MCMaëlle Coulon-Decorzens

Maëlle Coulon-Decorzens

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Maëlle Coulon-Decorzens has a total of 23 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Evaluation and improvement of the cold pool parameterization in the LMDZ climate model against Large Eddy Simulations

    by Mamadou Lamine Thiam, Frédéric Hourdin, Jean-Yves Grandpeix, Catherine Rio, Maëlle Coulon-Decorzens

    Cold pools, formed under clouds by the evaporation of precipitation, play a central role in maintaining and organizing atmospheric convection. It is suspected that their absence in climate models may lead to significant errors in the representation of convection, such as the premature convection extinction after sunset. The introduction of a cold pool parameterization into the LMDZ climate model has significantly improved the representation of convection, in particular its diurnal cycle. Thanks to this indirect evaluation, the parameterization of cold pool was retained in the further versions of LMDZ. However, no evaluation of the representations of cold pools themselves was done so far. This work provides for the first time such an evaluation based on Large Eddy Simulation (LES). The evaluation is done both on a case of radiative–convective equilibrium that allows the study of convective processes in a well-controlled and quasi-stationary framework, without the influence of large-scale dynamics, and on the time-dependent continental case AMMA (African Monsoon Multidisciplinary Analysis), representative of typical Sahelian conditions with local initiation during the afternoon. First, we evaluate the physical relationships underlying the cold pool model in the LES, then, in a second step, its behavior when coupled with the deep convection scheme in the single-column version of LMDZ. The analyses carried out demonstrate the relevance of the assumptions underlying the parameterization. The initial version actually captures the main characteristics of LES cold pools but also exhibits some biases. We show how substantial modifications to the cold pool scheme and a readjustment of certain free parameters helped reduce those biases significantly. The remaining flaws could be corrected by adding convective mixing through thermal plumes within the cold pools and by modeling the evolution of cold pools number density rather than imposing it.

  • Open Access

    Investigating Cloud–Radiation Error Compensations Stemming From Climate Model Calibration

    by M. Coulon‐Decorzens, F. Hourdin, N. Villefranque

    Compensating errors are an obstacle to the development of climate models. We wonder if systematic errors in simulated cloud properties might result from forced error compensation due to targeting top‐of‐atmosphere radiative fluxes in the tuning process while using an inaccurate radiative transfer (RT) parameterization. Here, we investigate the extent to which parametric and structural errors in state‐of‐the‐art RT models might be offset by errors relating to cloud properties, and under what circumstances. For this we run tuning experiments with a Single‐Column version of a climate Model (SCM), with and without a parameterization of 3D radiative effects of clouds, targeting reference solar fluxes calculated on Large‐Eddy Simulation cloud fields. In a first set of tuning experiments, the parameters that control cloud geometry in the radiative transfer code are tuned assuming perfect cloud profiles. Compensation mechanisms are found between cloud geometry effects, notably between cloud vertical overlap and 3D effects. In a second set of experiments, cloud geometry parameters are set to nominal values, and free parameters of the boundary layer and cloud parameterizations are tuned, still targeting radiative metrics. We found that, when 3D effects are neglected in these experiments, reasonable fluxes are obtained only at the expense of overestimated cloud fractions, partly compensating for underestimated cloud reflectivity at low sun. Targeting mean solar‐angle fluxes instead of detailed ones removes this mechanism entirely. This study both enlightens some compensation errors specific to cloud–radiation interaction and illustrates more generally how the tuning tool can be used to study and quantify compensating errors.

  • L'art et la science de l'ajustement des modèles de climats : de la théorie à la pratique

