Search

International audience

River Flow 2020 -10th Conference on Fluvial Hydraulics, DELFT, PAYS-BAS, 07-/07/2020 - 10/07/2020; As a result of climate change, we believe that extreme floods will become more frequent and more intense. The flows through obstacles representing an urbanized floodplain is realised in a laboratory at Re = 4766 and 8580. The purpose of the present study is first to validate a numerical model using Large Eddy Simulation (LES) against experimental data regarding the average longitudinal velocity and the secondary vortices formation. Then, additional data can be obtained and quantified. Therefore the geometrical parameters can be changed and their influence analysed and compared to the literature. A good agreement was found between simulated and experimental vertical profiles of time averaged velocity for the flow through emerged or slightly submerged obstacles. The additional simulations highlight that the flow structure is mostly driven by the transverse aspect ratio using the distance between two adjacent obstacles.

International audience; Urban flood laboratory experiments usually consider that the flow is mostly concentrated in streets and does not enter the building areas. However, urban blocks might modify the flow patterns at local scale due to the inflows or/and outflows through openings such as gates, windows, doors, etc. To investigate the influence of urban block porosity on flow distribution, flow depth and velocity in the surrounding streets, experiments were performed on a physical model. The results show that the urban block porosity does not substantially affect the discharge distribution in the streets located downstream of the urban block. However, flow depth, velocity and flow rate in the surrounding streets are more strongly modified.

International audience; Steady uniform and non-uniform flows through an array of emergent and slightly submerged square cylinders are experimentally investigated with a specific focus on transverse seiche waves induced by vortex shedding. The study is first and foremost aimed at assessing the effect of streamwise flow non-uniformity on seiche waves. Its secondary purpose is to investigate the change in seiche magnitude, when initially emerged cylinders become slightly submerged. Thirdly and lastly, the effect of seiche waves on mean velocities and velocity fluctuations is quantified. The lock-in process between waves and vortex shedding is unaltered by flow non-uniformity and by a change from cylinder emergence to submergence. For non-uniform flows, this results in the coexistence of two differently oscillating transverse waves close to each other. Relative wave amplitude is found to be mainly influenced by relative submergence in the case of submerged cylinders, and by Froude number and oscillation mode in the case of emergent cylinders. Finally, seiche waves modify the streamwise mean velocity, when cylinders are emergent.

International audience; The structure of free-surface flows in a straight compound channel was investigated in a laboratory flume, consisting of a central smooth-bed main channel (MC) and two adjacent rough-surface floodplains (FPs). The experiments covered both uniform and non-uniform flow conditions, with the latter generated by imposing an imbalance in the discharge distribution between MC and FPs at the flume entrance. The non-uniform cases involved transverse currents directed from MC to FPs and vice versa. The focus of the study was on assessing the effects of transverse currents on: (i) transverse shear layer and horizontal Kelvin-Helmholtz-type coherent structures (KHCSs) forming at the interfaces between MC and FPs; (ii) helical secondary currents (SCs) developing across the channel due to topography-induced flow heterogeneity; and (iii) turbulent large- and very-large-scale motions (VLSMs). Transverse currents can entirely displace the shear layer over FP or in MC, but they do not alter the KHCSs to the same degree, resulting in a mismatch between shear layer extent and KHCS length scales. KHCSs emerge once dimensionless velocity shear exceeds a critical value above which KHCS length scales increase with the shear. Three well-established SC cells, which are induced by turbulence anisotropy, are observed in uniform flow and non-uniform flow with transverse currents towards FP. They are replaced by a single cell in the presence of a transverse mean flow towards MC. The spectral signatures of VLSMs are visible at the upstream section of the flume but they quickly disappear along the flow, being suppressed by simultaneous development of KHCSs and SCs.

International audience; The study of flooding in urban environments has some complexity, mainly due to the presence of buildings, among others (e.g., vehicles, street furniture, drainage system, underground transportation system). These buildings and their facilities are grouped into blocks that are normally assumed to be impervious in urban flood studies. However, water enters these blocks through different openings such as e.g., windows, doors, gates, fences, and these lateral flow exchanges between streets and blocks can locally alter the flow pattern. In order to study these exchange processes and their influence on both flow depth and velocity in the streets during flood events, experiments were conducted on a devoted physical model termed MURI (Urban Model for the study of Inundation Risk), at Irstea, Lyon-Villeurbanne, France. The present study focused on two geometric configurations, a straight street flanked by two lateral cavities, and a straight street flanked by one-side opening. These flows were also compared to a quasi-uniform flow (reference flow without openings). The experiments have shown that a detailed description of the building blocks is required if flow depths and velocities (flood hazard) has to be locally estimated.

