Advances in Hydroinformatics

As waterways engineering consultant and expert, the Federal Waterways Engineering and Research Institute (BAW) supports the German Federal Waterways and Shipping Administration (WSV) and the Federal Ministry of Transport and Digital Infrastructure (BMVI) in the development, operation and maintenance of the waterways in Germany. To sustain and improve the quality of its consulting services, the BAW conducts research and development projects in the entire spectrum of waterways engineering. From the daily project work the R&D topics are derived, which in turn specify the development of the numerical methods. A brief introduction is given on the simulation techniques used for the meso-scale modelling of river reaches, in which the time-dependent shallow water equations are solved and morphological effects modelled in a specific way. Further, special features established for well-adapted resolutions of the flow field in hydraulic structures and around ships are emphasized. Application examples are given, ranging from flows and sediment transport in geometrically complex situations with dikes and groynes to hydraulically complex flows as e.g. in fish passages, where turbulent motion plays a dominant role. The use of a ship handling simulator to identify the circumstances and causes for the accident of an inland tank barge is presented.

Nice, located in the South–East of France, has constantly been developing on an economical and social level for the last decades. Thus, the urbanization process, implying the development of new infrastructures (Promenade des Anglais, roads and railways), has triggered an important erosion of the beach. Nice’s neighboring rivers, Var and Paillon, were the primary sources regarding the natural supply of beach gravels. But the combination of the airport construction, at the Var river mouth and the structures all along the streams, has stopped the gravel supplies. Those anthropological factors associated to the natural configuration of the beach (steep inner shoreface and narrow zone of concentrated wave breaking) contribute to an important loss of gravel. Studies have demonstrated that the erosion of Nice beaches reaches 15,000 m

3

/year. In this context, a modeling approach has been started to characterize the solid transport along the Nice shore. The TELEMAC-MASCARET system has been chosen for this study. It is a powerful integrated modeling tool which is composed of different simulation modules such as TOMAWAC and TELEMAC2D. The model created on TOMAWAC has provided results regarding the wave propagation in the bay. The simulations demonstrate the existence of high values of wave heights near the coast due to sub-marine caps. TELEMAC2D coupled with TOMAWAC has been used to simulate the current effects along the shore. During South-East storm events, the center part of the bay is exposed to high velocity values, whereas the west part is less vulnerable to this swell. There are two explanations: the orientation of the bay and the groynes. In order to validate the result of this model, we started a field measurement campaign. We will present its methodology and its first results in this paper.

Suspended sediment transport modeling has been of great concern to scientists and engineers interested in understanding the complex dynamics of sediment motion in fluid flows. Traditional models based on simplistic hypothesis have been applied in the past in the design of sediment handling and control structures. However, most often the spatially heterogeneous nature of the flow is not properly accounted for, despite the importance that such aspect may have in practical applications. This paper presents a one-dimensional model that incorporates the dispersion transport component in suspended sediment transport, in order to consider the effect associated with the cross-sectional heterogeneity of the velocity and concentration distributions. The longitudinal dispersion coefficient has been evaluated considering different parameterization alternatives. A sensitivity analysis of the model has been developed using dimensionless parameters. The model has been applied to the design of the intake silt basins of the Coca Codo Sinclair hydropower station in Ecuador based on the sediment removal capability of such structures. It has been concluded that accounting for dispersion results in smaller sediment removal efficiency and, therefore, longer silt basins. The comparison made with a simplistic method whose application is widespread for the dimensioning of silt basins, shows that it greatly overestimates their sediment removal capacity.

For the elaboration of flood risk prevention schemes (“Plan de Prévention du Risque Inondation” in France), the characterization and cartography of flood hazard can be based on different approaches. From historical approaches, consisting in the reconstitution of flood limits and heights from historical information and testimonies for some well-known major events to hydro-geomorphological methods, aiming at defining the limits of different envelops that the flood can mobilize, based on topographical and terrain analysis, or numerical modelling, with different modes of representation of the floodplain and varying degrees of precision. In urbanized areas, where the most vulnerable places reside due to the density and variety of the stakes exposed to the flood, classical hydraulic modelling technics such as 1D, 2D or simple 1D/2D coupled models often have important limitations regarding the complexity of the flow, due to the modification of the river streams and the presence of numerous artificial obstacles in the floodplain. The case study aims at showing the gain of integrating small-scale urban obstacles, like buildings, in the model, both as elements deflecting the flow, increasing the flow speed and hazard in the streets, and as places that can be flooded, taking into account the volume stored in the buildings. The study compares different approaches and shows the gain in terms of precision for the stakeholders of the building and using large models for integrating small-scale elements that can produce accurate results, usable both for the production of hazard maps and in the context of flood management plans.

Mexico City is facing problems of flooding in some areas at certain times of the year, causing important losses and damages on properties and residents including some casualties. Therefore, it is important to carry out a flood risk assessment in the catchment of Mexico City and estimate damages of probable flood events. However, limited data of observed discharges and water depths in the main rivers of the city are available, and this represents an obstacle for the understanding of flooding in Mexico City. The premise of this study is that with the limited data and resources available, the catchment can be represented to an acceptable degree by the construction of a deterministic hydrological model of the Mexico City basin. The objective of the developed tool is to provide an efficient support to management of the flood processes by predicting the behavior of the catchment for different rainfall events and flood scenarios. The capability of a model based on MIKE SHE modeling system for the Mexico City catchment was evaluated by comparing the observed data and the simulation results during a year after a careful development based on the most important parameters for characterizing the processes. Significant and operational results (>0.75 for Nash Sutcliffe coefficient) have been obtained on one of the major sub-catchments of the Mexico basin. These results demonstrate the interest to implement a deterministic hydrological model for assessing flood risks in a dense urban environment where data availability is limited. The developed model can be used for assessing the risks and designing some protection measures.

