Publications

2022

• Multifractal intermittency of forced and free turbulence studied with the gyroscope cascade model
  
  • Li Xin
  , 2022. Turbulence is one of the fundamental unresolved problems of classical physics, despite its manifestation in many fields, including engineering, for example in wind energy. This is linked to our lack of knowledge of properties of the deterministic 3D Navier-Stokes (NS) equations as basic as the existence and uniqueness of its solutions. This doesn't prevent researchers and engineers from using it. With the help of statistical methods, the mechanism of turbulence has been partially revealed, such as the energy transfer process. For instance, turbulence closure models, such as the Eddy Damped Quasi-Normal Markovian model, have been introduced to partially account for the infinite hierarchy of moment equations caused by the non-linear term in the NS equation. They have highlighted the possibility of backscattered energy from the energy spectrum peak to the largest eddies and therefore modification of the energy decay law of turbulence.However, these advancements do not take into account the fundamental characteristic of turbulence: intermittency, which means that turbulence is extremely heterogeneous and leads to a large discrepancy between the empirical evidence and these models.We therefore chose the deterministic Scaling Gyroscope Cascade (SGC) model, to investigate the multifractality of intermittency. SGC is based on a parsimonious discretisation of the Bernoulli's form of the NS equations in Fourier space that well preserves the triad interaction of a parent eddy and its child eddies, generating step by step a strong intermittency.Firstly, the Python codes for the three explicit numerical simulation methods - the Euler method, the fourth order Runge-Kutta method and the slaved Adams-Bashforth method - are presented and tested to determine the most efficient numerical simulation approach for the SGC model.It comes out that the Euler method is the most effective numerical simulation method by comparing the running time and maximum memory.Besides, the spatial structure of the SGC model suggests that the computing complexity increases exponentially with the number of cascade steps.Then, the intermittency of SGC model at large cascade steps is investigated by injecting various forcings. The existence of spatial-temporal fluctuations is confirmed with the help of a statistical analysis of the energy flux in the inertial range. The probability distribution of these fluctuations has tails that are much heavier than those of a Gaussian distribution. To get more detailed insights, the analysis is pursued in the Universal Multifractal (UM) framework, based on stochastic multiplicative cascades that are both stable and attractive. These cascades are determined by only a few UM parameters that are physically meaningful for any cascade models , including the SGC.Among the various obtained results, we demonstrate that the key multifractality index is significantly less than 2 thus questioning the log-normal model still often used for hydrodynamic turbulence.Last but not least, we revisit with the help of SGC the energy decay of a free turbulence taking into account the intermittency. Due to the latter, the evidence of the energy backscatter term is more complex to demonstrate, as well as its impacts on the energy decay law. But the phenomenology remains the same, although with intermittency effects, e.g., energy is stored at large scales by puffs, no longer in a continuous manner. (10.70675/a422b221z1eb5z4232zaf37zdbc2f2507b20)
  DOI : 10.70675/a422b221z1eb5z4232zaf37zdbc2f2507b20
• A three-year evolution and comparison of the bla genes in pathogenic and non-pathogenic Escherichia coli isolated from young diarrheic and septicaemic calves in Belgium
  
  • Guérin Virginie
  • Farchi Alban
  • Cawez Frédéric
  • Mercuri Paola
  • Lucas Pierrick
  • Blanchard Yannick
  • Saulmont Marc
  • Mainil Jacques
  • Thiry Damien
  Research in Veterinary Science, 2022, 152, pp.647-650. Escherichia coli producing Extended-Spectrum-β-Lactamases (ESBL) are a major public health hazard worldwide. The most frequent ESBL belong to the CTX-M family. This study follows their prevalence in pathogenic and non-pathogenic ESBL-producing E. coli isolated from young diarrheic and septicaemic calves over three calving seasons. The triplex PCR targeted three main groups: CTX-M-1, CTX-M-2 and CTX-M-9. Of the 394 isolates studied, 388 (98.5%) were positive, with a majority of CTX-M-1 (243, 61.7%), following by CTX-M-9 (74, 18.8%) and CTX-M-2 (64, 16.2%). The progressive decrease of ESBL-resistance of pathogenic E. coli is not linked to any shift in genetic background, blaCTX-M genes still present in 99% of the isolates, or to the proportion of the three CTX-M groups. Moreover, no significant difference was observed in the CTX-M content between pathogenic and non-pathogenic E. coli. (10.1016/j.rvsc.2022.09.037)
  DOI : 10.1016/j.rvsc.2022.09.037
• development of a CFD time scheme for indoor airflow applications
  
