage, pièce 417

Lien zoom: https://cnrs.zoom.us/j/98530085830?pwd=dHdXbEpEV29LU1hGUlZFZlJuSitGUT09