Next Article in Journal
Scale and Reasons for Changes in Chemical Composition of Waters During the Spring Freshet on Kolyma River, Arctic Siberia
Previous Article in Journal
Hydraulic Performance of an Angled Oppermann Fine Screen with Guidance Wall
Previous Article in Special Issue
Bedload Dynamics in a Partially Glaciated Catchment: Insights from over One Decade of Measuring Bedload Transport Processes and Future Perspectives Under Climate Change
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Water
Volume 17
Issue 16
10.3390/w17162399
water-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Simulation of Sediment Dynamics in a Large Floodplain of the Danube River

by
Dara Muhammad Hawez
1,2,*,
Vivien Füstös
3,4,
Flóra Pomázi
1,4,
Enikő Anna Tamás
5,6 and
Sándor Baranya
1,4
1
Department of Hydraulic and Water Resources Engineering, Faculty of Civil Engineering, Budapest University of Technology and Economics, Műegyetem rkp. 3., 1111 Budapest, Hungary
2
Civil Engineering Department, University of Raparin, Ranya 46012, Iraq
3
HUN-REN–BME Water Management Research Group, Műegyetem rkp. 3, 1111 Budapest, Hungary
4
National Laboratory for Water Science and Water Security, Department of Hydraulic and Water Resources Engineering, Faculty of Civil Engineering, Budapest University of Technology and Economics, Műegyetem rkp. 3, 1111 Budapest, Hungary
5
Faculty of Water Sciences, Ludovika University of Public Service, Bajcsy-Zsilinszky u. 12–14, 6500 Baja, Hungary
6
National Laboratory for Water Science and Water Security, Faculty of Water Sciences, Ludovika University of Public Service, Bajcsy-Zsilinszky u. 12–14, 6500 Baja, Hungary
*
Author to whom correspondence should be addressed.
Water 2025, 17(16), 2399; https://doi.org/10.3390/w17162399 (registering DOI)
Submission received: 26 June 2025 / Revised: 25 July 2025 / Accepted: 29 July 2025 / Published: 14 August 2025
(This article belongs to the Special Issue Advances in River Restoration and Sediment Transport Management)
Browse Figures
Versions Notes

Abstract

This study presents a two-dimensional (2D) hydro-morphodynamic simulation of sediment dynamics in the Gemenc floodplain, a critical ecological zone along Hungary’s Danube River. The 60 km study area has a mean discharge of approximately 2300 m3/s, with peak floods exceeding 8000 m3/s. The objective was to analyze sediment transport, deposition, and flood hydrodynamics to support future floodplain restoration. The HEC-RAS 2D model was calibrated using water levels (Baja station), 2024 flood discharges, suspended sediment measurements, and visual stratigraphy surveys conducted after the event. A roughness sensitivity analysis was conducted to optimize Manning’s n values for various land covers. The hydrodynamic model showed strong agreement with observed hydrographs and discharge distributions across multiple cross-sections, capturing complex bidirectional flow between the main River and side branches. Sediment dynamics during the September 2024 Danube flood were effectively simulated, with SSC calibration showing a decreasing concentration trend, highlighting the floodplain’s function as a sediment trap. Predicted deposition patterns aligned with field-based visual stratigraphy, confirming high sediment accumulation near riverbanks and reduced deposition in distal zones. The model reproduced deposition thickness with acceptable variation, demonstrating spatial reliability and predictive strength. This study underscores the value of 2D modeling for integrating hydrodynamics and sediment transport to inform sustainable floodplain rehabilitation.