  • Galante Amino Hector
  , 2022. Characterising indoor air flow is an important stake in a reglementary and social context of energy and thermal comfort optimisation of buildings. Concomitantly, the indoor air quality is responsible for a consequent number of deaths world wide and arouse a growing interest of the scientific community. The important number of studies related to the COVID virus propagation emphasises the importance of studying such environments. In this purpose, numerical simulations are a powerful tool to predict the indoor physical phenomena while being cost-less compared to experimental measurements. This thesis focuses on the development of a local scale simulation (CFD) scheme for indoor air flow in order to perform residential (indoor air quality and environment design) and industrial (nuclear safety, sport facilities ventilation) studies.After presenting the aforementioned context in the first chapter, an identification of the main physical phenomena driving the indoor air flow is made in the second chapter. The existing convection models are presented and a choice of the governing equations to be used is made.In the third chapter, to meet the modelling challenges presented, a second order conservative time scheme for variable density flow is proposed, for smooth and singular solutions. First written for dry air, the time scheme is implemented in the CFD open source solver code_saturne. The latter falls within the class of theta pressure correction schemes. The second order time convergence is reached by a time staggered variable arrangement. Moreover, the total energy is conserved thanks to the solving of the internal energy equation completed with a source term based on the kinetic energy discrete equation. Finally, the pressure variation is accounted by linearizing the equation of state, leading to a Helmholtz equation for the pressure correction. The pressure related terms are implicited, leading to faster calculations and avoiding any stability condition related to the acoustic waves. An analysis on the positivity of the thermodynamic variables is made, leading to new CFL and Fourier conditions which are studied in the manuscript. The scheme is verified and validated on test cases going from zero to three dimensions, for incompressible and compressible flows, chosen to represent the different indoor modelling stakes. Furthermore, it is verified that the scheme is compatible with first and second order turbulent approaches (RANS, LES).In the fourth chapter, the dry air time scheme is extended to moist air applications, including variable properties, another equation of state and phase change. Thermodynamic equilibrium is considered and the water mass fraction (in both states) is transported as a scalar. In order to use the set of equations chosen previously, the phase change is accounted using the Newton method related to the solved internal energy. A numerical analysis, verification and validation are made as well.Finally, in the fifth chapter, the numerical tool is applied on the study of the Pierre de Coubertin handball stadium, in the context of the Paris 2024 Olympic Games. The numerical mesh is generated from a three dimensional cloud of points, created from laser measurements. First simulations are performed to identify the stadium dynamic and thermal interest zones and a protocol is proposed for an experimental campaign. Then, a numerical validation is made on the evolution of particles during the french league handball final, where fog sources were ignited outside the stadium, promoting a particle concentration peak inside the system. (10.70675/40c3536az0edaz4a64za999z9b62350e230d)
  DOI : 10.70675/40c3536az0edaz4a64za999z9b62350e230d
• Non-local metrics applied to the comparison of CO2 plumes and their sensitivities to mesoscale meteorology
  
  • Vanderbecken Pierre J
  • Dumont Le Brazidec Joffrey
  • Farchi Alban
  • Bocquet Marc
  • Roustan Yelva
  • Potier Élise
  • Broquet Grégoire
  , 2022.
• French ARGONAUT project: Inferring pollutants (NOx, CO, NMVOCs) and CO2 emissions at high resolution over France using Sentinel-5P and the CIF-CHIMERE inversion system
  
  • Dufour Gaëlle
  • Broquet Grégoire
  • Bocquet Marc
  • Colette Augustin
  • Berchet Antoine
  • Coman Adriana
  • Descombes Gaël
  • Dumont Le Brazidec Joffrey
  • Farchi Alban
  • Fortems‐cheiney Audrey
  • Pison Isabelle
  • Plauchu Robin
  • Potier Élise
  • Roustan Yelva
  • Savas Dilek
  • Siour G.
  • Vanderbecken Pierre J
  , 2022.
• Study of atmospheric dispersion under low wind conditions in an urban environment, first results
  