According to the World Risk Report released by the United Nations University Institute for Environment and Human Security, the Philippines is ranked third globally in terms of disaster risk. One of those disaster risks is flooding which poses a serious challenge to development and the lives of the people. Public health risks and social vulnerability are usually overlooked and undermined and only very little attention is given. Thus, this study focuses on these aspects. This study was an exploratory step towards assessing vulnerability particularly to fluvial flooding, it was a rapid assessment of the Knowledge, Attitudes, and Practices (KAP) of the community people including their socio-demographic profile, physical environment, exposure to microorganisms such as

E.coli

, Liptospirosis and the Dengue Fever mosquito, and local indicators were formulated and developed. These are important factors to be assessed in order to establish correlations and relationships in understanding social vulnerabilities and its indicators which can be incorporated in the hydroinformatics. The survey was done from March 2013 to July 2013. A total of 361 household respondents from the 12 communities and 30 respondents from the LGU and NGO were surveyed. Results of the study revealed an overall Flood Vulnerability Index (FVI) of 39.34 %. Barangay Tabuc-tubig (53.39 %) topping from all the 12 communities surveyed using the local indicators of the five major components namely; hydro-climatic, social, economic, socio-behavioral and the politico-administrative component. This study also reveals the most vulnerable communities from each of those 5 major components surveyed. It is interesting to note that Flood Vulnerability Index remains low in spite that the exposure indicators are high. The low FVI can be attributed to the community’s high resilience in its coping and adaptation strategies. In this study, the Flood Vulnerability Index is significantly sensitive to susceptibility and flood resilience variables.

The urban areas have been rapidly developed in Taiwan in recent years. The expanding urban areas and the increasing population, especially in the metropolitan Taipei city, result in surface runoff discharge during typhoons or rainstorms. When the surface runoff exceeds the capacity of drainage systems of the city, the urban inundation and property losses occur. The urban flooding risk assessment is a useful tool for the decision-making in flood damage mitigation. In general, hazard and vulnerability are two main factors for the risk assessment. In the present study, the Center Taipei City (CTC) is chosen as the study area. The A1B scenario simulations proposed by IPCC are used to compare the flood risks between the present situation and future condition which is under climate change scenario. The ArcGIS is used to yield the flood potential information and the flood risk for the return period of 10, 25, 100 and 200-year flood. The simulated results revealed that the increasing water storage to meet the regulation of the Taipei City government would effectively reduce the effects of climate change on the decreasing inundation areas. The adaptation strategies will also reduce the high-risk areas in the CTC according to risk assessment. The adaptation strategies composed of increasing the water storage in urban areas and reducing the social vulnerability of flooding area are the effective way for the flood risk reduction in urban area.

The calculation of the increase of mean sea level during hurricanes is a major issue in ocean modeling. Consequently, the wave setup is an important aspect of research which can reach significantly high values. Storm events in particular, when wave setups generated by wind storms are able to increase the mean sea level, enhance the risks of damage to coastal infrastructure. SWAN is a third-generation wave model, developed at the Delft University of Technology that computes random, short-crested wind-generated waves in coastal regions and inland waters. The optional command SETUP in the SWAN configuration activates the wave-induced set-up computation, modifying the water level taken into account by the model. Concerning SWAN-SWAN nesting, SWAN reads the wave boundary conditions in the relevant output file of the parent grid that are then transferred to the child grid. By default, the wave setup information is not contained in this file. When the wave setup is involved, the user can impose only one value of wave setup in the child grid. This imposed value will not cover the open boundaries, but will be applied at the deepest point in the computational grid. The latest SWAN version 40.91 was improved to facilitate the transfer of wave setup simulated on the parent grid into the child grid without the intervention of the user, and as a consequence of increasing the accuracy of wave setup simulations. This new version was validated through a particular violent wind storm that impacted on the French coast on the 28th of February 2010 (Xynthia). This specific event generated significant wave setup, accumulating offshore throughout the period in question and propagating toward the Pertuis coast.

In the 2000s, CEREMA and EDF R&D have been collaborating to build two continuous wave databases through numerical hindcast simulations: one covers the Atlantic Ocean, the other the Mediterranean Sea. These databases are called ANEMOC. Over the last three years, new versions of the two numerical atlases have been created, in collaboration with Saint-Venant Laboratory. Several improvements have been made in the construction of ANEMOC-2: the Atlantic model covers a larger area, the temporal coverage of the atlases is larger (32 years from 1979 to 2010), direction and frequency discretization is finer, wind forcing is finer both in time (1 h resolution) and space (0.312) and the computation meshes are refined to reach 800–1000 m along the French coast. The simulations are performed with the numerical wave model TOMAWAC, a third generation spectral model, which is a module of the TELEMAC-MASCARET open source suite. The databases are calibrated with altimeters measurements, and validated in a second step against uncorrelated buoys measurements. The databases provide several wave parameters: significant wave height, mean, peak and energy period, mean direction, angular wave spreading, and wave power. More results regarding calibration and validation are presented for the Mediterranean wave model. Results of both Mediterranean and Atlantic databases are then presented. Their analyses by comparison with altimeter and buoy measurements provide an assessment of many of their characteristics. Finally, ANEMOC-2 ability to reproduce intense wave conditions is highlighted by the study of two storm events.