1. Introduction

Floodplains are vital ecosystems that provide sediment storage, habitat connectivity, and flood mitigation. Many extensive river floodplains, like the Danube [1], have been altered by anthropogenic activities, such as channelization, damming, and changing how land is used. These changes disrupt sediment transport and reduce lateral connectivity, leading to habitat loss and increased flood risk. To address these challenges, a variety of restoration strategies have been applied globally, such as re-meandering, channel reconfiguration, floodplain reconnection, habitat enhancement, dam removal, fish passage restoration, and natural flood management (NFM) [2,3,4,5,6]. Planting native plants in marshes and along riparian buffers is another way to help stabilize ecosystems [7,8]. Assessing the effectiveness of such measures requires robust modeling of discharge and sediment dynamics. While numerous modeling studies exist, challenges remain in simulating long-term sediment behavior under dynamic flow regimes.
Numerical models serve as powerful tools to assess and design floodplain restoration strategies by analyzing the complex interactions between hydrology, geomorphology, and ecology [9,10]. Hydrodynamic and sediment transport models can simulate floodplain deposition and erosion, helping to predict future changes in morphology, water quality, and hydrological processes. Such capabilities are essential for assessing the role of floodplains in supporting aquatic and riparian ecosystems [11]. To support these efforts, a range of modeling approaches has been used to evaluate the effectiveness of restoration measures across diverse floodplain settings.
A hydrodynamic model, incorporating an empirical erosion rate relationship, was applied to predict morphological changes over ten years, following the approach of Brennan, et al. [12], which supports river restoration by lowering erosion in the primary channel. This study’s importance lies in its ability to evaluate restoration alternatives for mitigating erosion in estuaries like Elkhorn Slough, despite potential limitations in the general applicability of its empirical erosion rate relationship. Clilverd, et al. [13] assessed removing levees by employing a coupled hydrological/hydraulic modeling method to simulate pre- and post-restoration conditions for rehabilitation plans for rivers and floodplains for Hunworth Meadow on the River Glaven. This research is important for demonstrating how embankment removal enhances river–floodplain connectivity, although its generalizability may be limited to similar lowland systems. Likewise, Acreman, et al. [14] examined removing river embankments and restoring the river channel in the River Cherwell between Oxford and Banbury to its pre-engineered dimensions (narrower and shallower) scenarios with a one-dimensional hydraulic model, looking for the peak of flood in each scenario hydrograph. This study highlighted that such restorations can reduce downstream flood peaks, despite using hypothetical scenarios and a simplified model. Willis, et al. [15] analyzed the hydrodynamic impacts of floodplain modifications, such as canal restoration and levee construction, emphasizing how these restoration efforts alter floodwater distribution, peak flows, and recession dynamics. This work underscores the sensitivity of extensive floodplains to modifications. It is crucial for climate-resilient health systems, though it faces challenges due to sparse hydrological data in the large African floodplains of the Barotse Floodplain of the Upper Zambezi, Zambia.
Integrating hydraulic modeling with other techniques has been verified as valuable for restoration planning, mainly in determining suitable sites and assessing the impacts of restoration efforts. Leyer, et al. [16] adopt an iterative approach that combines ecological and hydraulic modeling to identify suitable sites for floodplain forest restoration and evaluate the potential hydraulic impacts of reforestation. Their study, conducted along the Middle Elbe River in Germany, demonstrated the potential for softwood forest restoration without exacerbating flood risks, achieving a balance between ecological benefits and flood protection. However, the study was limited by its focus on steady-state hydraulic conditions, which may not fully capture dynamic flood events, and it did not account for long-term climate change impacts on flood regimes. Gourevitch, et al. [17] incorporated hydraulic modeling, financial analysis, and multi-objective optimization to identify cost-effective and equitable locations for floodplain rehabilitation interventions. Their study, set in the Lewis Creek watershed in Vermont, USA, simulated woodland and marsh restoration by altering Manning’s roughness coefficients, increasing landscape roughness to slow floodwaters, and enhancing flood hazard mitigation. The study’s reliance on 1D hydraulic modeling simplified complex floodplain dynamics, and the small sample size of at-risk properties limited the generalizability of equity-focused conclusions.
In complex case studies, two-dimensional (2D) hydro-morphodynamic modeling can show the interaction between floodplain restoration, hydraulics, and sediment dynamics. Such modeling approaches allow for detailed analysis of how restoration measures influence flow patterns, sediment transport, and deposition over time, particularly in ecologically sensitive or morphologically dynamic environments. Maaß and Schüttrumpf [18] used a 2D (depth-averaged) hydrodynamic model to investigate the long-term outcomes of reactivated floodplains on sediment deposits of remote rivers. The model focuses on depositing fine deposits on floodplains over decades in two cases: elevating the riverbed level and digging the floodplain surface. This study is important for understanding the long-term implications of reactivating floodplains on the mobility of floodplain sediment deposits, particularly in terms of erosion and deposition processes. A limitation noted is that the numerical model’s length should ideally extend until the maximum extent of floodplain inundation is reached. Ahilan, et al. [19] applied a 2D hydro-morphodynamic model to study the influence of floodplain rehabilitation on flood events and sediment dynamics with varied flood events. The study’s primary focus is on an urban river, specifically exploring the impact of floodplain restoration on its hydro-morphological processes. Limitations include the general challenges associated with modeling complex natural systems, especially the interactions between flow, sediment, and vegetation in urban settings, which can introduce uncertainties. González-Sanchis, et al. [20] designed a 2D finite volume numerical model to simulate actual flood events and propose restoration strategies. A limitation is that the accuracy of the model depends on precise input data for real flood events, and generalization to other regions might require calibration. Poulsen, et al. [21] also utilized 2D modeling to explore floodplain hydraulics and sediment deposition in a restored river area, concentrating on the consequences of re-meandering on sediment dynamics. This study is crucial for understanding the linkages between floodplain hydraulics and sedimentation patterns in restored river channels, particularly the effects of re-meandering on sediment deposition. The study area is the restored River Odense in Denmark. Limitations could involve the inherent complexities of accurately modeling sediment transport and deposition in dynamic river systems, and potential site-specific factors that may not be universally applicable.
Despite their potential in floodplain rehabilitation, 2D hydro-morphodynamic models face substantial challenges. Sediment transport remains complex and uncertain due to the dynamic nature of sediment processes [22]. Model accuracy depends heavily on high-quality input data such as bathymetry, topography, and sediment characteristics [23,24] and is affected by simplifications and parameter uncertainties [25]. Field data collection is crucial for model calibration and validation [26], yet acquiring long-term datasets is often expensive and time-consuming [27]. Thus, model outcomes should be analyzed cautiously, fully aware of inherent uncertainties. A combined modeling and field monitoring technique is recommended to improve reliability and understanding [24], primarily through long-term programs that support continuous calibration [28]. These models remain vital for evaluating restoration strategies in heavily altered floodplains like Gemenc, where human interventions have significantly reshaped morphology.
The Gemenc floodplain, a unique wetland ecosystem of the Danube River in Hungary (the Gemenc Region of the Danube-Drava National Park, also protected under the Ramsar Convention), underwent significant morphological changes predominantly due to intensive river regulation works initiated in the 19th century. These interventions, intended for flood control and navigation, involved shortening the Danube’s riverbed by cutting through numerous meanders, which consequently increased the river’s gradient and led to the isolation of former river bends, forming oxbow lakes [29]. The construction of embankments further decreased the connection of the floodplain area, restricting the river’s natural inundation patterns [30]. Above all, river channelization increased erosion within channels and resulted in significant deepening of the central bed of the river, where the decreasing of the water levels is still ongoing and has reached 1.8 m in water levels at the Paks hydrographic station. It directly affected the floodplain, the depth of surface waters and water table fell and sedimentation hastened [31,32]. These morphological changes have triggered biological issues in the Gemenc floodplain. Reduced inundation frequency [33] and duration and drying out of water bodies resulted in the significant loss and degradation of critical wetland habitats [29]. Furthermore, the reduced lateral connectivity between the river and the floodplain disrupts important ecological processes like nutrient exchange and the migration of organisms. Consequently, the Gemenc floodplain faces substantial ecological challenges that endanger its biodiversity and the long-term sustainability of its natural values.
Conservation of ecosystems and biodiversity on the Gemenc floodplain is very crucial. It necessitates exhaustive and detailed studies that indicate solutions for existing issues and support restoration plans. Floodplain restoration should consider ecological, hydrodynamic, and sediment transport components together. These aims could be achieved through hydro-morphological modeling, whereby one can determine the complex relationship between water dynamics, sediment transport, and habitat structures. A few earlier modeling efforts have explored the Gemenc area, although they remain preliminary. For example, Füstös, et al. [34] validated a 2D hydro-morphodynamic model using historical elevation data from 1990 and 2009, but their model was limited by the resolution and quality of data available for validation. Józsa, et al. [35] applied a depth-averaged numerical flow model to investigate steady-state hydraulics in secondary branches of the floodplain to improve water exchange without affecting navigation or sediment conditions; however, their fixed-bed approach excluded sediment dynamics. In this context, the present study introduces a novel hydrodynamic and sediment transport model calibrated and validated using measured suspended sediment concentration (SSC) and discharge data from a real flood event, as well as observed post-flood sediment deposition layers. This represents the first integrated model in the Gemenc floodplain calibrated for both hydraulics and sediment transport, offering essential support for future restoration planning across ecological and hydraulic dimensions. When considering model types, one-dimensional (1D) models are usually unacceptable since they cannot capture spatial variability of such complex floodplain. While conceptually ideal, three-dimensional (3D) models require massive computational efforts and very detailed input data that are typically unavailable. Therefore, 2D modeling represents the best compromise between accuracy and practicality. Therefore, this paper aims to use unsteady 2D hydrodynamic and sediment transport modeling for simulating flow conditions, sediment transport, and channel–floodplain interactions. This system will not only enhance a better understanding of floodplain functioning but also support calibration work and provide the basis for future restoration planning in the ecological and hydraulic dimensions of Gemenc.

2. Materials and Methods

2.1. Study Area

The Gemenc floodplain, located along the lower reaches of the Hungarian Danube, illustrates one of the most extensive contiguous floodplain ecosystems in Europe, an area of approximately 18,000 hectares [29,36]. This ecologically significant area stretches 60 km in length and 4 to 5 km in width [29]. Geographically, it lies along the meandering lowland section of the Danube River, on the southwestern edge of the Great Hungarian Plain (see Figure 1).
This reach of the Danube River is characterized by low flow velocities and limited lateral movement from the main channel [29]. The landscape of the Gemenc floodplain is further characterized by numerous fluvial landforms, including side branches, oxbow lakes, and various other lentic water bodies [29,37]. The floodplain is predominantly forested (see Figure 1, land cover map) and typically experiences flooding during summer [38]. However, the duration of flooding has significantly declined in this reach, likely due to riverbed incision, historical human interventions, and climate change [39].
The Gemenc floodplain features extensive hardwood forests and a diverse array of aquatic habitats, including meanders, oxbow lakes, and marshlands [37,40]. The result of historical river regulation (e.g., cut-offs, bank stabilization) has deepened the primary riverbed, lowering the water supply to the floodplain. This has led to lower groundwater levels, augmented sedimentation in side branches., and shorter overflow durations, limiting discharge dynamics [31].

2.2. Model Setup and Simulation Approach

The model developed for the Gemenc floodplain is unsteady and simulates both hydrodynamic and sediment transport processes using HEC-RAS 2D. Water surface elevation (WSE) and recorded cross-sectional flow rate data are used for hydrodynamic verification. The validation involved a detailed sensitivity analysis of roughness parameters in the main river and the forested floodplain areas. During a field survey, suspended sediment concentration (SSC) was measured at multiple locations within both the Danube River and its side branches, and these data were used to calibrate sediment boundary conditions. The hydrodynamic model uses a depth-averaged shallow water equation and integrates the total load equation by Wu, et al. [41] to simulate sediment transport. With variable cell resolution and strategic break lines, the computational mesh design accurately represents channel–floodplain interactions while maintaining computational efficiency. These integrated approaches and field-based adjustments reinforce the robustness of the Gemenc model for simulating floodplain hydro-morphodynamics processes.

2.2.1. Hydrodynamic Modeling

The hydrodynamic component of the research was simulated using HEC-RAS 2D, which solves the depth-averaged shallow water equations (SWEs). The SWEs are widely accepted for modeling floodplains and river hydrodynamics where vertical acceleration is negligible compared to horizontal flow. This approach ensures computational use while maintaining accuracy for large-scale floodplain and sediment transport applications [42,43]. HEC-RAS 2D’s implementation of the shallow water equations (SWEs) for sediment transport has been validated through both the official technical reference manual [44] and applied studies such as the Woodbridge Creek flood simulation study, which successfully coupled HEC-RAS 2D with WASP to model sediment transport under extreme flooding [45]. Additionally, HEC-RAS 2D was effectively implemented to simulate sediment transport and deposition in the Pare Reservoir, Central Lombok, Indonesia. This study estimated the impact of sedimentation on reservoir storage capacity and functional efficiency [46].