  • Bounouas Hanane
  • Roupsard Pierre
  • Dupont Eric
  • Roustan Yelva
  • Chardeur Johann
  • Connan Olivier
  • Hebert Didier
  • Laguionie Philippe
  • Maro Denis
  • Renard Hugo
  • Rozet Marianne
  • Regi Thaddé
  • Carissimo Bertrand
  • Faucheux Aurélien
  • Lefranc Yannick
  , 2022. All atmospheric conditions must be considered in the impact calculation of industrial facilities. In low wind conditions (wind speed below 2m.s-1), the meandering (low frequency horizontal wind oscillation) becomes one of the predominant physical processes which drive atmospheric dispersion of pollutants. Experimentally, it can be identified by analyzing the autocorrection function of the horizontal wind speed components. However, modeling these situations is more complicated, and most models are unable to correctly reproduce the turbulent flow structure and the resulting plume dispersion. The CFD models could overcome these limits by adapting the existing modeling approaches to low wind speed situations. In this study, we present the first analyses of datasets from an experimental campaign that has been performed at SIRTA in the south of Paris in 2020 as well as first simulation results obtained with a CFD model using two different modelling approaches: stationary and pseudo stationary conditions. The analysis of wind data acquired during one selected time period of the campaign allowed to identify the presence of meandering and to estimate its period. Compared to concentration measurements, the first modelling approach seems to lower lateral dispersion while the second approach seems to give closer results.
• Simulations of street-canyon air-quality using fluid dynamics and aerosol modelling
  
  • Wang Yunyi
  • Flageul Cédric
  • Lin Chao
  • Ooka Ryozo
  • Kikumoto Hideki
  • Kim Youngseob
  • Sartelet Karine
  , 2022. High concentrations of nitrogen dioxide and particulate matter are often observed locally in streets. Because of the spatial resolution limit, regional-scale chemical transport models cannot reproduce these high concentrations. Traditional local-scale methods such as computational fluid dynamics (CFD) often neglect chemical reactions and aerosol dynamics, which leads to inaccuracy in the simulation of local air quality. In this study, 2D CFD simulations performed by Code_Saturne and OpenFOAM and coupled with the chemical aerosol module SSH-Aerosol are used to model pollutant dispersion, chemical reactions and aerosol dynamics during a period of 12 hours (from 4 a.m. to 4 p.m., local time, GMT+2h) in a street of Greater Paris. For both CFD codes, the setup is validated by comparing the simulated NO2 and PM10 concentrations with measurements. The impact of turbulence model and coupling strategy on reactive and non-reactive pollutant concentrations is assessed by comparing the concentrations simulated by two codes. This comparison of the CFD tools provides a qualitative estimation of the uncertainty associated with the modelling of the atmospheric flow and of the coupling between dispersion, chemistry and aerosol dynamics. In order to understand the impact of chemical processes on aerosol formation, sensitivity tests concerning gas chemistry and aerosol dynamics are conducted. A non-neglectable under-estimation of some pollutant concentrations is observed when gas chemistry and aerosol dynamics are not taken into account. Gas chemistry significantly increases NO2 concentrations in the street, which is underestimated by 41% on average when gas chemistry is not considered. Although the impact of gas chemistry on inorganic and organic condensables is limited, inorganic and organic aerosol concentrations in the street are largely impacted by aerosol dynamics. For inorganic aerosols the concentrations increase because of the formation of ammonium nitrate, partly due to the ammonia emission by traffic and partly due to the lack of thermodynamic equilibrium between gas and aerosols in the background regional concentration with high concentrations of nitric acid. For organic aerosols, the concentrations are strongly influenced by the increase of inorganic aerosols.
• Summary of recent progress and recommendations for future research regarding air pollution sources, processes, and impacts in the Mediterranean region
  
  • Dulac François
  • Hamonou Eric
  • Sauvage Stéphane
  • Kanakidou Maria
  • Beekmann Matthias
  • Desboeufs Karine
  • Formenti Paola
  • Becagli Silvia
  • Di Biagio Claudia
  • Borbon Agnès
  • Denjean Cyrielle
  • Gheusi François
  • Gros Valérie
  • Guieu Cécile
  • Junkermann Wolfgang
  • Kalivitis Nikolaos
  • Laurent Benoît
  • Mallet Marc
  • Michoud Vincent
  • Nabat Pierre
  • Sartelet Karine
  • Sellegri Karine
  , 2022, pp.543-571. Mediterranean atmospheric pollution sources, processes, and impacts are summarized in this chapter. The companion Volume 1 describes the context and the distribution of gaseous and particulate pollutants. The present volume is composed of six sections that make the synthesis of our knowledge on air pollutant sources (Part V), atmospheric chemical processes (VI), aerosol properties (VII), atmospheric deposition fluxes (VIII), and the impacts of air pollution on the chemical composition of precipitation and climate (IX) and on human health and ecosystems (X). The conclusions presented in this volume demonstrate large variety of impacts of air pollution on the environment in the Mediterranean, including the impacts on human health, ecosystems, and climate. Accurate emission evaluation of gaseous and particulate pollutants and understanding of their chemical transformation are critical for representation and prediction of atmospheric pollution and climate change in this region. Long-term observations of physical and chemical parameters describing atmospheric composition together with climate variables are essential to constrain models and to better quantify interactions between climate change and air quality. Addressing the impacts of pollution and supporting mitigation measures require holistic and integrated approach to the related studies, putting together scientific communities working on atmosphere, on health, as well as on both terrestrial and marine ecosystems. (10.1007/978-3-030-82385-6_25)
  DOI : 10.1007/978-3-030-82385-6_25
• Sources and processes influencing black carbon over Alaska during wintertime
  