Within the Technical Commission for the Study and Evaluation of Maritime Submersions in the Seine Estuary (CTeeSMES), whose aim is to improve the collective knowledge on physical processes linked to maritime surge levels, a numerical model of the Seine estuary based on TELEMAC2D has been used to study the evolution of surge levels from the ocean to the harbour area of Le Havre and, in particular, evaluate the amplification of the global signal and the apparition of seiches inside René Coty’s basin. The bathymetry of the model was partially provided by Le Havre and Rouen Harbours for the north-eastern part of the model. For the rest of the area, EMODNET was used. The numerical model was calibrated on the Johanna (March 2008) and Xynthia (February 2010) storm events. The global signal (tide + surge levels) was calibrated using measurements available on seven outputs of the Seine Estuary and provided by ports of Le Havre and Rouen to optimize the friction coefficient and the coefficient for wind influence. Winds and pressure fields came from CFSR data. Once the numerical model of the Seine Bay had been calibrated, it was possible to draw the evolution of surge levels from the ocean to Le Havre (quai Meunier) and then compare the signal obtained inside René Coty harbour basin. As shown by measurements, numerical results stress out the apparition of an oscillating signal which is added to the signal at the entry of the harbour. The process of amplification of seiche inside the port near the René Coty harbour basin is still underestimated by the model and further investigations must be realised.

Several studies have investigated the physics of squall lines in many parts of the world, but very few studies have examined the characteristics of Sumatra squalls and specifically the wind waves generated by them. Sumatra squalls as other tropical squall lines comprise of a convective region forming at the leading edge and a region of trailing stratiform rain to the rear. Wind speed increases gradually from the rear to the gust front before dropping rapidly afterwards. The study used Weather Research & Forecasting (WRF) model to successfully downscale and hindcast a squall event in April 2012, and the resulting wind field was applied to run a MIKE21 Spectral Wave model. The simulation results reproduced the properties of waves created by the squall wind. This paper further proposed shape description to parameterize the squall line. Synthetic wind fields were produced by varying three parameters: squall speed (

V

f

), peak wind intensity (

U

p

) and squall width (

W

s

). A matrix of such wind field was used to force the same spectral wave model. The outcome helps to inspect the influence of these parameters in affecting the wind-wave generation. Important findings from the assessment include: (1) The sensitivity level of the parameters in descending order:

U

p

,

V

f

,

W

s

; (2) For fixed values of the remaining two parameters,

W

s

is directly proportional to the maximum significant wave height (

H

s

max). This too applies for

U

p

; (3) For fixed values of

U

p

and

W

s

,

H

s

max increases gradually as a function of

V

f

till a peak is reached, after which

H

s

max decreases considerably.

The SPH method (smoothed-particle hydrodynamics) is a numerical meshless, particle and lagrangian method. It is used in a lot of fields of engineering and science such as solids mechanics, hydraulics, and astrophysics. The medium is represented, thanks to a set of particles which interact with each other. Nowadays, the SPH method is still under development but is able to deal with a wide range of problems in hydraulics. This article focuses especially on open channel quasi-incompressible flows. While implementing a SPH code, a programmer can face up some difficulties such as the neighbors search, the boundary conditions, the speed of sound, or the initialization of the particles. We have drawn some unexpected conclusions concerning the compressibility of the fluid and the way the particles are initialized. This paper presents also a list of test cases that can be performed in order to validate an SPH code. It includes: (a) a tank of still water, (b) a spinning tank, and (c) a dam break on a dry bed. These test cases allowed us to highlight some undesired effects. Finally, a new test case is developed. It is based on new experimental results of a flow on a spillway. For this test case, open boundaries have been implemented. The results presented in this paper are based on a 3-D code implemented during a master thesis.

Uncertainty in modeling results can be introduced via different sources including: the input data, the modeling assumptions, simulations based on hypothetical scenarios, etc. In this paper the uncertainty in modeling results of 1D and 1D/2D hydrodynamic Sobek models of flooding in the Niger River are analyzed. The models were set up with discharge data as upstream boundary conditions and tidal water level data as downstream boundary conditions. The models were run for the years 1998, 2005, 2006, and 2007. Data available for 1998, 2006, and 2007 were for flooding, while 2005 data represents normal flow data. The model setup included 48 cross sections located between Lokoja and the two ends of the rivers Forcados and Nun. The boundary conditions were varied downstream at the mouths of rivers Forcados and Nun using sea level rise (SLR) values adopted from the Rahmstorf predicted values; the simulations were projected for the years 2030 and 2050. Five modeling scenarios were set up to simulate the interaction of river flooding with downstream rise in sea levels. The scenarios were: sea level rise with normal year flow from upstream, sea level rise with a flooding year flow from upstream, sea level rise with flash floods from upstream, sea level rise with subsidence and flooding year flow from upstream, and sea level rise with subsidence and flash floods from upstream. The use of predicted SLR values introduces uncertainties in the model outputs. Another source of uncertainty was the value for land subsidence (25 mm/yr) adopted from estimates by local experts (the exact value is not yet known and might vary within the area). Uncertainty analysis of the modeling results were carried out using probability-based sampling methods in order to determine the uncertainties in modeling results for effects of downstream SLR on flooding extent, flooding time, and change in water depth in the Niger delta.

Crue is a central modelling tool for CNR. Thanks to a recent major update to enhance long-term maintainability, the system has been totally redesigned. It mainly relies on the Fudaa-Crue modelling software developed in Java and Crue10 hydraulic computational core redeveloped from Fortran to C++. It is also connected to the GeoGAMA-Crue GIS modelling software, SYSSIH simulation component repository and the hydraulic model repository. Four notable characteristics of the new system are presented, as examples of useful features simulation software may offer nowadays. First, the general design of the system relies on a combination of a computational core, strict property of CNR, and open source data model and modelling software, developed on the open Fudaa framework. This way, the modelling software can be easily shared within the hydraulic computational community. The second feature relies on the central object-oriented data modelling, based on an interlinked dictionary file. It holds the extensive variable description and configuration (for data and results) shared between the different components of the Crue system. The third characteristic is the built-in comparison system able to detect only significant differences between two scenarios (on both data and results). It is fully configurable and can also run a test-case library for non-regression checking. The fourth feature is an open documentation system (on a wiki model) gathering all the documentation available around Crue (user’s manual, hydraulic modelling, numerical analysis, collaborative user’s documentation, etc.). It is fully integrated on the modelling software and can react to the core’s log content, for example.