2.2.2. Sediment Transport Modeling

Sediment transport in this study was modeled using the total load equation proposed by Wu et al. [41], which accounts for bedload and suspended load. Wu’s equation is well-suited for natural rivers characterized by mixed size, nonuniform sediments, making it appropriate for complex fluvial systems. The equation has been validated across various sediment sizes, from fine silt (0.01 mm) to very coarse gravel (28.7 mm), encompassing uniform and nonuniform mixtures. Validation efforts included laboratory experiments (e.g., Samaga, et al. [47], Kuhnle [48], Wilcock and McArdell [49]), field measurements from major river systems (e.g., the Yellow River, Mississippi River, and Rio Grande), and comparisons with specified datasets such as Brownlie’s uniform sediment data [41].
In the Gemenc floodplain study site, sediment transport states are near those for which Wu’s equation was validated. The bed material is mainly sandy gravel, while the suspended sediment primarily consists of silty and sandy fractions (see Figure 2). These characteristics support Wu’s total load approach in this modeling context.

2.2.3. Computational Mesh and Numerical Scheme

The numerical scheme employed in the simulation was based on an implicit finite-volume approach, as used by the HEC-RAS 2D unsteady-flow solver. The simulation based on the Courant–Friedrichs–Lewy (CFL) was dynamically adjusted throughout the computational time step to maintain numerical stability and consistency.
A detailed sensitivity analysis of grid resolution determined that a mesh configuration incorporating approximately ten computational cells across the main river channel and four cells across the side branches provided stable and reliable results. A flexible 2D mesh was employed to ensure a balance between model accuracy and computational efficiency, with cell sizes varying from 25 m in the vicinity of river channels to 100 m across broader floodplain areas.
Break lines were strategically implemented within the HEC-RAS 2D environment to enhance spatial resolution and control mesh refinement further. This technique allowed for targeted refinement in areas with complex topography or hydraulic features, thereby improving model accuracy without incurring excessive computational costs.

2.2.4. Boundary and Initial Conditions

The model domain extends from Dombori to Mohács, between rkm 1506.8–1446.9. Boundary and initial conditions were carefully defined to capture the hydrologic and sediment transport processes. Discharge data from Dombori are imposed at the upstream boundary, while a WSE boundary condition at Mohács station governs the downstream side (as shown in Figure 3). The Baja station, located near the center of the model domain along the Danube River (see Figure 1 for its location), is used for model validation. WSE data from Baja were used to validate the December 2023 event, while the September 2024 event served to validate WSE, discharge, and sediment dynamics. All discharge and water surface elevation (WSE) data were obtained from the General Directorate of Water Management of Hungary, which maintains an online database of all hydrological stations (accessible at: https://www.vizugy.hu, accessed on 15th December 2024). The landcover spatial distribution is specified based on multiple categories such as forest, water surface, agriculture, and pasture, which are detailed in Figure 1. A uniform bed material is applied to all main river reaches for sediment simulation. Additionally, suspended sediment characteristics at Dombori are defined using the previously measured size distribution of suspended sediment (see Figure 2), thus, the hydrodynamic and sediment transport components were effectively integrated to simulate realistic floodplain conditions.
For the sediment load and flow rate time series in the model, the relationship illustrated in Figure 4 was applied. The figure shows that the sediment load increases with the rising flow rate. This trend indicates that higher discharges are associated with significantly greater sediment transport rates, consistent with typical sediment dynamics in fluvial systems. The peak sediment load was determined through model trials until calibration was achieved for the point record of SSC during the September 2024 event.

2.2.5. Model Calibration and Data Sources

The Digital Terrain Model (DTM), the foundation for the model, was developed from LiDAR data obtained in 2013, with a spatial resolution of 0.5 m and covering an area of about 220 km2 (refer to Figure 1). Complementary bathymetric data were obtained through cross-sectional surveys along the main river channel, and associated side branches were merged into the DTM.
A roughness sensitivity analysis was conducted by applying Manning’s n values to different land cover classes, including side arms/streams/oxbow lakes, groin/guide banks (wing dams), agricultural fields/pastures, settlements, forests, infrastructure (e.g., pillars, roads, railways), and the main riverbed (see Figure 1). Initial Manning’s n values were assigned based on existing publications and expert estimation. These values were afterward refined through a calibration procedure. The final calibrated model specified forested areas (n = 0.095) and the main riverbed (n = 0.032) as the most sensitive land cover types influencing the hydrodynamic flow and sediment transport behavior.
Hydrodynamic model calibration was carried out using water level data from the Baja hydrometric station and discharge measured at the September 2024 flood event, covering the main river channel and its anabranches. Additionally, field measurements of suspended sediment concentration (SSC) collected at several locations during the same event on 23–24 September were used to calibrate the model’s sediment transport.
The spatial distribution of recent sediment deposition was evaluated during a field campaign conducted on 10 December 2024 using a visual stratigraphy survey. This method, which relied on identifying sediment layers laid down based on observable differences in color, texture, and vegetation cover, offered a rapid and flexible way of delineating depositional zones across the floodplain. Although the technique is advantageous in terms of efficiency and ease of implementation, it lacks the quantitative precision offered by more established methods, such as sediment traps, core sampling, or radioisotopic dating (e.g., using 210Pb). During the survey, five test pits were excavated in the main floodplain to determine the sediment layers attributable to the most recent flood event. These observations contributed to the model calibration effort by providing spatial indicators of sediment deposition.

3. Results and Discussion

3.1. Hydrodynamic Model Performance

Simulation of Gemenc floodplain sediment transport needed calibration and robust testing of the hydrodynamic parameters with a well-defined flow hydrodynamic model. Therefore, the hydrodynamic model performance was tested by executing roughness sensitivity analysis and comparing observed water levels and discharge at different Danube River points and lateral channels in the Gemenc floodplain. A comparison of simulated flow characteristics indicates that the model can accurately simulate the complex flow dynamics of the Danube River and branches under flood conditions.

3.1.1. Sensitivity Analysis of Roughness

The roughness sensitivity analysis revealed that Manning’s n values significantly influenced water surface elevations and flow distribution, particularly within forested floodplain areas and the main riverbed. Changes in the main river and forest roughness yielded the most noticeable variations in water surface profiles during the September 2024 flood event. In contrast, adjustments to roughness within agricultural and urban land uses had relatively minor hydrodynamic impacts. Under conditions where Manning’s n value is low in the main river channel and high in the adjacent forested region, the volume of water reaching the floodplain is reduced. In contrast, a high n value in the main river and a lower one in the forest (floodplain) promotes water propagation over the floodplain. Significant changes in WSE were primarily driven by roughness variation in the main channel, which is highly sensitive due to its higher flow velocity. In contrast, the forested floodplain has lower flow velocities, making WSE less sensitive to changes in roughness. This shows that a high Manning’s n indicates greater surface roughness, slowing down water velocity. This reduction in velocity decreases the overall flow rate through a channel or floodplain. As a result, the water level rises to compensate for the reduced flow capacity and maintain energy balance, which can significantly affect floodplain hydraulics and sediment transport dynamics.
The validation process began with assumed Manning’s n values of 0.2 for forested areas and 0.235 for the main river, based on Füstös et al. [34]. These values provided a good fit at lower WSE levels but underestimated peak water levels. Therefore, adjustments were made to better align the simulated WSE with the flood crest. The final calibrated values, 0.095 for the forest and 0.032 for the main river channel, best matched the observed data.