International audience; The study of flooding in urban environments has some complexity, mainly due to the presence of buildings, among others (e.g., vehicles, street furniture, drainage system, underground transportation system). These buildings and their facilities are grouped into blocks that are normally assumed to be impervious in urban flood studies. However, water enters these blocks through different openings such as e.g., windows, doors, gates, fences, and these lateral flow exchanges between streets and blocks can locally alter the flow pattern. In order to study these exchange processes and their influence on both flow depth and velocity in the streets during flood events, experiments were conducted on a devoted physical model termed MURI (Urban Model for the study of Inundation Risk), at Irstea, Lyon-Villeurbanne, France. The present study focused on two geometric configurations, a straight street flanked by two lateral cavities, and a straight street flanked by one-side opening. These flows were also compared to a quasi-uniform flow (reference flow without openings). The experiments have shown that a detailed description of the building blocks is required if flow depths and velocities (flood hazard) has to be locally estimated.

International audience; This paper investigates the physical and numerical modellings of unsteady overbank flows in a compound open-channel, which consists in a main channel (MC) and one adjacent floodplain (FP). The experiments were performed in an 18 m long and 2 m wide flume. Simulations using the 1D code MAGE coupled with an improved one-dimensional (denoted as 1D+) method termed ISM (Independent Subsections Method, Proust et al. 2009) are compared to the experimental data. The originality of the ISM lies in its solving of the momentum conservation equation in each of the channel sub-sections (MC and FP). The ISM explicitly models the depth-averaged Reynolds stress at the MC/FP interface, and the transverse exchanges of mass and momentum by the mean flow between MC and FP. It also accounts for the upstream discharge distribution between MC and FP. As this method has been validated only for steady non-uniform flows, the present study aims at validating it under unsteady flow conditions. Hydrographs are injected at the flume entrance in the MC and FP, with 100 runs in order to compute ensemble averages of the flow parameters. The ratio of FP discharge to total discharge equals to 7 % at baseflow and 15 % at peakflow, while the ratio of FP flow depth to MC flow depth ranges from 0.14 to 0.30. It was found that the ISM could accurately predict the temporal variations in: (i) the flow depth along the flume; (ii) the depth-averaged streamwise and transverse velocities at the MC/FP interface; and subsequently (iii) the interfacial lateral discharge per unit length.

According to the European Flood Directive 2007, the flood hazard must be accurately assessed in high-risk areas over the river floodplains, during extreme flood events. However, the prediction of the very high flows is not an easy task due to the lack of field data and to the strong link between flow resistance and land occupation of the floodplain (Figure 1). With increasing return period, the confinement decreases and the heterogeneity in lateral and longitudinal directions of the hydraulic roughness elements (e.g. trees, houses) strongly increases. The physical processes are complex, still largely unexplored, and the assumptions linked to numerical modeling cannot be validated without field data. The Flowres ANR project (https://flowres.irstea.fr/en/) aims at improving the flood hazard assessment in floodplains, by relying on experimental data base and on one field case (floods at Besançon, France).; Selon la directive européenne sur les inondations, l'aléa hydraulique (hauteurs, vitesses) doit être évalué avec précision dans les zones à haut risque lors des crues extrêmes. Or, la prédiction des écoulements dans les plaines inondables n'est pas aisée en raison de l'absence de données de terrain et du lien étroit entre résistance à l'écoulement et occupation du sol (Figure 1). Le confinement et l'hétérogénéité des rugosités hydrauliques (comme des arbres, des maisons, ...) dans les directions latérales et longitudinales varient fortement avec l'augmentation de la période de retour de la crue. Les processus physiques sont complexes, encore largement inexplorés, et les hypothèses liées à la modélisation numérique de ces écoulements ne peuvent être validées sans données de terrain. Le projet Flowres (https://flowres.irstea.fr/en/) vise à améliorer l'évaluation de l'aléa dans les plaines inondables en s'appuyant sur les données expérimentales et sur un cas de terrain (les inondations à Besançon).

