Search

The correct choice of interface conditions and effective parameters for coupled macroscale free-flow and porous-medium models is crucial for a complete mathematical description of the problem under consideration and for accurate numerical simulation of applications. We consider single-fluid-phase systems described by the Stokes–Darcy model. Different sets of coupling conditions for this model are available. However, the choice of these conditions and effective model parameters is often arbitrary. We use large-scale lattice Boltzmann simulations to validate coupling conditions by comparison of the macroscale simulations against pore-scale resolved models. We analyse three settings (lid-driven cavity over a porous bed, infiltration problem and general filtration problem) with different geometrical configurations (channelised and staggered distributions of solid grains) and different sets of interface conditions. Effective parameters for the macroscale models (permeability tensor, boundary layer constants) are computed numerically for each geometrical configuration. Numerical simulation results demonstrate the sensitivity of the coupled Stokes–Darcy problem to the location of the sharp fluid–porous interface, the effective model parameters and the interface conditions., Bundesministerium für Bildung und Forschung, Deutsche Forschungsgemeinschaft, Projekt DEAL

A benchmark test for Euler simulations of a high-lift configuration vortex wake is presented. Four different partners have used their in-house grid generators and Euler-codes within the framework of the European research program C-Wake for calculating the nearfield vortex wake of the DLR-F11 model in take-off configuration. Compared are nearfield results up to a half span downstream of the model wing tip obtained on structured and unstructured grids, employing 2nd, 3rd and 4th order accurate schemes. For validation 5-hole probe and pressure distribution measurements carried out in the DNW-NWB are used.

The solution of Poisson-like equations defined on complex geometry is required for gyrokinetic simulations, which are important for the modelling of plasma turbulence in nuclear fusion devices such as the ITER tokamak. In this paper, we compare three solvers capable of solving this problem, in terms of the accuracy of the solution, and their computational efficiency. The first, the Spline FEM solver, uses C 1 polar splines to construct a finite elements method which solves the equation on curvilinear coordinates. The resulting linear system is solved using a conjugate gradient method. The second, the GmgPolar solver, uses a symmetric finite differences method to discretise the differential equation. The resulting linear system is solved using a tailored geometric multigrid scheme, with a combination of zebra circle and radial line smoothers, together with an implicit extrapolation scheme. The third, the Embedded Boundary solver, uses a finite volumes method on Cartesian coordinates with an embedded boundary scheme. The resulting linear system is solved using a multigrid scheme. The Spline FEM solver is shown to be the most accurate. The GmgPolar solver is shown to use the least memory. The Embedded Boundary solver is shown to be the fastest in most cases. All three solvers are shown to be capable of solving the equation on a realistic non-analytical geometry. The Embedded Boundary solver is additionally used to attempt to solve an X-point geometry, highlighting the problems with concave boundaries.

International audience; The availability of large eddy simulations (LES) models for the prediction of soot production in complex configurations is essential to the design of the new generation of combustion systems. This requires a good understanding of the processes leading to soot production and the development and validation of the corresponding models. This motivates the present calculations carried out for a laboratory scale burner (EM2Soot) designed at the EM2C laboratory to study soot production in turbulent swirled flames operating under fully-premixed rich conditions. The laser induced incandescence (LII) imaging data for the soot volume fraction (SVF) are used to test the validity of state-of-the-art numerical LES models for soot prediction. The retained models aim at achieving a good level of accuracy for a reasonable computational cost. The gaseous species kinetics is described with an analytically-reduced chemistry approach while the soot solid phase is computed using a three-equation model. This numerical strategy qualitatively reproduces the SVF in laminar premixed flames. In the considered turbulent burner, the numerical strategy captures the SVF spatial distribution but the predicted yield is notably overestimated with respect to the LII results. An extensive analysis of the flame structure as well as of the soot source terms indicates that this is due to an underestimation of soot oxidation reactions. A parametric study on rich laminar and turbulent flames is then carried out to examine the effect of soot oxidation model on the SVF prediction. This indicates that it is worth testing soot models in turbulent flame conditions to examine their capacities and reveal their shortcomings. It is also concluded that the EMSoot configuration offers an interesting situation for developing and validating soot oxidation models.

Over the last 70 years, extreme heat has been increasing at global scale [1,2], with a rapid rate in several regions including Western Europe [3]. Climate models broadly capture heat trends globally [1], but exhibit systematically weaker extreme heat trends than observations in Western Europe [4-6], together with a weaker summer warming [7,8]. The causes of this mismatch, confirmed here by the analysis of 273 latest generation coupled climate simulations, among which only a handful of them overpass observed trends, are not well understood. Here we use a circulation analogue approach [9,10] to identify the dynamical contribution to daily maximum temperature trends [11-12], and show that a substantial fraction (0.8°C [0.2°-1.4°C] of 3.4°C per global warming degree) of the trend is due to circulation changes, largely due to increases in southerly flows over Western Europe. Their rapid increase in frequency (+43% per global warming degree [10%-76%] since 1950) and persistence are largely underestimated in the 32 climate model flow simulations analyzed, as well as their overall dynamical contributions to temperature trends. The few simulations reaching the observed warming trends in extreme heat have weaker and non-significant dynamical changes, indicating compensating biases in dynamical and thermodynamical trends. These model biases in circulation trends can be due to a systematically underestimated or erroneous representation of the circulation response to external forcing, or to a systematic underestimation of interdecadal variability, or both. The former implies that future projections are too conservative, the latter that we are left with deep uncertainty regarding the pace of future summer heat in Europe: the current strong trend could weaken or increase in future decades. This calls for caution when interpreting climate projections of heat extremes over Western Europe, in particular in view of adaptation to heat waves.

Studies of emergent constraints have frequently proposed that a single metric can constrain future responses of the Earth system to anthropogenic emissions. Here, we illustrate that strong relationships between observables and future climate across an ensemble can arise from common structural model assumptions with few degrees of freedom. Such cases have the potential to produce strong yet overconfident constraints when processes are represented in a common, oversimplified fashion throughout the ensemble. We consider these issues in the context of a collection of published constraints and argue that although emergent constraints are potentially powerful tools for understanding ensemble response variation and relevant observables, their naïve application to reduce uncertainties in unknown climate responses could lead to bias and overconfidence in constrained projections. The prevalence of this thinking has led to literature in which statements are made on the probability bounds of key climate variables that were confident yet inconsistent between studies. Together with statistical robustness and a mechanism, assessments of climate responses must include multiple lines of evidence to identify biases that can arise from shared, oversimplified modelling assumptions that impact both present and future climate simulations in order to mitigate against the influence of shared structural biases.

In atmospheric chemistry retrievals and data assimilation systems, observation errors associated with satellite radiances are chosen empirically and generally treated as uncorrelated. In this work, we estimate inter-channel error covariances for the Infrared Atmospheric Sounding Interferometer (IASI) and evaluate their impact on ozone assimilation with the chemistry transport model MOCAGE (Modèle de Chimie Atmosphérique à Grande Echelle). The method used to calculate observation errors is a diagnostic based on the observation and analysis residual statistics already adopted in many numerical weather prediction centres. We used a subset of 280 channels covering the spectral range between 980 and 1100 cm−1 to estimate the observation-error covariance matrix. This spectral range includes ozone-sensitive and atmospheric window channels. We computed hourly 3D-Var analyses and compared the resulting O3 fields against ozonesondes and the measurements provided by the Microwave Limb Sounder (MLS) and by the Ozone Monitoring Instrument (OMI). The results show significant differences between using the estimated error covariance matrix with respect to the empirical diagonal matrix employed in previous studies. The validation of the analyses against independent data reports a significant improvement, especially in the tropical stratosphere. The computational cost has also been reduced when the estimated covariance matrix is employed in the assimilation system, by reducing the number of iterations needed for the minimizer to converge.