3.1.2. Hydrodynamic Validation

Model validation was performed using measured hydrographs from the Baja gauging station for two flood events, December 2023 and September 2024, and cross-sectional discharge data for the latter flood event. As shown in Figure 5, the model accurately reproduced the observed discharge peaks, demonstrating strong temporal reliability. The RMSE for WSE was 0.39 m for the December 2023 event and 0.38 m for the September 2024 event, further supporting the model’s performance.
A more detailed analysis of Figure 5 for the September 2024 flood event reveals that during the initial stages of inundation and recession limb (lower water surface elevations) the simulated WSE was higher than observed. This indicates that the adjusted Manning’s n coefficients were comparatively high during the initial stages of flooding, especially within the main channel and early floodplain areas. At this point water depths are generally shallow and hydraulic resistance is greater, making the flow more responsive to variations in surface roughness. As the hydrograph peaked, the agreement between the simulated and measured SWEs became more precise, suggesting that the chosen Manning’s n values represented hydraulic resistance more effectively during high-flow situations. This finding supports the idea proposed by Domhof, et al. [50] that roughness coefficients alter in response to variations in discharge and water surface elevation. Although HEC-RAS allows for flow-dependent variable Manning’s n values, implementing this feature may introduce numerical instability and increase computational demands. Therefore, a constant Manning’s n value was used for each land cover category. Furthermore, this study focuses primarily on peak flow conditions, where complete inundation occurs across the floodplain. Under these high-flow scenarios, the calibrated Manning’s n values agreed well with the observed WSEs and discharge distributions in auxiliary channels. Therefore, the selected roughness coefficients are considered appropriate and effective for the objectives of this study. This approach aligns with the study’s objectives, which focus on simulating flow conditions and sediment transport during flood events and analyzing sediment deposition on the floodplain. However, calibrating Manning’s n values only for these high-flow conditions may introduce limitations. At high flows, vegetation is often submerged, which decreases its contribution to flow resistance and lowers the effective roughness. In contrast, the same vegetation can significantly increase roughness during low-flow circumstances by obstructing the flow. Additionally, high water levels cause hydraulic connectivity between the main river channel and the floodplain, making it difficult to distinguish and accurately represent their individual roughness characteristics.
The comparison of modeled and recorded flow rates at multiple Danube River and side-branch cross-sections (Figure 6A–E and Figure 7A–D) further confirms the spatial and temporal accuracy of the hydraulic simulation. These comparisons include measurements taken at strategically important cross-sections along the Danube River and its key tributaries and side branches, specifically the Sió, Grébeci-Duna, and Rezéti-Duna (covering upstream (inflow) and downstream (outflow) confluence points). Field-based cross-section discharge measurements were carried out at these locations, allowing for the direct validation of simulated discharges and the directionality of flow exchanges between the main river and floodplain system. The ADCP measurements began with the first cross-section located in the upper part of the Danube floodplain and continued into the Sió channel, as shown in Figure 6B and Figure 7A. The following measurements were conducted in the Grébec side branch and upstream from it in the Danube: these two cross-sections are also shown in Figure 6C and Figure 7B. Downstream of Grébec lies the Rezéti side branch, which receives water from an upstream inlet from the Danube and drains water to the Danube from it through an outlet. Discharge was measured in this side branch upstream, and a cross-section of the Danube just upstream of the Rezéti inlet is illustrated in Figure 6D and the corresponding graph in Figure 7C. The final cross-section on the Danube was recorded just upstream of the Rezéti outlet, and the discharge measurements downstream of this outlet are presented in both Figure 6E and the graph in Figure 7D.
Figure 6A–E illustrate the spatial arrangement of cross-sections and shows both observed and modeled discharge values, indicating where water enters or exits side branches. For example, the Sió channel receives flow from the Danube River (inflow), while at other locations, such as the downstream Rezéti-Duna confluence, flow drains back from the floodplain into the Danube River. These directional patterns reflect the natural flood dynamics, where side branches alternately function as distributaries or return flow channels depending on the stage and flood timing.
Figure 7A through D compare recorded and time-series-simulated discharges for the main channel and side branches. Notably, discharge simulations at the Grébeci-Duna and Rezéti-Duna branches (upper inflows and lower outflows) strongly agree with the observed hydrographs, including peak timings and magnitudes. These patterns demonstrate the model’s ability to capture dynamic flood conditions: floodplain inflows from the Danube River and subsequent return flows. Figure 7A–D present time-series comparisons between the modeled and observed discharges at key inflow and outflow locations within the Gemenc floodplain system. This provides further insight into the model’s ability to reproduce flow partitioning and directional dynamics between the Danube River and its side branches.
In Figure 7A, the model captures the magnitude and timing of flow into the Sió channel, which functions as a controlled inflow from the Danube River. The observed discharge at the Sió during the flood peak reaches approximately 52.2 m3/s, while the modeled value closely follows at 51.5 m3/s, resulting in an absolute error of −0.7 m3/s and a relative error of −1.3%. Concurrently, the Danube River’s main channel at this cross-section reaches a peak observed discharge of 6729.46 m3/s, with the model simulating 6722.36 m3/s, indicating excellent agreement, showing an absolute error of −7.1 m3/s and a relative error of just −0.1%.
In Figure 7B, the flow at the Danube River upstream of Grébec shows strong agreement between observed and modeled values, with the model capturing both the shape and peak of the hydrograph. The observed peak discharge at Grébec is approximately 5900 m3/s, while the model slightly underestimates it at 5540.39 m3/s, yielding an absolute error of −359.61 m3/s and a relative error of −6.1%. However, the adjacent Grébec branch shows better alignment as the model simulates 91.47 m3/s compared to the observed 88.21 m3/s, resulting in an absolute error of 3.26 m3/s and a relative error of 3.7%.
Figure 7C focuses on the Danube River upstream of the Rezéti-Duna inflow, where the observed discharge is 6030.2 m3/s. The model simulates 5544.13 m3/s, producing a larger deviation with an absolute error of −486.07 m3/s and a relative error of −8.1%. For the Rezéti-Duna upstream inlet, the model reaches 117.79 m3/s compared to the observed 121.26 m3/s, leading to a smaller absolute error of −3.47 m3/s and relative error of −2.9%. This suggests minor underestimation, although the values still lie within the observed confidence intervals.
Figure 7D evaluates the downstream section of the Rezéti-Duna confluence, where the Danube River’s main channel shows agreement again, though with some deviation. The observed discharge is 6170.3 m3/s, while the model estimates 5327.33 m3/s, resulting in a notable absolute error of −842.97 m3/s and a relative error of −13.7%. The Rezéti-Duna branch at the downstream confluence reveals an overprediction, with the model simulating 107.15 m3/s against an observed 93.01 m3/s, amounting to an absolute error of 14.14 m3/s and a relative error of 15.2%. This overestimation likely reflects modeling challenges in capturing downstream floodplain outflow behavior and the delayed storage-release effects characteristic of such zones.
The results validate the hydrodynamic reliability of the 2D HEC-RAS model in simulating flow partitioning and channel–floodplain interactions during the September 2024 flood event. This provides a solid foundation for extending the model toward sediment transport and deposition analysis in the subsequent phase.
However, as shown in Figure 6B–E, while the modeled and measured flows align well at an upstream cross-section of the main channel, increasing discrepancies are observed at downstream cross-sections. As shown in Figure 6B, the upper measurement section on the Danube River closely matches the model results; for instance, at Sió channel the observed discharge of Danube River’s main channel is 6729.46 m3/s, while the model simulates 6722.36 m3/s, resulting in an absolute error of −7.1 m3/s and a relative error of −0.1%. However, this agreement diminishes in the last cross-section measurement on the Danube River, located just upstream of the Rezéti-Duna outflow (see Figure 6E). At this location, the observed discharge is 6170.3 m3/s, while the model underestimates it at 5327.33 m3/s, producing an absolute error of −842.97 m3/s and a relative error of −13.7%. This discrepancy may stem from outdated bathymetry and terrain data, collected over a decade ago, and insufficient calibration of velocity-related turbulence parameters. These limitations can significantly impact the representation of lateral flow exchanges between the main channel and the adjacent floodplain, particularly in geomorphologically dynamic areas.