In this study, we address the difficult problem of flash flood prediction in Caribbean watersheds for which current hydrological system are not well suited. These basins have small surface areas with steep slopes due to their volcanic origin, and they are subjected to tropical rainfall conditions such as massive and localized precipitations. We propose a data-driven solution whose main originality is two-folded: (1) the predictive model is defined as a set of aggregate variables that act as classifiers, (2) an evolutionary algorithm is implemented to find best juries of such classifiers. The design of this solution was guided by the necessity to reach three main objectives: precision, readability, and flexibility. Indeed, a flood forecasting solution should not only provide accurate prediction performances, but it should also give clear explanations about how and why an alert is triggered or not on one hand, and be easily adaptable on similar catchments on the other hand. The concept of aggregate variables allow to reach the objective of readability by using simple rules based on threshold over-passing of aggregated values, while the data-driven nature of the solution and the use of combinations of aggregate variables allow to reach the objective of flexibility. The results obtained for the case study of a typical Caribbean river, for which runoff data are available at three locations, demonstrate the efficiency of the solution.

In this work, we are interested in the derivation of a new shallow water model with a diffusion source term. Analytical solutions for steady flow regimes are first presented to validate a numerical method designed to solve this new model. Then this model is applied on real data and seems to give better results than the classical shallow water system.

The paper presents an integrated tool for dam-break hazard modelling. It is based on a two-dimensional, depth-averaged hydraulic model that uses a conservative and shock-capturing finite volume scheme on a Cartesian grid. The hydraulic model is designed so that different zones within the computation domain can be modelled with different spatial resolutions and/or different model enhancements (e.g. pressurised flows, sediment transport, etc.). The two-dimensional model can be coupled with lumped models that compute the water stage in reservoirs and the outflow discharge through hydraulic structures, in case of normal operating, failure of a valve or breaching processes. These features make hydraulic modelling versatile and computationally efficient. They enable the definition of different failure scenarios, which is of prominent importance given the uncertainty of such a phenomenon. If several hydraulic structures are involved, the procedure takes the behaviour of each structure into account. The sensitivity of the results with respect to the interactions between the flow and the terrain (roughness coefficient, collapse of buildings, breaching process) can be analysed. The results of the hydraulic model are handled thanks to a graphical user interface that provides one-, two- and three-dimensional views and animations of the unsteady flow-field and enables the understanding and verification of the results. Danger maps are generated based on the results of one or more ‘worst-case’ scenarios. In case of a complete risk analysis, the danger maps are combined with data on the exposure and vulnerability of elements at risk for the computation of the corresponding damage.

Following the December 2003 floods on the Rhône, the public authorities embarked on a far-ranging flood prevention plan called the “Rhône Plan”, The reinforcement works for the embankments foreseen by Symadrem, the management authority overseeing the embankments of the Rhône; aim to move the banks back along certain sections and develop overflow zones in order to ensure overflows without breaches in the event of water levels higher than the reference water levels. The hydraulic study was carried out using two-dimensional modelling with Telemac-2D software and Infoworks RS 2D software. Four two-dimensional models were built: three models, each featuring approximately 300,000 elements, were built for the channel of the petit Rhône with its embankments, and the protected beds on the right bank and the left bank (the insular part of the Camargue), one model covers the left bank of the Rhône, with about 150,000 elements. The linear distances modelled cover a length of approximately 60 km of the Rhône and the Petit Rhône between Arles and the sea. The surface area modelled is approximately 1800 km

2

. The objective of the modelling is to determine the initial and intermediate state without breaches in the riverbed within the embankments, adjust the works, determine the final state in the riverbed within its embankments, determine the hydrographs of the overflowing high water levels for propagation in the floodplain, and determine several scenarios of embankment breaches for the purpose of the hazard study.

In recent years, along with the impact of global climate change, the city’s rapid development, and the backward of disaster prevention systems, the megacities and medium-sized cities in China have suffered an increasingly serious “water” problem. Especially, for the frequent urban local flooding, it considerably threatens sustainable human development, disrupts continuous delivery of urban infrastructure service to citizen, endanger public security, and leads to considerable economic losses. All these issues contribute to the necessity and urgency of enhancing the resilience of the urban local flooding management to natural hazards and human activity. In this article, the definition of resilience is described in the field of urban local flooding management. Simultaneously, along with specific case study of flood resilience strategy—Service Risk Framework of Beijing—this study demonstrates the current advanced resilience methodologies particularly for flood hazards, which provide valuable reference for proposing the idea of resilience and formulating relevant industry regulation standards.

The overall objective of the present paper is to propose a statistical approach to downscale the precipitation process at an ungauged location in the context of climate change. More specifically, the proposed approach consists of a combination of three components: (i) a regionalization approach for identifying the homogeneous groups of observed daily precipitation series available at different raingauges; (ii) a stochastic model for constructing daily rainfall events at an ungauged location within a homogeneous group; and (iii) a statistical downscaling model (SDRain) for describing the linkage between the constructed daily precipitation series and the large-scale climatic predictors given by the GCM simulation outputs. The feasibility of the proposed stochastic approach has been assessed using the available daily precipitation data for the period 1973–2001 from a network of 63 raingauge stations in South Korea and the NCEP reanalysis climate predictors. Results of the numerical application have indicated that it is feasible to estimate the missing precipitation data at an ungauged site based on the data available at other sites within the same homogeneous region. Furthermore, the proposed SDRain was able to generate daily precipitation sequences for an ungauged site with comparable statistical characteristics as those given by the application of SDRain for a gauged site with available observed precipitation data.