Two new NEMO-SAS (stand alone surface module) configurations are derived from a global ORCA grid. A FORTRAN tool is developed to transform the ORCA input files and define the OASIS weights & addresses to map fluxes and surface variables between one global and two hemispheric grids. A tri-component coupled system is set up, including an ocean on the ORCA2 global grid and two hemispheric NEMO-SAS with SI3 sea-ice. Removing North Pole folding communications in ice model helps to improve computing performance, since our ORCA2 coupled configuration is twice faster and two times cheaper than its single executable counterpart. The North Pole folding free hemispheric grid discretisation simplifies the coding of the future neXtSIM model, while relying on the same grid points than the ocean, thus avoiding inaccurate interpolation between ice and ocean grids

International audience; Computational prediction of thermoacoustic instabilities arising in gas turbine and aero-engine combustors still remains a challenge especially if fuel is injected in a liquid spray form. This study shows that, in LES of such a combustor, the treatment of the liquid fuel film created on the walls of the injection system affects the mean flame weakly, but modifies the flame dynamics strongly. The configuration used for this work is the experimental setup SICCA-spray available at EM2C laboratory in Paris. First steady spray flame measurements are used to validate the LES Euler-Lagrange approach. Two modelling strategies for the interaction between the liquid fuel and the injector walls are tested with a negligible impact on the flame shape and structure. In the second part the same comparison is applied to another operating condition where a self-sustained thermo-acoustic limit-cycle is experimentally observed. In that case resonant coupling is achieved with LES, confirming the adequacy of the approach but only when the film layer is taken into account. Indeed, contrarily to the stable configuration, the difference between the two Lagrangian boundary conditions is shown to have a major impact on the feedback mechanism leading to the thermoacoustic oscillation.

The present work aims at better understanding regional climate–aerosol interactions by studying the relationships between aerosols and synoptic atmospheric circulation over the Euro-Mediterranean region. Two 40-year simulations (1979–2018) have been carried out with version 6.3 of the Centre National de Recherches Météorologiques (National Centre for Meteorological Research) – Aire Limitée Adaptation dynamique Développement InterNational (CNRM-ALADIN) regional climate model, one using interactive aerosols and the other one without any aerosol. The simulation with aerosols has been evaluated in terms of different climate and aerosol parameters. This evaluation shows a good agreement between the model and observations, significant improvements compared to the previous model version and consequently the relevance of using this model for the study of climate–aerosol interactions over this region. A first attempt to explain the climate variability of aerosols is based on the use of the North Atlantic Oscillation (NAO) index. The latter explains a significant part of the interannual variability, notably in winter for the export of dust aerosols over the Atlantic Ocean and the eastern Mediterranean, and in summer for the positive anomalies of anthropogenic aerosols over western Europe. This index is however not sufficient to fully understand the variations of aerosols in this region, notably at daily scale. The use of “weather regimes”, namely persisting meteorological patterns, stable at synoptic scale for a few days, provides a relevant description of atmospheric circulation, which drives the emission, transport and deposition of aerosols. The four weather regimes usually defined in this area in winter and in summer bring significant information to answer this question. The blocking and NAO+ regimes are largely favourable to strong aerosol effects on shortwave surface radiation and near-surface temperature, either because of higher aerosol loads or because of weaker cloud fraction, which reinforces the direct aerosol effect. Inversely, the NAO− and Atlantic Ridge regimes are unfavourable to aerosol radiative effects, because of weaker aerosol concentrations and increased cloud cover. This study thus puts forward the strong dependence of aerosol loads on the synoptic circulation from interannual to daily scales and, as a consequence, the important modulation of the aerosol effects on shortwave surface radiation and near-surface temperature by atmospheric circulation. The role of cloud cover is essential in this modulation as shown by the use of weather regimes.

International audience; A methodology is developed where the acoustic response of a swirl stabilized flame is obtained from a reduced set of simulations. Building upon previous analytical flame transfer functions, a parametrization of the flame response is first proposed, based on six independent physical parameters: a Strouhal number, the mean flame angle with respect to the main flow direction, the vortical structures convection speed, a swirl intensity parameter, a time delay between acoustic and vortical perturbations, as well as a phase shift between bulk and local velocity signals. It is then shown how these parameters can be deduced from steady and unsteady simulations. The methodology is applied to a laboratory scale premixed swirl stabilized flame exhibiting features representative of real aero-engines. In this matter, cold and reactive flow Large Eddy Simulations are first validated by comparing results with reference data from experiments. The high fidelity simulations are seen to be able to capture the flame structure and velocity profiles at different locations while forced flame dynamics for the frequency range of interest also match the experimental data. From the same analytical transfer function model, three methodologies of increasing complexity are presented for the determination of the model parameters, depending on the available data or computational resources. A first estimation of the flame acoustic response is obtained by evaluating parameters from a single stationary flame simulation in conjunction with analytical estimations for the acoustic-convective time delay. Flame dynamics and swirl related parameters can then be determined from a series of robust treatments on pulsed simulations data to improve the model accuracy. It is shown that good qualitative agreement for the flame transfer function can be obtained from a single non-forced simulation while quantitative agreement over the frequency range of interest can be obtained using additional reactive or non-reactive pulsed simulations at one single forcing frequency corresponding to a local gain minimum. The method also naturally handles different perturbation levels.

Bridging physics and deep learning is a topical challenge. While deep learning frameworks open avenues in physical science, the design of physically consistent deep neural network architectures is an open issue. In the spirit of physics-informed neural networks (NNs), the PDE-NetGen package provides new means to automatically translate physical equations, given as partial differential equations (PDEs), into neural network architectures. PDE-NetGen combines symbolic calculus and a neural network generator. The latter exploits NN-based implementations of PDE solvers using Keras. With some knowledge of a problem, PDE-NetGen is a plug-and-play tool to generate physics-informed NN architectures. They provide computationally efficient yet compact representations to address a variety of issues, including, among others, adjoint derivation, model calibration, forecasting and data assimilation as well as uncertainty quantification. As an illustration, the workflow is first presented for the 2D diffusion equation, then applied to the data-driven and physics-informed identification of uncertainty dynamics for the Burgers equation.

International audience; Ensemble variational methods are being increasingly used in the field of geophysical data assimilation. Their efficiency comes from the combined use of ensembles, which provide statistics estimates, and a variational analysis, which handles nonlinear operators through iterative optimization techniques. Taking model error into account in four-dimensional ensemble variational algorithms is challenging because the state trajectory over the data assimilation window (DAW) is no longer determined by its sole initial condition. In particular, the control variable dimension scales with the DAW length, which yields a high numerical complexity. This is unfortunate since accuracy improvement is expected with longer DAWs. Building upon the work of [P. Sakov and M. Bocquet, Tellus A, 70 (2018), 1414545], this paper discusses how to algorithmically construct and numerically test an iterative ensemble Kalman smoother with additive model error (IEnKS-Q) which is thought to be the natural weak constraint generalization of the IEnKS [M. Bocquet and P. Sakov, Quart. J. Roy. Meteorol. Soc., 140 (2014), pp. 1521--1535], as well as the generalization of IEnKF-Q [P. Sakov, J. Haussaire, and M. Bocquet, Quart. J. Roy. Meteorol. Soc., 144 (2018), pp. 1297--1309] to general DAWs. The number of model evaluations per cycle of the IEnKS-Q is also examined. Solutions based on perturbation decomposition are proposed to dissociate those numerically costly evaluations from the control variable dimension.