3.2. Sediment Transport Simulation Outcomes

3.2.1. Suspended Sediment Dynamics

SSC was simulated and compared with field measurements collected during the September 2024 flood event. Figure 8 and Figure 9A,B illustrate SSC’s spatial and temporal distribution along the Danube River within the Gemenc floodplain. The initial approach to estimate SSC relied on empirical discharge–SSC relationships reported by Vas and Tamás [51] which were derived from three historical flood events and were based on measurements at the Baja station for water levels up to approximately 6.7 m, corresponding to discharges below 5000 m3/s. However, the September 2024 flood event significantly exceeded this threshold, with peak discharges reaching approximately 7000 m3/s. As a result, the regression curves and empirical models proposed by Vas and Tamás [51] were not directly applicable to this event.
Given this limitation, a complete model calibration was undertaken to determine an appropriate upstream SSC boundary condition. A series of simulation iterations identified an optimal upstream SSC value of 231 mg/L as the maximum SSC at 8000 m3/sec of flow rate at the upstream boundary condition. This value yielded results that closely matched the field measurements recorded during the peak flood days of 23–24 September 2024.
Figure 8 and Figure 9A present the recorded and modeled suspended sediment concentration (SSC) for 23 September, comparing values at four locations. Danube_DunaST represents the upstream point in the floodplain area of the Danube River. Danube_Grébec is located at the inflow of the Grébec side branch, while the remaining two points (RezétF and RezétA) are situated at the inflow and outflow of the Rezét side branch (see Figure 8). Danube SSC values recorded upstream of the Sió channel on the Danube River (at point DunaST) during the first day of the field campaign were reasonably consistent with those predicted by the model. An underestimation was observed only at the inflow of the Grébec side branch (at point Grébec) (Figure 8 and Figure 9A). This discrepancy may result from the complex processes of deposition and resuspension in low-energy conditions, which are not modeled explicitly. Factors such as grid resolution in narrow anabranches and simplifications inherent in the hydrodynamic–sediment transport coupling likely contributed to this underestimation [52]. The remaining points (RezétF and RezétA) show good agreement between the modeled and recorded SSC values (Figure 8 and Figure 9A).
Further discrepancies are evident in Figure 8 and Figure 9B, which show that the modeled SSC does not closely follow the observed trend upstream and downstream of the Vén-Duna side branch (at points VDF and VDA). It shows that these points are near the peak of the modeled SSC. This suggests that SSC peaked on 24th September during the falling limb of the hydrograph. This delayed peak aligns with patterns documented by Vas and Tamás [51], particularly during the 2021 flood event, and reflects a typical sediment transport behavior in the Gemenc region during high-magnitude floods. Similar delayed peaks during the recession phase of the hydrograph have also been reported for the Drava River by Stajnko, et al. [53], who attributed this phenomenon to factors such as bed material resuspension and upstream erosion processes.
The observed decoupling between SSC and peak discharge and peaked SSC at the falling limb of the hydrograph is consistent with the findings of Vas and Tamás [51], who emphasized that SSC peaks in significant flood events are often governed by event sequencing, sediment storage, remobilization, and particle size distribution rather than instantaneous flow magnitude. This behavior contrasts the more commonly observed pattern where SSC peaks during the rising limb or peak discharge. For example, Haimann, et al. [54] reported that in large rivers a substantial proportion of suspended sediment is transported during the rising stage of floods. Similarly, Guan, Ahilan, Yu, Peng and Wright [11], Varvani, et al. [55], and Rosen and Xu [56] found that SSC typically peaks during the rising limb, coinciding with increased stream power and proximity to sediment sources.
These differing dynamics of SSC are further illustrated in Figure 4, which shows the relationship between suspended sediment load (SSL) and discharge. While the model generally exhibits an increasing SSC trend with increasing flow, this does not fully capture the delayed peak behavior observed in the Gemenc floodplain. This is particularly evident at the point VDF and VDA, where the modeled SSC does not align with the measured values.
A closer examination of both observed and simulated SSC data revealed a clear spatial trend, which is that in the upstream section (DunaST, upstream of Sió inlet) SSC exceeded 300 mg/L, while progressively lower concentrations were recorded downstream at various branch outlets—183 mg/L at Grébec, 105 mg/L at upstream inlet Rezéti-Duna (RezétF), and below 100 mg/L at and outlet Rezéti-Duna (RezétA) (Figure 8). This trend indicates a steady decline of SSC along the floodplain, implying that the floodplain serves as a depositional zone for suspended sediment, effectively trapping suspended sediments and acting as a natural filter. A similar trend was observed on the final day of the field survey (24 September 2024). SSC measured at the upstream inlet of the Vén-Duna (Old Danube side branch) was higher than at the downstream outlet of the same anabranch. This further supports the notion that sediment deposition within the floodplain is crucial in shaping SSC dynamics during flood events [11].

3.2.2. Sediment Deposition Patterns

The sediment deposition pattern observed in the Gemenc floodplain following the September 2024 flood event demonstrates pronounced spatial heterogeneity, effectively captured by the HEC-RAS 2D sediment transport simulation. Field data were collected at five strategically selected locations (PT1–PT5) across the inundated area (Figure 10A). Due to logistical constraints during data collection, a visual stratigraphy survey was employed as an alternative to more quantitative methods (see Figure 10C). While such visual assessments are commonly used for preliminary reconnaissance [57], they inherently limit the precision of spatial and volumetric estimations of sediment deposition. Future research should incorporate high-resolution topographic differencing techniques or direct measurements of sedimentation rates to improve model calibration and validation. Such data would allow for a more robust assessment of depositional dynamics and improve the representation of fine-scale variability.
The locations of the five measurement points were as follows: PT1 was located near Báta settlement adjacent to the Danube River; PT2 was downstream of the sluice gate on the Báta Oxbow; PT3 was near Road 55, beneath the bridge farthest upstream from the Danube; PT4 was at the confluence of the Old Danube (Vén-Duna) and the Danube; and PT5 was on Road 55 beneath the bridge closest to the Danube, also on the upstream side. Model outputs were compared to field measurements within 100 m-diameter circular zones centered on each point. The maximum and minimum simulated sediment deposition values within each zone were extracted for comparison with observed values (Figure 10B).
The spatial pattern revealed by the simulation (Figure 10A) indicated that the most significant sediment accumulation occurred adjacent to the main river channel, especially in low-lying, concave overbank areas. This distribution is consistent with well-documented sedimentation dynamics in alluvial floodplain systems [58,59]. Quantitative comparisons show that the model reproduces the general trend of sediment deposition with reasonable accuracy at most locations. At PT1 the observed deposition was 1–1.5 cm, whereas predicting 0.1–0.8 cm and slightly underestimating deposition. This discrepancy could be attributed to unresolved microtopographic variability or vegetation-induced sediment trapping, which were not fully represented in the model. At PT2 the measured deposition was 3–3.5 cm, closely matched by the model which predicted a maximum of 3.5 cm, which indicated strong model performance in this area. At PT3 both field observations and simulations indicated no significant deposition, indicating that the model accurately predicted the conditions of this area, which is either an erosive environment or shows a lack of sediment deposition. At PT4 the simulation yielded a maximum of 7.6 cm, although field measurements documented only 4–5 cm deposition. Despite this overestimation, the estimated value lies within the computational span, suggesting valid agreement. The location of PT4 near the main river contributed to the higher deposition observed. PT5 showed low deposition in the field (2 cm) and in the simulation (0–0.4 cm), consistent with its more distal position. Overestimation and underprediction of sediment deposition in the floodplain could be attributed to outdated terrain data, the simplified representation of vegetation roughness that affects flow deceleration and sediment settling, and the coarse spatial resolution of the computational mesh, which may overlook small-scale depositional zones.
Overall, the model effectively captures the spatial gradient of sediment deposition, from high values near the channel to lower values in more distal floodplain areas (Figure 10B). This trend aligns with established conceptual models of floodplain sedimentation, emphasizing deposition near levees and a progressive decline in thickness with increasing distance from the channel [60,61]. Although minor discrepancies appear at certain locations, the differences typically remain within a few centimeters (i.e., less than ten centimeter). Incorporating additional data on sediment deposits could enhance model fidelity and better represent the dominant patterns observed in the aftermath of the flood event.

4. Conclusions

This research demonstrates that 2D hydro-morphodynamic modeling is a robust approach for analyzing sediment dynamics and channel–floodplain interactions in extensive riverine floodplains. The calibrated HEC-RAS 2D model for the Gemenc floodplain successfully reproduced water levels, discharge, SSC, and deposition patterns during a significant flood event. The main takeaways include the following:
Roughness calibration is crucial: The forested floodplain zones and the main channel exhibited the highest sensitivity to Manning’s n values, significantly influencing water flow patterns and water surface elevations (WSEs).
The floodplain acts as a natural sediment filter: Suspended sediment concentrations diminished downstream, and deposition patterns indicated higher sedimentation close to the riverbanks, enhancing the floodplain’s trapping efficiency.
Hydrograph phase impacts sediment behavior: The study revealed that peak suspended sediment concentration (SSC) occurred not during the peak discharge, but on the falling limb of the hydrograph. This highlights the importance of accounting for temporal variations in sediment transport dynamics. Such delayed sediment peaks are likely driven by processes like sediment resuspension, exhaustion of upstream sources, and changes in flow velocity during recession stages. Restoration planning should therefore consider not only peak flows but also the timing of sediment mobilization and deposition phases, as these influence sediment trapping efficiency and floodplain functionality over the course of an entire flood event.
Practical Implications:
The results provide crucial guidance for developing nature-based restoration strategies, including enhancing lateral connectivity, regulating vegetation to influence roughness, and pinpointing critical sediment retention areas. These results support the expansion of floodplain management strategies that facilitate biodiversity, improve water quality, mitigate excessive sediment deposition in the floodplain, and reduce incising in the main river.
Model Limitations and Future Work:
While the model effectively simulates overall trends, limitations include simplified sediment input assumptions, limited availability of sediment data, and reliance on visual stratigraphy for deposition validation. Future work should integrate higher-resolution sediment measurements and explore 3D modeling approaches in critical zones. Long-term monitoring will be vital to validate model predictions and guide adaptive management for floodplain restoration in complex, regulated river systems like the Danube River.

