ABSTRACT. Numerical flow models have become indispensable tools for many issues in hydraulic engineering practice over the past decades and play a crucial role in decision-making processes. Thanks to rapid developments – especially regarding data availability and computational power – results and appealing visualizations can be produced increasingly faster even for the most complex hydraulic scenarios. However, this can easily lead to overconfidence in the models and obscure errors, while at the same time making also the assessment of the robustness of the models more challenging.

To ensure high quality in modelling and its results, quality assurance and verification are becoming increasingly essential. This paper presents the results of an online survey of more than 200 stakeholders in numerical flow modelling from the field of hydraulic engineering, which examined current quality assurance methods and requirements for new systematic concepts.

The survey revealed that measures for ensuring quality are rarely standardized in modelling processes and existing strategies for quality management are not applied in practice. The resulting strong dependence of modelling quality on the experience and individual approach of the modelers could be reduced by systematic quality management concepts. The outcomes of the survey can be used to derive objectives and framework conditions for their development that must be met for successful implementation in modelling practice.

This work is conducted within the context of the DWA (German Association for Water, Wastewater and Waste) working group WW-1.7 “Qualitätssicherung und -management beim Einsatz mehrdimensionaler Strömungsmodelle”, aiming to develop methods for quality assurance and quality management for all participants in the process of flow modeling.

ABSTRACT. The flood hydrograph is a key tool for representing the hydrological response of a watershed following a rainfall event. Its construction typically relies on precipitation data, discharge measurements, and watershed characteristics. However, reconstructing the hydrograph and analyzing sediment dynamics in the Rouyonne River, presents a significant challenge due to limited hydrological datasets availability. No continuous measurements are possible in a context of limited resources and general insecurity. However, it can be taken advantage of occasional research projects to capture some events with the best available accuracy in order to improve the knowledge of precipitation-discharge mechanisms. Some measurements are easy to implement. For example, in the frame of a research program between the State University of Haiti (UEH) and the University of Louvain (UCLouvain), it was possible to study some flood events such as the June 2–3, 2023 flood, in Léogâne, event that caused bank failure of the river, inundations in the city, with serious damages and large sediment deposition inside and outside the river. Precipitation data were recorded at a one-minute interval as well as water levels in the river. However, only direct measurement of small flows was possible when access to the river was dangerous for larger flows. As a result, the amount of discharge data was not sufficient to build a reliable stage-discharge relationship, nevertheless essential to establish the hydrograph in the event of a flood. Therefore, a hydrological model specifically adapted to data-scarce environments is required. Various approaches were employed in this study to estimate the flood hydrograph, including the development of a stage-discharge relationship and the application of rainfall-runoff transformation models such as ATHYS. This model estimates watershed response based on observed precipitation while integrating soil characteristics, vegetation cover, and morphological conditions through the SCS-CN production function. The calibration and validation hydrological model yielded Kling-Gupta Efficiency (KGE) values of 0.92 and 0.90, respectively. These were performed using past flood events and short-duration direct observations. The peak discharge during the June 2023 flood event was twice the maximum capacity of the river at the measurement site, leading to a violent inundation of Léogâne city, which resulted in over twenty fatalities. These recurrent and overflowing floods are often accompanied by significant sediment transport, which drastically alters and destabilizes the riverbed. Between 2010 and 2023, major morphological changes were observed at the river mouth, with the formation of a wave-dominated delta extending over 200 meters, covering an estimated deltaic plain area of approximately 68 000 square meters, thus testifying the substantial sediment transport processes occurring within the Rouyonne River. This study underscores the urgent need for improved hydrometric monitoring, flood risk mitigation strategies, and sediment management interventions to enhance resilience in Léogâne and other flood-prone regions of Haiti.

ABSTRACT. Urban pluvial flooding have a significant hazard, particularly in densely built environments where impermeable surfaces such as roofs contribute to rapid runoff generation. While dual drainage models offer robust flood simulations, their reliability depends on accurate input data, which is often incomplete for large urban areas. This study proposes and evaluates alternative roof modelling strategies that address data limitations while maintaining simulation accuracy. Two case studies were used: idealized urban catchment (IUC), representing a controlled environment with complete data, and a real urban catchment (RUC), representing a complex, highly urbanized area with limited data availability. The 2D-1D coupled drainage model Iber-SWMM was applied to simulate runoff behavior under three rainfall scenarios. The study assessed roof geometry modelling strategies (RGMS) based on width estimations and roof-to-manhole connection modelling strategies (RMCS) that approximate drainage connections on incomplete data scenarios. The sensitivity analysis of RGMS showed that approximating roof width using aspect ratio (AR) or block ratio (BR) resulted in minor discrepancies compared to direct measurements. For AR values between 1 and 2 and BR values between 10 and 20, the Nash-Sutcliffe efficiency (NSE) coefficient remained above 0.95, indicating strong agreement with benchmark models. Similarly, RMCS results demonstrated that for manholes with a roof contribution area (RCA) greater than 10.000 m², the differences in discharge hydrographs between strategies were negligible. This suggests that simplified connectivity assumptions are viable for large-scale urban models. The study highlights practical strategies for roof modelling in large-scale urban drainage simulations where data availability is limited. The proposed RGMS and RMCS approaches significantly reduce model setup time while maintaining hydrodynamic accuracy. These findings provide a framework for improving urban flood simulations in cities lacking detailed roof geometry and drainage network data.

ABSTRACT. The objective of sensor networks is to obtain data of the highest quality that is useful for real-time and delayed use in mathematical analysis and prediction models. The quality of the data from these networks is related to the mission of the sensors, the variables to be measured, the design of the network, the measurement points chosen, the sensors and the technology used, the computational infrastructure required to receive, store, process and display the measured data, the managerial and administrative aspects regarding the maintenance of the network, the application of international standards and norms, the planning of the people who will perform the tasks necessary for the control of the network. Having a general knowledge of the development of these concepts allows a manager or designer to plan and implement a network in the most efficient way possible, and in the case of a data user it will allow to select the sensor network to be used in the most appropriate way, and to weigh the quality by anticipating the impact that the sensors will have when implemented in a model, and then evaluate the results and conclusions obtained from it. This work seeks to provide the basic information and concepts for understanding sensor networks.

ABSTRACT. The most common cause of floods is heavy rain events, which can occur over short temporal and spatial scales. To better capture such events and their variability, spatially distributed hydrological models have been used at hourly or sub-hourly time steps. High-resolution hydrological models can potentially make it easier to integrate spatially-distributed, more locally relevant precipitation datasets (e.g. meteorological radar data) or high-resolution numerical weather forecasts, which are often provided at the pixel scale. However, they face the challenge of being computationally demanding and often not fast enough for operational applications such as flash-flood forecasting, where real-time run times have to be fast enough to enable actionable forecasts. In this study, we investigate the impact on streamflow simulation of varying spatial resolution in a semi-distributed hydrological model applied to a large sample of catchments in France. The main question that guides this work is: to what extent does the spatial distribution adopted in hydrological modelling affect the ability of the model to accurately capture the hydrological response of a catchment to heavy rain events? Our methodology employs the GRSD semi-distributed model across multiple spatial resolution (mesh of 10, 50, 100, 150, 200, or 250 km²) to evaluate its representativeness at the catchment outlet. For comparative purposes, we also apply the model with a lumped configuration. The model runs with hourly precipitation from the COMEPHORE (COmbination for Best Estimation of Hourly Precipitation) reanalysis produced by Météo-France, which merges radar and rain gauge data. Data is available at 1x1 km² resolution over the period 1997-2022 and is spatially aggregated according to the resolution of the hydrological model. Results obtained in terms of gain in performance of simulated flows, using the Kling-Gupta Efficiency (KGE) and the percentage bias (PBIAS) criteria, and of peak discharges for selected events are discussed. We also reflect on how the outcomes of this study could provide insights into the trade-offs between computational efficiency and model accuracy in flood prediction.

ABSTRACT. Floods are one of the most damaging natural or human-induced phenomena in the world. In the last 20 years, they have caused economic losses of more than USD 600 million and affected more than 2 billion people. To prevent and minimise this hazardous situations, flood hazard and risk maps are developed around the world. These studies are commonly based on the application of hydrodynamic simulation tools, being feed by geospatial data for the construction of the numerical model such as meteorology, topography and land use. Additionally, some of these data could require a statistical treatment, obtaining different results depending on the approach carried out or statistic applied. The accessibility and updating of these data is becoming more and more frequent. However, the timing of these works does not always coincide, covering parts of the territory for some products in a particular time-gap and no for others. This can generate situations in which the model is not representative of the physical process to be simulated, either due to the spatiotemporal mismatch of the data or the lack of it. This situation is exemplified through a hydrologic-hydraulic study in an area near to Salou and a hydraulic study in Llobregat River near Sant Boi de Llobregat (both NE Spain). The study aimed on identify processes that help to define a clear and homogeneous methodology through the analysis of flooding caused by rainwater in urban areas. To that end, an analysis in deep was done on all available data (e.g. study area domain, rainfall, topography, land uses, drainage network). In terms of data acquisition, there is a wide range of web-based tools that provide free distributed data, the use of it should be limited to particular regions. The verification, contrast and quality of the available data can sometimes be difficult to carry out due to the lack of other data or the difficulty of obtaining them through field campaigns, being still essential to carry out flood studies with warranties. The use of high spatial resolution topography and the incorporation of transversal and longitudinal drainage elements is postulated as an essential point to correctly define the flow patterns. Although the urban drainage network, if it exists and is incorporated into the numerical model, would allow for a detailed evaluation of various urban runoff and propagation processes, the acquisition and processing of these data can be complex. In addition, the reliability and representativeness of the system may, depending on the area, be reduced due to the lack of specific data, inconsistencies in the network, etc., and, in particular, be subject to expert knowledge of the functioning of the system.

