Predicting coastal wave overtopping is a significant challenge, exacerbated by climate change, increasing the frequency of severe flooding and rising sea levels. Digital twin technologies, which utilise artificial intelligence to mimic coastal processes and dynamics, may offer new opportunities to predict coastal wave overtopping and flooding reliably and computationally efficiently. This study investigates the effectiveness of training various artificial intelligence models using wave buoy, meteorological, and recorded coastal wave overtopping observations to predict the occurrence and frequency of overtopping at 10-minute intervals. These models have the potential for future large-scale global applications in estimating wave overtopping and flood forecasting, particularly in response to climate warming. The model types selected include machine-learning random forests, extreme gradient boosting, support vector machines, and deep-learning neural networks. These models were trained and tested using recorded observational overtopping events, to estimate wave overtopping and flood forecasting in Dawlish and Penzance (Southwest England). The random forests performed exceptionally well by accurately and precisely estimating coastal wave overtopping and non-overtopping 97% of the time within both locations, outperforming the other models. Moreover, the random forest model outperforms existing process-based and EurOtop-based models. This research has profound implications for increasing preparedness and resilience to future coastal wave overtopping and flooding events by using these random forest models to predict overtopping and flood forecasting on wider global and climate scales. These trained random forests are significantly less computationally demanding than existing process-based models and can incorporate the important effect of wind on overtopping, which was neglected in existing empirical approaches.

Rip currents are the single largest cause of beach safety incidents globally, but where an estuary mouth intersects a beach, additional flows are created that can exceed the speed of a typical rip current, significantly increasing the hazard level for bathers. However, there is a paucity of observations of surfzone currents at estuary mouth beaches, and our understanding and ability to predict how the bathing hazard varies under different wave and tide conditions are therefore limited. Using field observations and process-based XBeach modelling at an embayed, estuary mouth beach, we demonstrate how surfzone currents can be driven by combinations of estuary discharge and wave-driven bathymetric and boundary rip currents under various combinations of wave and tide forcing. While previous studies have demonstrated the high hazard that rip currents pose, typically during lower stages of the tide, here we demonstrate that an estuary mouth beach can exhibit flows reaching 1.5 m s(-1) - up to 50 % stronger than typical rip current flows - with a high proportion (> 60 %) of simulated bathers exiting the surfzone during the upper half of the tidal cycle. The three-dimensional ebb shoal delta was found to strongly control surfzone currents by (1) providing a conduit for estuary flows that connect to headland boundary rips and (2) acting as a nearshore bar system to generate wave-driven "river channel bathymetric rips". Despite significant spatio-temporal variability in the position of the river channels on the beach face, it was possible to hindcast the timing and severity of past bathing incidents from model simulations, providing a means to forewarn bathers of hazardous flows.

Observations of the depth integrated and time averaged sediment transport on a mixed sand and gravel (MSG) beach are presented and analysed to examine the performance of a new portable streamer trap. Measurement of the longshore sediment transport rate in the surf zone remains one of the great challenges in coastal engineering and coastal sciences. Sediment traps for sand beaches have proven useful in the past, but are not suitable for MSG beaches. This paper describes a portable depth-integrated streamer trap designed to measure the depth-integrated combined bed load and suspended longshore sediment transport on MSG beaches. The device consists of a polyester sieve cloth mounted into a rectangular holding frame. The stability of the device is achieved by gravity: the combined weight of the device and the operator, who is standing on and down-current of the device. The device has been tested in the field under moderate wave conditions at Minsmere, UK. We show that the observed suspended and bed load sediment transport are proportional to the wave energy flux, as formulated in the standard theoretical model, CSHORE. The data suggest that the empirical efficiency of wave breaking and bed load parameter are several orders of magnitude larger than that previously observed for uniform fine sand values.

Most of rip-current field experiments have focused on persistent rips along rip-channeled sandy beaches or transient rips along reasonably alongshore-uniform surf-zone morphology, while experiments on rip flowing against structures are scarce. In October 2018, a 3-week field experiment was performed at Anglet beach, SW France, aiming at examining the dynamics of high-energy rip currents in complex settings. The beach is barred with prominent inherited geology, characterized by the presence of a 500-m headland and a natural submerged reef. A large array of in-situ instruments was deployed to capture the temporal and spatial variability of rip flow circulations, including ADCPs, surf-zone drifters and video monitoring. The latter allowed to identify a wide range of rip-flow patterns. Among these patterns, a high-intensity rip current flowing against the headland was a dominant feature for obliquely incident waves. Such a boundary rip current was driven by the deflection of the longshore current against the headland, peaking at 0.7 m/s (5-min time- and depth-averaged) 800-m offshore in 12-m depth for a moderate storm event with 4-m obliquely incident waves. Very-low-frequency (O(1h) and O(30min)) fluctuations of this rip current were observed around low tide. Measurements of the vertical structure of the rip reveal that the deflection rip was more vertically-sheared as the water depth increases, with higher velocities near the surface, which is typical of a theoretical rip head structure.