    by Maëlle Coulon--Decorzens

    L'ajustement paramétrique, aussi appelé tuning ou calibration, est une étape essentielle du développement des modèles de circulation générale atmosphérique, utilisés tant pour la compréhension du climat que pour la réalisation de projections climatiques. Autrefois réalisée sur la base d'essais-erreurs, cette étape d'ajustement est bouleversée depuis une dizaine d'années par l'arrivée de nouvelles méthodes, ouvrant la voie à de nouvelles questions scientifiques : l'ajustement devient peu à peu une science à part entière. L'équipe développant le modèle atmosphérique LMDZ s'est depuis approprié la méthode d'History Matching with Iterative Refocusing (HMIR), et propose, en 2021, une première stratégie de tuning semi-automatique de son modèle basée sur cette approche. Cette stratégie repose sur un pré-conditionnement du tuning du modèle global, visant les flux radiatifs observés par satellite, par un tuning en mode unicolonne. Elle vise à limiter les compensations d'erreurs en garantissant le bon comportement du modèle à l'échelle des processus sur un certain nombre de cas tests. Cette thèse a pour objectif d'explorer les potentialités de la méthode d'HMIR pour l'ajustement et la modélisation du climat en général, afin notamment d'améliorer la simulation du climat global et/ou régional, de limiter les compensations d'erreurs dans le processus d'ajustement, et d'aider plus largement au développement des modèles de climat. Dans un premier temps, la thèse explore le potentiel d'une configuration dite guidée-zoomée pour l'ajustement du modèle LMDZ, dans laquelle la dynamique du modèle est contrainte par une technique de guidage. Avec une telle configuration, les paramètres du modèle peuvent être contraint par des observations journalières in situ (ici celles du site SIRTA en région parisienne) tout en évitant, grâce au guidage, les erreurs dues à une mauvaise simulation de la situation synoptique par le modèle, et ainsi se focaliser uniquement sur les biais des paramétrisations physiques du modèle. Des difficultés liées notamment à la surestimation de la variabilité des précipitations convectives par le modèle nous ont menées à développer une configuration de LMDZ dans laquelle la boucle de rétroaction de l'hydrologie du sol est coupée. Avec cette configuration, nous avons pu mener la première expérience de tuning de LMDZ visant des observations sur site. Elle pourra servir de base pour tuner conjointement les paramètres libres des modèles atmosphériques et de surface continentale. La thèse aborde ensuite la question des compensations d'erreurs entre paramétrisations des nuages et du transfert radiatif susceptibles d'émerger lors de l'ajustement des flux radiatifs du modèle global. Elle propose une méthodologie originale permettant une étude quantitative de ces compensations d'erreurs, rendue possible grâce à la faculté de la méthode d'HMIR à explorer et quantifier l'erreur paramétrique des modèles. Nous montrons ainsi qu'inclure une paramétrisation des effets radiatifs 3D, proposée dans le code de rayonnement ecRad, permet d'éviter que le tuning à l'échelle des processus ne mène à une surestimation des fractions nuageuses pour compenser un manque de réflectivité intrinsèque aux modèles radiatifs 1D. Cette étude illustre notamment comment la méthode d'HMIR peut guider des choix de modélisation. Cette thèse a coïncidé avec l'appropriation de la méthode d'HMIR dans la communauté scientifique française de modélisation du climat. Elle a été l'occasion de pratiquer cette nouvelle méthode pour les expériences de cette thèse et également pour un panel plus large d'expériences auxquelles elle a contribué. Les réflexions et prises de recul nourries de cette pratique sont partagées dans la dernière partie de cette thèse.

  • Open Access

    Comment on egusphere-2025-5329

    by Mamadou Lamine Thiam, Frédéric Hourdin, Jean-Yves Grandpeix, Catherine Rio, Maëlle Coulon-Decorzens

    Cold pools, formed under clouds by the evaporation of precipitation, play a central role in maintaining and organizing atmospheric convection. It is suspected that their absence in climate models may lead to significant errors in the representation of convection, such as the premature convection extinction after sunset. The introduction of a cold pool parameterization into the LMDZ climate model has significantly improved the representation of convection, in particular its diurnal cycle. However, this parameterization had not yet been accurately evaluated in terms of representing the cold pool properties. This work provides for the first time such an evaluation based on Large Eddy Simulation (LES). First, we evaluate the physical relationships underlying the cold pool model in the LES, then, in a second step, its behavior when coupled with the deep convection scheme in the single-column version of LMDZ. The analyses carried out demonstrate the relevance of the assumptions underlying the parameterization. The initial version actually captures the main characteristics of LES cold pools but also exhibits some biases. We show how substantial modifications to the cold pool scheme and a readjustment of certain free parameters helped reduce those biases significantly. The remaining flaws could be corrected by adding convective mixing through thermal plumes within the cold pools and by modeling the evolution of cold pools number density rather than imposing it.

  • Open Access

    Comment on egusphere-2025-5329

    by Mamadou Lamine Thiam, Frédéric Hourdin, Jean-Yves Grandpeix, Catherine Rio, Maëlle Coulon-Decorzens

    Cold pools, formed under clouds by the evaporation of precipitation, play a central role in maintaining and organizing atmospheric convection. It is suspected that their absence in climate models may lead to significant errors in the representation of convection, such as the premature convection extinction after sunset. The introduction of a cold pool parameterization into the LMDZ climate model has significantly improved the representation of convection, in particular its diurnal cycle. However, this parameterization had not yet been accurately evaluated in terms of representing the cold pool properties. This work provides for the first time such an evaluation based on Large Eddy Simulation (LES). First, we evaluate the physical relationships underlying the cold pool model in the LES, then, in a second step, its behavior when coupled with the deep convection scheme in the single-column version of LMDZ. The analyses carried out demonstrate the relevance of the assumptions underlying the parameterization. The initial version actually captures the main characteristics of LES cold pools but also exhibits some biases. We show how substantial modifications to the cold pool scheme and a readjustment of certain free parameters helped reduce those biases significantly. The remaining flaws could be corrected by adding convective mixing through thermal plumes within the cold pools and by modeling the evolution of cold pools number density rather than imposing it.