The fuel films that can be generated during the injection process in gasoline direct injection engines are the most important factor in carbon particle mass and number. They also have an influence on combustion chamber and injector tip deposits. A model experiment was set up to study a liquid film, its evaporation, and combustion with soot generation on a metal plate in realistic engine conditions. The experiment was conducted in a dedicated constant volume vessel. A liquid fuel injection system (with injection pressures up to 100 bar) directs the spray onto a plate with a controlled temperature in the range of 80 °C–200 °C. The resulting liquid film and vaporization process were studied when subjected to interaction with a laminar spherical flame. A blend of four components was used as a gasoline surrogate. The liquid film spreading, thickness, and evaporation rates were initially measured in ambient conditions. Mie scattering and schlieren measurements in the chamber conditions returned a qualitative correlation of the vaporized area with the surface temperature. Fluorescence of the light and heavy fuel components was used to quantify the influence of the vaporization process on soot production. Simultaneous measurements of natural luminosity and KL factor were analyzed to understand the process of soot production. The results showed a critical wall temperature of 120 °C at which the maximum quantity of soot is generated, which can be due to the quantity and composition of fuel film in interaction with the entrainment flows generated during combustion.

61 p.; Le projet Aqui-FR vise à construire un outil de suivi, de prévision et de projections de la ressource en eau souterraine en France en s’appuyant sur les modélisations hydrogéologiques régionales existantes et les outils de suivi et prévisions météorologiques existants. La première phase du projet Aqui-FR a permis de mettre en place et d’exploiter une plateforme de modélisation hydrogéologique nationale rassemblant trois modèles (codes numériques) hydrogéologiques, sur 19 applications (dont 6 bassins karstiques) couvrant 149 000 km2 et représentant 57 couches aquifères ou aquitards. Cette deuxième phase a permis de faire évoluer la plateforme, notamment par l’extension des applications (Tarn & Garonne), recalage avec les flux de la dernière version de SURFEX, et évolution des pré- et post- traitements. Une étape importante a été l’évaluation des simulations sur une longue période de la ré-analyse (60 ans) qui a conduit à la publication de Vergnes et al., 2020. Cette évaluation montre que si des biais importants peuvent persister entre observations et simulations de la piézométrie, ils sont fortement réduits lorsque l’indice de piézométrie standardisé (SPLI) est utilisé. Globalement, Aqui-FR représente bien les évolutions observées. Cette analyse permet d’identifier des variabilités pluriannuelles avec des périodes de plusieurs années en hautes ou basses eaux. Un effort particulier a été mené pour évaluer la reconstitution des basses eaux et plus particulièrement des sécheresses depuis 1958 (stages masters Nalivaev, 2020; Mounier, 2021). Cette évaluation est rendue difficile par le fait que peu de piézomètres ont des longues chroniques observées. L’utilisation du SPLI est alors impossible. L’évaluation de l’occurrence de niveaux piézométriques en dessous du 1er décile sec permet de montrer une bonne adéquation entre observations et simulations sur la période de plus de 60 ans, avec cependant une tendance à sous-estimer ces périodes sèches depuis 2015. Les causes ne sont pas explicitées, mais, une mise à jour des données de prélèvements semble être un bon point de départ pour tenter d’améliorer ce biais. L’étape majeure franchie durant cette phase par Aqui-FR est l’évaluation puis le passage en mode opérationnel de prévisions saisonnières de la ressource en eau souterraine. L’évaluation sur une période prospective de 25 ans a été menée lors du post-doc de Delphine Leroux financée d’abord par la convention service climatique, puis par ce projet. L’évaluation de ces prévisions d’ensemble est réalisée en comparant les prévisions à la ré-analyse. Cette évaluation montre un fort potentiel des prévisions saisonnières, même à 6 mois d’échéance, pour les prévisions débutant en début d’année civil jusqu’à Juillet, et plus particulièrement à partir de Mars, période à laquelle la plus grande partie de la recharge des nappes a déjà eu lieu. Les scores se dégradent sensiblement lorsque les prévisions intègrent les périodes de fortes recharges, i.e. les mois de novembre à janvier. La plus-value des prévisions saisonnières météorologiques, alors réalisées par la version 6 du système de prévisions saisonnières de Météo-France est apparente, comparée à des prévisions climatiques basées sur des années historiques (prévision climatique), ainsi que comparée à des prévisions naïves considérant une persistance de l’anomalie par rapport à la normale sur les prochains mois. La capacité d’évaluer l’extension des sécheresses restituées par la ré-analyse a également été évaluée, et les scores sont bons, avec une grande majorité des membres de la prévision se situant en état de sécheresse. Ce travail a fait l’objet d’une publication en préparation, (Leroux et al., s.d.) qui n’a pu être finalisé à ce stade suite à des aléas administratifs ayant conduits à une fin prématurée du post-doc. Un effort doit être mené pour finaliser la valorisation de ce travail. Face aux résultats encourageants obtenus lors de la validation, les prévisions ont été portées en opérationnel par Météo-France, en utilisant, cette fois, les prévisions du dernier système de prévisions saisonnières (Météo-France system7). Ces prévisions sont ainsi réalisées chaque mois depuis Janvier 2020. Un site web dédié permettant d’accéder à ces prévisions a été créé, et les accès ont été transmis à une quarantaine d’acteurs de l’eau, fin 2020. Un retour d’expérience (REX) de cette première année de prévision a été réalisé (Mounier, 2021). Ce REX a réalisé une vérification des prévisions, via des comparaisons directes entre observations et prévisions des indices standardisés. Il faut souligner qu’il ne peut pas s’agir d’une validation, car les prévisions d’ensemble ne peuvent s’évaluer sur une seule année. Le REX a montré la difficulté de comparer directement les prévisions Aqui-FR avec les observations, tout d’abord du fait de la disponibilité de ces dernières. En effet, alors que l’indice standardisé d’Aqui-FR est calculé sur la même période de référence de 30 ans, cet indice est calculé sur des périodes variables et généralement plus courtes et plus récentes dans les observations. Cela tend à rendre un signal plus humide dans les observations que les simulations, et donc, les prévisions. Il est donc nécessaire, pour ces comparaisons, de recalculer les SPLI sur la même période. Les scores alors obtenus sont qualitativement cohérents avec ceux obtenus lors de la validation sur la période rétrospective, avec des meilleurs résultats sur les prévisions réalisées de Janvier à Juillet. Cependant, ces scores sontenviron deux fois plus faibles. Cela est lié à la comparaison directe aux observations, la comparaison à la ré-analyse permettant de retrouver des scores équivalents lors de la vérification que de la validation rétrospective. Cela conduit à envisager qu’une méthode une adaptation statistique des prévisions basées sur un apprentissage des erreurs entre ré-analyse et observation pourrait être appliquée aux observations et ainsi améliorer sensiblement les comparaisons entre prévisions et observations. Aqui-FR a également été utilisée pour projeter l’impact du dérèglement climatique sur la ressource en eau souterraine. Un ensemble de projections climatiques issues du 5ième rapport du GIEC, régionalisées par Dayon et al., 2018, sous trois scénarios possibles d’évolution de gaz à effet de serre a été analysé. Les projections indiquent une diminution de la ressource en eau souterraine et une augmentation de l’intensité et de la durée des sécheresses en nappe et en rivière sur le domaine pour les scénarios les plus émetteurs (RCP8.5 et RCP4.5). Ces impacts sont fortement atténués avec le scénario le moins émetteur (RCP2.6).