Technologies such as aerial photogrammetry allow production of 3D topographic data including complex environments such as urban areas. Therefore, it is possible to create High-Resolution (HR) Digital Elevation Models (DEM) incorporating thin above-ground elements influencing overland flow paths. Although this category of “big data” has a high level of accuracy, there are still errors in measurements and hypothesis under DEM elaboration. Moreover, operators look for optimizing spatial discretization resolution in order to improve flood model computation time. Errors in measurement, errors in DEM generation, and operator choices for inclusion of this data within 2D hydraulic model, might influence the results of flood model simulations. These errors and hypothesis may influence significantly the flood modeling results variability. The purpose of this study is to investigate uncertainties related to (i) the own error of high-resolution topographic data and (ii) the modeler choices when including topographic data in hydraulic codes. The aim is to perform a Global Sensitivity Analysis (GSA) which goes through a Monte-Carlo uncertainty propagation, to quantify impact of uncertainties, followed by “Sobol” indices computation, to rank influence of identified parameters on result variability. A process using a coupling of an environment for parametric computation (Promethée) and a code relying on 2D shallow water equations (FullSWOF_2D) has been developed (P-FS tool). The study has been performed over the lower part of the Var river valley using the estimated hydrograph of a 1994 flood event. HR topographic data has been made available for the study area, which is 17.5 km

2

, by Nice municipality. Three uncertain parameters were studied: the measurement error (var. E), the level of details of above-ground element representation in DEM (buildings, sidewalks, etc.) (var. S), and the spatial discretization resolution (grid cell size for regular mesh) (var. R). Parameter var. E follows a probability density function, whereas parameters var. S and var. R are discrete operator choices. Combining these parameters, a database of 2,000 simulations has been produced using P-FS tool implemented on a high-performance computing structure. In our study case, the output of interest is the maximal water surface reached during simulations. A stochastic sampling on the produced result database has allowed to perform a Monte-Carlo approach. Sensitivity index have been produced at given points of interest, enhancing the relative weight of each uncertain parameters on variability of calculated overland flow. Perspectives for Sobol index maps production are brought to light.

The management of navigation canals is complex. The most important requirement apart form flood protection and maintenance of ecological flow is to keep the water level within a certain range around the normal navigation level. Due to geographical reasons, some navigation reaches are connected by locks. When these locks overcome big level differences the disturbance created by a lock operation can influence the water level in the reach and the possibility of navigation. There can be known and unknown, controllable and uncontrollable inputs influencing the water level. In order to plan the management and control of such a system a model is needed. An example for such reach is the Cuinchy-Fontinettes reach in the north of France. The reach is bounded by two locks, and the (desired) operation of the downstream lock causes a disturbance that does not allow keeping the water level in the navigation range. In this work, the model of the system is built, and the estimation and the approximation of the unknown inputs is presented with uncertainties.

A real-time operation method of a multi-purpose reservoir for flood management considering an ensemble streamflow prediction (ESP) is investigated in this study. The ESP is derived by a distributed rainfall-runoff model from an operational ensemble prediction of precipitation. Japan Meteorological Agency’s One-week Ensemble Forecast of precipitation, which is provided every day and has 51 ensemble members of six-hour precipitations for the coming eight days, is employed here. ESPs with 51 members for the coming eight days are calculated from the ensemble predictions of basin precipitation by use of Hydrological River Basin Environment Assessment Model (Hydro-BEAM), a distributed rainfall-runoff model. Reservoir states such as release or storage are then estimated for each ensemble member of the streamflow predictions to support preliminary release operation. Chance and the expected amount of recovery in storage water at the end of the flood event are also estimated for each scenario of reservoir operation to estimate impacts of the preliminary release operation on water supply operation in the following period, in order to help reservoir manager with making a decision on preliminary release considering the prediction and its uncertainty. The presented method was applied to Nagayasuguchi Reservoir in the Naka River basin in Japan, demonstrating the effectiveness and potential to provide useful information for real-time preliminary release operation of reservoirs.

In 2010, hydraulic energy represented 80

% of renewable energy production in France. It has the advantage of being highly adaptable to the demand placed on the network, in contrast to solar and wind energy. However, most potential production sites are already equipped and it is now necessary to improve their operation in order to increase productivity.

The goal of the PENELOP2 project is to study the influence of different factors contributing to the flow disturbance through low-head turbines hydraulic passages. As they typically operate on a run-of-river basis, the turbines (which are generally of the bulb type) are vulnerable to flow disturbances, especially upstream. The aim is to devise a method for optimizing the shape of the built structure around the turbine, depending on inflow and outflow conditions.

Our approach consists of reproducing the flow disturbances and evaluates their influences on global productivity of the turbines, meaning under non-ideal conditions. We have already done several stages with the instrumentation of a unit at Vaugris plant on the Rhone river. These measurements were performed under several hydraulic conditions, in order to collect a maximum amount of data exploited by numerical and physical models built subsequently. These models (the power dam and a part of the river) have tested the various possible modifications for improving the uniformity of flow along the flow path. We are now building a more general optimization tool by interpolating/extrapolating the results obtained on the numerical and physical models and by constructing a simulator to evaluate the loss of energy (unrecovered production part operated by turbine groups), which depends on the encountered conditions on the river and the power plant: upstream disturbance, through flow separation with several turbines in different operating conditions, and downstream disturbance.

A revision of the hydrology study for the Galaube dam catchment area has been made as part of the dam safety report. This study highlights an increase of the design flood. A 3D CFD free surface hydraulics model of the Galaube dam is built with ANSYS CFX in order to verify the discharge capacity for the new design flood. The 3D CFD model is validated on two elements: the results of the initial studies and the model’s sensitivity on the roughness coefficients and the mesh size. The different attempts on the model’s sensitivity do not lead to important modifications of the spillway rating curve. Moreover, it appears that, in a range of flow, the rating curve calculated with the 3D computational model is smaller than the rating curve obtained in the initial studies. This reduction is due to a drowning of the weir. This phenomenon is highlighted with the 3D CFD model. Furthermore, the model permits the identification of the limit flow rate before the saturation of the spillway and the representation of the transition in flow characteristics. Otherwise, the results are used to define some works to insure the dam safety: the dam freeboard and the wave wall safety are checked, and the height of the spillway training wall is increased.