Author Contributions

Conceptualization, D.M.H. and S.B.; methodology, D.M.H., E.A.T. and S.B.; software, D.M.H.; validation, D.M.H. and S.B.; formal analysis, D.M.H.; investigation, D.M.H., V.F., F.P. and E.A.T.; resources, D.M.H., F.P. and S.B.; data curation, D.M.H. and S.B.; writing—original draft preparation, D.M.H. and S.B.; writing—review and editing, F.P., V.F., E.A.T. and S.B.; visualization, D.M.H. and S.B.; supervision, S.B.; project administration, S.B.; funding acquisition, S.B. All authors have read and agreed to the published version of the manuscript.

Funding

The research presented in the article was carried out within the framework of the Széchenyi Plan Plus program with the support of the RRF 2.3.1 21 2022 00008 project.

Data Availability Statement

Raw data underlying the findings of this study are available from the authors upon reasonable request.

Conflicts of Interest

The authors declare no conflicts of interest. The funders had no role in the design of the study in the collection, analysis or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

References

  1. Habersack, H.; Baranya, S.; Holubová, K.; Vartolomei, F.; Skiba, H.; Babic-Mladenovic, M.; Cibilic, A.; Schwarz, U.; Krapesch, M.; Gmeiner, P.; et al. Danube Sediment Management Guidance. Interreg Danube Transnational Project Danube Sediment co-funded by the European Commission Vienna. 2019. Available online: https://dtp.interreg-danube.eu/uploads/media/approved_project_output/0001/39/ee566924f1764d4798dc7bb9b59537ce84d98101.pdf (accessed on 15 December 2024).
  2. Dadson, S.J.; Hall, J.W.; Murgatroyd, A.; Acreman, M.; Bates, P.; Beven, K.; Heathwaite, L.; Holden, J.; Holman, I.P.; Lane, S.N.; et al. A restatement of the natural science evidence concerning catchment-based ‘natural’ flood management in the UK. Proc. R. Soc. A Math. Phys. Eng. Sci. 2017, 473, 20160706. [Google Scholar] [CrossRef]
  3. Lenhart, C.F. A Preliminary Review of NOAA’s Community-Based Dam Removal and Fish Passage Projects. Coast. Manag. 2003, 31, 79–98. [Google Scholar] [CrossRef]
  4. Lorenz, A.W.; Jähnig, S.C.; Hering, D. Re-Meandering German Lowland Streams: Qualitative and Quantitative Effects of Restoration Measures on Hydromorphology and Macroinvertebrates. Environ. Manag. 2009, 44, 745–754. [Google Scholar] [CrossRef] [PubMed]
  5. Opperman, J.J.; Galloway, G.E.; Fargione, J.; Mount, J.F.; Richter, B.D.; Secchi, S. Sustainable Floodplains Through Large-Scale Reconnection to Rivers. Science 2009, 326, 1487–1488. [Google Scholar] [CrossRef] [PubMed]
  6. Roni, P.; Hanson, K.; Beechie, T. Global Review of the Physical and Biological Effectiveness of Stream Habitat Rehabilitation Techniques. N. Am. J. Fish. Manag. 2008, 28, 856–890. [Google Scholar] [CrossRef]
  7. Naiman, R.J.; Décamps, H. The Ecology of Interfaces: Riparian Zones. Annu. Rev. Ecol. Evol. Syst. 1997, 28, 621–658. [Google Scholar] [CrossRef]
  8. Zedler, J.B.; Kercher, S. Wetland Resources: Status, Trends, Ecosystem Services, and Restorability. Annu. Rev. Environ. Resour. 2005, 30, 39–74. [Google Scholar] [CrossRef]
  9. Anees, M.T.; Abdullah, K.; Nawawi, M.N.M.; Rahman, N.N.N.A.; Piah, A.R.M.; Zakaria, N.A.; Syakir, M.I.; Omar, A.K.M. Numerical modeling techniques for flood analysis. J. Afr. Earth Sci. 2016, 124, 478–486. [Google Scholar] [CrossRef]
  10. Dixon, S.J.; Sear, D.A.; Odoni, N.A.; Sykes, T.; Lane, S.N. The effects of river restoration on catchment scale flood risk and flood hydrology. Earth Surf. Process. Landf. 2016, 41, 997–1008. [Google Scholar] [CrossRef]
  11. Guan, M.; Ahilan, S.; Yu, D.; Peng, Y.; Wright, N. Numerical modelling of hydro-morphological processes dominated by fine suspended sediment in a stormwater pond. J. Hydrol. 2018, 556, 87–99. [Google Scholar] [CrossRef]
  12. Brennan, M.L.; May, C.L.; Danmeier, D.G.; Crooks, S.; Haltiner, J.H. Numerical Modeling of Restoration Alternatives in an Erosional Estuary. In Proceedings of the 10th International Conference on Estuarine and Coastal Modeling, Newport, RI, USA, 5–7 November 2007. [Google Scholar]
  13. Clilverd, H.M.; Thompson, J.R.; Heppell, C.M.; Sayer, C.D.; Axmacher, J.C. Coupled Hydrological/Hydraulic Modelling of River Restoration Impacts and Floodplain Hydrodynamics. River Res. Appl. 2016, 32, 1927–1948. [Google Scholar] [CrossRef]
  14. Acreman, M.C.; Riddington, R.; Booker, D.J. Hydrological impacts of floodplain restoration: A case study of the River Cherwell, UK. Hydrol. Earth Syst. Sci. 2003, 7, 75–85. [Google Scholar] [CrossRef]
  15. Willis, T.D.M.; Smith, M.W.; Cross, D.E.; Hardy, A.J.; Ettrich, G.E.; Malawo, H.; Chalo, C.; Sinkombo, M.; Thomas, C.J. Hydrodynamic Modeling of Inundation Patterns of a Large African Floodplain Indicates Sensitivity to Waterway Restoration. Water Resour. Res. 2022, 58, e2021WR030107. [Google Scholar] [CrossRef]
  16. Leyer, I.; Mosner, E.; Lehmann, B. Managing floodplain-forest restoration in European river landscapes combining ecological and flood-protection issues. Ecol. Appl. 2012, 22, 240–249. [Google Scholar] [CrossRef]
  17. Gourevitch, J.D.; Singh, N.K.; Minot, J.; Raub, K.B.; Rizzo, D.M.; Wemple, B.C.; Ricketts, T.H. Spatial targeting of floodplain restoration to equitably mitigate flood risk. Glob. Environ. Chang. 2020, 61, 102050. [Google Scholar] [CrossRef]
  18. Maaß, A.-L.; Schüttrumpf, H. Reactivation of Floodplains in River Restorations: Long-Term Implications on the Mobility of Floodplain Sediment Deposits. Water Resour. Res. 2019, 55, 8178–8196. [Google Scholar] [CrossRef]
  19. Ahilan, S.; Guan, M.; Sleigh, A.; Wright, N.; Chang, H. The influence of floodplain restoration on flow and sediment dynamics in an urban river. J. Flood Risk Manag. 2016, 11, S986–S1001. [Google Scholar] [CrossRef]
  20. González-Sanchis, M.; Murillo, J.; Latorre, B.; Comín, F.; García-Navarro, P. Transient Two-Dimensional Simulation of Real Flood Events in a Mediterranean Floodplain. J. Hydraul. Eng. 2012, 138, 629–641. [Google Scholar] [CrossRef]
  21. Poulsen, J.B.; Hansen, F.; Ovesen, N.B.; Larsen, S.E.; Kronvang, B. Linking floodplain hydraulics and sedimentation patterns along a restored river channel: River Odense, Denmark. Ecol. Eng. 2014, 66, 120–128. [Google Scholar] [CrossRef]
  22. Sabatine, S.M.; Niemann, J.D.; Greimann, B.P. Evaluation of Parameter and Model Uncertainty in Simple Applications of a 1D Sediment Transport Model. J. Hydraul. Eng. 2015, 141, 04015002. [Google Scholar] [CrossRef]
  23. Papanicolaou, A.T.N.; Elhakeem, M.; Krallis, G.; Prakash, S.; Edinger, J.; Papanicolaou, A.T.N.; Elhakeem, M.; Krallis, G.; Prakash, S.; Edinger, J. Sediment Transport Modeling Review—Current and Future Developments. J. Hydraul. Eng. 2008, 134, 1–14. [Google Scholar] [CrossRef]
  24. Roni, P.; Hall, J.E.; Drenner, S.M.; Arterburn, D. Monitoring the effectiveness of floodplain habitat restoration: A review of methods and recommendations for future monitoring. WIRESs Water 2019, 6, e1355. [Google Scholar] [CrossRef]
  25. Clare, M.C.A.; Kramer, S.C.; Cotter, C.J.; Piggott, M.D. Calibration, inversion and sensitivity analysis for hydro-morphodynamic models through the application of adjoint methods. Comput. Geosci. 2022, 163, 105104. [Google Scholar] [CrossRef]