ABSTRACT. Climate resilience starts with adequate water management. Yet, managing its extremes—too much, too little, too murky—remains a pressing challenge. One way to mitigate and adapt to climate extremes is by enhancing the natural water retention capacity of landscapes, acting like a sponge. Adict Solutions is part of the Horizon Europe SpongeWorks project (funded by the European union and UKRI) whose purpose is to advance scientific knowledge and implement “sponge measures” at large scale in 3 European regions, thus enhancing soil and water health, and climate resilience. Nature Based Solutions (NBS) help regulate water storage and release but face challenges in large-scale implementation. One key barrier is the simultaneous evaluation of their benefits, particularly in addressing floods, droughts, and water quality.

We will show how the agro-hydrological model SWAT (Soil and Water Assessment Tool) may contribute to these objectives since it integrates: hydrology, water quality, agricultural practices, climate change, and simulates the effects of Nature-Based Solutions on hydrological regulation and pollution reduction, from field to watershed scale.

The SWAT model has been implemented on the Lèze watershed (France), one of the three demonstrators of the SpongeWorks project. The model accounts for crop rotations (representative of the area), water extractions including for irrigation, reservoir releases, effluent discharges from wastewater treatment plants. The Lèze river basin faces three major issues: i) floods and mudslides (intense storms causing frequent flooding); ii) summer droughts (with water levels dropping drastically, relying on upstream reservoirs for irrigation); iii) water quality degradation (low summer flows limiting pollutant dilution, leading to cyanobacterial blooms). The magnitude of overflowing floods will increase, and their frequency could double by 2040-2070 whereas summer low flows may decrease by 12–50%, worsening drought impacts. Several Nature Based Solutions have been evaluated on the Lèze watershed, and it will be shown that SWAT proves to be a valuable decision support tool for identifying the most effective sponge measures for climate resilience, by guiding the selection of locally adapted solutions for sustainable watershed management in an era of global changes.

ABSTRACT. The interlinkages between water and energy are crucial, as water sustains life while energy drives development. Achieving the UN Sustainable Development Goals requires reliable renewable energy, and India is no exception. With rapid urbanization, India's per capita electricity demand has surged from 500 kWh in 2000 to 1400 kWh in 2024. In response, installed power capacity has reached 462 GW, including 10.17% from large hydro and 1.1% from small hydro. Over 15 years, renewable energy capacity has grown from 58 GW to 209 GW, while large hydro has increased only slightly from 45 GW to 47 GW. India aims to expand total power generation capacity to 2042–3100 GW to meet a projected demand of 6273 TWh by 2047. This paper examines challenges in large hydro development, particularly environmental and resettlement concerns. A benefit-cost analysis highlights the importance of large hydro in achieving India's ambitious growth targets. Expanding large hydro from 47 GW to 75 GW will require $42 billion in capital expenditure and generate 75 billion kWh annually, demanding an additional 277.5 billion cubic meters of water. Declining per capita water availability necessitates recycling water with reversible turbines. The paper analyzes the economic feasibility of this approach and critiques how financial constraints hinder implementation. To achieve "Viksit Bharat" by 2047 and net-zero emissions by 2070, India must prioritize large hydro development. A broader developmental perspective is needed to ensure energy security, sustainability, and long-term national benefits.

ABSTRACT. Flash floods are relatively common in semi-arid and Mediterranean regions. They are expected to be a growing hazard, due to climate change, with increasing frequency and magnitude. Additionally, their incidence is expected to extend into new regions. The recent catastrophic flash floods in Valencia (Spain) in October 2024 are a clear example of this, with their very high toll in terms of lost lives and destruction of infrastructure and property. Despite this being a flash flood prone region, early warning and first response systems were deficient. Although the exact causes of the deficiencies are yet to be identified, we argue that part of the reason for the suboptimal performance of the early warning systems (EWS) is that they still do not rely on high resolution, physically-based 2D models. Until recently, the main barrier to the adoption of 2D shallow water solvers by EWS has been the large computational effort required, especially when attempting to simulate catchment-scale flash floods at very high resolution. In this work, we explore the use of the SERGHEI model as a proof-of-concept for simulations informing EWS, and study the flash flood event affecting the Rambla del Poyo catchment (500km2) in Valencia (Spain). We generate multiple models at different resolutions (up to 1m) based on DTM at the highest resolutions available from the Spanish geoportals. Infiltration parameters and hydraulic roughness are generated from openly available land cover, land use and soil data. We rely on radar precipitation and meteorological forecasts for rainfall input. To offset the large computational effort, we leverage the GPU-readiness and high scalability of SERGHEI. We run simulations on the JUWELS supercomputer, using different sets of computational resources as a function of resolution. Our results show a proof-of-concept of the application of highly-scalable, and highly-performing shallow water solvers for high resolution catchment scale flash flood simulation, with runtimes that allow sufficient lead times for early warning. Our proof-of-concept shows the maturity of the solver, and the opportunities that supercomputing resources enable for these types of problems. This approach promises to deliver more quantitative, more informative and more actionable information for flood risk managers.

ABSTRACT. Erosive phenomena play a critical role inecosystem preservation and land management, as they directly impact the fertile soil layer, with significant implications for agriculture. In the context of climate change, extreme rainfall events are expected to become more frequent, increasing risks to ecosystems and communities. Accurate simulation of these events can provide valuable information to regional authorities for developing mitigation strategies and reducing associated risks. In this work, we will be studying the risk of erosion in catchments due to rainfall, modeling the phenomena and using numerical methods to simulate Soil-Loss scenarios. High-performance computing (HPC) techniques are increasingly used to simulate large-scale catchments with acceptable computational speeds. The SERGHEI framework enables the simulation of entire catchments, providing an efficient solution to modeling complex hydro-erosive processes. The implemented model adds transport equations and bed exchange to the Shallow Water Equations in order to simulate erosion in catchments. Furthermore, the model leverages the increasing availability of satellite imagery, cartographic data, and detailed field measurements provided by experimental soil research teams , including soil type, infiltration rates, linear landscape elements, and soil depth. To validate the model results, we use soil-loss measurements associated with a 50-year period in an experimental catchment (based on the FRN Cs-137 method). The catchment is endorheic lacking permanent streams. To achieve this, we simulate the most extreme rainfall events over recent years and compute the sediment movement (soil loss). Given that computing simulations and data centers produce emissions comparable with aviation, this study also examines the energy efficiency of numerical simulations by comparing the computational performance of HPC approaches with non-parallelized methods. The results highlight the benefits of using HPC for large-scale rainfall-induced erosion modeling, not only in terms of computational speed but also in reducing the energy cost of simulations.

ABSTRACT. The catastrophic Kerala flood of 2018 imposed unprecedented damage throughout the region, highlighting the need for advanced numerical framework to comprehend its hydrodynamics and develop resilient flood management strategies. This study targets the south-central region of Kerala state in India, utilizing two hydraulic models, HEC-RAS and IROMS (Integrated River and Ocean Modeling System), to analyze the flood dynamics of the event thoroughly. The HEC-RAS is a freely downloadable two-dimensional (2D) model, whereas IROMS is indigenously developed model suit in the Department of Civil Engineering, Indian Institute of Technology Madras, India. High-resolution topographic and hydrological datasets were employed to simulate key characteristics of the flood, including its extent, depth, and duration. The local-inertial 2D model in IROMS, known for its computational efficiency in large-scale floodplain simulations, captured the inundation pattern across the entire basin. Meanwhile, HEC-RAS offered detailed insights into localized flood propagation. Observational data were incorporated to validate the models, ensuring the accuracy and reliability of the results. The analysis on flow through river-networks, canals, city and then into a large lake and ocean provides critical insights into the spatial and temporal evolution of the flood, identifying inundation hotspots and peak discharge pathways. These findings enhance our understanding of the 2018 flood event, demonstrating the complementary strengths of both modeling approaches in assessing complex flood dynamics. This work advances our knowledge of extreme flood events in tropical monsoon-dominated basins and offers actionable recommendations for risk mitigation and adaptation planning in the context of climate change. Based on the study's insights, the recommendations can be implemented practically to mitigate the impact of future flood events, empowering policymakers and practitioners to effect meaningful changes in their communities.

ABSTRACT. Water resources in Western France are increasingly under strain, with demand rising and conflicts over resource allocation multiplying. The Authion watershed is a 100 km² complex hydraulic system involving surface and groundwater interactions and withdrawals from external sources in the Loire River and the Rillé Lake. It is divided into 10 Management Units (MU), water withdrawals having exceeded supply in 2 of them in the last decade. The irrigation demand is high in summer, when surface water resources are scarce, forcing farmers to rely on groundwater pumping. The area has also become increasingly dependent on external withdrawals from the Loire River. Summer droughts affect local ecosystems. The project presented in this article aims to better understand the current and future water management challenges and to evaluate the impact of a water storage solution in these Management Units. Surface water would be stored in winter to be used in summer, reducing stress on groundwater. The watershed is represented coupling a 1D MIKE + hydraulic model including rivers and canals and a MIKE SHE hydrological model simulating aquifer recharge and 3D groundwater dynamics. Because of climate change, groundwater recharge and summer flow in the Loire River are expected to decrease, increasing the strain on the resource. Potential storage sites are identified and evaluated through a Multi-Criteria Analysis (MCA) including regulatory, land use, technical, and environmental criteria. Besides seasonal storage, the management scenario also includes wetland restoration projects and nature-based solutions aimed at infiltrating rainwater. Two storages options are conceived, both showing increased groundwater and surface water levels, with almost no droughts and a reduced dependence on withdrawals from the Loire River.