Even though the coarsening strategy relevance in production mode could be jeopardised by the disk access cost, and even though, because of the load imbalance, the cost of coarsening operation involving a large number of computing nodes and performed at each time step becomes as expensive as the whole BGC computations on the coarsened grid, the OASIS based coupled configuration is two time cheaper than the full resolution ORCA025 ocean-BGC model. In a context of energy saving and carbon footprint reduction, this result cannot be neglected. But the importance of the coupling cost revealed in this study emphasises the reality of the MPI parallelism limits, that one can not only notice in the exchange of NEMO variable boundaries, but that must be lowered each time than other tasks, usually neglected, have to be operated in addition to those needed for the core model equation resolving: coupling, diagnostics, resolution coarsening or resolution refinement

International audience; This paper presents the broadband noise analysis of a Counter Rotating Open Rotor (CROR) configuration. The numerical study is based on a hybrid approach: a Large Eddy Simulation (LES) code solves the near-flow field of the CROR configuration and the near-to-far-field propagation is then predicted by a Ffowcs Williams-Hawkings analogy either based on a solid or porous formulation. The LES solver uses a phase-lagged assumption with a proper orthogonal decomposition for the data storage. The numerical approach is validated against wind tunnel experimental data of a CROR configuration (AI-PX7) at three different operating points focused on the broadband noise. The analysis of the numerical simulations shows the predominant effect of the rear rotor suction side on the radiated broadband noise with an increasing contribution with span location. This source is related to the impingement of front rotor wakes and the development of leading edge vortices that induce large pressure fluctuations close to the leading edge on the suction side.

11 pages, 6 figures, supplementary information https://doi.org/10.1038/s41612-021-00188-5.-- Data availability; The data generated and analyzed during the current study are available from the corresponding author YRR on reasonable request, Atlantic multidecadal variability (AMV) has been linked to the observed slowdown of global warming over 1998–2012 through its impact on the tropical Pacific. Given the global importance of tropical Pacific variability, better understanding this Atlantic–Pacific teleconnection is key for improving climate predictions, but the robustness and strength of this link are uncertain. Analyzing a multi-model set of sensitivity experiments, we find that models differ by a factor of 10 in simulating the amplitude of the Equatorial Pacific cooling response to observed AMV warming. The inter-model spread is mainly driven by different amounts of moist static energy injection from the tropical Atlantic surface into the upper troposphere. We reduce this inter-model uncertainty by analytically correcting models for their mean precipitation biases and we quantify that, following an observed 0.26 °C AMV warming, the equatorial Pacific cools by 0.11 °C with an inter-model standard deviation of 0.03 °C, Y.R.-R. was founded by the European Union’s Horizon 2020 Research and Innovation Program in the framework of the Marie Skłodowska-Curie grant INADEC (Grant agreement 800154). E.M.-C. acknowledges funding from the European Commission’s Horizon 2020 projects PRIMAVERA (Grant Agreement 641727). X.L. has received funding from the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement H2020-MSCA-COFUND-2016-754433. A.B. and D.N. acknowledge funding from the European Commission’s Horizon 2020 project EUCP (Grant agreement 776613). F.C. and G.D. were supported by the US National Science Foundation (NSF) under the Collaborative Research EaSM2 Grant OCE-1243015 to NCAR and by the US National Oceanic and Atmospheric Administration (NOAA) Climate Program Office under the Climate Variability and Predictability Program Grant NA13OAR4310138. NCAR is a major facility sponsored by the US NSF under Cooperative Agreement 1852977. [...] R.E., N.D., L.H., and D.S. were supported by the Met Office Hadley Center 522 Climate Program funded by BEIS and Defra and by the European Commission Horizon 2020 EUCP 523 project (GA 776613). J.L.-P. was funded by the European Union’s Horizon 2020 Research and Innovation Program in the framework of the PRIMAVERA project (Grant Agreement 641727). J.R. and D.H. were funded by NERC via NCAS and the ACSIS project (NE/N018001/1), and JR was also funded by the NERC SMURPHS project (NE/N006054/1). M.M.-R. was funded by the European Union’s Horizon 2020 Research and Innovation Program in the framework of the Marie Skłodowska-Curie grant FESTIVAL (Grant agreement 797236). E.T. has received funding from the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement No. 748750 (SPFireSD project), With the funding support of the ‘Severo Ochoa Centre of Excellence’ accreditation (CEX2019-000928-S), of the Spanish Research Agency (AEI)

L’impact du changement climatique sur les ressources en eau est une préoccupation majeure, tant pour les scientifiques que pour les opérationnels. L’état de l’art sur les modifications du cycle de l’eau dans ce contexte d’influence anthropique met en avant deux points : i) l’influence humaine sur la circulation atmosphérique à large échelle, les précipitations, les débits et les niveaux piézométriques est difficile à caractériser car la variabilité naturelle est importante et peut compenser l’impact anthropogénique; ii) bien que dans certains scénarios, les projections CMIP5/6 convergent sur les tendances, elles sous ou sur-estiment les changements sur le cycle de l’eau. Les projections de débits et niveaux de nappes sont affectées par le biais supplémentaire lié au filtre du bassin versant. Les incertitudes liées à la variabilité naturelle, aux biais des modèles, et au filtre du bassin versant sont différentes selon les échelles de temps considérées, impliquant i) que les projections de précipitation, servant d’entrée pour la simulation de débits et niveaux de nappes, sont sous/sur-estimées seulement pour certaines composantes (i.e. oscillations à certaines échelles de temps) ; ii) que la réponse des débits et niveaux de nappe aux projections de précipitation biaisées est modulée par le bassin versant. Dans ce volet de la phase 8, nous étudions la réponse des débits et niveaux piézométriques sur le bassin de la Seine aux précipitations dont certaines composantes ont été modifiées. La méthodologie repose sur la modification spectrale de chroniques de précipitation SAFRAN, en augmentant ou diminuant l’amplitude de certaines composantes par méthodes spectrales, puis en utilisant ces chroniques modifiées en entrée de la plateforme de modélisation CaWaQS afin d’établir une régionalisation de la sensibilité des débits et niveaux piézométriques aux modifications dans les précipitations.

As a variety of quantum computing models and platforms become available, methods for assessing and comparing the performance of these devices are of increasing interest and importance. Despite being built of the same fundamental computational unit, radically different approaches have emerged for characterizing the performance of qubits in gate-based and quantum annealing computers, limiting and complicating consistent cross-platform comparisons. To fill this gap, this work proposes a single-qubit protocol (Q-RBPN) for measuring some basic performance characteristics of individual qubits in both models of quantum computation. The proposed protocol scales to large quantum computers with thousands of qubits and provides insights into the distribution of qubit properties within a particular hardware device and across families of devices. The efficacy of the Q-RBPN protocol is demonstrated through the analysis of more than 300 gate-based qubits spanning eighteen machines and 2000 annealing-based qubits from one machine, revealing some unexpected differences in qubit performance. Overall, the proposed Q-RBPN protocol provides a new platform-agnostic tool for assessing the performance of a wide range of emerging quantum computing devices., Comment: 10 pages, 8 figures, 1 table

International audience; Abstract West Africa is one of the regions the most concerned with structural food and nutrition security. Consequently, agricultural development pathways and scenarios are under high scientific and political scrutiny in this region. Rice, as a rapidly growing staple plays a key role in the West African diet representing close to 40% of the total volume of cereal consumed in the region. In the context of the 2008 food price crisis several West African countries have since proclaimed rice self-sufficiency as a target. Here, we show that rice yields tend to be, on average over the entire region, less stable (by a range of 15%–30%) than that of alternative crops, possibly substitutable in diets. The regions where yields of alternative crops are more stable than those of rice correspond to the main climatic regions where these crops are grown: sorghum, millet in the Sahelian and Sudanian regions and tubers in the Guinean region. Rice yields are significantly less stable for 33% of the comparisons considered and are significantly more stable than any alternative crop for 15% of the comparisons in few areas without clear longitudinal patterns. Models accounting for climate variability explain up to 17% of the variance of the data and reveal that yield variability differences between rice and alternative crops tends to widen in the areas where the monsoon precipitation is more variable between-years. The highest levels of variability of rice yields compared to alternative crops are recorded in regions where the monsoon varies strongly between years. Our analysis advocates for an explicit account of yield stability in West African rice expansion scenarios and supply strategies.