  26. Simons, R.K.; Canali, G.E.; Anderson-Newton, G.T.; Cotton, G.K. Sediment Transport Modeling: Calibration, Verification, and Evaluation. Soil Sediment Contam. 2000, 9, 427–434. [Google Scholar] [CrossRef]
  27. Pomázi, F.; Baranya, S. Comparative Assessment of Fluvial Suspended Sediment Concentration Analysis Methods. Water 2020, 12, 873. [Google Scholar] [CrossRef]
  28. Williams, J.J.; Esteves, L.S. Guidance on Setup, Calibration, and Validation of Hydrodynamic, Wave, and Sediment Models for Shelf Seas and Estuaries. Adv. Civ. Eng. 2017, 2017, 5251902. [Google Scholar] [CrossRef]
  29. Guti, G. Water bodies in the Gemenc floodplain of the Danube, Hungary (A theoretical basis for their typology). Opusc. Zool. 2001, 33, 49–60. [Google Scholar]
  30. Loczy, D. The Changing Geomorphology of Danubian Floodplains in Hungary. Hrvat. Geogr. Glas. 2007, 69, 5–19. [Google Scholar] [CrossRef]
  31. Berczik, Á.; Buzetzky, G. Realistic restoration of the Gemenc region of the Danubian floodplain based on hydroecological priorities. Int. Ver. Theor. Angew. Limnol. Verhandlungen 2006, 29, 1595–1599. [Google Scholar] [CrossRef]
  32. Tamás, E.A.; Kovács, P. Changes in the Water Regime of the River Danube and their Implications on the Safe Operation of Water Management Installations and the Sustainable Management of Floodplains. In Környezeti Biztonság; Ludovika Egyetemi Kiadó: Budapest, Hungary, 2020; Available online: https://www.researchgate.net/publication/349088955_Changes_in_the_Water_Regime_of_the_River_Danube_and_their_Implications_on_the_Safe_Operation_of_Water_Management_Installations_and_the_Sustainable_Management_of_Floodplains (accessed on 15 December 2024).
  33. Tamás, E.A.; Djordjević, D.; Kalocsa, B.; Vujanović, A. Hydrological indicators of the riverbed incision along the free-flowing Danube River reach from Budapest to Slankamen relevant for the lateral connectivity between the river channel and floodplains. In Proceedings of the Conference: Rivers and Floodplains in the Anthropocene—Upcoming Challenges in the Danube River Basin: 43rd IAD Conference, Neuburg an der Donau, Germany, 9–11 June 2021. [Google Scholar]
  34. Füstös, V.; Baranya, S.; Kutassy, E.; Tamás, E.A.; Erős, T.; Józsa, J. Validating a Two-dimensional Sediment Transport Model on a Large Danubian Floodplain. Period. Polytech. Civ. Eng. 2024, 68, 812–820. [Google Scholar] [CrossRef]
  35. Józsa, J.; Gáspár, C.; Szél, S. Numerical modelling of flow patterns to assist the revitalization of secondary river branches in Gemenc area. Period. Polytech. Civ. Eng. 1994, 38, 187–203. [Google Scholar]
  36. Schöll, K.; Kiss, A. Checklist of the planktonic rotifer fauna in the active floodplain area of the Danube (1843–1806, 1669 and 1437–1489 rkm). Opusc. Zool. 2009, 40, 63–73. [Google Scholar]
  37. Kiss, A.; Ágoston-Szabó, E.; Dinka, M.; Berczik, Á. Microcrustacean diversity in the Gemenc-Béda-Karapancsa Floodplains (Danube-Drava National Park, Hungary): Rare and alien species. Opusc. Zool. 2015, 46, 183–197. [Google Scholar] [CrossRef]
  38. Schoor, M.M. The Relation between Vegetation and Hydrology/Geomorphology in the Gemenc Floodplain Forest, Hungary. Water Sci. Technol. 1994, 29, 289–291. [Google Scholar] [CrossRef]
  39. Tadić, L.; Tamás, E.A.; Mihaljević, M.; Janjić, J. Potential Climate Impacts of Hydrological Alterations and Discharge Variabilities of the Mura, Drava, and Danube Rivers on the Natural Resources of the MDD UNESCO Biosphere Reserve. Climate 2022, 10, 139. [Google Scholar] [CrossRef]
  40. Kalcheva, H.; Dinka, M.; Ágoston-Szabó, E.; Berczik, Á.; Kalchev, R.; Tarjanyi, N.; Kiss, A. Bacterioplankton from Two Hungarian Danube River Wetlands (Beda-Karapancsa, Danube-Drava National Park) and its Relations to Environmental Variables. Transylv. Rev. Syst. Ecol. Res. 2016, 18, 1–12. [Google Scholar] [CrossRef]
  41. Wu, W.; Wang, S.S.Y.; Jia, Y. Nonuniform sediment transport in alluvial rivers. J. Hydraul. Res. 2000, 38, 427–434. [Google Scholar] [CrossRef]
  42. Brocchini, M.; Dodd, N. Nonlinear Shallow Water Equation Modeling for Coastal Engineering. J. Waterw. Port Coast. Ocean Eng. 2008, 134, 104–120. [Google Scholar] [CrossRef]
  43. Xing, Y. Numerical Methods for the Nonlinear Shallow Water Equations. Handb. Numer. Anal. 2017, 18, 361–384. [Google Scholar] [CrossRef]
  44. Sanchez, A.; Gibson, S. HEC-RAS 2D Sediment Technical Reference Manual; U.S. Army Corps of Engineers, Institute for Water Resources, Hydrologic Engineering Center: Davis, CA, USA, 2023. [Google Scholar]
  45. Shabani, A.; Woznicki, S.A.; Mehaffey, M.; Butcher, J.; Wool, T.A.; Whung, P.Y. A coupled hydrodynamic (HEC-RAS 2D) and water quality model (WASP) for simulating flood-induced soil, sediment, and contaminant transport. J. Flood Risk Manag. 2021, 14, e12747. [Google Scholar] [CrossRef]
  46. Pratama, F.; Wulandari, S.; Rohmat, F.I.W. Modeling sediment accumulation in Pare Reservoir using HEC-RAS 2D: Assessing storage capacity over a 10-year period. Results Eng. 2025, 25, 104333. [Google Scholar] [CrossRef]
  47. Samaga, B.R.; Raju, K.G.R.; Garde, R.J. Bed Load Transport of Sediment Mixtures. J. Hydraul. Eng. 1986, 112, 1003–1017. [Google Scholar] [CrossRef]
  48. Kuhnle, R.A. Fluvial transport of sand and gravel mixtures with bimodal size distributions. Sediment. Geol. 1993, 85, 17–24. [Google Scholar] [CrossRef]
  49. Wilcock, P.R.; McArdell, B.W. Surface-based fractional transport rates: Mobilization thresholds and partial transport of a sand-gravel sediment. Water Resour. Res. 1993, 29, 1297–1312. [Google Scholar] [CrossRef]
  50. Domhof, B.C.A.; Berends, K.D.; Spruyt, A.; Warmink, J.J.; Hulscher, S.J.M.H. Discharge and location dependency of calibrated main channel roughness: Case study on the River Waal. In Proceedings of the River Flow 2018: 9th International Conference on Fluvial Hydraulics, Lyon, France, 5–8 September 2018. [Google Scholar]
  51. Vas, L.; Tamás, E.A. Surrogate Method for Suspended Sediment Concentration Monitoring on the Alluvial Reach of the River Danube (Baja, Hungary). Appl. Sci. 2023, 13, 5826. [Google Scholar] [CrossRef]
  52. Erikson, L.H.; Wright, S.A.; Elias, E.; Hanes, D.M.; Schoellhamer, D.H.; Largier, J. The use of modeling and suspended sediment concentration measurements for quantifying net suspended sediment transport through a large tidally dominated inlet. Mar. Geol. 2013, 345, 96–112. [Google Scholar] [CrossRef]
  53. Stajnko, J.K.; Jecl, R.; Perc, M.N. Advances in Monitoring and Understanding the Dynamics of Suspended-Sediment Transport in the River Drava, Slovenia: An Analysis More than a Decade-Long. Appl. Sci. 2023, 13, 9036. [Google Scholar] [CrossRef]
  54. Haimann, M.; Hauer, C.; Tritthart, M.; Prenner, D.; Leitner, P.; Moog, O.; Habersack, H.; Haimann, M.; Hauer, C.; Tritthart, M.; et al. Monitoring and modelling concept for ecological optimized harbour dredging and fine sediment disposal in large rivers. Hydrobiologia 2016, 814, 89–107. [Google Scholar] [CrossRef]