ABSTRACT. The Etablissements Publics Territoire de Bassin (EPTB) Seine Grands Lacs and Entente Oise Aisne are responsible for flood protection and prevention, and as such operate hydraulic flood management structures. In order to anticipate and optimise the management of these facilities and, more generally, crisis management in the event of flooding, they rely on real-time hydrological and hydraulic modelling tools based on :

-semi-distributed hydrological models fed by measurement and forecast data supplied by Météo France, -frugal hydraulic models built by combinig 1D/2D meshing and ponctual headloss modeling flood strucutre with their management logic.

These models are setup in a supervisory WebGIS application, in order to fullfill a special function in each area: -monitor forecasts at reference hydrometric stations and alert levels, -for Seine Grands Lacs, the ability to test different management scenarios for the regulation structures, and thus adapt their management to the available forecasts, -for the Entente Oise Aisne mapping of predicted flooded areas at 24, 48 and 72 hour intervals (including an increased scenario). These maps are then exploited by a dedicated crisis management tool.

These applications form part of an open and scalable global ecosystem developed around the hydra calculation engine, with the aim of bringing together real-time business applications (hydra live) and hydraulic modelling tools (hydra QGIS) used in study projects (sizing structures, mapping flooding areas, understanding flood generation and propagation phenomena, ....). Spatial databases in PostgreSQL/PostGIS are the spine of these applications. Real crossroads for data, the databases used by real-time applications can also be accessed by hydra from a desktop GIS via its QGIS plugin for use in ‘study’ mode. In this way, the webGIS platform is dedicated to the business application for which it has been developed, while the advanced model exploitation functions are directly accessible via hydra under QGIS. Interaction between the different applications is made easier by the ability to retrieve real-time scenarios for re-running simulations or analyzing simulations carried out in QGIS in greater detail. The modularity of the real-time applications means that they can be enhanced with new functionalities, both in terms of hydrological calculations (calling up other models upstream of the hydraulic models) and in terms of the interface.

ABSTRACT. In the context of climate change, there is an increasing evidence of extreme flooding events that affect urban areas. The FAVOUR project aims to enhance cities resilience by considering specific urban characteristics. In particular the project analyses affectations of street furniture (vehicles, waste containers, etc.) during flooding events. This project aims to develop a 3D numerical model to simulate the stability of waste containers in urban floods scenarios.This project was been build upon prior research conducted at the Flumen Institute at the Technical University of Catalunya (UPC), which already included mathematical and experimental studies. Consequently, a series of three-dimensional computational fluid dynamics (CFD) models will complement earlier findings. This will involve comparing experimental and numerical results while investigating the behaviour of waste containers after they lose stability. The numerical simulations will be conducted using the commercial software Flow-3D. Acknowledgments This work was supported by the Spanish National Agency of Research of the Ministry of Science, Innovation and Universities with the project PID2019-110965RB-I00 “Analysis of flood hazard considering vehicles and other floating elements” acronym FAVOUR.

ABSTRACT. The accumulation of cohesive fine sediments in reservoirs leads to the formation of slopes that are prone to instability and potential collapse. A progressive increase in deposited mass can destabilize sediment slopes in the reservoir, leading to possible slides and ultimately to locally elevated sediment transport rates. Dredging operations are carried out to lower the reservoir bed and stabilize the slopes. However, these operations are expensive and require careful planning logistics. Therefore, the goal of this study is to propose a practical tool to be able to predict and quantify the potential collapse of slopes. Our analysis focuses on the modeling and application of a geomechanical analysis aimed at quantifying the risk of circular slip failure of a sedimentary slope. The calculation method used to solve the circular failure problem is based on the Bishop formula via the slice method. To ensure an accurate representation of the physical processes, the model accounts for the cohesive mechanical behavior of the sediments, which is one of the key parameters linking hydrodynamics and geomechanical processes in reservoirs.

ABSTRACT. The Somme River, the Authie River and the Bresle River flow into the English Channel forming adjacent estuaries in Northern France. This coastal zone is exposed to both coastal submersion and river flooding due to the presence of large areas of lowlands far behind the coastline. During the implementation of the Flood Prevention and Action Programme, a numerical model which covers approximately 300 km2 of the floodplain has been developed to simulate scenarios of coastal submersion at a regional scale. Modelled water levels are calculated under the influence of the combined effects of atmospheric contribution to the mean sea level or storm surge, wave setup, tidal propagation, wind setup and river flow during extreme hydrological events. The system of coastal protection (140 km in length) is included to represent the flooding due to wave overtopping, overflow and formation of breach depending on the state of the structure. A system of 115 hydraulic structures composed of culverts, sluice gates, bridges and the sea lock of the Somme canal is also integrated as it contributes to water propagation and retention in the lowland area. At first, the model is calibrated in nominal conditions using observed tidal water levels in spring tide et neap tide at five locations in the Somme Bay and 3 locations in the Authie Bay. Then the model is validated in extreme conditions using observations from the Viviane storm of February 1990 and the Xynthia storm of February 2010. The modelled water levels agree with the observed water height of 30 cm above the quay at Saint-Valery-sur-Somme during the storms.

ABSTRACT. The design of an integrated fish migration facility within the powerhouse structure of a hydropower plant requires optimization to meet both technical project requirements and energy production goals, while also ensuring a safe and functional migration path for fish species native to the Northern Highlands of the Lower Mekong Basin. Given the complexity of the problem and the limited applicability of other methods, a computational fluid dynamics (CFD) approach was identified as the most viable solution to address these multifaceted challenges. A series of three-dimensional CFD simulations were conducted using advanced algorithms to accurately track fluid interfaces, enabling precise modelling of surface shapes and fluid interactions. This paper discusses the challenges of balancing project requirements using CFD tools and presents insights into achieving an optimal design that harmonizes hydropower generation with ecological conservation.

ABSTRACT. The analysis of the results of the measurement of local losses of hydraulic energy in a physical device installed in the laboratory, in which sargassum is used as a flow filter medium, is presented. As part of a project that considers giving use to the material that reaches the Mexican Caribbean coast year after year, it is necessary to evaluate the local energy losses that occur in a filter that is considered to have a load of the order of 3 column meters. . water. The evaluations are presented considering the Reynolds number, which considers the relationship between inertial forces and viscous forces present in a fluid. This relates the density, viscosity, velocity and typical dimension of a flow in a dimensionless expression, which is involved in numerous fluid dynamics problems. Electronic pressure cells are used in the device to have greater precision in the evaluations of hydraulic behavior.

ABSTRACT. The researchers were interested in studying the impact of the bank effect on the hydrodynamic properties of the ship and thus mitigating the negative effects that it could cause on navigation in canals and waterways, in order to ensure the safety of navigation for ships that cross these paths and which accordingly are exposed to the forces resulting from this effect, whose intensity varies depending on the type of the ship and the shape of its hull. Navigation in canals and waterways is of great importance because of its role in shortening distances and thus reducing transportation time and costs, but it carries many risks to the safety of ships, including crew and cargo, which has led to strict restrictions and instructions for maneuvering in these lanes. This research aims to analyze the bank effect, determine its relationship with the ship’s speed, propose the optimal distance for the ship from the bank, identify the factors affecting this phenomenon, and determine how the intensity of its impact varies on two ships of different types. To achieve this, five main hypotheses were formulated. This research relied on the use of CFD (computational fluid dynamics) to collect data by studying two models of ships, one of which is a JBC bulk carrier and the other equipped with an X-Bow. The study and data analysis were carried out using the ANSYS Fluent program and based on the principle of reflection of flow. A number of conclusions have been reached by this research, the most important of which are: There is a linear relationship between the bank effect and the ship's speed and its distance from the bank. The effect of the bank is primarily determined by the ship's draft to water depth ratio, as the latter determines the direction of the transverse force acting on the ship. Under certain conditions, the sloped bank reduces the forces and moments of the bank effect. The X-bow has a great impact on reducing the bank effect.

ABSTRACT. Earth embankment failures are most often caused by overtopping. This type of event is expected to become more frequent in the near future due to climate change. When overtopped, the earthen dike suffers from erosion caused by high velocity flows. Eventually, a breach is created that first deepens and then widens as the flow continues to enter the protected area. The dynamic of the breach opening plays an important role in the way a given area is flooded after a dike failure. This type of event is usually simulated using the shallow-water equations combined with the Exner equation and a bank failure operator to reproduce the progressive erosion of the embankment and the growth of the breached area. The parameters of the bank failure operator are in general tuned to better reproduce the effects of apparent cohesion that causes the dike material to hold almost vertically in the breached area. However, such a bank failure operator is a simplification that overlooks the precise effects of pore pressure on soil shear strength. The evolution of the degree of saturation is classically neglected, the past hydrological history is not considered in the initial strength of a given embankment. In addition, embankments do not usually consist entirely of non-cohesive uniform material, and the shear resistance of the soil depends not only on the angle of friction but also on cohesion. Complex behaviors can be observed in the geomaterials, such as softening of the shear strength, which cannot be accounted for with a simplified bank failure operator. In this work, we present a model that attempts to encompass a variety of complex phenomena occurring throughout a dike failure event. The main idea lies in the combination of a coupled surface-subsurface flow solver with a large displacement geomechanical model. The coupled surface-subsurface flows solver, part of the Watlab environment (https://sites.uclouvain.be/hydraulics-group/watlab/index.html), provides information on the evolution of the saturation degree in the dike, water depths and velocities during a dike overtopping event. The surface flow is evaluated using the shallow-water equations, while the 2D Richards equation is simultaneously solved for the flow through the dike. The key novelty in this work is the combination of this latter solver with a geomechanical solver to better simulate the breaching process. The Particle Finite Element Method (PFEM) is employed to reproduce the large displacements that can occur during an earthen dike failure. Complex constitutive laws can be employed and the saturation degree variation in time is accounted for. Several numerical tests were conducted to demonstrate the validity of the proposed framework. The effects of initial saturation degree and physical parameters on the dike failure process are studied, highlighting their importance in the evaluation of dike resistance to overtopping flows.