Investigations indicate freshwater sources are dwindling day-to-day and becoming contaminated throughout the world due to environmental problems, and fast growing population. Therefore, flows in the dam reservoir using proper realistic water modeling should be well defined. In this study, three-dimensional hydrodynamic simulation model of an actual dam reservoir for a season is created. The density differences between inflow river and ambient dam reservoir water can create stratified and circulation flows in the real dam reservoirs. The density differences can be due to the discrepancies in temperatures, concentration of dissolved or suspended substances, or a combination of both. The numerical model is developed using nonlinear and unsteady continuity, momentum, energy, and

k-ε

turbulence model equations. In order to include the Coriolis force effect on the flow in a dam reservoir, Coriolis force parameter is also added to the model equations. These equations are constructed using actual dimensions, shape, boundary, and initial conditions of the dam and reservoir. The 3-D mathematical model developed is capable of simulating the flow and thermal characteristics of the reservoir for using season. Model simulations results are compared with field measurements obtained from gauging stations. The results are found to be in accordance with the field measurements.

FullSWOF_2D (Full Shallow Water equation for Overland Flow in two dimensions) is a free software designed for shallow water flow simulations. The shallow water equations are solved thanks to a well-balanced finite volume scheme (based on the hydrostatic reconstruction), which is adapted to the properties of the model considered (in particular conservative laws, hyperbolic system, and steady states). The sources of this software (in C++) are available from

https://sourcesup.renater.fr/projects/fullswof-2D/

. This software has been validated on several analytical test cases integrated in SWASHES library and on rainfall overland flow simulations. Because of the simulations on big data necessity, this software has been parallelized with two different strategies (MPI and SKELGIS) in the framework of the CEMRACS 2012. Our purpose is to continue the comparison and the validation of these two versions of FullSWOF_Paral on realistic test cases. Our methodology will consist in comparing these two approaches on classical test cases such as Malpasset’s dam break with 2D hydraulic softwares such as MIKE 21, MIKE 21 FM, and TELEMAC 2D. The two strategies are presented in this paper. As this work is still in progress, only results from MPI version are presented here. More results will be given in future works.

Flows in the lakes are driven in most cases by shear stresses resulted from wind and freshwater inflow. The effect of wind-induced current has been studied through field observations, laboratory small-scale physical models, and numerical models to evaluate the impacts on hydraulic structures, mixing, and stratification. Typhoon-induced inflow, especially in Taiwan, is an important factor to affect the mixing in the water column. In the present study, a three-dimensional, time-dependent hydrodynamic model was performed and applied to the alpine Yuan-Yang Lake in northeastern region of Taiwan. The model was driven with discharge inflow, heat, and wind stress to simulate the hydrodynamics of lake. The model was validated with measured water surface elevation, current, and temperature in 2008. The overall model simulation results are in quantitative agreement with the observed data. The validated model was then used to investigate mean circulation and residence time in the YYL. The simulated mean current reveals that the surface currents flow toward the southwest direction and form a clockwise rotation. The calculated residence time is strongly dependent on the inflows and wind effects. The calculated residence time is approximately 2~2.5 days under low inflow with wind effect.

The present study has been carried out to analyze the hydraulic behavior of the PAC-UPC canal, located in the Campus Nord of the Technical University of Catalonia. This constitutes a model of a real irrigation canal and, since its construction in 2003, it has been used for different studies and PhD dissertations related to irrigation canals and control algorithms. The snake-shaped canal, with the aim to minimize the space to be occupied, generates some bends along itself, where pronounced upraising water surface levels are produced. Even with a subcritical regime it is clearly observable, whose details require a complete analysis. In order to do these, different types of approaches considering a 1D, 2D, and 3D analysis, respectively, have been proposed. Codes considered were: Hec-Ras (1D), Iber (2D), and Flow 3D performing a comparative study of the results of each code. This comparison highlights the limitations of each one. As it was known previously, 3D results offer much more information about the flow behavior, even enabling the analysis of the recirculation zones and eddies formed in the z direction. The final results, after flow analysis, suggest possible improvement in the canal for future works and studies, regarding for instance the best location of the instrumentation (Level Sensors) or modifications about the geometry of the structure.

During storm events inducing extreme sea levels, coastal floods may occur. Water volumes that flood low-lying coastal zones may be important and significantly impact the sea levels along the coast. This effect is expected to be particularly significant in tidal bays or estuaries. This phenomenon was recently illustrated during storm Xynthia (February 2010) on the French Atlantic coast; volumes that flooded Charente-Maritime and Vendée departments corresponded to a non-negligible fraction of the tidal prisms of the Pertuis Charentais Sea (area between Ré and Oléron Islands and the continent); the order of magnitude is 10 %. The impact of the flooding associated with Xynthia is investigated by means of numerical modelling in the Pertuis Charentais carried out by LIENSs. Within the framework of risk mapping, CREOCEAN conducted flood studies in two sites of the eastern English Channel: Authie Bay and Dives Estuary, for which simulations of assessed future extreme events were run. Although these studies used different modelling systems at sites displaying distinct features, they all suggest that flooding can impact water level along the coastline up to more than 0.5 m. Limitation of high tide levels, related to flooding phenomena, are mapped and analyzed. Recommendations are provided to define a relevant methodology, accounting for such a limitation process, for submersion and risk studies.