International audience; A large-eddy simulation is used to assess the Reynolds-stress budgets of a round impinging jet in the context of a second-moment closure of turbulence. The present work focuses on the stagnation region where no data are available in the literature except in the wall vicinity. Inside the stagnation region, it is shown that the pressure terms are dominant in the budgets. They balance the Reynolds-stress production and the convective fluxes. A visualization of this equilibrium through a specific indicator reveals that impingement effects extend to less than one nozzle diameter in the wall-normal direction and to about one diameter radially. This study underlines the role of the pressure diffusion term that conveys energy to the wall, balancing the high production rates (both positive or negative). Finally, the failure of turbulence models is explained by the absence of appropriate modeling of this pressure diffusion term leading to excessive Reynolds-stress values inside the impingement region.

We consider a second-order, elliptic partial differential equation (PDE) discretized by the Hybrid High-Order (HHO) method. HHO is a polyhedral method that handles arbitrary polynomial orders, and for which globally coupled unknowns are located at faces. To efficiently solve the linear system arising after static condensation, this work proposes novel, skeleton-based multigrid methods. One is geometric, the other is algebraic. The geometric algorithm is an h-multigrid method that conserves the polynomial degree at every level. It handles non-nested, unstructured, polyhedral meshes. Numerical tests on homogeneous and heterogeneous diffusion problems show fast convergence, scalability in the mesh size and polynomial order, and robustness with respect to heterogeneity of the diffusion coefficient. The algebraic multigrid method (AMG) applies to the lowest order hybrid methods. It leverages the uncondensed matrix to extract the connectivity graph between element and faces, and subsequently implements an element-defined aggregation-based coarsening strategy. Used as a preconditioner, this AMG conserves the performance and scalability of standard plain aggregation AMGs that directly work on the condensed system, while exhibiting notable improvement on anisotropic problems with Cartesian meshes.; On considère une équation aux dérivées partielles (EDP) elliptique du second ordre discrétisée par la méthode Hybrid High-Order (HHO). HHO est une méthode polyédrique qui supporte les ordres polynomiaux arbitraires, et pour laquelle les inconnues globalement couplées sont situées aux faces. Afin de résoudre efficacement le système linéaire obtenu après condensation statique, ce travail propose de nouvelles méthodes multigrilles basées sur le squelette du maillage. L'une est géométrique, l'autre algébrique. L'algorithme géométrique est un h-multigrille qui conserve le degré polynomial à tous les niveaux. Il gère les maillages polyédriques non structurés et non imbriqués. Les tests numériques sur des problèmes de diffusion homogènes et hétérogènes montrent une convergence rapide, un comportement asymptotique optimal par rapport à la taille du problème et les ordres polynomiaux, ainsi qu'une grande robustesse aux discontinuités du coefficient de diffusion. Le multigrille algébrique (AMG) s'applique aux méthodes hybrides d'ordre bas. Le graphe de connectivité entre éléments et faces est extrait de la matrice non condensée, ce qui permet l'implémentation du méthode de coarsening basée sur l'agrégation des éléments. Utilisé comme préconditionneur, cet AMG conserve les performance et scalabilité des AMG basés sur l'agrégation simple qui travaillent directement sur le système condensé, tout en affichant une amélioration notable sur les problèmes anisotropiques avec maillage cartésien.

We are interested in the quantification of uncertainties in discretized elliptic partial differential equations with a random coefficient field. In sampling-based approaches, this relies on solving large numbers of symmetric positive definite (SPD) linear systems with different matrices. In particular, we consider the case in which these operators are sampled by Markov chain Monte Carlo, which leads to sequences of correlated matrices. We investigate recycling Krylov subspace strategies for the iterative solution of sequences of linear systems formed with such matrices. The linear systems are solved using initialized conjugate gradient (Init-CG) methods, where approximate eigenvectors of the previously sampled operator are used to set an initial guess, and deflated conjugate gradient (Def-CG) methods, where the Krylov subspace is augmented with these vectors. The following aspects of eigenvector approximation, and their effect on deflation and initialization, are investigated in this context: (i) projection technique, and (ii) refreshing strategy of the eigen-search space. Our numerical experiments show that, when not using a preconditioner, these aspects only impact convergence behaviors of Def-CG at the early stages of the sampling sequence. Second, unlike sequences with multiple right-hand sides and a constant operator, our experiments with multiple matrices show that, even for highly correlated matrices, Init-CG does not reproduce the convergence behavior of Def-CG. Finally, the limits of deflation used as a means to compensate for the inefficiency of block-Jacobi (bJ) preconditioners are investigated. For small systems, using a bJ preconditioner while deflating with at least as many approximate eigenvectors as the number of bJ blocks achieves similar convergence behaviors to PCG with a constant algebraic multigrid (AMG) preconditioner. For larger systems, although the effect of deflation is improved when using the right refreshing strategy of the eigen-search space, the combination of deflation with bJ preconditioners does not scale as well as using PCG with a constant AMG preconditioner based on the median realization of the coefficient field.; Nous nous intéressons à la quantification des incertitudes dans les équations aux dérivées partielles elliptiques discrétisées avec un champ de coefficients aléatoires. Dans les approches basées sur l'échantillonnage, celle-ci repose sur la résolution d'un grand nombre de systèmes linéaires symétriques définis positifs (SPD) avec des matrices différentes. En particulier, nous considérons le cas où ces opérateurs sont échantillonnés par une méthode de Monte Carlo par chaînes de Markov, ce qui produit des suites de matrices corrélées. Nous étudions des stratégies de recyclage de sous-espaces de Krylov pour la résolution itérative de suites de systèmes linéaires provenant de telles matrices. Les systèmes linéaires sont résolus en utilisant des méthodes de gradient conjugué initialisé (Init-CG), où les vecteurs propres approximatifs de l'opérateur précédemment échantillonné sont utilisés pour définir une estimation initiale, et des méthodes de gradient conjugué déflaté (Def-CG), où le sous-espace de Krylov est augmenté de ces vecteurs propres. Les aspects suivants de l'approximation des vecteurs propres, et leur effet sur la déflation et l'initialisation, sont étudiés dans ce contexte : (i) la technique de projection, et (ii) la stratégie de rafraîchissement de l'espace de recherche des vecteurs propres. Nos expériences numériques montrent que, lorsqu'on n'utilise pas de préconditionneur, ces aspects n'ont un impact sur les comportements de convergence de Def-CG qu'au début de la suite d'échantillonnage. Deuxièmement, contrairement aux séquences avec des second-membres multiples et un opérateur constant, nos expériences avec des matrices multiples montrent que, même pour des matrices fortement corrélées, Init-CG ne reproduit pas le comportement de convergence de Def-CG. Enfin, nous étudions les limites de la déflation utilisée comme moyen de compenser l'inefficacité des préconditionneurs bloc-Jacobi (bJ). Pour les petits systèmes, l'utilisation d'un préconditionneur bJ tout en déflatant avec au moins autant de vecteurs propres approximatifs que le nombre de blocs bJ permet d'obtenir des comportements de convergence similaires à ceux de PCG avec un préconditionneur multigrille algébrique (AMG) constant. Pour les systèmes plus grands, bien que l'effet de la déflation soit amélioré en utilisant la bonne stratégie de rafraîchissement de l'espace de recherche des vecteurs propres, la combinaison de la déflation avec les préconditionneurs bJ ne s'adapte pas aussi bien que l'utilisation du PCG avec un préconditionneur AMG constant basé sur la réalisation médiane du champ de coefficients.