  • Open Access

    Evaluation and improvement of a cold pool parameterization against Large Eddy Simulations

    by Mamadou Lamine Thiam, Frédéric Hourdin, Jean-Yves Grandpeix, Catherine Rio, Maëlle Coulon-Decorzens

    Cold pools, formed under clouds by the evaporation of precipitation, play a central role in maintaining and organizing atmospheric convection. It is suspected that their absence in climate models may lead to significant errors in the representation of convection, such as the premature convection extinction after sunset. The introduction of a cold pool parameterization into the LMDZ climate model has significantly improved the representation of convection, in particular its diurnal cycle. However, this parameterization had not yet been accurately evaluated in terms of representing the cold pool properties. This work provides for the first time such an evaluation based on Large Eddy Simulation (LES). First, we evaluate the physical relationships underlying the cold pool model in the LES, then, in a second step, its behavior when coupled with the deep convection scheme in the single-column version of LMDZ. The analyses carried out demonstrate the relevance of the assumptions underlying the parameterization. The initial version actually captures the main characteristics of LES cold pools but also exhibits some biases. We show how substantial modifications to the cold pool scheme and a readjustment of certain free parameters helped reduce those biases significantly. The remaining flaws could be corrected by adding convective mixing through thermal plumes within the cold pools and by modeling the evolution of cold pools number density rather than imposing it.

  • Open Access

    Tunability of ice supersaturation in the upper troposphere and lower stratosphere in the LMDZ atmospheric model

    by Sidiki Sanogo, Olivier Boucher, Audran Borella, Frédéric Hourdin, Nicolas Bellouin, Étienne Vignon, Ionela Musat, Maëlle Coulon-Decorzens, Adriana Sima, Frederique Cheruy

    Ice supersaturation (ISS) is an atmospheric phenomenon that is necessary for the in situ formation of cirrus clouds and persistent contrails. Including ISS in climate models is essential for more realistic modelling of ice clouds, but this remains difficult because the mechanisms involved are not yet fully understood. In this study, we present a parameterisation of ISS in the LMDZ atmospheric component of the Institut Pierre-Simon Laplace climate model and address the question as to whether this parameterisation can be tuned to fit observational constraints on ISS while maintaining the realism of many other LMDZ features. To this end, temperature biases in the upper troposphere and lower stratosphere were corrected using a nudging approach prior to the tuning phase. Tuning was performed using a statistical tool based on history matching with iterative refocusing and Gaussian processes to adjust 16 free parameters of ISS and large-scale condensation and convection parameterisations in LMDZ. This method identified three LMDZ configurations that best match the observed frequency of occurrence of ISS regions (ISSRs) at 250 hPa in the mid-latitudes of the Northern Hemisphere while also meeting global (top-of-atmosphere radiative budget) and regional (cloud fraction and radiative effects) observational constraints. The main shortcoming of these model configurations is their underestimation of the frequency of occurrence of ISSRs at 200 hPa. This suggests that a realistic representation of tropopause dynamics is an important consideration when developing and choosing climate models to study the radiative forcing of contrails.

  • Open Access

    Designing a fully-tunable and versatile TKE-l turbulence parameterization for atmospheric models

    by Étienne Vignon, Khadija Arjdal, Frederique Cheruy, Maëlle Coulondecorzens, Clément Dehondt, Thomas Dubos, Sébastien Fromang, Frédéric Hourdin, Lucas Lange, Lea Raillard, et al.

    This study presents the development of a TKE-l parameterization of the diffusion coefficients for the representation of turbulent diffusion in neutral and stable conditions in large-scale atmospheric models. The parameterization has been carefully designed to be completely tunable in the sense that all adjustable parameters have been clearly identified and their number minimized as much as possible to help the calibration and to thoroughly assess the parametric sensitivity. We choose a mixing length formulation that depends on both static stability and wind shear to cover the different regimes of stable boundary layers. We follow a heuristic approach for expressing the stability functions and turbulent Prandlt number in order to guarantee the versatility of the scheme and its applicability for planetary atmospheres composed of an ideal and perfect gas such as that of Earth and Mars. Particular attention has also been paid to the numerical stability at typical time steps used in General Circulation Models. Test, parametric sensitivity assessment and preliminary tuning are performed on single-column idealized simulations of the weakly stable boundary layer. The robustness and versatility of the scheme are also assessed through its implementation in the LMDZ General Circulation Model and the Mars Planetary Climate Model and by running simulations of the Antarctic and Martian nocturnal boundary layers.