<p>The boundary layer is the part of the atmosphere that is in direct contact with the ground via turbulent motion. At mid-latitudes the boundary layer is usually one or a few kilometers deep, while in the Arctic it is much more shallow, typically a few hundred meters or less. The reason is that here the absolute temperature increases in the lowest kilometer, making the boundary layer semi-permanently stably stratified. The exchange of heat, momentum and tracers between the atmosphere, ocean and ground under stable stratification is discussed from an observational, modeling and climate-change point of view. A compilation of six observational datasets, ordered by the Richardson number (rather than the widely used Monin-Obukhov length) reveals new information about turbulence in the very stably stratified regime. An essentially new turbulence closure model, based on the total turbulent energy concept and these observational datasets, is developed and tested against large-eddy simulations with promising results. The role of mesoscale motion in the exchange between the atmosphere and surface is investigated both for observations and in idealized model simulations. Finally, it is found that the stably stratified boundary layer is more sensitive to external surface forcing than its neutral and convective counterparts. It is speculated that this could be part of the explanation for the observed Arctic amplification of climate change.</p>

The boundary layer is the part of the atmosphere that is in direct contact with the ground via turbulent motion. At mid-latitudes the boundary layer is usually one or a few kilometers deep, while in the Arctic it is much more shallow, typically a few hundred meters or less. The reason is that here the absolute temperature increases in the lowest kilometer, making the boundary layer semi-permanently stably stratified. The exchange of heat, momentum and tracers between the atmosphere, ocean and ground under stable stratification is discussed from an observational, modeling and climate-change point of view. A compilation of six observational datasets, ordered by the Richardson number (rather than the widely used Monin-Obukhov length) reveals new information about turbulence in the very stably stratified regime. An essentially new turbulence closure model, based on the total turbulent energy concept and these observational datasets, is developed and tested against large-eddy simulations with promising results. The role of mesoscale motion in the exchange between the atmosphere and surface is investigated both for observations and in idealized model simulations. Finally, it is found that the stably stratified boundary layer is more sensitive to external surface forcing than its neutral and convective counterparts. It is speculated that this could be part of the explanation for the observed Arctic amplification of climate change.

L’observation des couches limites atmosphériques au dessus du plateau antarctique a mis en évidence les plus fortes inversions de température proches de la surface de la planète. Bien paramétriser ces couches limites extrêmes dans un modèle de circulation générale est essentiel pour represéntercorrectement l’inversion climatologique de température au dessus du plateau, mais également pour reproduire des vents catabatiques réalistes en aval du plateau et de surcroit, une circulation atmosphérique correcte dans l’hémisphère sud. Les conclusions des précédentes "Gewex AtmosphericBoundary Layer Studies" (GABLS) ont conduit au constat que la paramétrisation des couches limites stables dans les modèles climatiques est une des priorités pour la communauté des modélisateurs.Ceci est dû au fait que la nature même des processus physiques en jeu est mal connue mais aussi parce que les lois de similitudes, sur lesquelles les paramétrisations du mélange turbulent sont fondées, ne sont pas applicables en condition très stable. L’objectif de ces travaux de thèse est d’évaluer et d’améliorer la représentation des couches limites sur le plateau antarctique dans le modèle français de circulation générale Laboratoire de MétéorologieDynamique-Zoom (LMDZ), composante atmosphérique du modèle de climat IPSL. Avant l’évaluation même du modèle, une étude approfondie de la couche limite de surface et de la structure de la couche limite stable a été conduite à partir de l’analyse de mesures in situ au Dôme C. Il en a résulté une caractérisation de la hauteur de rugosité aérodynamique, une estimation des flux turbulents de surface sur une année entière ainsi que la mise en évidence de sursaturations de la vapeur d’eau par rapport à la glace. L’analyse des mesures de température et de vent le long d’une tour de 45 m a aussi montré que la couche limite se comporte tel un système dynamique à deux régimes distincts. La relation