1. Introduction
Coastal erosion poses a paramount challenge for coastal communities globally. Besides being valuable natural resources and recreational areas, beaches play a crucial role in protecting against extreme events, maintaining biodiversity, and supporting various economic activities such as tourism and fishing. However, coastal erosion continually threatens these areas, resulting in significant land, infrastructure, and habitat losses [
1,
2].
While traditional structures like seawalls, groins, and breakwaters may protect specific stretches, they often exacerbate erosion by disrupting natural sediment transport processes, leading to negative impacts elsewhere [
3,
4]. Despite ongoing efforts to mitigate the effects of coastal erosion, these approaches often prove insufficient in the face of growing pressures from climate change, sea level rise, and human activities. On the other hand, these structures, especially groins, significantly influence longshore sediment transport dynamics, often leading to morphological changes along coastlines. A common consequence of interrupting natural sand transport is the accretion of beaches updrift and erosion downdrift, exacerbating shoreline recession and threatening coastal communities and ecosystems at those locations [
5,
6].
This downdrift erosion, a focal point of coastal degradation, poses challenges for effective erosion management strategies, highlighting the urgency to comprehend its processes and magnitudes.
Existing studies, including those by Pelnard-Considère (cited in [
7]) and Bruun [
7], have conceptualized the evolution of downdrift erosional shorelines, emphasizing their nonlinear expansion and characteristic “S” shape. However, predicting the spatial and temporal behaviours of downdrift erosion remains challenging.
When considering the complex interactions between waves, sediment transport, and coastal morphology, it becomes evident that a nuanced understanding of these processes is crucial for effective coastal management. For medium- to long-term studies, one-dimensional numerical models, commonly referred to as one-line models, offer a complementary perspective to empirical and detailed numerical approaches. These models provide valuable insights by simplifying the representation of coastal dynamics along either the alongshore or cross-shore direction. They focus on key processes such as wave propagation, sediment transport, and morphological changes within a computationally efficient framework [
8,
9,
10,
11].
Globally, current management strategies tend to compartmentalize sand beach usage, often failing to integrate tourism and conservation effectively, especially in high-energetic coastlines [
1]. Conflicts of interest arise, particularly in regions where diverse activities coexist within limited coastal spaces. Historically, beach management has prioritized recreational aspects over ecological preservation. However, recent approaches signify a paradigm shift towards more comprehensive beach management frameworks [
12]. In this regard, artificial sand bypass solutions have been employed globally, aiming to restore natural sediment transport patterns and mitigate the negative impacts of coastal structures [
3,
6].
Despite lacking specific policies in some regions, integrated coastal management is gaining affiliates as a viable approach to bridge the gap between recreational demands and ecological conservation. Previous studies have delineated management strategies into protection, regulation, and restoration categories, with an increasing emphasis on the latter, particularly through beach nourishment initiatives [
13]. Climate change has exacerbated existing challenges, influencing shoreline dynamics through erosion, flooding, and long-term shifts in weather patterns [
14,
15,
16,
17]. As such, research endeavours to refine beach management strategies by incorporating a Threat Index (TI), which integrates erosion rates and shoreline recession due to sea level rise, alongside physical, environmental, and socio-economic factors [
1].
Moreover, innovative solutions, such as artificial structures designed to induce longshore drift reversal, are being explored, mimicking natural sediment accumulation processes. These structures aim to mitigate erosion and enhance coastal stability while considering hydrodynamic and morphodynamic implications [
4].
Achieving sustainable solutions for coastal management necessitates a proactive approach that anticipates local morphodynamics, especially in the vicinity of coastal defence structures, under storm conditions as a primary driver of beach morphodynamics.
