Climate regionalization with the LMDZ model : methodological study

• Main Author: Li, Shan
• Other Authors: Laboratoire de Météorologie Dynamique (UMR 8539) (LMD), Université Pierre et Marie Curie - Paris 6 (UPMC)-Institut national des sciences de l'Univers (INSU - CNRS)-École polytechnique (X)-École des Ponts ParisTech (ENPC)-Centre National de la Recherche Scientifique (CNRS)-Département des Géosciences - ENS Paris, École normale supérieure - Paris (ENS-PSL), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-École normale supérieure - Paris (ENS-PSL), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL), Université Pierre et Marie Curie - Paris VI, Hervé Le Treut, Laurent Li
• Format: Doctoral or Postdoctoral Thesis
• Language: French
• Published: HAL CCSD 2017
• Subjects: Regionalization; OWN/TWN; Newtonian relaxation; Régionalisation; Master versus slave; Relaxation newtonienne; One-Way nesting; Two-Way nesting; Raffinement de maille; [SDU.STU]Sciences of the Universe [physics]/Earth Sciences; Arctic; North Atlantic
• Online Access: https://theses.hal.science/tel-01941400; https://theses.hal.science/tel-01941400/document; https://theses.hal.science/tel-01941400/file/2017PA066451.pdf

The work developed in this thesis explores through methodological modelling studies the current techniques of climate regionalization. In this case, the regionalization focuses on a geographical domain covering from the North Atlantic to Eastern Europe longitudinal wise, and from the Sahel to the Arctic as a latitudinal interval. The aim of this thesis is not the improvement of regional climate modelling per se, but tackling three key questions that are commonly met by all attempts when trying to improve climate regionalization. Firstly, the choice and advantages of the nesting scheme: one-way nesting (OWN) versus two-way nesting (TWN). Secondly, the evaluation of the nesting method, which is generally a Newtonian relaxation operation added to the prognostic equations of the model. And finally, the consequences of the mesh refinement in Regional Circulation Models (RCM). The objective of this manuscript consists in conceptualizing and carrying out numerical simulations to answer these three questions by isolating each individual effect and quantifying the consequences of each of the effects. The general circulation model LMDZ is used for all experiments. It is able to play the role of the General Circulation Model (GCM) and the RCM, keeping the same physical parameterizations and the same dynamical configuration, as well as the same external forcings and model parameters. Our experimental set-up, referred as “Master versus Slave”, consists on two related protocols: “DS-300-to-300” and “DS-300-to-100”. The former implies the downscaling of the GCM at 300 km of horizontal resolution while the RCM has the identical resolution of 300 km. The latter implies the downscaling from 300 km (GCM) to 100 km (RCM). We have assumed the “DS-300-to-300” as an idealized framework, particularly appropriate to evaluate the relaxation operation effect. In parallel, the “DS-300-to-100” protocol, subtracted from the “DS-300-to-300”, allows assessing the effect of the increased resolution for the RCM. In each protocol, two ...