  • Ioannidis Eleftherios
  • Law Kathy S.
  • Raut Jean-Christophe
  • Marelle Louis
  • Onishi Tatsuo
  • Bekki Slimane
  • d'Anna B.
  • Temime-Roussel Brice
  • Barret B.
  • Roberts Tjarda J
  • Andrews Elisabeth
  • Ohata Sho
  • Mori Tatsuhiro
  • Kondo Yutaka
  • Kim Youngseob
  • Taketani Fumikazu
  • Kanaya Yugo
  • Granier Claire
  • Quinn Patricia K.
  • Pratt Kerri A.
  • Cesler‐maloney Meeta
  • Mao Jingqui
  • Simpson William R.
  , 2022.
• The December 2016 extreme weather and particulate matter pollution episode in the Paris region (France)
  
  • Foret Gilles
  • Michoud Vincent
  • Kotthaus Simone
  • Petit J.-E.
  • Baudic A.
  • Siour G.
  • Kim Y.
  • Doussin J.-F.
  • Dupont J.-C.
  • Formenti P.
  • Gaimoz C.
  • Ghersi V.
  • Gratien Aline
  • Gros Valérie
  • Jaffrezo Jean-Luc
  • Haeffelin Martial
  • Kreitz M.
  • Ravetta François
  • Sartelet Karine
  • Simon Laura
  • Té Y.
  • Uzu Gaëlle
  • Zhang S.
  • Favez O.
  • Beekmann Matthias
  Atmospheric Environment, Elsevier, 2022, 291 (15 December), pp.119386. Highlights • Study of an intense winter episode of particulate pollution in the Paris region. • Exceptional stagnant conditions explain the highest concentrations. • Organic matter of local origin dominates the chemical composition of the particles with also a strong nitrate component from traffic. • Significant values of oxidative potential are observed. (10.1016/j.atmosenv.2022.119386)
  DOI : 10.1016/j.atmosenv.2022.119386
• N, H and C-atoms Density in Flowing Afterglows of Microwave R/N2-H2 and R/N2-CH4 Discharges with R=N2, He, Ar and Applications to TiO2 Surface Nitriding
  
  • Ricard A.
  • Amorim J.
  • Abdeladim M.
  • Sarrette Jean-Philippe
  • Kim Y.
  , 2022, pp.61-117. Variations of and -atoms density have been determined along the reduced pressure flowing afterglows of microwave and discharges with , He and Ar. Density of and -atoms and other nitrogen active species such as were obtained from that of -atoms calibrated by titration, using the method of band intensity ratios in several conditions (between early and late afterglows). It has been obtained in addition the density of O-atoms and NO molecules coming from air impurity. It has been deduced the wall destruction probability of and -atoms on the quartz afterglow tube , and . The effects of and -atoms on -atoms inclusion inside surfaces are reported. (10.9734/bpi/mono/978-93-5547-643-2/CH13)
  DOI : 10.9734/bpi/mono/978-93-5547-643-2/CH13
• Non-local metrics applied to the comparison of CO2 plumes and their sensitivities to mesoscale meteorology
  
  • Vanderbecken Pierre J
  • Dumont Le Brazidec Joffrey
  • Farchi Alban
  • Bocquet Marc
  • Roustan Yelva
  • Potier Élise
  • Broquet Grégoire
  , 2022.
• Car fleet synthesis for agent-based mobility models
  