This study comprehensively analyses coastal erosion issues during storm events around an innovative semicircular coastal defence structure using the XBeach model [
18]. The effectiveness in reversing sediment drift is evaluated, which is a promising method to strategically accumulate sediments downdrift of these structures to stabilize beaches and reduce erosion. Furthermore, this study explores how the gap spacing between these structures affects coastal dynamics, particularly in relation to sediment accumulation and erosion rates along specific coastline sections. This analysis considers sedimentary factors and the complex interactions between waves and their breaking patterns. These are essential for understanding beach behaviour and proposing proper design rules for sustainable coastal engineering practices.
The sedimentation and erosion dynamics analysed for the gap spacing of semicircular structures shed light on critical factors that have significant implications for both scientific understanding and practical applications by elucidating the relationships between structure gap spacing and sedimentation/erosion rates in different sections of the beach.
The observed trends highlight the importance of considering spatial configurations of structures in coastal management initiatives. Specifically, the findings underscore the need for a nuanced approach that accounts for the varying responses of different coastal sections to alterations in structure gap spacing. Such insights are invaluable for optimizing the design and implementation of coastal restoration projects, ensuring their effectiveness in mitigating sedimentation and erosion while preserving ecological integrity.
Furthermore, this study underscores the dynamic nature of coastal systems and the importance of adaptive management strategies to prevent undesirable outcomes during storm events.
By investigating these aspects, this study seeks to advance scientific knowledge and develop more efficient and sustainable options for managing coastal erosion, with the ultimate goal of preserving vital ecosystems and protecting communities dependent on them. This study aims to inform future coastal management practices and sustainable shoreline management through proposing and analysing a new set of coastal defence structures with the aid of comprehensive simulation.
2. Materials and Methods
This study applied the XBeach model to analyse the influence of gap spacing between semicircular structures on significant wave height, Hs, wave direction, wave energy dissipation, sediment volume variation, and final bed configuration. This model allowed for a comprehensive analysis, considering general behaviour and specific sections, establishing relationships among the scenarios under study.
2.1. Validation of the Model
The validation of the model aimed to assess its qualitative performance, particularly as the proposed semicircular structure had only undergone numerical testing. As the primary objective was to assess the impact of the new shape within a generic bathymetric context, the validation of the groins employed this generic bathymetry, focusing specifically on sedimentation/erosion patterns. Measured morphodynamic data sources were incorporated into the validation process to ensure that the model accurately reflected real-world conditions. This included measured bathymetric data at a coastal stretch influenced by a coastal defence structure and consistent local wave patterns, allowing for a sound assessment of the coastal domain’s morphology and wave climate. Real-world data from a groin (
Figure 1) located at Castelo de Trancoso beach in Gafanha da Encarnação, Aveiro, Portugal, was compared with a modelled groin with similar characteristics.
The bathymetric data, obtained through topo-hydrographic surveys provided by the COSMO Programme [
19], supplied essential information instrumental in validating the model. Although the available data spanned one year for the Castelo de Trancoso beach groin, a continuous 48 h storm simulation was chosen due to the significant computational demands of more extended periods. The wave climate parameters (H
s, peak wave period, T
p, and dominant wave direction) were based on data from a previous study conducted by the authors that characterized local wave patterns along the western coast of the Iberian Peninsula. This approach ensured both consistency and reliability in the model’s representation of wave conditions [
20]. The reliability and strength of sediment dynamics simulations in this study were bolstered by the thorough validation of the model in analogous coastal research works [
21,
22,
23,
24,
25].
After calibrating the flow parameters, the validation process involved qualitatively comparing bathymetric variations over their respective time frames. Bathymetric changes for the real groin were assessed using ArcGIS [
26] software, and then qualitatively compared with the simulation results. The bathymetric variations included the sedimentation and erosion trends surrounding a groin. The modelled groin presented height, length, slope, beach bathymetry, and topography similar to the real groin (
Figure 2). Adopting consistent parameters in XBeach between the real groin and the innovative structure facilitated model validation.