  55. Varvani, J.; Khaleghi, M.R.; Gholami, V. Investigation of the Relationship between Sediment Graph and Hydrograph of Flood Events (Case Study: Gharachay River Tributaries, Arak, Iran). Water Resour. 2019, 46, 883–893. [Google Scholar] [CrossRef]
  56. Rosen, T.; Xu, Y.J. A Hydrograph-Based Sediment Availability Assessment: Implications for Mississippi River Sediment Diversion. Water 2014, 6, 564–583. [Google Scholar] [CrossRef]
  57. Conroy, E.; Turner, J.N.; Rymszewicz, A.; Bruen, M.; O׳Sullivan, J.J.; Kelly-Quinn, M. An evaluation of visual and measurement-based methods for estimating deposited fine sediment. Int. J. Sediment Res. 2016, 31, 368–375. [Google Scholar] [CrossRef]
  58. Asselman, N.E.M. Suspended sediment dynamics in a large drainage basin: The River Rhine. Hydrol. Process. 1999, 13, 1437–1450. [Google Scholar] [CrossRef]
  59. Walling, D.E.; He, Q. The spatial variability of overbank sedimentation on river floodplains. Geomorphology 1998, 24, 209–223. [Google Scholar] [CrossRef]
  60. Nicholas, A.P.; Walling, D.E. Modelling flood hydraulics and overbank deposition on river floodplains. Earth Surf. Process. Landf. 1997, 22, 59–77. [Google Scholar] [CrossRef]
  61. Pizzuto, J.E. Sediment diffusion during overbank flows. Sedimentology 1987, 34, 301–317. [Google Scholar] [CrossRef]
Water 17 02399 g001
Figure 1. Location and characteristics of the Gemenc floodplain, Hungary: (A) map of Hungary showing the study area; (B) land cover distribution; (C) terrain elevation. (The white dot indicates the Baja Hydrographic Station.)
Figure 1. Location and characteristics of the Gemenc floodplain, Hungary: (A) map of Hungary showing the study area; (B) land cover distribution; (C) terrain elevation. (The white dot indicates the Baja Hydrographic Station.)
Water 17 02399 g001
Water 17 02399 g002
Figure 2. Sediment-grain-size-distribution curves.
Figure 2. Sediment-grain-size-distribution curves.
Water 17 02399 g002
Water 17 02399 g003
Figure 3. Boundary condition time series for the December 2023 and September 2024 flood events. (Left vertical axis is discharge, and right is water surface water elevation.)
Figure 3. Boundary condition time series for the December 2023 and September 2024 flood events. (Left vertical axis is discharge, and right is water surface water elevation.)
Water 17 02399 g003
Water 17 02399 g004
Figure 4. Suspended sediment load rating curve. (SSL: suspended sediment load, ton per day.)
Figure 4. Suspended sediment load rating curve. (SSL: suspended sediment load, ton per day.)
Water 17 02399 g004
Water 17 02399 g005
Figure 5. Measured vs. model water surface elevation time series for December 2023 and September 2024 floods (at Baja station).
Figure 5. Measured vs. model water surface elevation time series for December 2023 and September 2024 floods (at Baja station).
Water 17 02399 g005
Water 17 02399 g006
Figure 6. Measured (m) and simulated (sim) discharges at cross-sections of the Danube River and its side branches in the Gemenc floodplain during the September 2024 flood. (A) Overall flow direction in the main river and side branches; (B) discharge magnitude and direction in the Sió channel and Danube main channel upstream of the Sió; (C) discharge magnitude and direction in the Danube and Grébec branch upstream of Grébec; (D) discharge magnitude and direction upstream of the Rezéti-Duna inflow; (E) discharge magnitude and direction downstream of the Rezéti-Duna confluence.
Figure 6. Measured (m) and simulated (sim) discharges at cross-sections of the Danube River and its side branches in the Gemenc floodplain during the September 2024 flood. (A) Overall flow direction in the main river and side branches; (B) discharge magnitude and direction in the Sió channel and Danube main channel upstream of the Sió; (C) discharge magnitude and direction in the Danube and Grébec branch upstream of Grébec; (D) discharge magnitude and direction upstream of the Rezéti-Duna inflow; (E) discharge magnitude and direction downstream of the Rezéti-Duna confluence.
Water 17 02399 g006
Water 17 02399 g007
Figure 7. Time series of observed and simulated flow discharges at key cross-sections along the Danube River and its side branches during the September 2024 flood in the Gemenc floodplain. (A) Sió channel and Danube main channel upstream of the Sió; (B) Danube upstream of Grébec and Grébec branch; (C) upstream of the Rezéti-Duna inflow and Rezéti branch; (D) downstream of the Rezéti-Duna confluence and branch.
Figure 7. Time series of observed and simulated flow discharges at key cross-sections along the Danube River and its side branches during the September 2024 flood in the Gemenc floodplain. (A) Sió channel and Danube main channel upstream of the Sió; (B) Danube upstream of Grébec and Grébec branch; (C) upstream of the Rezéti-Duna inflow and Rezéti branch; (D) downstream of the Rezéti-Duna confluence and branch.
Water 17 02399 g007
Water 17 02399 g008
Figure 8. Modeled vs. measured SSC.
Figure 8. Modeled vs. measured SSC.
Water 17 02399 g008
Water 17 02399 g009
Figure 9. Time-series comparison of modeled vs. measured SSC in the Gemenc area. (A) DunaST, Grébec, Rezéti-F (RezetF), and Rezéti-A (RezetA); (B) VDF and VDA.
Figure 9. Time-series comparison of modeled vs. measured SSC in the Gemenc area. (A) DunaST, Grébec, Rezéti-F (RezetF), and Rezéti-A (RezetA); (B) VDF and VDA.
Water 17 02399 g009
Water 17 02399 g010
Figure 10. Measured and simulated sediment deposition thickness during the September 2024 flood event. (A) Locations of sediment sample collection; (B) comparison of measured and simulated sediment deposition thickness; (C) recent deposition layer assumed to have formed during the September 2024 flood.
Figure 10. Measured and simulated sediment deposition thickness during the September 2024 flood event. (A) Locations of sediment sample collection; (B) comparison of measured and simulated sediment deposition thickness; (C) recent deposition layer assumed to have formed during the September 2024 flood.
Water 17 02399 g010
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Hawez, D.M.; Füstös, V.; Pomázi, F.; Tamás, E.A.; Baranya, S. Simulation of Sediment Dynamics in a Large Floodplain of the Danube River. Water 2025, 17, 2399. https://doi.org/10.3390/w17162399

AMA Style

Hawez DM, Füstös V, Pomázi F, Tamás EA, Baranya S. Simulation of Sediment Dynamics in a Large Floodplain of the Danube River. Water. 2025; 17(16):2399. https://doi.org/10.3390/w17162399

Chicago/Turabian Style

Hawez, Dara Muhammad, Vivien Füstös, Flóra Pomázi, Enikő Anna Tamás, and Sándor Baranya. 2025. "Simulation of Sediment Dynamics in a Large Floodplain of the Danube River" Water 17, no. 16: 2399. https://doi.org/10.3390/w17162399

APA Style

Hawez, D. M., Füstös, V., Pomázi, F., Tamás, E. A., & Baranya, S. (2025). Simulation of Sediment Dynamics in a Large Floodplain of the Danube River. Water, 17(16), 2399. https://doi.org/10.3390/w17162399

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Article metric data becomes available approximately 24 hours after publication online.
Back to TopTop