  • Lannes Marjolaine
  • Coulombel Nicolas
  • Roustan Yelva
  , 2022. As various countries seek to improve air quality by enforcing vehicle fleet regulation measures, better understanding the determinants of household car ownership and car type choice is critical. This would provide better inputs for agent-based mobility models, which can compute traffic-related daily emission profiles based on a synthetic population and a synthetic vehicle fleet. Usually, car type choice and ownership are estimated from household characteristics using discrete choice models (Jong and al. 2004). But recent studies point out the contribution of machine learning methods for the estimation of car ownership (Paredes and al. 2017; Dixon and al. 2021). This work investigates the performance of several classification models in the prediction of car type and ownership at the household level. We compare a discrete choice model against various machine learning classification methods (e.g. Gradient Boosting, Random Forest) for the estimation of household car ownership, fuel type and car pollutants emissions standards. Explanatory variables include household socio-economic characteristics as well as local and metropolitan accessibility variables (parking availability, public transport accessibility). The methodology is applied to the Paris area, using the “Enquête Globale de Transport” mobility survey. Considering the Matthew Correlation Coefficient, F1 score and Cohen’s kappa as evaluation metrics, we conclude that logistic regression slightly outperforms AI models for car ownership whereas Gradient Boosting classifier gets the best results for vehicle type estimation. Our results show a strong relationship for car ownership prediction and a slight agreement for fuel type and emission standard predictions, with a major importance of household composition and accessibility variables. This work emphasizes that discrete choice and AI models should rather be brought together than opposed to improve the performance of vehicle fleet synthesis in agent-based modeling.
• State, global, and local parameter estimation using local ensemble Kalman filters: Applications to online machine learning of chaotic dynamics
  
  • Malartic Q.
  • Farchi A.
  • Bocquet M.
  Quarterly Journal of the Royal Meteorological Society, Wiley, 2022, 148 (746), pp.2167-2193. Abstract In a recent methodological article, we showed how to learn chaotic dynamics along with the state trajectory from sequentially acquired observations, using local ensemble Kalman filters. Here, we investigate more systematically the possibility of using a local ensemble Kalman filter with either covariance localisation or local domains, in order to retrieve the state and a mix of key global and local parameters. Global parameters are meant to represent the surrogate dynamical core, for instance through a neural network, which is reminiscent of data‐driven machine learning of dynamics, while the local parameters typically stand for the forcings of the model. Aiming at joint state and parameter estimation, a family of algorithms for covariance and local domain localisation is proposed. In particular, we show how to update global parameters rigorously using a local‐domain ensemble Kalman filter (EnKF) such as the local ensemble transform Kalman filter (LETKF), an inherently local method. The approach is tested with success on the 40‐variable Lorenz model using several of the local EnKF flavors. A two‐dimensional illustration based on a multilayer Lorenz model is finally provided. It uses radiance‐like nonlocal observations. It features both local domains and covariance localisation, in order to learn the chaotic dynamics and the local forcings. This article addresses more generally the key question of online estimation of both global and local model parameters. (10.1002/qj.4297)
  DOI : 10.1002/qj.4297
• Modèle désagrégé de choix de voiture par les ménages : comparaison des modèles de choix discret et d'apprentissage supervisé
  