entrevitesse du vent et inversion de température décrit un "S renversé", suggérant une transition de régime suivant un hystérésis. Une étude complémentaire a révélé que ce comportement dynamique à deux régimes est une caractéristique générale et robuste des couches limites stables, qui peuvent transiter,selon l’intensité des forçages, d’un régime ’turbulent’ peu stable à un régime ’radiatif’ très stable et vice et versa.Le modèle LMDZ a ensuite été évalué en configuration 1D sur un cycle diurne d’été dans le cadre de la quatrième expérience GABLS. Des tests de sensibilité aux paramètres de surface et à la paramétrisation du mélange turbulent ont été réalisés. Ils ont conduit à de nettes améliorations des performancesdu modèle ainsi qu’à la mise en place d’une configuration adaptée aux conditions antarctiques. Des simulations complémentaires en 3D ont par la suite soulevé l’importance du transfert radiatif infrarouge et de la paramétrisation des flux turbulents de surface pour la modélisation de la couchelimite sur le plateau pendant la nuit polaire. Enfin, les travaux de thèse ont été étendus à la modélisation des couches limites stables continentales. Les paramétrisations locales de turbulence ont en effet tendance à sous-estimer le mélange sous-maille continental, en raison de la multitude des processusde mélange en jeu. Un réflexion a donc été portée sur la façon de palier ce manque de mélange, avec comme idée directrice de transférer la perte d’énergie cinétique grande échelle perdue lors du freinage de l’écoulement par les ondes de gravité, vers de l’énergie cinétique turbulente. / Observation of the Atmospheric Boundary Layers (ABL) above the Antarctic Plateau has revealed the strongest near-surface temperature stratifications on the Earth. A correct parametrization of the very stratified Antarctic ABLs in General Circulation Models (GCM) is critical since they exert a strongcontrol on the continental scale temperature inversion, on the coastal katabatic winds and subsequently on the Southern Hemisphere circulation. The previous Gewex Atmospheric Boundary Layer Studies (GABLS) highlighted that the parametrization of the very stratified, or very stable, ABLs isone of the most critical challenge in the atmospheric modelers community. Indeed, the nature of the mixing processes are not completely understood and the commonly used similarity laws, on which the model’s parametrization are usually based, are no longer valid. The aim of this PhD work is to evaluate and improve the modelling of the ABL over the Antarctic Plateau by the Laboratoire de Météorologie Dynamique-Zoom (LMDZ) GCM, the atmospheric component of the IPSL Earth System Model in preparation for the sixth Coupled Models Intercomparison Project. Before the model evaluation itself, an in-depth study of the dynamics of the atmospheric surface layer and of the stable ABL over the Antarctic Plateau was carried out from in situ measurements at Dome C. The analysis enabled the first estimations of the roughness length and of the surface fluxes during the polar night at this location as well as the characterization of very frequent occurences of near-surface moisture supersaturations with respect to ice. Investigation of meteorological measure-ments along a 45 m tower also revealed two distinct dynamical regimes of the stable ABL at this location. In particular, the relation between the near surface inversion amplitude and the wind speed takes a typical ’reversed S-shape’, suggesting a system obeing with an hysteresis. A further analysisshowed that this is a clear illustration of a general and robust feature of the stable ABL systems, corresponding to a ‘critical transition’ between a steady turbulent and a steady ‘radiative’ regime. LMDZ was then run on 1D simulations during a typical clear-sky summertime diurnal cycle in the framework of the fourth GABLS case. Sensitivity tests to surface parameters, vertical grid and turbulent mixing parametrizations were performed leading to significant improvements of the model and to a new configuration better adapted for Antarctic conditions. 3D simulations were then carried outwith the ’zooming capability’ of the horizontal grid and with nudging. These simulations enabled a further evaluation of the model over a full year and extending the analysis beyond Dome C. In particular, this study raised the importance of the radiative scheme and of the surface layer scheme forthe modelling of the ABL during the polar night over the Plateau. Finally, the PhD work extented toward the modelling of the stable ABL over the other continents, assessing how the frequently underestimated subgrid mixing of momentum and heat can be compensated by a transfer of large scalekinetic energy toward turbulent kinetic energy when the flow is slowed down by orographic gravity wave drag.