Steady uniform and non-uniform flows through an array of emergent and slightly submerged square cylinders are experimentally investigated with a specific focus on transverse seiche waves induced by vortex shedding. The study is first and foremost aimed at assessing the effect of streamwise flow non-uniformity on seiche waves. Its secondary purpose is to investigate the change in seiche magnitude, when initially emerged cylinders become slightly submerged. Thirdly and lastly, the effect of seiche waves on mean velocities and velocity fluctuations is quantified. The lock-in process between waves and vortex shedding is unaltered by flow non-uniformity and by a change from cylinder emergence to submergence. For non-uniform flows, this results in the co-existence of two differently oscillating transverse waves close to each other. Relative wave amplitude is found to be mainly influenced by relative submergence in the case of submerged cylinders, and by Froude number and oscillation mode in the case of emergent cylinders. Finally, seiche waves modify the streamwise mean velocity, when cylinders are emergent.

The chapter starts by describing the physical processes typical of floods, particularly, the ones that cannot be observed at low river flows. All these latter processes are linked to the various kinds of floods described above. In order to simulate these processes for flood management, a relevant spatial scale should be selected. For each kind of flood and spatial scale, one - dimensional and two - dimensional classical hydrodynamic models are explained. Limits and requirements of these latter approaches permit to introduce the recent evolutions in modelling and in available data, which permits to obtain more realistic and accurate results., Le chapitre commence par une description des phénomènes physiques propres aux inondations, en particulier, ceux qui les distinguent des écoulements ordinaires en rivière. Ces différents phénomènes sont rattachés aux différents types d'inondations qui ont été énumérés auparavant. Pour modéliser ces phénomènes en vue de la gestion des inondations, il est, entre autres, nécessaire de définir l'échelle spatiale pertinente. En fonction de cette échelle et du type d'inondation, on explique les modélisations hydrauliques classiques 1D et 2D, leurs limitations et leurs besoins ainsi que les dernières évolutions des modèles et surtout des données disponibles qui permettent d'obtenir des résultats toujours plus réalistes et parfois plus précis.

International audience; The simulation of open-channel flows through emergent rigid vegetation is carried out, using the open-source TELEMAC-2D system. The aim of this study is to find an optimal representation of the flow resistance caused by emergent rigid vegetation for a wide range of flow rates in absence of calibration data for the highest flows. Two strategies of flow resistance representation are compared: (1) the vegetation, in addition to the bed friction, is implicitly considered in a flow resistance coefficient; or (2) the vegetation is explicitly considered by being excluded from the computation domain. Results emphasize that the explicit representation is the more accurate solution for computing both flow depth and velocity, but field data are generally not available. The uncertainty observed on the water depth with the implicit representation is higher than for the explicit representation, but it can be reduced by using more evolved resistance coefficient.

International audience; no abstract

International audience; Convergent hydraulic structures are found in different open-channel flow configurations, particularly in the courier of the surface spillway and in some stilling basins. In some cases, the convergent transition is followed by a prismatic part with the same slope and steps with a different slope. The work that is presented here focuses on the physical modeling of this flow configuration type under steady flow conditions. It has been performed in a laboratory flume, located in the hydraulic laboratory of the National Polytechnic School of Algiers. As far as we are concerned, there has not been any systematic scientific research for the conception of this configuration type so far. In particular, the slope change effect between the convergent transition and the stepped spillway has not been explored yet. According to the slope change and the flowrate, we are interested in the transverse wave propagation that arises at the convergent downhill and the maximum height of the transverse wave, We have measured water heights and instantaneous velocities (transverse and longitudinal profiles) by varying the flowrate from 2 to 20 l/s, the main channel slope (convergent transition and prismatic part) from 0° to 24.4° and the stairs slope from 0° to 45°.