  • Lannes Marjolaine
  • Coulombel Nicolas
  • Roustan Yelva
  , 2022. Alors que divers pays et métropoles cherchent à améliorer la qualité de l'air en appliquant des mesures de régulation du parc automobile, il semble de plus en plus important de comprendre les déterminants de la possession de voitures et du choix du type de véhicule par les ménages. En effet, l'Union Européenne a adopté des normes d'émissions de polluants pour les constructeurs automobiles et de nombreux pays européens ont également mis en place des zones à faibles émissions (ZFE) limitant le trafic routier dans une zone restreinte. Les ZFE et leurs impacts ont été largement évalués grâce à l'utilisation de modèles de mobilité multi-agents (e‧g. Adnan et al. 2021). La représentation d'un parc de véhicules désagrégé au niveau des ménages fournirait donc de meilleures entrées pour les modèles de mobilité, avec la perspective de calculer des profils d'émissions journalières liées au trafic sur la base d'une population et d'un parc de véhicules synthétiques et d'évaluer des scénarios prospectifs pour les ZFE. Malgré l'abondante littérature sur la modélisation de la possession de voitures, peu de recherches ont été menées sur le type de carburant des voitures possédées et encore moins sur leur ancienneté, laquelle définit leur norme d'émission. En outre, les choix de posséder une ou des voitures et de leurs types sont généralement estimés à partir des caractéristiques des ménages à l'aide de modèles de choix discrets (Purvis 1994). Mais des études récentes soulignent la contribution des méthodes d'apprentissage automatique à la modélisation des choix de transport (van Cranenburgh et al. 2021) et en particulier pour l'estimation du choix de possession de voitures (Paredes et al. 2017; Dixon et al. 2021). Cette étude vise à modéliser les choix relatifs à la possession de voitures et à leurs émissions au niveau des ménages et à comparer la performance de plusieurs modèles de classification. Pour ce faire, nous avons construit un modèle à deux étapes au niveau des ménages : estimation (1) du nombre de voitures possédées par le ménage, préalable à l’estimation (2) du type de ces véhicules, c'est-à-dire leur type de carburant et leur norme européenne d’émission (qui est donnée par l’ancienneté du véhicule). Le modèle représente ainsi trois niveaux de décisions des ménages : le nombre de voitures possédées, leur type de carburant et leur norme européenne d’émission de polluants. Nous comparons ainsi un modèle de choix discret à diverses méthodes de classification par apprentissage supervisé (par exemple Gradient Boosting, ou Random Forest) pour la prédiction de ces choix par les ménages. Les variables explicatives incluent les caractéristiques socio-économiques des ménages, par exemple la composition du ménage ou le niveau de revenu, ainsi que des variables d'accessibilité locales et métropolitaines comme la présence d’un parking à domicile ou l’accessibilité de la commune de résidence en transports en commun. La méthodologie est finalement appliquée à la région Île-de-France, à partir de l'Enquête Globale de Transport (EGT 2018). En considérant le coefficient de corrélation de Matthew, le score F1 et le kappa de Cohen comme métriques d'évaluation, nous concluons que la régression logistique surpasse légèrement les modèles d'intelligence artificielle pour la possession de voitures, tandis que le classificateur Gradient Boosting obtient les meilleurs résultats pour l'estimation du type de véhicule. Nos résultats montrent une forte relation pour la prédiction de la propriété des voitures et une légère concordance pour les prédictions du type de carburant et des normes d'émission, avec une importance prépondérante des variables de composition du ménage et d'accessibilité.
• A TIME-STAGGERED SECOND ORDER SCHEME FOR MOIST AIR VARIABLE DENSITY FLOW
  
  • Amino Hector
  • Flageul Cédric
  • Carissimo Bertrand
  • Ferrand Martin
  • Hérard Jean-Marc
  , 2022. (10.23967/eccomas.2022.032)
  DOI : 10.23967/eccomas.2022.032
• Car fleet synthesis for agent-based mobility models: a comparison of machine learning and discrete choice methods
  
  • Lannes Marjolaine
  • Coulombel Nicolas
  • Roustan Yelva
  , 2022. As various countries and metropolises seek to improve air quality by enforcing vehicle fleet regulation measures, better understanding the determinants of household car ownership and vehicle type choice is increasingly important. The European Union has as a matter of fact adopted pollutant emission standards for car manufacturers and many European countries have also established Low Emission Zones (LEZ) which limit traffic within a restricted zone. LEZs and their impacts have been extensively assessed through the use of agent-based mobility models (e.g. Dias, Tchepel, and Antunes 2016; Fosset et al. 2016; de Bok, Tavasszy, and Thoen 2022), including MATSim (e.g. Adnan et al. 2021). The representation of a disaggregated vehicle fleet at the household level would thus provide better inputs for MATSim, with the prospect of calculating traffic-related daily emission profiles based on a synthetic population and a synthetic vehicle fleet and assessing prospective scenarios for LEZs. Despite the extensive literature on modeling car ownership, little research has been conducted on the type of fuels used in cars and even less on their age or emissions standards. Moreover, car ownership and vehicle type choice are usually estimated from household characteristics using discrete choice models (Purvis 1994; Jong et al. 2004). But recent studies point out the contribution of machine learning methods for transportation choice modeling (van Cranenburgh et al. 2021) and in particular for the estimation of car ownership (e.g. Paredes et al. 2017; Kaewwichian, Tanwanichkul, and Pitaksringkarn 2019; Dixon et al. 2021). This work investigates the performance of several classification models in the prediction of vehicle ownership and vehicle type at the household level. We compare a discrete choice model against various machine learning classification methods (e.g. Gradient Boosting, Random Forest) for the estimation of household car ownership, fuel type and car pollutant emissions standards. Explanatory variables include household socio-economic characteristics (household type, income) as well as local and metropolitan accessibility variables (parking availability, public transport accessibility). The methodology is applied to the Paris region, using the “Enquête Globale de Transport” mobility survey. Considering the Matthew Correlation Coefficient, F1 score and Cohen’s kappa as evaluation metrics, we conclude that logistic regression slightly outperforms artificial intelligence models for car ownership whereas Gradient Boosting classifier gets the best results for vehicle type estimation. Our results show a strong relationship for car ownership prediction and a slight agreement for fuel type and emission standard predictions, with a preponderant importance of household composition and accessibility variables.
• Hétérogénéité des modes de transferts convectifs au sein des centrales solaires photovoltaïques
  