This work is based on the seminar titled ``Resiliency in Numerical Algorithm Design for Extreme Scale Simulations'' held March 1-6, 2020 at Schloss Dagstuhl, that was attended by all the authors. Naive versions of conventional resilience techniques will not scale to the exascale regime: with a main memory footprint of tens of Petabytes, synchronously writing checkpoint data all the way to background storage at frequent intervals will create intolerable overheads in runtime and energy consumption. Forecasts show that the mean time between failures could be lower than the time to recover from such a checkpoint, so that large calculations at scale might not make any progress if robust alternatives are not investigated. More advanced resilience techniques must be devised. The key may lie in exploiting both advanced system features as well as specific application knowledge. Research will face two essential questions: (1) what are the reliability requirements for a particular computation and (2) how do we best design the algorithms and software to meet these requirements? One avenue would be to refine and improve on system- or application-level checkpointing and rollback strategies in the case an error is detected. Developers might use fault notification interfaces and flexible runtime systems to respond to node failures in an application-dependent fashion. Novel numerical algorithms or more stochastic computational approaches may be required to meet accuracy requirements in the face of undetectable soft errors. The goal of this Dagstuhl Seminar was to bring together a diverse group of scientists with expertise in exascale computing to discuss novel ways to make applications resilient against detected and undetected faults. In particular, participants explored the role that algorithms and applications play in the holistic approach needed to tackle this challenge., Comment: 45 pages, 3 figures, submitted to The International Journal of High Performance Computing Applications

International audience

International audience; This work aims at improving Lagrangian particle injectors for the simulation of sprays. In such simulations, primary atomization is not resolved and Lagrangian particles are directly injected as a dispersed phase in the flow. Two main challenges arise in such methodology: i) the prescription of the correct droplet size and velocity distributions at injection, ii) ensuring the proper coupling of the dispersed phase with the gas phase to have the correct gas flow field after the injection. The proposed approach relies on an improved Lagrangian injector model and on resolved primary atomization simulations to feed the injector model parameters. The resolved atomization simulations are performed using a sharp-interface approach (ACLS/GFM) on unstructured grids [1, 2]. The validation test case is a high-pressure, non-reactive kerosene jet in crossflow (JICF) atomizer configuration [3], which is representative of complex injection systems. Resolved simulations of atomization for this configuration are performed and validated against the experimental correlation for the jet trajectory, showing good accordance. These simulations are then post-processed to feed the Lagrangian injector model. Finally, the injectors are applied to the same configuration and compared to experimental data.

International audience; Extreme scale simulation requires fast and scalable algorithms, such as multigrid methods. To achieve asymptotically optimal complexity it is essential to employ a hierarchy of grids. The cost to solve the coarsest grid system can often be neglected in sequential computings, but cannot be ignored in massively parallel executions. In this case, the coarsest grid can be large and its efficient solution becomes a challenging task. We propose solving the coarse grid system using modern, approximate sparse direct methods and investigate the expected gains compared with traditional iterative methods. Since the coarse grid system only requires an approximate solution, we show that we can leverage block low-rank techniques, combined with the use of single precision arithmetic, to significantly reduce the computational requirements of the direct solver. In the case of extreme scale computing, the coarse grid system is too large for a sequential solution, but too small to permit massively parallel efficiency. We show that the agglomeration of the coarse grid system to a subset of processors is necessary for the sparse direct solver to achieve performance. We demonstrate the efficiency of the proposed method on a Stokes-type saddle point system. We employ a monolithic Uzawa multigrid method. In particular, we show that the use of an approximate sparse direct solver for the coarse grid system can outperform that of a preconditioned minimal residual iterative method. This is demonstrated for the multigrid solution of systems of order up to 1+e11 degrees of freedom on a petascale supercomputer using 43 200 processes.

International audience; Recent adverse weather events have questioned the stability of crop production systems. Here, we assessed the vulnerability of eleven major crops in France between 1959 and 2018 as a function of climate, crafting a novel hazard framework that combines exposure and sensitivity to weather-related hazards. Exposure was defined as the frequency of hazardous climate conditions. Sensitivity of crops was estimated by the yield response to single and compound hazards, using observed yields available at département (county) level. Vulnerability was computed as the exposure-weighted average of crop sensitivities. Our results do not reveal any evidence for historically increased vulnerability of French crop production. Sensitivity to adverse weather events, and thus the overall vulnerability, has significantly decreased for six of the eleven crops between 1959 and 2018, and shown no significant decline or remained stable for the other five. Yet compound hazards can induce yield losses of 30% or more for several crops. Moreover, as heat-related hazards are projected to become more frequent with climate change, crop vulnerability may rise again in the future.

International audience; This chapter deals with the prediction of turbulent jet simulations, which are seen to be efficient flow cases to investigate further compressible LES solvers regarding their ability to simulate jet noise.

We address the numerical solution of linear systems arising from the hybrid discretizations of second-order elliptic partial differential equations (PDEs). Such discretizations hinge on a hybrid set of degrees of freedom (DoFs), respectively defined in cells and faces, which naturally gives rise to a global hybrid system of linear equations. Assuming that the cell unknowns are only locally coupled, they can be efficiently eliminated from the system, leaving only face unknowns in the resulting Schur complement, also called statically condensed matrix. We propose in this work an algebraic multigrid (AMG) preconditioner specifically targeting condensed systems corresponding to lowest order discretizations (piecewise constant). Like traditional AMG methods, we retrieve geometric information on the coupling of the DoFs from algebraic data. However, as the condensed matrix only gives information on the faces, we use the uncondensed version to reconstruct the connectivity graph between elements and faces. An aggregation-based coarsening strategy mimicking a geometric coarsening or semi-coarsening can then be set up to build coarse levels. Numerical experiments are performed on diffusion problems discretized by the Hybrid High-Order (HHO) method at the lowest order. Our approach uses a K-cycle to precondition an outer flexible Krylov method. The results demonstrate similar performances, in most cases, compared to a standard AMG method, and a notable improvement on anisotropic problems with Cartesian meshes.