3rd European Conference on Flood Risk Management, Lyon, FRA, 17-/10/2016 - 21/10/2016; International audience; Flood hazards (flow depth and velocity) must be accurately assessed in high-risk areas during extreme flood events. However, the prediction of the very high flows is not an easy task due to the lack of field data and to the strong link between flow resistance and the land occupation of the floodplain. Confinement and inhomogeneity in lateral and longitudinal directions of hydraulic roughness strongly vary with return period T. The physical processes are complex, some still largely unexplored, and the assumptions linked to numerical modelling cannot be validated without field data. The FlowRes project (2015-2018), funded by the French National Research Agency (ANR), aims at improving the flood hazard assessment in floodplains in: 1) investigating in laboratory the hydrodynamic structure associated with extreme flood flows for various land occupations and flow discharge magnitudes; 2) assessing if the existing numerical modelling practices used for T ~ 100 years are still valid for extreme events with T > 1000 years, relying on the experimental

My research works focus on the physical and numerical modeling of river floods, in the case of streamwise non-uniform overbank flows in compound open channels. These flows are characterized by longitudinal changes in flow depth and by a lateral time-averaged discharge between the main channel of the river and the adjacent floodplains. The longitudinal non-uniformity of flow is mostly caused by the changes in topography and land occupation of floodplains. Non-uniformity can also be related to the influence of the boundary conditions associated with the reach under consideration. These various sources of non-uniformity have been investigated throughout this work. The physical processes induced by flow non-uniformity have been studied in laboratory flumes. The analysis is carried out at the reach scale, but also at a smaller scale, with a three dimensional description of the mean flow and turbulent quantities, especially in the mixing layer developing at the junction between the main channel and the floodplains. The numerical modeling work is focused on the approaches 1D, 2D and 1D+ (development of a new method). In particular, we assess their ability to take into account the physical processes that are specific to non-uniforms flows, and eventually, to predict the flow depth and velocity in the floodplains. This is done for small overbank flows, but also for very high flows associated with extreme flood events.; Mes travaux de recherche portent sur la modélisation physique et numérique du débordement des rivières en régime non-uniforme. Ce régime est caractérisé par des variations longitudinales de hauteur d'eau, et par un débit d'échange latéral entre le lit mineur de la rivière et les plaines d'inondations adjacentes. La non-uniformité de l'écoulement est principalement due aux variations de topographie et d'occupation du sol des plaines d'inondation, mais également à l'influence des conditions limites du bief de rivière considéré. Ces différentes sources de non-uniformité ont été explorées tout au long de ces travaux. Les processus physiques induits par la non-uniformité de l'écoulement ont été étudiés dans des canaux de laboratoire. L'analyse est faite à l'échelle du bief, mais également à une échelle plus fine (description tridimensionnelle du champ de vitesse moyenne et des quantités turbulentes, notamment dans la couche de mélange entre lit mineur et lit majeur). Le travail de modélisation numérique porte sur les approches numériques 1D, 2D et 1D+ (développement d'une nouvelle méthode). En particulier, on évalue leurs capacités à prendre en compte les phénomènes physiques spécifiques aux écoulements non-uniformes, et in fine, à simuler hauteur et vitesse de l'écoulement en plaine d'inondation, pour des faibles débordements, comme pour des crues extrêmes.

When dealing with flood issues, compound channels flows are often encountered in the field. This paper investigates the errors that can be expected when using 2-D modelling for compound channel flows, comparing the simulations with experiments. Three flow configurations are analyzed: uniform, gradually varied and rapidly varied flows. The last configuration is obtained by setting a transverse embankment on the flood plain. Errors are estimated on the sub-section mean velocity, discharge andwater depth, on the mixing layer width and on the depth-averaged stream-wise velocity and lateral shear stress. Depending on the flow configuration and on the studied parameter, relative errors significantly vary from nearly zero to 50%. The influence of the 0th order turbulence closure on the mean flow and the dimensions of the recirculation zone behind the embankment is also investigated, using either constant eddy viscosity or Elder’s model., Pour le cas des inondations, les écoulements en canal composé sont fréquents. Cet article recherche les erreurs faites lors d'une modélisation 2D en comparant expériences en laboratoire et résultats numériques. Trois configurations sont analysées : régimes uniforme, graduellement varié et fortement varié. les erreurs sont estimées sur les variables hydrauliques de base. les erreurs peuvent aller jusqu'à 50%. L'effet du terme de fermeture turbulente (coefficient constant ou modèle de Elder) est analysé sur les dimensions de la recirculation et l'écoulement moyen.