2.2. Semicircular Innovative Coastal Defence Structure
The innovative structure geometry (
Table 1) made out of armour units had previously been presented and tested [
4], exploring how its shape and dimensions influence the reversal of littoral drift, facilitating greater sediment accumulation downdrift. This structure is connected to the shoreline and features a rounded ledge with a slope of 1:10 and a crest level of 0 m, which offers the potential for safeguarding urban areas along the coast by attenuating or inducing a reversal of the longitudinal drift that is responsible for erosion in many coastal stretches. The adopted length (509 m) and width (217 m) dimensions have been specifically chosen and adjusted to ensure the drift reversal effect. Additionally, this geometry aids in sediment accumulation near the shoreline, and the low crest level contributes to mitigating environmental visual impact concerns since the structure will only be visible along the low tide period. The 1:10 slope aids in reducing the abruptness of wave breaking compared to other slopes (wall and 1:2). While the sedimentary distribution of this structure is generally satisfactory, some scouring may occur at its extremities. However, stability can be ensured by placing the foundation deeper than a safer scour depth.
While this structure has some limitations, it offers notable improvements in sediment accumulation, particularly when compared to groins. Unlike groins, which hinder sediment transport from up- to downdrift, causing sediment buildup only at a considerable distance downdrift, the proposed structure facilitates sediment accumulation both up- and downdrift. This feature stands out as its primary advantage.
In this study, the varying gap spacings between a pair of these structures are analysed to assess their influence on sediment transport near the shoreline. Gap spacing between structures is determined according to DEFRA [
27] guidelines, ensuring a consistent and reliable setup.
Although these guidelines are specifically recommended for the design of detached breakwaters, they can also provide a framework for sizing the current structures. These gaps are calculated based on the allowable maximum shoreline erosion, which can be derived from existing design curves [
28]. Equation (1) represents the maximum permissible shoreline erosion.
In this equation, G represents the gap spacing between the structures, and X is the distance from the beach to the centreline of the structure. Considering these structures’ large size and asymmetry along the Y-axis (unlike typical detached breakwaters), it has been determined that X should be measured from the beach to the edge of the structure. The value of X is calculated by adding the width of the structure to the distance from the structure to the shoreline (217 m in the present case).
Selecting appropriate gap spacing is crucial to prevent extensive erosion zones.
Table 2 outlines the chosen values for each maximum shoreline erosion under conditions of certain erosion, no erosion, and possible erosion. The gap spacings for the selected scenarios are also presented. Additional information regarding the calculations supporting these estimations can be found in another source [
4].
2.3. XBeach Model and Numerical Model Conditions
XBeach is an open-source numerical model with a process-driven morphodynamic numerical model, specialized for nearshore and coastal dynamics. Initially designed to replicate hydrodynamic and morphological processes during storm waves on sandy island systems, its application has been expanded to include the interaction between moderate wave conditions and coastal structures. Widely recognized for its effectiveness, XBeach has been successfully used in numerous studies across sandy coastlines [
4,
21,
22,
23,
24,
25].
The input conditions for the model (
Table 3), comprising mean values of H
s, T
p, dominant wave direction and storm duration, were based on a previous study regarding local wave patterns that characterize the NW Portuguese coast [
20]. The parameters associated with flow (roughness and viscosity), tide, wave spectrum, sediment density, and morphology updating (avalanching and bottom friction) were predetermined based on previous simulations conducted by the authors [
4,
21]. These parameters are further detailed in the manuals provided by the respective model developers [
18]. A morphological acceleration factor (morfac) of 10 was selected to expedite calculations, allowing for assessing a prolonged storm lasting 24 h. A zero sediment layer thickness was specified at the structure’s location, indicating non-erodibility, corresponding to using artificial concrete armour blocks.
The coastal domain must possess dimensions that mimic real-world scenarios accurately. However, since the primary focus was to examine a generic coastal segment for implementing a coastal protection solution rather than analysing a specific location’s bathymetry, this study focused on a simplified bathymetry known as reference bathymetry (S0). The chosen dimensions for the domain (
Table 3) prevented computational boundary instabilities that could have impacted the areas where the structures are located. The selected grid cell size was previously validated as suitable for similar numerical simulations [
4,
21]. The reference bathymetry domain featured a maximum offshore depth of −31.4 m and a maximum coastal elevation of +2.0 m, considering the reference vertical datum coincident with the mean sea level. This elevation of +2.0 m enabled a thorough examination of the beach’s hydro- and morphodynamic impacts, especially since the tidal level was fixed at 0 m for all the presented scenarios.