  • Amiot Baptiste
  • Ferrand Martin
  • Le-Berre Rémi
  • Giroux-Julien Stéphanie
  , 2022, pp.183-190. Dans le cadre des applications de centrales photovoltaïques, déterminer la température d’opération des cellules demande d’estimer le mode de déperdition thermique privilégié par chaque module. Un des verrous de cette estimation concerne l’influence du champ de vitesse du vent qui modifie localement le mode de convection dominant. Dans cette étude, l’évolution du nombre de Nusselt pour chaque module d’une centrale numérique est calculée. Un schéma périodique qui permet de simplifier la simulation est également élaboré. (10.25855/SFT2022-047)
  DOI : 10.25855/SFT2022-047
• How can we assimilate geoscientific data according to Wasserstein metric?
  
  • Vanderbecken Pierre J
  • Dumont Le Brazidec Joffrey
  • Farchi Alban
  • Bocquet Marc
  • Roustan Yelva
  • Potier Élise
  • Broquet Grégoire
  , 2022.
• Comparison of non-local metrics towards the assimilation of pollutant plumes without the double penalty
  
  • Vanderbecken Pierre J
  • Farchi Alban
  • Bocquet Marc
  • Dumont Le Brazidec Joffrey
  • Roustan Yelva
  • Broquet Grégoire
  • Potier Élise
  , 2022. (10.5194/egusphere-egu22-1754)
  DOI : 10.5194/egusphere-egu22-1754
• Intermittency, stochastic Universal Multifractals and the deterministic Scaling Gyroscope Cascade model
  
  • Li Xin
  • Schertzer Daniel
  • Roustan Yelva
  • Tchiguirinskaia Ioulia
  , 2022. <p>Intermittency is a fundamental feature of turbulence and more generally of geophysics, where its ubiquity is increasingly recognized. It corresponds to the concentration of the activity of a field, e.g. the vorticity of a flow, into very small fractions of the physical space. This induces strongly non-Gaussian fluctuations over a wide range of space-time scales. Multifractality corresponds to the fact that this concentration for increasing level of activity, in fact increasing singular behaviour, is supported by fractal sets of decreasing dimensions (and increasing codimensions). This is a general outcome of the (stochastic) multiplicative cascade models and of the universal multifractals, which statistics are defined with the help of two physically meaningful parameters:</p><ul><li>the ‘mean codimension’ C<sub>1</sub> ≥ 0 measures the mean concentration of the activity (C<sub>1</sub> = 0 for a non-intermittent field);</li> <li>the ‘multifractality index’ α ∈ (0, 2) measures how fast increases the concentration of the activity with the activity level (α=0 correspond to the monofractal case with a unique singularity / codimension C<sub>1</sub>, α= 2 corresponds to another exceptional case, the so-called ‘Log-normal’ model)</li> </ul><p>Multifractal analysis of various turbulence data, especially from lab experiments and atmospheric in-situ/remotely sensed data, have rather constantly yielded estimates of α ≈ 1.5 and C<sub>1</sub> ≈ 0.25 , although error bars are difficult to assess. However, the relation between stochastic cascades and the deterministic Navier-Stokes equations have often been brought into question. We therefore analysed in more details the relation between stochastic multiplicative cascades, namely their universality case, and the deterministic Scaling Gyroscope Cascade (SGC, [1]), whose philosophy is rather different: it is based on a parsimonious discretisation of the Fourier transform of the Bernoulli’s form of the Navier-Stokes equations:</p><p>(∂/∂t -vΔ)u(x,t)=u(x,t)∧w(x,t)-grad(α), w(x,t)=curl(u(x,t)).</p><p><br>The discretization of the Bernoulli’s form is performed along a dyadic tree structure in a 2D cut: each eddy of velocity u<sup>i</sup><sub>m</sub>has two interacting sub-eddies of velocities u<sup>2i−1</sup><sub>m+1</sub>and u<sup>2i</sup><sub>m+1</sub>, where m indexes the cascade level of wave-number k<sub>m</sub> = 2<sup>m</sup>, i ∈ [1, 2<sup>m</sup>] being the eddy location. This discretization preserves many symmetries, including the most important one: the non trivial ‘detailed energy conservation’, i.e. that nonlinearly transferred within the triad of a parent eddy and its two children.</p><p>We have performed numerous SGC simulations with a constant forcing at a low wave number, a number of cascade levels as high as N = 15 and a duration of 150 largest eddy turnover times. All these simulations display an extreme space-time intermittency. Their multifractal analysis confirms in a very robust manner the estimated α ≈ 1.5 , which is a very important result: it brings into question more than ever the relevance of the often used of the log-normal model, at least for hydrodynamic turbulence. We will present at the conference a similarly robust estimate of C<sub>1 </sub>after having clarified a recently noted, unexpected sensibility to simulation details.</p><p><strong>Keywords</strong>: intermittency; the SGC model; multifractal </p><p><strong>Reference:</strong></p><p>[1]Chigirinskaya Y, Schertzer D. Cascade of scaling gyroscopes: Lie structure, universal multifractals and self-organized criticality in turbulence[M]//Stochastic Models in Geosystems. Springer, New York, NY, 1997: 57-81.</p><p> </p> (10.5194/egusphere-egu22-7783)
  DOI : 10.5194/egusphere-egu22-7783
• Effect of vehicle fleet composition and mobility on outdoor population exposure: A street resolution analysis in Paris
  