ABSTRACT. Extreme flood events occurring in the last years demonstrated the need for hydraulic models to evolve, to adequately represent also the residual risks which may exacerbate the effects of flooding. Large wood transport, for example, was included in the last years in some 2D hydrodynamic models, to simulate the afflux at bridge piers its consequences on flooded areas. Such methodological evolution requires not only to develop adequate approaches for the considered phenomenon, but also to adopt the most suitable solutions to limit the computational burden. This is even more necessary if the focus of hydraulic modelling is shifted from single river reaches to the watershed-scale. To this aim, this contribution presents the results of the coupling of a 2D HPC Hydrodynamic Model with a model that simulates the displacement of large wood due to hydrodynamic forces. Such a combined model can help in evaluating which areas of the watershed generate a greater mobilization of large wood, verifying the possibility of recruitment, entrainment, transport to and within the main channel, also depending on the triggering rain event. The 2D HPC Hydrodynamic Model is a Rain On Grid model, based on Finite Volume (FV) method, that can provide the hydrodynamic response to either river flooding or rain event at watershed-scale. It is written in Fortran, CPU parallelized, based on MPI approach. The wood transport model is a Discrete Element Model (DEM) that considers each log as a single cylinder. The DEM is part of ORSA2D_WT, in which it is coupled with a 2D-SWEs model, not parallelized and written in Fortran. It computes the hydrodynamic forces in four points for each log, making it burdensome in terms of computational costs, especially if several logs are present. The parallelization of the DEM code has required some adjustments. For example, to make it suitable for MPI approach, the use of information from mesh cells not contiguous to those of interest for each point on the log was removed. The model is tested on flume experiments for single log transport, to evaluate how the changes made to the original DEM have impacted the computation of the log trajectory. The results show that the parallelized code can replicate the experimental trajectory with an accuracy comparable to ORSA2D_WT. Minor differences appear in those areas where the flow gradients are stronger. The application to real-scale tests will allow for further comparison between the two codes, to check the results accuracy in a more realistic domain and to assess the computation time savings.

ABSTRACT. The transport of material by water flows, particularly in rivers, plays a crucial role in natural and environmental disasters, from the movement of vehicles and large debris during flash floods to the widespread distribution of pollutants such as macro- and microplastics. Moreover, recent studies reveal alarming concentrations of microplastics in freshwater systems and even domestic water supplies, having significant public health risks due to potential ingestion and associated health effects on humans and animals. For this reason, the development of tools that can provide more information about the behavior of these pollutants is essential to identify solutions to reduce the plastic concentration in fluvial ecosystems. However, addressing these hazards requires advanced mathematical modeling and computational solutions capable of capturing the complexity of transport dynamics and environmental interactions. This work presents the development and integration of a Lagrangian model to study the temporal and spatial evolution of plastics transport in shallow water flows. While the flow is modelled by an Eulerian hydrodynamic model based on the Shallow Water Equations (SWE), the plastic particles are modeled as discrete entities transported by hydrodynamic forces computed using physical coefficients in a kinematic approach. These approaches provide a robust basis for computing flow evolution while offering detailed insights into material transport mechanisms. The newly developed Lagrangian Particle Tracking (LPT) module is integrated into the SERGHEI (Simulation Environment for Geomorphology, Hydrodynamics, and Ecohydrology in Integrated form) model. SERGHEI enables comprehensive investigations into plastic dynamics by accounting for key processes such as advection and dispersion, along with plastic-specific processes including deposition, flotation, degradation, and biofouling for microplastics; and drag forces, degradation or bedload transport for macroplastics. The model provides valuable information about the plastics behavior in rivers, aiding in the identification of potential mitigation strategies. The use of a 2D hydrodynamic model reduces the computational cost compared to 3D hydrodynamic approaches. This efficiency is further improved through the implementation of both plastic transport and hydrodynamic models in a high-performance computing (HPC) framework, enabling realistic simulations of complex scenarios. Additionally, the integration of plastic-specific processes ensures realistic time evolution of particle positions. To validate the model, analytical and experimental results are simulated, while realistic cases are studied to analyze whether the model is able to provide useful information in more complex problems. Future work will focus on extending the model to simulate larger debris, such as vehicles, waste containers, or boulders, expanding its applicability to environmental risk assessment.

ABSTRACT. Sodium polyacrylate is extensively utilized across various industries, necessitating an understanding of its environmental fate upon release into aquatic systems. This study examines the behavior of sodium polyacrylate over a 7.4 km long reach of the Seine River through laboratory experiments, analytical solutions, and numerical modeling. Laboratory tests were conducted to determine key input parameters for the numerical model, with a focus on the adsorption of sodium polyacrylate onto suspended sediments. Simplified one-dimensional analytical solutions, derived under certain assumptions, closely approximated the comprehensive model, which comprises a system of three coupled partial differential equations for the dissolved and adsorbed polymer in both the water column and riverbed. The comparison revealed excellent concordance between the numerical and analytical solutions in a simplified flume-like setup. In the river reach, the analytical solution showed good agreement near the discharge point, although transport exhibited significant two-dimensional characteristics further downstream. The study also explores the long-term accumulation of sodium polyacrylate in the riverbed, highlighting the critical role of flow and sediment dynamics in its transport.

ABSTRACT. Membrane based separation processes such as membrane distillation (MD) systems hold promise in water and wastewater treatment technologies due to its feasibility of using waste heat and low-grade energies. However, fouling and wetting phenomena often impedes the efficiency of the membrane-based separation processes. One of the common techniques to combat membrane fouling/wetting is using feed spacers to enhance the turbulence of the flow in feed and permeate channels near the membrane surface. In this study, which is part of the iWAYS project funded by the European Union’s program Horizon 2020 (grant agreement: 958274), we investigate the impact of the addition of spacers to the MD system using physical and computational fluid dynamics (CFD) models. The goal is to test and optimize the design of feed spacers. A specific configuration of spacers was designed and (i) 3D printed to be implemented in the lab-scale membrane module, (ii) included in the CFD numerical experiment using the code OpenFOAM. The membrane module has an active area of 450 cm2, an overall length of 30 cm, width of 15 cm and height of 5 mm for each channel. In each channel two layers of spacers with cylindrical geometry and a diameter of 2.5 mm are placed on top of each other with a 45-degree angle to the incoming feed and permeate flow. Then a set of experiments were performed for saline water as a feed solution with and without the presence of spacers in the membrane module to assess the impact of spacers on the permeate flux. Finaly, the obtained results from CFD study were compared with the experimental results to validate the model. The results confirm that the addition of spacers to the membrane module improves the turbulence along the membrane surface which will enhance the membrane performance regarding fouling/wetting leading to a more stable operation.

ABSTRACT. Reservoirs can play a significant role in water resources management to mitigate impacts of water-related disasters such as floods or droughts. More integrated operation of multi-purpose reservoirs, which usually have storage capacities for both flood control and water use purposes, is especially important for more effective management of those water-related disasters. Prior release operation, in which water storage in the reservoir is released just in advance of flood arrival considering real-time flood predictions to secure increased empty storage volume for flood control, can be considered as an effective way to derive more capability of a multi-purpose reservoir for both flood management and water supply. However, it has not yet been established to determine how much water storage needs to be recovered after the flood control from the view point of drought management for the following weeks or months. Considering the circumstances described above, a method for adaptive operation of multi-purpose reservoirs for both of flood and drought management is developed in this study. Short-term (for the coming 39 hours) and medium-term (for the coming 264 hours) ensemble forecast of precipitation provided by Japan Meteorological Agency (JMA) are considered to estimate the possibility of flood occurrence in order to determine in real time how much of prior release needs to be conducted. At the same time, JMA’s long-term ensemble forecast of precipitation (for up to the coming three months) is considered to carry out real-time drought risk assessment for the following months to decide how much water storage needs to be recovered in the reservoir after flood control. These decisions for prior release and water recovery are made based on ensemble prediction of reservoir inflow for the coming months which are estimated from the ensemble forecasts of precipitation described above by use of the Hydro-BEAM (Hydrological River Basin Environment Assessment Model), a distributed rainfall-runoff model. The optimized timing and amount of prior release from the reservoir are estimated from the ensemble inflow prediction by the simulation of reservoir operation, maximizing the effect of prior release operation on flood control while minimizing long-term risk in water use by securing water recovery after the flood event based on the accuracy analysis of past ensemble forecast data. The proposed method was applied to the reservoir system in a river basin in Japan, demonstrating its usefulness for mitigating both of flood and drought risks through effective prior release operation by the reservoirs.