Particle deposition in turbulent flows is a phenomenon which can lead to fouling and affect normal operating conditions of key components of industrial processes. To explain the deposition mechanisms and predict the deposition rate, several models have been proposed in the literature. The model presented in this paper is based on a stochastic Lagrangian approach, where each particle is explicitly tracked, and where the velocity of the flow seen by particles is modeled by a stochastic process which depends on the mean fluid properties at particle locations. The interactions between particles and near-wall coherent structures are taken into account. Recent developments have shown that the model is not only able to reproduce single-particle deposition and resuspension but can also be applied to simulate the formation and the growth of multilayer deposits. Such deposits result from the competition between particle–fluid, particle–surface, and particle–particle interactions. Different morphologies of the deposit (monolayer, dendrites, multilayer) can exist according to the chemical properties of the particles and wall. A porous medium approach is used to take into account the effect of the deposit formed on the flow to obtain more realistic evolution.

The NEPTUNE project constitutes the thermal-hydraulic part of the long-term EDF-CEA-AREVA-IRSN joint research and development program for the next generation of nuclear reactor simulation tools. The project aims at developing high modeling capabilities for advanced two-phase flow thermal-hydraulics covering the whole range of modeling scales. The CFD scale for flow description is covered with NEPTUNE_CFD code. The multiphase approach, developed in the NEPTUNE_CFD code for nuclear engineering, is based on separate Eulerian transport equations for mass, momentum, energy, and turbulent quantities of the different fluids, which are coupled through inter-phase transfer terms. This model is primarily dedicated to the simulation of multiphase flows containing one continuous fluid always present, which carries dispersed fluids present in the form of bubbles, droplets, particles, whose dimensions are much smaller than the spatial resolution length of the model. The simulation of all range of multiphase flow situation, such as dispersed and liquid/gas stratified (separated) flows, which can be encountered in nuclear PWR circuits and pipes under nominal or incidental conditions, remain challenging cases for multiphase volume averaged flow models. The paper deals with a short presentation of NEPTUNE_CFD model, dedicated to incompressible, weakly compressible, unsteady, and turbulent 3D two-phase flow computations. Some modeling strategies will be detailed through the examples of two validations of semi-integral cases.

Boiling crisis and flows occurring in a steam generator or during the dewatering step of the cold leg of the reactor pressure vessel remain a major limiting phenomenon for the analysis of operation and safety of both nuclear reactors and conventional thermal power systems. Firstly, the choice is made to investigate a hybrid modeling of the flow, considering the gas phase as two separated fields, each one being modeled with different closure laws. In so doing, the small and spherical bubbles are modeled through a dispersed approach within the two-fluid model, and the distorted bubbles are simulated with an interface locating method. Secondly, this approach is generalized to simulate free surfaces (large bubbles considered as continuous gas phase degenerate into large interfaces) and heat and mass transfer. The present work gives an overview of the multifield approach implemented in NEPTUNE_CFD code at EDF R&D.

The Reynolds averaged Navier-Stokes equation model (RANS) is state-of-the-art for numerical simulations of cavitating throttle and injector flows. RANS models are based on time-averaged Navier-Stokes equations, and the computational costs are much lower than those for the more advanced Large Eddy Simulations (LES). The principle of LES is low-pass filtering of the Navier-Stokes equations to eliminate the small scales of the solution. In general LES requires higher numerical resolution in space and time, and higher order discretization schemes. In the recent years, clustered processing units provide increased computational resources, and therefore LES simulations became, also for two-phase flows, more and more interesting. The current paper presents Large Eddy Simulations of the cavitating two-phase flow in a rectangular micro-scale throttle operated with Diesel fuel and compares them with RANS simulations. The LES shows interesting new details which cannot be resolved by RANS simulations in general, such as the transition from laminar to turbulent flow in the channel or the phase change caused by turbulent pressure fluctuations in the shear layer. A cavitation erosion model predicts the zones with highest damage probability. All simulations are performed with the commercial CFD code AVL FIRE

®

. Time-averaged results of the numerically predicted velocity profiles and the liquid–vapor distributions are compared with already published optical measurements performed with the Laser-Induced Fluorescence (LIF) and the light transmission techniques.

The Swiss national research project Hydronet 2 gathers a consortium of industrial and academic partners supported by the Competence Center of Energy and Mobility (CCEM) and Swiss Electric Research (SER) in order to improve the hydropower plants. One of the research topic focuses on the cavitating tip vortex. Such a vortex takes place in axial turbines as Kaplan turbines used for producing hydroelectricity. This phenomenon drives a lot of drawbacks such as erosion, unsteady flow rate and a decrease of the turbine efficiency. To better understand the behaviour of the tip vortex, computations of a simple test case are performed. The test case consists in a NACA profile mounted in a channel with a gap between the NACA tip and the lateral wall. The computations are carried out with the OpenFOAM solver both in one-phase and two-phase flows. The turbulent motion is modelled with a RANS approach. For the two-phase flow computations, the phase change between liquid and vapour is achieved with the model proposed by Kunz. The results will be compared in cavitating and non-cavitating cases with the experimental data provided by the EPFL Laboratory for Hydraulic Machines. The comparisons deal with global picture of the flow, the trajectory of the tip vortex and the velocity field downstream the NACA profile.