The sediment composition of the seabed (
Table 3) was characterized by fine sand, with particle sizes (D
50 = 0.2 mm and D
90 = 0.3 mm) conforming to the Unified Soils Classification of Sediments [
29]. Boundary conditions were set to ensure stability: the Neumann boundary maintained elevation and velocity, ensuring continuity without instabilities near the north and south boundaries. An absorbing condition was applied to the east boundary to minimize reflections, while the west boundary remained open to simulate wave entry into the domain.
Table 3 summarises the main numerical model conditions selected for this study.
The morphodynamic analysis focused on the volume variations of the sedimentation/erosion outcomes for the entire nearshore domain and targeted areas for a 24 h storm. This aspect held significant importance for analysing the influence of gap spacing of 304 m, 261 m, and 152 m, particularly in understanding the formation of erosion and accretion patterns near the shoreline. The variation in the bed level for selected profiles was also analysed.
The hydrodynamics analysed the location of wave breaking and demonstrated how the structure aids in dissipating wave energy across different gap spacings.
4. Discussion
4.1. Significance of the Results
The study underscored the pivotal role of structural design and gap spacing of semicircular structures in mitigating coastal wave impacts during storm events using XBeach. While coastal structures generally diminish wave energy, the gap spacing between them critically influences localized wave height changes due to diffracted and reflected waves. Hence, achieving optimal gap spacing is vital to maximize protective benefits while minimizing adverse effects within protected zones.
The observed sediment volume variations revealed the intricate interplay between gap spacing in coastal protection structures and sedimentation-erosion dynamics. Optimal gap spacing of 304 m struck a balance between sedimentation and erosion, supporting natural coastal sediment transport processes. However, site-specific factors such as wave climate, sediment characteristics, and coastal morphology must also be considered in designing effective coastal protection strategies.
Data analysis indicated that the gap spacing between semicircular structures significantly influences sedimentation and erosion. Wider gaps generally reduce sedimentation and erosion by altering wave dynamics around the structures. However, exceptions in certain stretches may suggest the influence of local topography and hydrodynamic conditions.
Firstly, it was observed that as the gap spacing between structures decreased, sedimentation increased significantly in sections N, B1, B2, and S. This phenomenon can be attributed to reduced wave energies caused by the closer proximity of structures, which allows the conditions for sediment deposition.
Conversely, a contrasting trend was noted in the stretch between structures, G, where sedimentation increased with wider gap spacing between structures. This observation suggests that in wider inter-structure spaces, there is a greater opportunity for sediment to accumulate due to decreased flow disturbance and enhanced settling conditions.
Moreover, the findings revealed a notable inverse relationship between structure gap spacing and erosion rates in sections N and B1. It was found that despite lower sedimentation levels, wider gap spacing between structures correlates with reduced erosion at these locations. This can be explained by the increased wave energy dissipation resulting from wider spacing, which promotes a reduced erosive force acting on the coastline, and a reduced wave reflection and refraction towards the coastline.
On the contrary, wider gap spacing between structures corresponded to higher erosion rates in section G. This observation was consistent with the increased wave energy and reduced obstruction to sediment transport associated with wider spacing, which exacerbates erosive processes.
Remarkably, sections B2 and S exhibited amplified erosion with a specific gap spacing of 261 m between structures. This indicates a threshold effect potentially due to wave hydrodynamic changes and channel morphology feedback.