ABSTRACT. The planning and management of multiple dams within a river basin are critically important for the effective use of water resources and energy generation. Under the growing pressures of climate change, Türkiye, classified as a water-stressed region, faces significant challenges in balancing water availability with population growth, environmental sustainability, and energy demands. The sustainable operation of multiple dams is a vital step toward ensuring the efficient utilisation of water resources for the country's future, while addressing environmental considerations and climate resilience. The Euphrates-Tigris Basin spans a semi-arid area of 762,000 km², covering six countries: Türkiye, Iran, Iraq, Jordan, Syria, and Saudi Arabia. The two major rivers in the basin, the Tigris and Euphrates, originate from the mountains in eastern Türkiye. These rivers are primarily fed by snowmelt stored during the winter, while during the dry summer months, they rely heavily on groundwater, making the region particularly vulnerable to climate change. Approximately 60 million people depend on these rivers for irrigation, energy production, and other water-related needs. This study focuses on the portion of the Euphrates-Tigris Basin located within Türkiye, covering an area of 18,500 km² (Esit et al., 2023; Rateb et al., 2021). Türkiye has built 19 hydropower plants and 22 dams in this region over recent decades to store water for irrigation, energy generation, and flood control (SAPRDA, 2009). However, the basin faces increasing challenges due to climate change, including reduced precipitation, rising temperatures, and greater variability in seasonal water availability. These changes exacerbate flood risks during extreme rainfall events while also intensifying drought conditions in dry seasons. The stored water in dams is essential for hydropower production, contributing to Türkiye’s renewable energy targets, yet evaporation losses and reduced inflows pose threats to long-term sustainability. Addressing these interconnected issues is critical for maintaining water security, energy production, and ecosystem stability in the region. Currently, dams in Türkiye are operated individually, often without coordination or consideration for downstream interdependencies, population growth, or the effects of a changing climate. Using the Euphrates- Tigris basin as a case study, this study seeks to explore the impacts of multiple dam operations in water management. The research will analyse the implications of uncoordinated dam operations on water allocation, seasonal water availability, and hydropower production. Furthermore, it will assess the potential benefits of integrated dam management strategies for improving water resource efficiency. By identifying key challenges and opportunities, this study aims to contribute to the sustainable management of the Euphrates-Tigris Basin in the face of evolving climatic and socio-economic pressures.

ABSTRACT. Natural dams are one of the most hazardous phenomena in mountainous regions worldwide. Predicting and simulating their failure processes, the induced floods, and the resulting morphological changes are critical for effective hazard mitigation. Over the past decades, several numerical models have been developed to simulate these events, primarily considering uniform sediments. However, natural dams and their local riverbeds typically consist of non-uniform, multi-layered sediments, which significantly influence sediment erosion, transport, and deposition. These processes, in turn, affect the dam failure mechanisms, flood intensity, and downstream morphological evolution. Therefore, incorporating sediment heterogeneity into numerical models is essential for accurately simulating these events.

Here we present a 2D numerical model that integrates non-uniform and multi-layered sediments, based on the shallow water equations, the modified Exner equation, and Hirano’s active layer equation. The model is implemented using a weakly coupled finite volume scheme on an unstructured triangular mesh. The model’s performance is evaluated through three cases: a synthetic test involving sediment feeding under aggradation and degradation conditions, and two experimental tests: a dam-break flow test over a movable bed and a dam breaching test. Additionally, the model is applied to simulate the failure of a real natural dam, with an emphasis on analyzing the effects of inflow discharge and material composition on the dam failure process.

The results demonstrate that the proposed model reliably reaches equilibrium states under both uniform and non-uniform sediment feeding conditions, whether aggradation or degradation is present. Regarding the first test about dam-break flow, the model can well simulate the flow and morphological changes, as well as the surface coarsening and fining phenomena related to multi-layered and non-uniform sediments. The numerical results for the failure process and associated morphological changes show good agreement with experimental data, though the model tends to underestimate the peak discharge and bed elevation when finer materials are involved. In the real-case simulation, the model effectively replicates the dam failure process and related parameter variations, closely aligning with observed field data. Notably, it captures the evolution of surface grain size distribution during the failure. While an increased inflow rate upstream has a limited impact on downstream flood magnitude, lower inflow rates provide more time for evacuation and engineering interventions. The findings also highlight that the distinction between uniform and non-uniform sediments is negligible under high flow rates, as these are capable of eroding all sediment sizes of the initial dam material.

Overall, the proposed model provides a robust tool for analysing natural dam failures and their downstream impacts, offering valuable insights for hazard management in mountainous regions worldwide.

ABSTRACT. Understanding sediment dynamics in dam reservoirs is a major challenge, due to the complex interactions between various natural factors and management practices. For long dam reservoirs, the application of 1D modeling can be of real interest (Guertault et al., 2016). In this study, we present an analysis of fine sediment dynamics in a dam reservoir located in West Africa: the Lobo dam reservoir, in central-western Côte d’Ivoire. This study is based on a 1D modeling using Mage and AdisTS solvers, developed at INRAE Lyon. One-dimensional models are generally used to study hydrodynamic parameters in large-scale contexts or over long-term periods. Compared with two- or three-dimensional numerical models, they offer the advantage of requiring less input data, which is often an issue, especially in developing countries. Our first results demonstrate the models' ability to accurately simulate water levels, transported sediment fluxes, and morphological changes within the uncertainty ranges of experimental measurements. They were also useful for discussing data quality and improving the understanding of the system In addition, the model has been used to evaluate various reservoir management protocols, with the aim of establishing operating rules to limit reservoir siltation and reduce long-term environmental impacts. Finally, the results provide a sound basis for implementing sustainable management strategies to preserve the reservoir's ecological and functional equilibrium.

Guertault, L., Camenen B., Peteuil, C., & Paquier, A. (2018). A one-dimensional process-based approach to study reservoir sediment dynamics during management operations. Earth Surface Processes & Landforms, 43(2): 373-386. doi: 10.1002/esp.4249

ABSTRACT. Analyzing the impact of climate variables on operational planning processes is essential for the effective implementation of sustainable power systems. Climate and energy are closely intertwined, with each significantly influencing the other in complex ways. Energy production contributes to climate change through greenhouse gas emissions from fossil fuels such as coal, oil, and natural gas, which worsen global warming. At the same time, changing climate patterns affect energy resources by altering weather conditions and increasing the frequency of disruptive weather events, which can impact the performance and reliability of various energy generation methods.

Hydropower plants are a crucial component in the transition to renewable energy sources. They require a thorough understanding of how climatic factors influence their performance, especially for the run-of-river technology. Indeed, unlike traditional hydropower plants, run-of-river systems do not require large reservoirs, which reduces their environmental impact and minimizes the regulatory constraints associated with large reservoir plants. However, this also presents challenges, primarily the dependence of electricity production on climate variables, as water flow is directly influenced by precipitation, temperature, and solar radiation, making it difficult to control the output apart from shutting down the plants.

Accurate forecasting for run-of-river hydropower is crucial for optimizing energy production, effectively managing resources, and integrating with other renewable sources. To achieve this, our work tackles the complex issue of converting climate data into the capacity factor for run-of-river hydropower production. The challenge of this task arises from the necessity of capturing the intricate relationship between water availability and electricity generation. This complexity increases when considering large, interconnected regions, as the European countries considered in this study.

We aim to identify the best data-driven techniques for providing power system operators with accurate forecasts and operational guidelines. Hence, we propose and evaluate machine and deep-learning models for forecasting the run-of-river hydropower capacity factor based on climate variables such as temperature and precipitation on the European scale.

We compare the accuracy of models based on an ensemble of decision trees with deep learning models designed to take temporal relations into consideration. We also improve forecasting abilities by employing specialized methods that merge seasonal decomposition with some classes of recurrent neural networks, such as Long-Short-Term Memory.

Finally, we discuss the practical applicability of these learning models for medium-term forecasts and show that some very context-specific but influential events are hard to capture.

ABSTRACT. CNR is France’s leading producer of 100% renewable energy, operating and managing 19 hydroelectric power plants along the Rhône River. In the event of a sudden turbine shutdown due to mechanical or electrical failure, the abrupt disruption of flow at the power plant causes water to accumulate in the reservoir and generates waves – primary and secondary ones – that propagate upstream for several kilometers, both phenomena leading to a rise in water level unless the flow is promptly restored. To mitigate this effect, the standard response is to quickly operate the dam gates with a controlled opening gradient. If this response is delayed or insufficient, the uncontrolled rise in water level may lead to dike overtopping, potentially compromising its integrity and posing risks to surrounding areas. To address these risks, emergency structures must be operated following a strict protocol, leading CNR to investigate the maximum permissible water levels in the reservoir and the most effective way to operate these structures to ensure adequate discharge while protecting the facility. CNR chose to focus on the primary wave captured thanks to an adapted 1D modeling involving refined time step and spatial resolution to better capture transient hydraulic phenomena. These models are used for each development scheme to: - Simulate various turbine shutdown scenarios, including full or partial simultaneous turbine shutdown at different initial water levels. - Determine the maximum acceptable water levels in the reservoir and assess the hydrograph at the dam based on these critical water levels before dam intervention is required. - Define the optimal opening sequence for the dam to generate the assessed hydrograph, to ensure the protection of the dikes while preventing excessive scouring at the dam’s base. These models reasonably represent primary wave dynamics and reservoir filling following such a failure, providing reliable insights for an operational context. Their application, combined with internally developed automated tools such as Hydrogen and Chronogen, allows quick scenario testing and improve the efficiency of the analysis of multiple situations within a short computation time, making them well suited for operational challenges. Once numerical simulations are completed, an in-situ test is usually conducted to qualify and validate the 1D model, and to refine the study’s parameters if required. The 1D models and their automated tools assist in identifying the most relevant test scenario – such as the number of turbines to shut down, the initial water level, and the delay before dam intervention – based on hydrological conditions and equipment availability on the test day. This paper presents the hydraulic methodology applied from failure modeling to in-situ testing, illustrating how 1D models, along with Hydrogen and Chronogen, are used to evaluate various critical shutdown scenarios an enhance emergency response strategies.