International audience; Under uniform flow conditions, compound channel flows are characterised by an horizontal mixing layer developing at the junction between the flows in the main channel (deeper and faster) and in the floodplain (shallower and slower), e.g. Shiono & Knight (1991). Maintaining the total discharge corresponding to the uniform flow and imposing a disequilibrium in the upstream discharge distribution generates a non-uniform flow where lateral mass exchange occurs between channels along the flume. Two cases are distinguished: an excess and a deficit in the inlet floodplain discharge. The over- and under-feeding of the floodplain reduces and increases, respectively, the difference in velocity between both channels at the inlet, compared to uniform flow conditions. This paper investigates the effect of the magnitude of the upstream disequilibrium on the turbulent shear layer. The influence of lateral mass exchange on three-dimensional velocity field and on Reynolds stresses is assessed.

International audience; En utilisant des mesures en laboratoire sur des lits composés où la largeur du lit majeur varie, un calage du coefficient de frottement du lit majeur nfp utilisé dans des modélisations 1D a été effectué. Le calage direct sur le cas modélisé donne des valeurs de nfp très variables (1 à 5 environ pour un modèle donné) et ne permet souvent pas de donner à la fois une ligne d'eau et une répartition de débit acceptables. Si le calage est uniquement effectué sur un écoulement faiblement débordant, l'erreur résiduelle pour cet écoulement et pour les écoulements plus largement débordants dans la même géométrie peut dépasser 30%. Les modèles incluant une prise en compte de la turbulence à l'interface permettent d'obtenir une amélioration des performances dans certains cas mais restent insuffisants pour décrire l'ensemble des processus prédominants et, en particulier, le transfert de masse entre lits. / Using laboratory measurements of compound channels in which the flood plain width varies, the friction coefficient of the flood plain nfp used in 1-D modelling is calibrated. The direct calibration for the modelled case provides very varied values for nfp (about 1 to 5 for one given model) and often does not permit to obtain both convenient water level and flow distribution. If nfp is only calibrated on a slightly overbank flow, the error after calibration for the calibrated case and the cases with higher water depth in the same geometry can go beyond 30%. The models taking into account the turbulence at the interface between main channel and flood plain provide better results in some cases but cannot describe all the main processes and, particularly, the mass transfer between beds.

International audience; In May 2008, a 10-year return period flood occurred in the Arc River, France, with a maximum discharge over 500 m3/s. During this event, large standing waves (approximately 1 to 2 meter high and 5 to 10 meter long) were observed at several locations in the river and for a given discharge range (approximately 250 m3/s, for both rising and falling flood limbs). These standing waves appeared to be unstable (lasted 15 to 30 min), and induced large bedforms located at the edge of a gravel bar. These bedforms form typically an antidune system. Two other assumptions were proposed in order to explain the formation of these standing waves in a compound channel flow. First, the waves may be the consequence of the formation of large eddies in the interfacial zone between the main channel and the floodplain. Secondly, the waves may result from an instability due to the juxtaposition of a supercritical flow in the floodplain and of a subcritical flow in the main channel, which induces a small surge at the interface. Finally, a discussion is provided on the consequences of these standing waves on the bed morphology. First, these standing waves are directly associated with bedforms observed later on in the side of a gravel bar. Also, the location of transverse channels on the gravel bar appeared to be strongly influenced by the location of these bedforms during the ebb flow.

International audience; There has been recently increasing concern about the uniformity of flow at measuring sections in straight compound channel experiments. Experiments have been usually carried out with the same total head at the inlet of the main channel and floodplain and as a result the velocities entering into the main channel and onto the floodplain are also the same. As flow enters onto the floodplain, the floodplain discharge exceeds the discharge for a uniform flow condition, hence mass transfer towards the main channel progressively occurs along the flume until flow becomes uniform for both main channel and floodplain. This paper investigates the influence of relative depth on the length required for a uniform flow condition to be achieved in compound channel flumes. The quasi one-dimensional model, the Independent Sub-Sections Method (ISM), the two-dimensional model Telemac 2D and a three-dimensional numerical model have been used to simulate the experiments conducted at LMFA (France) and at UCL (Belgium). The ISM has been subsequently used to investigate more upstream conditions at the LMFA and their resulting flow developments in relation to the flume length. The results show that as the relative depth increases, the length required for uniform flow condition increases and can even exceed the actual length of some of the experimental flumes studied in literature.