The analysis demonstrates that gap spacing between semicircular structures profoundly affects sediment deposition and erosion patterns. Narrower gap spacings tend to induce more sediment accumulation behind the structures, likely due to wave energy dissipation. Conversely, wider gap spacings facilitate sediment transport, leading to reduced deposition immediately behind the structures, but lower erosion further away. Consistent erosion across all profiles relative to initial conditions suggests an overall destabilizing effect of the structures on bed level, highlighting the complexity of predicting structural impacts.
4.2. Limitations and Future Research
This study delves into optimizing the gap spacing of semicircular structures based on conditions favouring longitudinal drift reversal, using a generic bathymetry as a basis. Future scientific endeavours could complement this investigation by conducting specific fieldwork to gather historical data on bathymetry and sediment characteristics along a designated coastal stretch. Such efforts would yield a more comprehensive and site-specific evaluation of the optimized semicircular structure. Additionally, exploring variations in wave parameters through time series analysis of local wave climate regimes could provide valuable insights for comparison with the findings of this study.
To further elucidate these findings, future research could involve comparing the outcomes of this study with the impact of artificial sediment nourishment deposition in coastal segments, conducting a similar methodological approach. Such comparative analyses would broaden the understanding of diverse methodologies for mitigating erosive processes.
Also, it could be worthwhile to investigate the implementation of a porous structure constructed from blocks with strategically positioned apertures, forming pathways for sediment transfer from the accretion zone to erosion-prone areas. Furthermore, the impact of energy dissipation could be examined by introducing channels with significant geometric irregularity.
Additionally, exploring complementary aspects related to the economic sustainability of the semicircular structures could be a focal point for future research. Such investigations would offer valuable insights for decision-makers and stakeholders engaged in coastal protection projects by considering alternative innovative solutions for coastal defence in addition to the traditional ones.
5. Conclusions
The study underscores the critical role of semicircular structure design and gap spacing in effectively managing coastal wave impacts during storm events using the XBeach model. These structures are instrumental in attenuating wave energy, but their spacing significantly influences localized wave height changes, thus impacting sedimentation and erosion dynamics within protected zones.
This research provides valuable insights into coastal engineering design, particularly emphasizing wave-dissipating structures’ strategic placement and gap spacing. It highlights the necessity for ongoing research and adaptive management practices to safeguard coastal ecosystems against evolving environmental challenges. These findings are pivotal for managing erosion and sedimentation, stressing the importance of carefully considering structure gap spacing to achieve a balance in sediment transport and deposition in coastal engineering projects. However, it is important to note that the results and conclusions of this study are tied to the specific analysed scenarios, which are linked with the local beach profile, sediment size, structure dimensions, and limited tide and wave action conditions. The main findings should be further analysed for other hydro- and morphodynamic scenarios in future work.
Optimizing gap spacing is, therefore, paramount to maximize protective benefits while minimizing adverse effects. The analysis highlights that an optimal gap spacing of 304 m balances sedimentation and erosion, supporting natural coastal sediment transport processes. However, site-specific factors such as wave climate, sediment characteristics, and coastal morphology must be carefully considered when designing effective coastal protection strategies.
The significant influence of gap spacing between semicircular structures on sedimentation and erosion patterns shows that while narrower gap spacings induce more sediment accumulation behind the structures due to wave energy dissipation, wider gaps facilitate sediment transport, leading to reduced deposition immediately behind the structures, but lower erosion further away.
Notably, the analysis reveals complex interactions between gap spacing and sediment dynamics. Narrower gap spacings tend to exacerbate sediment accumulation in the gap section, while wider spacings promote sediment transport, affecting local and downdrift erosion patterns. The observed consistent sedimentation across all profiles relative to the reference bathymetry conditions suggests an overall stabilizing effect behind and at the gap section of the structures on bed level, highlighting a positive influence in predicting structural impacts.
In conclusion, achieving an optimal balance in gap spacing between coastal structures is essential for effective coastal protection strategies. This requires a comprehensive understanding of the intricate interplay between structural design, hydrodynamics, and sediment transport processes, ensuring sustainable coastal management practices in dynamic coastal environments.