ABSTRACT. Sukkur barrage is located in the Sindh Province of Pakistan. Commissioned in 1932, it was the first barrage constructed on the Indus River. Its primary purpose is to supply an extensive irrigation network through off-taking canals on both banks of the river. The barrage consists of 66 gated spans across the Indus with a total length about 1.4 km. Sukkur barrage faces several challenges, such as reduced flood capacity and severe sediment deposition in the right bank canals. The Government of Sindh has conducted several studies over the past decades and would like to ensure now that the foreseen solutions meet the requirements of flood capacity, sediment exclusion and flushing efficiency (without unacceptable risk of scouring). To this end, the Government of Sindh entrusted ARTELIA with consultancy services for Physical and Numerical Hydraulic Model Studies of Sukkur barrage. In response to the request, 3 models were developed: a physical model (hydraulics and movable bed), a 2D numerical model (hydraulics and movable bed), and a 3D numerical model (hydraulics). The physical model is constructed with a non-distorted geometrical scale λ = 1:75 and is based on the Froude similitude for hydraulics. This scale is selected as it allows for the reproduction of secondary currents, particularly those in the approach channel in front of a submerged weir, which governs sediment exclusion. Given the scale of the model and the small size of the sand in the Indus River, a movable bed made of light material is required to simulate suspended load. The 2D numerical model is developed to provide an estimation of the morphological evolution of the river bed during an annual hydrograph. It also provides a first assessment of flow conveyance and hydrodynamic analysis around the structures, but these objectives are more addressed with the 3D numerical model. Our presentation will outline the methodology and demonstrate how these different models complement each other. The numerical models will be more specifically described, and their results presented, particularly for the study of new layouts.

ABSTRACT. The digitalization of the urban water cycle has emerged as a key pillar for the sustainability and efficiency of modern cities. In this context, the city of Valencia has positioned itself as a benchmark in the implementation of smart technologies for water management, managing to reduce losses by more than 30%, increasing hydraulic efficiency by more than 86% and saving more than one billion liters annually. These advancements have been made possible through the integration of a digital twin and a holistic water management platform, two technologies that have revolutionized the operational efficiency of the city's water system.

The digital twin is a dynamic virtual replica of the Valencia's supply system, enabling real-time modeling, simulation, and optimization of the behaviour of the network in real time. By collecting data from IoT sensors, smart meters, and predictive models based on artificial intelligence, this system can anticipate leaks, optimize pressure levels, and manage water resources with unprecedented precision. Its implementation has transformed operational decision-making, reducing response times to incidents and minimizing water waste.

In parallel, the holistic management platform centralizes and analyzes vast amounts of data, providing a unified environment for real-time monitoring and comprehensive water cycle management. Through advanced analytics and machine learning tools, this platform facilitates early anomaly detection, efficient infrastructure planning, and energy optimization in water distribution and treatment. Moreover, its interoperability with other urban systems strengthens the smart city approach, enabling integrated management alongside environmental policies and other municipal services.

The results achieved in Valencia demonstrate that water digitalization not only enhances operational efficiency but also contributes to urban sustainability and resilience against climate change and water scarcity. The combination of digital twins and advanced management platforms represents a paradigm shift in the water sector, laying the groundwork for the evolution toward more intelligent, autonomous, and sustainable systems. This presentation will analyse the impact of these technologies, the challenges overcome and the lessons learned after more than 20 years of innovation in water management

ABSTRACT. Insufficient hydrodynamics and resulting water quality deterioration are key drivers of environmental evolution in plain river networks. In recent years, population growth and rapid urban industrialization have significantly increased pollution pressure on the river networks in the Hangjiahu region. Structural modifications to rivers and artificial regulation have further impacted water quality and the regional water environment. Yaozhuang in Jiashan County, Zhejiang Province, is a typical plain river network area characterized by flat terrain, sluggish flow, and severe sediment accumulation. The construction of urban flood control and drainage infrastructure, along with riverbank hardening, has reduced river connectivity and exacerbated water quality degradation. Understanding the coupling processes of hydrodynamics and water quality, as well as predicting dynamic changes under different regulation scenarios, is crucial for effective river and lake management. This study focuses on the Yaozhuang river network, utilizing field surveys and on-site investigations to collect long-term hydrological and water quality data. A two-dimensional hydrodynamic-water quality coupling model was developed using Mike21, based on measured topography, water levels, flow rates, and five key water quality indicators (DO, NH3-N, TN, TP, and COD). The model was validated against observed data and used to simulate different regulation scenarios, exploring the influence of natural and anthropogenic factors on water environment evolution. Key findings include: 1. The Mike21-based hydrodynamic-water quality coupling model for Yaozhuang achieved high accuracy, with average simulation errors for water level, DO, NH3-N, TN, TP, and COD at 4.68%, 6.57%, 8.34%, 4.82%, 7.12%, and 4.67%, respectively. The model effectively describes water flow, substance transport, and chemical processes, providing a reliable tool for evaluating hydrodynamic regulation and water quality management. 2. Simulation results indicate that wind speed is a critical natural driver affecting water quality, particularly DO. The most significant improvement in water quality occurs when the prevailing wind is from the south with a surface wind speed of approximately 2 m/s. Rainfall primarily influences DO and TN, with poorer water quality observed during flood seasons, whereas stable conditions are maintained in non-flood and normal flow periods. Anthropogenic regulation has a substantial impact on water dynamics and renewal frequency. When the typical river water renewal cycle extends from 3 to 7 days, significant water quality improvements are observed. Additionally, increasing water supplementation in dense, interconnected river networks enhances water quality in blind channels and stagnant sections.

ABSTRACT. Switzerland, often referred to as the "water tower of Europe", provides essential water resources to its neighboring countries through its river systems. While the main Alpine region is not currently experiencing significant water scarcity at the scale of large river basins, future challenges are expected due to glacier retreat, increasing water demand across various economic sectors, and growing pressure from neighboring countries, compounded by the effects of climate change. Moreover, water scarcity is already occurring locally during dry years. In this context, it is crucial to assess water resources for multiple uses, considering both synergies and conflicts between different demands. A key aspect of this assessment is the evaluation of available water resources, both the portion allocable to specific uses and the total availability for all purposes. However, direct monitoring of these resources remains limited, particularly at the scale of Alpine users such as villages, high-altitude reservoirs, ski resorts, and small mountain catchments. To address this gap, a multi-method approach was applied, integrating sparse water usage data from municipalities, climate data, and hydrological modeling to evaluate current water resources in the Sionne catchment (Valais). This approach enabled the estimation of monthly water availability, including groundwater and surface water, for a specific year, along with associated uncertainties. Additionally, regional climate and hydro-climatic scenarios, combined with small-scale hydrological modeling of ungauged catchments, were used to project future water resources. The adopted methodology provides insights into water availability for multiple uses, identifying where and for what purposes water resources can be utilized. These findings highlight future challenges in water resource assessment and support informed decision-making in water management.

ABSTRACT. The Vésubie Basin is one of the sub-catchment of the Var Catchment that covers almost the entire region of the Nice Cote d'Azur metropolitan area. The area of the Nice Côte d'Azur metropolis benefits from multiple water resources - surface and underground - which provides drinking water to the inhabitants, as well as for agricultural, industrial or leisure activities. It also contributes to supplying watercourses and rivers, or the soil. In 2022-2023, Nice experience extreme drought where water resource availability was heavily impacted. Also, climate change projections on the South East region of France indicate the possible increase in temperatures and dry periods. These developments could potentially affect the hydrological conditions of the watersheds and impact the territory's water resources quantitatively or qualitatively. To assess the possible impact of climate change on the water resource in the Vésubie watershed, reference climate scenarios produced by the IPCC and regionalized by the DRIAS were used as input data within a distributed hydrological model. The distributed hydrological model Mike SHE, which is developed as part of the AquaVar modeling project, is used to examine the evolution of the watershed's hydrology under various climate scenarios (RCP 8.5 and 2.6 in particular). This allows us to translate long-term changes in temperature and rainfall into the form of hydrological indicators specific to the Vésubie watershed. These indicators allow us to better perceive the changes induced from a water resource point of view in order to anticipate and reflect on the different strategies that could be implemented to ensure the future water supply of the municipalities in the NCA Metropolis territory.

ABSTRACT. Numerous dams worldwide are around 50 years old or older, and therefore, their useful life is nearing its end or has already been exceeded. The behavior of inflow volumes to the reservoirs experiences seasonal variability, which has also been modified due to changes in land type and use in upstream areas. Changes in precipitation events also contribute to variations in runoff, making it necessary to periodically review and update the design floods of the reservoirs, as well as to review the operation of their spillways. In this study, the method of the Institute of Engineering, UNAM, based on an analysis of maximum average flows for different durations, was applied to the updated data of the El Infiernillo dam, Michoacán, Mexico, to calculate its design floods for two seasons of the year and different return periods. Additionally, these floods were routed through the reservoir to evaluate different operation policies for the spillway to review safe operating conditions, as the tunnels of this reservoir are often subject to cavitation phenomena when high flows occur. Open-source software of simple application was used in most of the processes. It was found that for a flood with a return period close to 20 years, these tunnels suffered severe damage in the flood that occurred in 2024, with total discharges of around 6000 m³/s. However, if extractions are proposed well before the peak flow occurs, these discharge effects can be avoided or mitigated. On the other hand, the inflow hydrographs to the La Villita dam, located downstream, are subject to discharges from the El Infiernillo dam spillway, making their review necessary to ensure the safety and proper operation of the last dam in the Balsas River hydroelectric system.We thank the PAPIIT project IN102625.