  • Lugon Lya
  • Kim Youngseob
  • Vigneron Jérémy
  • Chrétien Olivier
  • André Michel
  • André Jean-Marc
  • Moukhtar Sophie
  • Redaelli Matteo
  • Sartelet Karine
  Atmospheric Pollution Research, Elsevier, 2022, 13 (5), pp.101365. (10.1016/j.apr.2022.101365)
  DOI : 10.1016/j.apr.2022.101365
• Modeling the Contribution of Aerosols to Fog Evolution through Their Influence on Solar Radiation
  
  • Al Asmar Lea
  • Musson-Genon Luc
  • Dupont Eric
  • Ferrand Martin
  • Sartelet Karine
  Climate, MDPI, 2022, 10 (5), pp.61. Aerosols and in particular their black carbon (BC) content influence the atmospheric heating rate and fog dissipation. Substantial improvements have been introduced to the solar scheme of the computational fluid dynamic model code_saturne to estimate fluxes and heating rates in the atmosphere. This solar scheme is applied to a well-documented case of a fog that evolves into a low stratus cloud. Different sensitivity tests are conducted. They show that aerosols have a major effect with an overestimation of the direct solar fluxes by 150 W m−2 when aerosols are not considered and a reduction of the heating of the layers. Aerosols lead to an increase of the heating rate by as much as 55% in the solar infrared (SIR) band and 100% in the Ultra-Violet visible (UV-vis) band. Taking into account the fraction of BC in cloud droplets also accentuates the heating in the layers at the top of the fog layer where water liquid content is maximum. When the BC fraction in cloud droplets is equal to 8.6 × 10−6, there is an increase of approximately 7.3 °C/day in the layers. Increasing the BC fraction leads to an increase of this heating in the layer, especially in the UV-vis band. (10.3390/cli10050061)
  DOI : 10.3390/cli10050061
• Metrics and optimal transport towards urban plume data assimilation
  
  • Vanderbecken Pierre J
  • Dumont Le Brazidec Joffrey
  • Farchi Alban
  • Bocquet Marc
  • Roustan Yelva
  • Potier Élise
  • Broquet Grégoire
  , 2022.
• A second order conservative time-staggered scheme for all speed flow
  
  • Amino Hector
  • Flageul Cédric
  • Benhamadouche Sofiane
  • Tiselj Iztok
  • Carissimo Bertrand
  • Ferrand Martin
  , 2022. A time-staggered scheme for variable density flow is presented. The compressible Navier-Stokes equations are used by this pressure correction scheme which is implemented in the collocated finitevolumes open-source computational fluid dynamics solver code saturne. The Helmholtz equation is solved for the pressure increment, taking the thermodynamic pressure into account and avoiding the acoustic time step limitation. The internal energy equation is used to compute the temperature and a numerical analysis providing conditions ensuring the positivity of the thermodynamic variables is proposed. The scheme accuracy is first verified on a 0-D pressure cooker system. Then, a discontinuous shock problem shows its capability to reproduce shocks correctly and finally, the scheme is validated on a natural convection numerical reference case.