International audience; In pollutant dispersion problems, mapping concentrations in the first tens or hundreds of meters from the source still remains a modelling challenge. Large-eddy simulations (LES) are able to represent time and space variability of turbulent atmospheric flow, which is of prime importance to assess public short-term exposure. However, they remain far from real time and subject to uncertainties, in particular to parametric uncertainties associated with the large-scale atmospheric forcing and the emission source position. In this work, we show that an efficient and accurate metamodel of the tracer concentration information provided by LES and encapsulating their associated uncertainties can be built using appropriate statistical tools combining machine learning and principal component analysis. We present a proof-ofconcept study based on a simplified but representative flow configuration (two-dimensional flow around a surfacemounted cube) using the AVBP LES solver and testing a variety of metamodels (linear regression, Gaussian processes, random forest, gradient boosting, etc.). Results reinforce the idea that for sufficiently statistically-converged quantities of interest and for a sufficiently large LES data set, a compound surrogate model can succeed in synthesizing information from the LES in the whole computational domain (with a Q 2 predictivity coefficient above 90 %). Downstream of the obstacle, the Q 2 coefficient of all metamodels reaches excellent results over 90%. Upstream, the tracer concentration is subject to strong discontinuities; combining metamodels allows to guarantee a good predictivity coefficient over 75%.

The Northern Hemisphere transient atmospheric response to Arctic sea decline is investigated in autumn and winter, using sensitivity experiments performed with the CNRM-CM6-1 high-top climate model. Arctic sea ice albedo is reduced to the ocean value, yielding ice-free conditions during summer and a more moderate sea ice reduction during the following months. A strong amplification of temperatures over the Arctic is induced by sea ice loss, with values reaching up to 25°C near the surface in autumn. Significant surface temperature anomalies are also found over the midlatitudes, with a warming reaching 1°C over North America and Europe, and a cooling reaching 1°C over central Asia. Using a dynamical adjustment method based on a regional reconstruction of circulation analogs, we show that the warming over North America and Europe can be explained both by changes in the atmospheric circulation and by the advection of warmer oceanic air by the climatological flow. In contrast, we demonstrate that the sea ice–induced cooling over central Asia is solely due to dynamical changes, involving an intensification of the Siberian high and a cyclonic anomaly over the Sea of Okhotsk. In the troposphere, the abrupt Arctic sea ice decline favors a narrowing of the subtropical jet stream and a slight weakening of the lower part of the polar vortex that is explained by a weak enhancement of upward wave activity toward the stratosphere. We further show that reduced Arctic sea ice in our experiments is mainly associated with less severe cold extremes in the midlatitudes.

A Large-Eddy Simulation of the tip leakage flow of a single airfoil is carried out. The configuration consists of a non-rotating, isolated airfoil between two horizontal plates with a gap of 10 mm between the tip of the airfoil and the lower plate. The Mach number of the incoming flow is 0.2, and the Reynolds number based on the chord is 9.3 × 105. The objective of the present study is to investigate the best way to compute both the aerodynamics and acoustics of the tip leakage flow. In particular, the importance of the inflow conditions on the prediction of the tip leakage vortex and the airfoil loading is underlined. On the other hand, the complex structure of the tip leakage vortex and its convection along the airfoil was recovered due to the use of a mesh adaptation based on the dissipation of the kinetic energy. Finally, the ability of the wall law to model the flow in the tip leakage flow region was proven in terms of wall pressure fluctuations and acoustics in the far-field.

State-of-the-art Earth system models, like the ones used in the Coupled Model Intercomparison Project Phase 6 (CMIP6), suffer from temporal inconsistencies at the ocean–atmosphere interface. Indeed, the coupling algorithms generally implemented in those models do not allow for a correct phasing between the ocean and the atmosphere and hence between their diurnal cycles. A possibility to remove these temporal inconsistencies is to use an iterative coupling algorithm based on the Schwarz iterative method. Despite its large computational cost compared to standard coupling methods, which makes the algorithm implementation impractical for production runs, the Schwarz method is useful to evaluate some of the errors made in state-of-the-art ocean–atmosphere coupled models (e.g., in the representation of the processes related to diurnal cycle), as illustrated by the present study. IPSL-CM6-SW-VLR is a low-resolution version of the IPSL-CM6 coupled model with a simplified land surface model, implementing a Schwarz iterative coupling scheme. Comparisons between coupled solutions obtained with this new scheme and the standard IPSL coupling scheme (referred to as the parallel algorithm) show large differences after sunrise and before sunset, when the external forcing (insolation at the top of the atmosphere) has the fastest pace of change. At these times of the day, the difference between the two numerical solutions is often larger than 100 % of the solution, even with a small coupling period, thus suggesting that significant errors are potentially made with current coupling methods. Most of those differences can be strongly reduced by making only two iterations of the Schwarz method, which leads to a doubling of the computing cost. Besides the parallel algorithm used in IPSL-CM6, we also test a so-called sequential atmosphere-first algorithm, which is used in some coupled ocean–atmosphere models. We show that the sequential algorithm improves the numerical results compared to the parallel one at the expanse of a loss of parallelism. The present study focuses on the ocean–atmosphere interface with no sea ice. The problem with three components (ocean–sea ice–atmosphere) remains to be investigated.

This paper presents a multi-chamber, multi-zone engine model to predict the ignition of a lean main chamber by a pre-chamber. The two chambers are connected by small cylindrical holes: the flame is ignited in the pre-chamber, hot gases propagate through the holes and ignite the main chamber through Turbulent Jet Ignition (TJI). The model original features are: (i) separate balance equations for the pre- and main chambers, (ii) a specific model for temperature and composition evolution in the holes and (iii) a DNS-based model to predict the ignition of the main chamber fresh gases by the burnt gases turbulent jets exiting the holes. Chemical reactions during TJI are the result of two competing mixing processes: (1) the hot jet gases mix with the fresh main chamber to produce heated zones and (2) at the same time, these hot gases cool down. (1) increases combustion and leads to ignition while (2) decreases it and can prevent ignition. The overall outcome (ignition or failure) is too complex to be modelled simply and the present model relies on recent DNSs of TJI which provided a method to predict the occurrence of ignition. Incorporating this DNS information into the engine model allows to predicts whether ignition will occur or not, an information which is not accessible otherwise using simple models. The resulting approach is tested on multiple cases to predict ignition limits for very lean cases, effects of H2 injection into the pre-chamber and optimum size for the holes connecting the two chambers as a function of equivalence ratio.

International audience; This paper describes a new boundary condition for subsonic inlets in compressible flow solvers. The method uses characteristic analysis based on wave decomposition and the paper discusses how to specify the amplitude of incoming waves to inject simultaneously three-dimensional turbulence and one-dimensional acoustic waves while still being non-reflecting for outgoing acoustic waves. The non-reflecting property is ensured by using developments proposed by Polifke et al. [1, 2]. They are combined with a novel formulation to inject turbulence and acoustic waves simultaneously at an inlet. The paper discusses the compromise which must be sought by the boundary condition formulation between conflicting objectives: respecting target unsteady inlet velocities (for turbulence and acoustics), avoiding a drift of the mean inlet velocities and ensuring non-reflecting performances for waves reaching the inlet from the computational domain. This well-known limit of classical formulations is improved by the new approach which ensures that the mean inlet velocities do not drift, that the unsteady components of velocity (turbulence and acoustics) are correctly introduced into the domain and that the inlet remains non-reflecting. These properties are crucial for forced unsteady flows but the same formulation is also useful for unforced cases where it allows to reach convergence faster. The method is presented by focusing on the expression of the ingoing waves and comparing it with the classical NSCBC approach [3]. Four tests are then described: (1) the injection of acoustic waves through a non reflecting inlet, (2) the compressible flow establishment in a nozzle, (3) the simultaneous injection of turbulence and ingoing acoustic waves into a duct terminated by a reflecting outlet and (4) a turbulent, acoustically forced Bunsen-type premixed flame.