ABSTRACT. The management of water resources is becoming increasingly complex due to the diverse factors affecting hydrological systems, leading to communication challenges among decision-makers and stakeholders. To address these issues, system dynamics models have been widely utilized as effective tools for representing complex water resource systems. These models simplify the representation of real-world complex water systems over time, thereby supporting deliberations and the establishment of alternatives. This study aims to build a shared vision model using a system dynamics approach and utilize it to assess drought coping capacity of the Yeong-Seom Basin from both supply and demand perspectives. Drought coping capacity was evaluated as a communication indicator during scenario simulations to establish alternatives, enabling the analysis of drought patterns affecting the target system. The assessment comprehensively considered the frequency, duration, and magnitude of droughts. The results indicated that dams supplying water to multiple mid-scale basins exhibited broader impacts under the same scenario, not only on the regions where the dams were located but also on a wider range of areas. Furthermore, the shared vision model proved useful in evaluating the effectiveness of existing measures and proposing new strategies for securing drought coping capacity. Through this study, it was demonstrated that establishing water resource allocation plans using a shared vision model can serve as an approach to resolving conflicts between decision-makers and stakeholders.

ABSTRACT. Le SMAVD a initié la démarche C3PO pour évaluer les effets du changement climatique sur la ressource en eau de la Durance, les usages socio-économiques et les milieux. Le programme national Explore2 fournit les débits futurs à l’échelle de la France sur plus de 4000 points. L’intégration de ces données à l’outil C3PO a permis de réaliser une simulation confrontant le territoire d’aujourd’hui (usages et règles de gestion actuels) avec la ressource en eau future projetée par Explore2. L’outil d’aide à la décision C3PO accompagne les travaux de la Commission Locale de l’Eau du SAGE Durance : il permet de mettre en débat le partage futur de l’eau de la Durance et de tester des leviers d’adaptation aux effets du changement climatique. Il s’agit de la première application opérationnelle des données Explore2 dans un modèle d’usages en France. Cet article présente une synthèse des principaux choix méthodologiques réalisés pour mener à bien cette première simulation : effectuer une sélection des futurs hydrologiques possibles ; définir et valider avec les acteurs du territoire les indicateurs d’évolution pertinents. Les résultats basés sur le scénario RCP8.5 sont présentés pour deux niveaux de réchauffement modéré (+2-3°C) et fort (+4-5°C). Ils mettent en évidence la mise en tension de la gestion multi-usages du fait de la baisse de la ressource en eau projetée notamment en période estivale : augmentation du besoin de soutien d'étiage par les retenues ; augmentation de la durée et de la fréquence des restrictions sur les usages agricoles notamment ; diminution de la valorisation énergétique des aménagements hydro-électriques ; diminution du taux de remplissage estival des grandes retenues pour le tourisme.

ABSTRACT. In recent years, accelerating impacts of climate change have exposed urban and rural areas to increasingly frequent and severe precipitation events which result in catastrophic losses of life, damage to infrastructure and property, disruption of ecosystems, and long-term socioeconomic complications. This highlights the urgent need for sustainable water management to mitigate risks and enhance preparedness for future climate-driven challenges. A key priority within these strategies is simplifying and improving the prediction and simulation of runoff during precipitation events. In this article we focus on developing an easy and efficient tool that couples morphological and hydrodynamic modeling to improve the prediction of surface runoff focusing on drainage basins at various scales (farm scale, catchment scale, regional scale). The article is divided into two parts. The first part introduces the development and testing of a novel "Multi-Sweep" morphological algorithm designed for efficient watershed and sub-catchment delineation. This algorithm, focused on Digital Elevation Models (DEMs), aims to accurately capture topographical features and the sub-catchments of a watershed by considering any pre-selected outlet pixel in the basin. The second part focuses on the development of a Lagrangian run-off model based on simplified Chezy 2D hydrodynamics. The initial version is based on quasi-uniform flow assumptions, and leads to simple water packet displacement for runoff simulation under a space-time distributed rainfall. The model is also enhanced to address and resolve boundary condition issues that may arise during the simulations. Additionally, it is used to cross-validate the results of the morphological algorithm. Further advances will focus on relaxing the uniform flow assumption, and in introducing infiltration models. The developed algorithms are tested on both synthetic DEM topographies (e.g., an egg-box topography) and on real DEM datasets. The results demonstrate promising outcomes, which can be further refined and extended to more complex and realistic scenarios.

ABSTRACT. Hydrological modeling is crucial for understanding water systems and guiding sustainable resource management. However, uncertainty quantification and sensitivity analysis continue to pose significant challenges. Although robust methodological frameworks are available for systematic parameter estimation and error assessment, their practical adoption has been limited due to computational demands, technical complexity, and a lack of integrated tools in commonly used models. The Water Evaluation and Planning System (WEAP), developed by the Stockholm Environment Institute (SEI), illustrates this gap. Although it has been cited in over 530 scientific articles since 2019, its dependence on the P.E.S.T. algorithm for calibration has limited its ability to conduct thorough uncertainty and sensitivity analyses, especially in complex systems. This limitation reflects broader trends in hydrological literature, where numerous studies overlook uncertainty quantification—a critical deficiency in contemporary modeling practices. To tackle these issues, this study integrates advanced calibration methodologies from OSTRICH (Optimization Software Toolkit) into WEAP’s framework. The open-source nature of the OSTRICH ensures reproducibility and adaptability, addressing user-identified limitations in WEAP’s accessibility for uncertainty studies. This tool can pursue more than single optimal parameter sets, this methodology characterizes complete probability distributions, providing a more robust foundation for water resource management decisions under uncertainty. This integration enables systematic sensitivity analysis to identify parameters with the strongest influence on model outputs—a crucial step for reducing computational burden and improving predictive reliability. An important improvement lies in enhancing computational efficiency for uncertainty analysis while maintaining compatibility with existing tools like P.E.S.T. By making prediction uncertainties explicit, the improved framework promotes transparency in water management decisions, avoiding the pitfalls of presenting deterministic results as definitive truths. We applied the integrated tool in six case studies, covering 16.5 % of the Colombian continental territory, within the framework of the project 'Support for the implementation of climate-smart rural landscapes in Colombia', funded by the European Union and managed by Fondo Acción. The results demonstrated that the incorporation of uncertainty, facilitated by OSTRICH, enabled a more realistic risk assessment, identifying scenarios that would not have been detected through deterministic calibration. This facilitated the design of more resilient water management strategies. Furthermore, the integration of OSTRICH significantly improved computational efficiency, reducing analysis times compared to P.E.S.T. and allowing for a broader exploration of scenarios and parameters. These technical improvements strengthen WEAP’s mathematical rigor and facilitate uncertainty analysis, directly benefiting water managers and vulnerable communities through enhanced risk assessments for floods, droughts, and water availability. This work strengthens WEAP’s position as a useful tool for climate-resilient water resource planning and policy development.

ABSTRACT. Mountain flash flood disasters are regarded as one of the most serious natural disasters worldwide, characterized by their acute spontaneity, swift escalation, immense destructive force, and high propensity for inflicting substantial casualties. Thus, for effective management and mitigation, it is crucial to identify potential high risk areas, and to adopt effective approaches. Located on the southeast coast of China, Fujian Province features a landscape dominated by continuous hills, with upland areas accounting for 90% of its territory, significantly surpassing the national average of 66.7%. Fujian Province presents complex terrain structures, frequent occurrences of heavy rainstorms, and densely populated areas. Mountain flash floods in Fujian have led to massive economic losses and considerable loss of life. This paper focuses on exploring the driving factors behind mountain flash flood disasters in Fujian Province, aiming to identify high-risk zones. The high risk areas are mainly concentrated in coastal areas with high typhoon frequency and inland mountainous areas with high rainstorm value. Meanwhile, through comparative analysis of historical flash flood disasters in Fujian Province, 80% of historical flash flood disasters fell in high-risk and medium-risk areas. The occurrence density of flash flood disaster in high-risk areas is 3 times than that in low-risk areas. For conclusions: (1) Flash flood risk identification based on small watershed can reflect the response of flash flood disaster to short-term heavy rainfall, the underlying surface and human activities. The risk assessment method based on watersheds has a certain physical mechanism. (2) The evaluation results show that flash flood disaster prone area in Fujian Province is concentrated in the coastal area affected by typhoon rainstorm and the high value area of severe convective rainstorm in inland mountainous area. The frequency of flash flood disaster is affected by both natural environment and human activities. (3) Through the verification of historical flash flood disaster data, the risk identification results are reliable, and risk assessment model can reflect the risks more accurately, which can provide reliable support for flash flood disaster prevention and accurate forecast and early warning.