Cavitation is a well-known physical phenomenon occurring in various technical applications such as hydraulic turbo-machines, pipe flows, and venturis. Coupling aeration in a cavitating flow is a recent technique to control the overall effect of the cavitation over the zone of interest. The aeration process is done by injecting spherical air bubbles into the fluid flow without having at the same time an interaction with it. The contact handling algorithm is based on the projection of the velocity field of the injected particles over the velocity field of the fluid flow, in such a manner that, at each time step the gradient of the distance between every two bubbles is kept non-negative as a guarantee of the non-overlapping. The collisions between the air bubbles are considered as inelastic. The differential equation system is composed of the Navier-Stokes equations, implemented with the Homogeneous Mixture Model. The latter accounts the three phases (liquid, vapor, and mixture) separately. In the mixture phase, the gas and liquid phases are considered in local thermodynamic equilibrium. A high-order Finite Volume solver based on Moving Least Squares approximations is used for this analysis. For the sake of the numerical simulations, structured and unstructured grids have been used. The code makes use of a SLAU-type Riemann solver for low Mach numbers in order to accurately calculate the numerical fluxes. To avoid any numerical oscillations in the zones of strong gradients, a slope limiter algorithm is coupled with a Moving Least Squares sensor detecting any discontinuities.

The results from analytical, numerical and experimental modelling of free-surface vortex flows are presented. Vortex flow is induced in a gravity-driven, open-channel flow chamber with a subcritical approach flow and is simulated using the ANSYS CFX steady Eulerian multiphase flow model in order to determine water surface- and velocity-field characteristics. Solution sensitivity to mesh type, density and various turbulence closure methods is considered. The water surface and tangential velocity profile are also modelled using the Vatistas (n = 2) analytical model. The numerical solution is validated using experiments conducted in a scaled physical model of the chamber which permits the investigation of the air/water interface and determination of the velocity fields using particle tracking velocimetry. The sensitivity analysis carried out presents a case for mesh independence and gives evidence that the baseline Reynolds stress model is most suited in simulating free-surface vortex flows. The predicted shape of the air core is in agreement with the physical model but the location of the resolved free-surface interface is under predicted. Concerning the velocity field, the Reynolds stress model makes a fair to moderate prediction of the tangential velocity field; however, the radial velocity field is typically underpredicted. It is concluded that unsteady flow features inherent in the vortex, namely, free-surface instabilities, are preventing the steady-state model from achieving the required accuracy, thus requiring further transient analysis.

Air entrained has become one of the main variables in the study of large spillways performance since it can help avoiding cavitation. Moreover, high rates of air concentration produce significant bulking of the flow as well as a water–solid friction reduction, generating flow acceleration and increasing maximum velocities at the inlet of the energy dissipation structure. Air entrained also affects turbulence inside the flow producing different energy dissipation rates. Aerated spillways physical models are affected by scale effects, with Weber and Reynolds numbers being usually too low to adequately reproduce observed flows. Alternatively, simulation of air–water flows can be carried out by means of Computational Fluid Dynamics techniques in 1:1 scale. However, 3D numerical simulations of spillway flows are time expensive and air–water interfaces need fine resolution meshes which would require extensive computing. Thus, the use of a subgrid scale in air entrainment models can be useful to predict the inception point and the air concentration profile of the flow along the spillway. Computational techniques can handle a more accurate momentum distribution over the spillway sections with affordable costs. In this research, FLOW-3D® routine for turbulent air entrainment is used, coupled with variable density evaluation. VOF and κ-ε RNG turbulence model are also employed. Over 200 spillway flow simulations have been carried out to obtain optimal values of the air-entrainment model parameters, which can be used in future spillway simulations. The calibration of the model is carried out employing prototype data. Interesting conclusions are obtained concerning air entrainment model performance.

Three-dimensional flow of a Newtonian fluid through porous media consisting of multisized perfectly spherical beads is numerically simulated. Porous media are virtually and randomly generated, in accordance with the particle size distribution. All the physical features of the porous media are fully represented and considered, in order to closely mimic the heterogeneity of these media. Simulations show that the smallest particles strongly affect the streamline shape by imposing small pores and small throats, and also by providing the shortest paths. The third dimension has a particular effect, which is to allow the streamlines to follow the path of least resistance, where the least amount of energy is required to produce the motion of a fluid particle. Once the three-dimensional images have been produced, the proposed model provides a three-dimensional representation of the streamlines and their velocity. Unfortunately, the results currently generated do not agree well with those proposed in the literature. Consequently, there is a pressing need to verify these results by conducting more numerical tests.

Particle deposition in turbulent flows is a phenomenon which can lead to fouling and affect normal operating conditions of key components of industrial processes. To explain the deposition mechanisms and predict the deposition rate, several models have been proposed in the literature. The model presented in this paper is based on a stochastic Lagrangian approach, where each particle is explicitly tracked, and where the velocity of the flow seen by particles is modeled by a stochastic process which depends on the mean fluid properties at particle locations. The interactions between particles and near-wall coherent structures are taken into account. Recent developments have shown that the model is not only able to reproduce single-particle deposition and resuspension but can also be applied to simulate the formation and the growth of multilayer deposits. Such deposits result from the competition between particle–fluid, particle–surface, and particle–particle interactions. Different morphologies of the deposit (monolayer, dentrites, multilayer) can exist according to the chemical properties of the particles and wall. A porous medium approach is used to take into account the effect of the deposit formed on the flow to obtain more realistic evolution.

Aircraft fuel systems are subject to icing at low temperatures. If the flow rate is increased, sudden releases of large quantities of ice may occur, called “snow showers”. They threaten the safety of flights and have been the subject of several investigations over past years. Jet engine fuel system components may be sensitive to clogging. When a snow shower happens, ice particles settle in seconds, forming a porous layer. Modelling such events involves transient hydraulics and solid dynamics. We propose to investigate numerically the dynamics of transient particle clogging. Equations of motion for the incompressible fluid phase are discretized in a high-order finite volume context and solved using a pressure-based algorithm. The discrete phase is modelled in a Lagrangian frame. Contacts between solids are handled by a dedicated algorithm. Solid volume fraction is calculated in regions occupied by particles. Finally, two-way coupling is achieved by source terms for momentum exchange, viscous and inertial loss. 2D simulation of the clogging of an ideal filter is performed.