International audience; The single element GCH 4 /GOx rocket combustion chamber developed at the Technische Universität München has been computed using Large Eddy Simulation. The aim of this work is to analyze the flow and combustion features at high pressure, with a particular focus on the prediction of wall heat flux, a key point for the development of reusable engines. The impact of the flow and flame, as well as of the model used, on thermal loads is investigated. Longitudinal distribution of wall heat flux, as well as chamber pressure, have been plotted against experimental data, showing a good agreement. The link between the heat released by the flame, the heat losses and the chamber pressure has been explained by performing an energetic balance of the combustion chamber. A thermally chained numerical simulation of the combustor structure has been used to validate the hypothesis used in the LES.

The Copernicus Atmosphere Monitoring Service (CAMS) reanalysis is the latest global reanalysis dataset of atmospheric composition produced by the European Centre for Medium-Range Weather Forecasts (ECMWF), consisting of three-dimensional time-consistent atmospheric composition fields, including aerosols and chemical species. The dataset currently covers the period 2003–2016 and will be extended in the future by adding 1 year each year. A reanalysis for greenhouse gases is being produced separately. The CAMS reanalysis builds on the experience gained during the production of the earlier Monitoring Atmospheric Composition and Climate (MACC) reanalysis and CAMS interim reanalysis. Satellite retrievals of total column CO; tropospheric column NO2; aerosol optical depth (AOD); and total column, partial column and profile ozone retrievals were assimilated for the CAMS reanalysis with ECMWF's Integrated Forecasting System. The new reanalysis has an increased horizontal resolution of about 80 km and provides more chemical species at a better temporal resolution (3-hourly analysis fields, 3-hourly forecast fields and hourly surface forecast fields) than the previously produced CAMS interim reanalysis. The CAMS reanalysis has smaller biases compared with most of the independent ozone, carbon monoxide, nitrogen dioxide and aerosol optical depth observations used for validation in this paper than the previous two reanalyses and is much improved and more consistent in time, especially compared to the MACC reanalysis. The CAMS reanalysis is a dataset that can be used to compute climatologies, study trends, evaluate models, benchmark other reanalyses or serve as boundary conditions for regional models for past periods.

In order to guarantee good re-ignition capacities in case of engine failure during flight, it is of prime interest for engine manufacturers to understand the physics of ignition from the spark discharge to the full burner lightning. During the ignition process, a spark plug delivers a very short and powerful electrical discharge to the mixture. A plasma is first created before a flame kernel propagates. The present work focuses on this still misunderstood first instants of ignition, i.e., from the sparking to the flame kernel formation. 3D Direct Numerical Simulations of propane-air ignition sequences induced by an electric discharge are performed on a simple anode-cathode set-up. An Analytically Reduced Chemistry (ARC) including 34 transported species and 586 irreversible reactions is used to describe the coupled combustion and plasma kinetics. The effect of plasma chemistry on the temperature field is found to be non-negligible up to a few microseconds after the spark due to endothermic dissociation and ionization reactions. However, its impact on the subsequent flame kernel development appears to be weak in the studied configuration. This tends to indicate that plasma chemistry does not play a key role in ignition and may be omitted in numerical simulations.

International audience; Crude and quasi Monte Carlo (MC) sampling techniques are common tools dedicated to estimating statistics (expectation, variance, covariance) of a random quantity of interest.We focus here on the uncertainty quantification framework where the quantity of interest is the output of a numerical simulator fed with uncertain input parameters.Then, sampling the output involves running the simulator for different samples of the inputs, which may be computationally time-consuming.To reduce the cost of sampling, a first approach consists in replacing the numerical simulator by a surrogate model that is cheaper to evaluate, thus making it possible to generate more samples of the output and therefore leading to a lower sampling error.However, this approach adds to the sampling error an unavoidable model error.Another approach, which does not introduce any model error, is the so-called multilevel MC (MLMC) method.Given a sequence of levels corresponding to numerical simulators with increasing accuracy and computational cost, MLMC combines samples obtained at different levels to construct an estimator at a reduced cost compared to standard MC sampling.In this paper, we derive and analyze multilevel covariance estimators and adapt the MLMC convergence theorem in terms of the corresponding covariances and fourth order moments. We propose a multilevel algorithm driven by a target cost as an alternative to typical algorithms driven by a target accuracy.These results are used in a sensitivity analysis context in order to derive a multilevel estimation of Sobol' indices, whose building blocks can be written as covariance terms in a pick-and-freeze formulation.These contributions are successfully tested on an initial value problem with random parameters.

A methodology to include flow–acoustic interactions in a Helmholtz solver is proposed and used to quantify their impact on the thermoacoustic modes of a real industrial combustor. This methodology ...

International audience; A Na-smectite clay mineral (Na-Mt) was exchanged with two concentrations of benzyldimethyltetradecyl ammonium chloride cationic surfactant up to one time the cation exchange capacity. Nonionic organoclay was prepared with polyoxyethylene (20) oleyl ether (Brij-O20) nonionic surfactant at one concentration. The resulting organoclays displayed lateral layer organization of the surfactants within their interlayer space.. The adsorption properties of these organoclays and the starting raw clay mineral were evaluated for three extensively used antibiotic pharmaceutical products: the amoxicillin (AMX), the sulfamethoxazole (SMX), and the trimethoprim (TRI), recognized as recalcitrant compounds to conventional water treatments and to display a complex behavior for different pH and temperature experimental conditions. Besides showing short half-life time with possible degradation by UV radiation, these antibiotics associated with mineral phases cause serious environmental issues of which the toxic effect can be exacerbated in the presence of other chemical compounds. From the set of data obtained by complementary techniques: UV and Fourier transform infrared spectroscopy, high-performance liquid chromatography coupled with mass spectrometry, and X-ray diffraction, it appears that the nonionic organoclay shows its versatility for the adsorption of individual molecules as well as a pool of antibiotics. The mixing of the three antibiotics showing different electric charged species (cations, anions, and zwitterions) mimics the natural context drives to a deep modification of the adsorption behavior onto the different materials that can act as possible carrier mineral phases in aquatic environment. These competition effects can be measured through the significant decrease of the K F Freundlich constants for AMX in the presence of other molecules (or electrolytes), whereas TRI and SMX, by their possible association, create a synergistic effect that favors their adsorption on the whole layered materials.

International audience; In an ever-increasing interest for Machine Learning (ML) and a favorable data development context, we here propose an original methodology for data-based prediction of two-dimensional physical fields. Polynomial Chaos Expansion (PCE), widely used in the Uncertainty Quantification community (UQ), has long been employed as a robust representation for probabilistic input-to-output mapping. It has been recently tested in a pure ML context, and shown to be as powerful as classical ML techniques for point-wise prediction. Some advantages are inherent to the method, such as its explicitness and adaptability to small training sets, in addition to the associated probabilistic framework. Simultaneously, Dimensionality Reduction (DR) techniques are increasingly used for pattern recognition and data compression and have gained interest due to improved data quality. In this study, the interest of Proper Orthogonal Decomposition (POD) for the construction of a statistical predictive model is demonstrated. Both POD and PCE have amply proved their worth in their respective frameworks. The goal of the present paper was to combine them for a field-measurement-based forecasting. The described steps are also useful to analyze the data. Some challenging issues encountered when using multidimensional field measurements are addressed, for example when dealing with few data. The POD-PCE coupling methodology is presented, with particular focus on input data characteristics and training-set choice. A simple methodology for evaluating the importance of each physical parameter is proposed for the PCE model and extended to the POD-PCE coupling.