ABSTRACT. The effective communication of flood risk is essential to improve public awareness, support preparedness, and enhance resilience to flooding events. Traditional flood maps with water depth and velocity distributions showed limitations in conveying intelligible and effective information to non-experts. On the other hand, hazard indexes, which combine water depth and velocity into a single parameter, are often based on oversimplified assumptions, such as a linear relationship between hazard and velocity. These assumptions can make these indexes unreliable for rating intermediate hazard levels, which is critical for assessing flood risk scenarios that typically occur in urban areas. This study introduces the Loss Probability (LP) of pedestrians as an alternative approach to enhance flood risk assessment and communication. LP represents the probability of a pedestrian to be swept away by floodwaters, and it integrates both hazard and vulnerability within a physics-based, data-consistent framework. By accounting for the non-linear interplay between water depth and velocity, LP can address the shortcomings of conventional hazard indexes, offering a more accurate picture of flood risk in urban areas. LP is computed using physically-based formulas that incorporate experimental data. Providing results on a scale of 0 to 100%, LP offers a comprehensible measure of risk that is easily interpretable by the general public. The spatial distributions of LP, represented through color maps, highlight areas of varying risk, distinguishing between low-risk nuisance flooding and high-risk situations caused by deep or fast-flowing waters. A real-world case study demonstrates the effectiveness of LP in identifying flood risk scenarios. In slow shallow waters, where traditional hazard indexes tend to overestimate risk, LP correctly predicts low risk levels. Conversely, LP identifies high-risk conditions in slow but deep waters, where other hazard indexes fail to capture the severity of the danger. As a contribution to bridge the gap between technical flood modeling and public understanding, LP maps provide a valuable tool for urban flood risk management. Their ability to communicate risk in a clear, yet still accurate way can be of help by fostering disaster preparedness and supporting decision-making.

ABSTRACT. Combined with advanced flash flood disaster prevention concepts and theoretical and technological advances in meteorology, hydrology and information technology, the overall framework of flash flood disaster monitoring, forecasting and early warning platform in Hainan Province was proposed, the overall framework, main functions and applications of the platform were designed, the application ways of big data on flash flood disaster investigation and evaluation were sorted out, and a multi-stage progressive flash flood disaster forecasting and early warning system and forecast were proposed The early warning model provides strong support for flash flood disaster prevention in Hainan Province. The provincial flash flood disaster monitoring and early warning platform was built in Hainan Province in 2021 on the principle of emergency first. The system is built from three aspects, i.e. data, models and business, including flash flood disaster prevention data integration and sharing, geographic information services through "one map" for flash flood disaster prevention, real-time weather forewarning service and social publishing support for flash flood disasters, functional maturation of flash flood disaster monitoring and early warning system, and multi-stage forecasting and early warning system construction for flash flood disasters. From May 15, 2022, 252 meteorologic early warnings have been cumulatively made based on the monitoring and early warning platform for flash floods in Hainan Province. Flash flood disaster survey and evaluation results of Hainan have been integrated, with the access to the 1km grid-based 24h rainfall forecasting data pushed by the Hainan Meteorological Service, and integrative validation of 999 automatic rainfall stations, 281 riverway water level stations, 14 riverway hydrological stations and 1118 reservoir hydrological stations in Hainan Province. 295 town- and village-based plans submitted by cities and counties have been collected, and the information of staff on duty in cities and counties has been updated, providing reliable data support and early warning basis for early warning of flash floods.

ABSTRACT. Intense convective precipitation events have been increasing in Switzerland since the early 2010s [1]. They may cause significant damages to people and infrastructure. A better forecasting of these sudden events can lead to a better safety of people and infrastructure. Typically, hydropower intakes can be opened in advance and protected. Today, it is very difficult to predict storm events, even tens of minutes in advance only. In particular, weather forecasting models are unable to predict thunderstorm cells with the required spatio-temporal accuracy for sound decision-making. Moreover, today's operational radar precipitation forecasting, based on deterministic forecasts up to 6 hours in advance, often misses the peaks of these events. A new technique based on radar ensembles, called “NowPrecip2.0 RZC-based”, is producing promising results (Germann et al., 2022, Sideris et al., 2020). Radar precipitation forecasts, updated every 10 minutes, are produced in the form of an 11-member ensemble. The algorithm for storm cell advection, growth and decay, has also been improved over the deterministic radar forecast NowPrecip1.0 (Sideris et al., 2014). These ensemble precipitation forecasts are finally introduced into the semi-distributed rainfall-runoff simulation model Routing System (Jordan, 2007), which has been calibrated on flow measurements in flood-affected catchments. The radar ensemble forecast is tested on several catchments where extreme floods have occurred in recent years. The ensemble flow forecast is then compared to the deterministic forecast and observations during these floods. The results show that ensembles produce some members capable of approximating very high peak flows as observed, in contrast to deterministic radar forecasts. These flash flood events can be predicted up to 3 hours in advance, thanks to catchment response times of the order of 1 to 2 hours. In urban settings with a response time of less than 30 minutes, anticipation is more likely to be on the order of 1 hour. These results demonstrate the need for further research into precipitation forecasting algorithms based on radar data, as the improvements achieved in recent years are significant and promising.

ABSTRACT. It was obtained through the review and examination of typical flash flood disaster events in China from 2019 to 2024 that extreme rainstorms, unfavorable underlying surface conditions and human activities, insufficient monitoring and forecasting capabilities, deficiencies in disaster warning when disasters are approaching, are the main causes of flash flood disasters in recent years.The prevention and control strategies of flash flood disasters under the background of climate change were discussed and proposed.They mainly include:enhancing the integration of structural and non-structural measures in flash flood prevention,implementing the management of dangerous areas,accelerating the construction of a "three-tiered defense line" for rainfall and water level monitoring,improving the capacity of forecasting and early warning,strengthening risk prevention and control management,reinforcing flood control governance of rivers in hilly and mountainous areas,and conducting thorough reviews and examinations as well as exploring innovations.

ABSTRACT. Nature-based Solutions (NBS) are increasingly being studied in local development plans to cope with runoff-based events, particularly hedgerows, retention ponds and wattle fences. This approach has been implemented in several French watersheds, including the basins of Noreade and Aa in Northern France, and Orge-Yvette and Beuvronne in the Paris region, among others. In terms of runoff management, the main goal of using such solutions is to promote rainwater infiltration and thereby reduce the volume of water and sediments reaching the rivers. While these developments must be studied from hydrological and hydraulic perspectives (direct impacts), their benefits extend beyond this framework: they enhance biodiversity, mitigate erosion, and provide urban cooling effects through the creation of cool islands. This paper is based on several studies conducted in watersheds of different scales, aiming to assess the impact of Nature-Based Solutions on flood risk. NBS were integrated into 3D hydrological models developed with the MIKE SHE modelling software. The tools simulate interactions between surface water, subsurface flow, and saturated zone dynamics on a grid cell-level. They therefore combine a hydrological balance and a surface flow model to accurately represent runoff events in heterogeneous landscapes. The impact of the NBS is assessed using flood risk mapping. The challenge is twofold: identifying the most suitable locations for NBS based on topography and land use, and sizing accurately the structures in a model constrained by the resolution of the Digital Elevation Model (DEM). This paper aims to present how these solutions have been integrated into 3D hydrological models and provides quantitative guidance for evaluating the impact of sub-grid scale structures at the design stage of NBS implementation for flood mitigation, as well as guidance for decision support tools in regional watershed planning.

ABSTRACT. In Mediterranean regions, which are particularly prone to flash floods, the difficulty of forecasting and measuring rainfall intensity, as well as the difficulty of identifying flood-generating processes, often leads to the use of conceptual models, which make few assumptions about the processes involved. Neural networks are one such model, and have proven their relevance for flood forecasting. However, in the absence of hydrometeorological coupling, current hydrological models are most often limited to a lead time equivalent to the response time of the basin, i.e. a few hours. The challenge is to find a way of increasing this lead time, which is often too short for crisis management. Since 2024, Météo France's AROME and ARPEGE weather forecasts have been freely available in real time, opening up new perspectives in operational hydrological forecasting for many stakeholders. A flood forecasting model for the Gardon de Mialet basin (Southern France) has been developed as part of the HydrIA joint laboratory funded by the French National Research Agency (ANR) and the Synapse company, which aims to develop a range of hydrometeorological forecasting services based on artificial intelligence techniques. The neural network model implemented (a Multilayer Perceptron) has been coupled with the AROME high rsolution model to produce forecasts with lead times of up to 24 hours, compared with 2 to 3 hours previously. Using the raw data predicted by AROME for the 49 rainfall events in the database produced perfectible flow results. We therefore began by identifying the sources of forecast error (hydrological model, coupling or meteorological model), and then qualifying and quantifying meteorological forecast errors. At the end of this process, an approach is proposed for making the best use of weather forecasts from AROME, taking into account errors in synchronization, location and intensity.

ABSTRACT. The Argens is a coastal river in the southeastern part of France. The Argens' catchment is subject to a Mediterranean climate and regularly experiences periods of drought and flooding. In order to improve the hydrological monitoring of watercourses in this catchment, hydrometric stations are to be installed on seven tributaries by the river basin authority (Syndicat Mixte de l'Argens). However, some monitoring stations will be installed on intermittent tributaries, while others will be located in broad floodplains or constrained locations. The aim of this study is to use 2D hydraulic models to establish beforehand a priori height-flow rating curves at each hydrometric station. For each station, two modeling approaches using the HEC-RAS and TELEMAC-2D software were compared. The topographic data used for these models was obtained from the open high-resolution LiDAR dataset provided by the French National Institute for Geographic and Forestry Information. The modeled flows were taken from the Shyreg discharge database, which covers events up to a thousand-year return period. The lessons learned highlight that, for the purpose of rating curve construction, the HEC-RAS software facilitates the integration of engineering structures. Moreover, its option for semi-automatic extraction of rating curves simplifies this type of application. In contrast, the TELEMAC-2D software does not allow for the efficient integration of engineering structures such as bridges. These structures must be represented in the models by a topographic adjustment in their location area. While this enables the modeling of bridge decks, it is less accurate than the solution proposed by HEC-RAS, which provides a more detailed representation of hydraulic structures. Significant differences appear in the rating curves obtained from the two approaches. TELEMAC-2D produces reliable results for low flows, whereas HEC-RAS provides a much more reliable height-flow relationship due to its ability to integrate engineering structures more effectively.