Coastal ecosystems, such as saltmarsh, produce a range of ecosystem services that underpin human well-being. In the UK, and globally, saltmarsh extent and quality is declining due to coastal squeeze, deteriorating water quality, and agricultural activities. Here, we develop a general framework to evaluate changes in coastal defence. Using this framework, we identify priority areas for saltmarsh re-alignment: re-creation of saltmarsh in areas that have been saltmarsh in the past - but that have been claimed for a variety of land uses, particularly agriculture. We base our re-alignment prioritisation on the ecosystem services provided by saltmarsh in the North Devon Biosphere Reserve: specifically carbon sequestration and recreational benefits, and the economic values of those services. We compare potential economic benefits with the economic costs of creating new saltmarsh areas - specifically lost agricultural output, property damages and direct re-alignment costs. We identify a number of priority areas for managed re-alignment that generate high recreational values in areas where properties would not be damaged. These findings provide a necessary and timely analysis for the managers of the North Devon Biosphere Reserve. Furthermore, we outline a comprehensive methodology to plan future management of coastal zones.

Observations of the depth integrated and time averaged sediment transport on a mixed sand and gravel (MSG) beach are presented and analysed to examine the performance of a new portable streamer trap. Measurement of the longshore sediment transport rate in the surf zone remains one of the great challenges in coastal engineering and coastal sciences. Sediment traps for sand beaches have proven useful in the past, but are not suitable for MSG beaches. This paper describes a portable depth-integrated streamer trap designed to measure the depth-integrated combined bed load and suspended longshore sediment transport on MSG beaches. The device consists of a polyester sieve cloth mounted into a rectangular holding frame. The stability of the device is achieved by gravity: the combined weight of the device and the operator, who is standing on and down-current of the device. The device has been tested in the field under moderate wave conditions at Minsmere, UK. We show that the observed suspended and bed load sediment transport are proportional to the wave energy flux, as formulated in the standard theoretical model, CSHORE. The data suggest that the empirical efficiency of wave breaking and bed load parameter are several orders of magnitude larger than that previously observed for uniform fine sand values.

ABSTRACT Poate, T.G.; Masselink, G; Austin, M.; Dickson, M.E., and Kench, P., 2016. Observations of Wave Transformation on Macro-Tidal Rocky Platforms. In: Vila-Concejo, A.; Bruce, E.; Kennedy, D.M., and McCarroll, R.J. (eds.), Proceedings of the 14th International Coastal Symposium (Sydney, Australia). Journal of Coastal Research, Special Issue, No. 75, pp. 602–606. Coconut Creek (Florida), ISSN 0749-0208. Correctly predicting the transformation of ocean waves across rocky platforms has direct implications for cliff stability modelling, coastal defences and long-term coastal evolution. Wave transformation across rocky intertidal platforms is dependent on the morphological characteristics of the platform, including platform width, slope and roughness, and forcing characteristics, including wave and tide conditions. In this paper we present early observations from four field studies providing detailed measurements of wave processes across contrasting rocky platform sites with wave conditions between Hs = 0.5 m and Hs = 1.9 m, water depths between h = 0.5 m and h = 6.8 m and variable platform morphology. Results show that the relative wave height in the surf zone H/h is generally larger than in previous studies (H/h is c. 0.6, instead of 0.3–0.5) and wave dissipation greater for sites with considerable roughness.

Video derived runup statistics from ten separate deployments at six field sites have been used to develop a new parameterisation for the prediction of runup of runup on gravel beaches. These data were collected over a 2-year period under energetic storm conditions with significant wave heights of Hs = 1–8 m from gravel beaches and barriers composed of fine gravel (D50 = 2 mm) to large pebbles (D50 = 160 mm). An additional data set was generated using the numerical model XBeach-G, developed specifically for gravel beaches, and this synthetic dataset was used to further explore the role of hydrodynamic and morphological parameters on wave runup. A runup equation was developed using the synthetic data set and validated using the field data. The four parameters in this equation are, in decreasing order of importance, significant deep water wave height (Hs), spectral mean period (Tm − 1,0), beach slope (tanβ) and grain size (D50). The new gravel beach runup equation was found to fit the synthetic data set and the field data extremely well (r2 = 0.97 and 0.89, respectively) and the new equation performs significantly better than existing runup equations, even those specifically developed for gravel beaches.

This paper presents an extension of the XBeach-G numerical model with a sediment transport and morphology module, which includes the effect of groundwater ventilation and flow inertia on sediment transport, to simulate the morphodynamic response of pure gravel beaches and barriers to storms. The morphodynamic XBeach-G model is validated by simulating the morphodynamic response of one laboratory and four natural gravel barriers to 10 separate storm events, where the observed morphodynamic response ranged from berm building to barrier rollover. Model results show that XBeach-G is capable of reproducing the type of morphodynamic response of the barrier well in qualitative and quantitative sense (median BSS 0.75), with higher skill for more energetic storm conditions. Inclusion of acceleration forces on coarse gravel beaches is shown to significantly increase model skill and may be essential in modelling these types of beaches. The effect of varying hydraulic conductivity within estimated and published ranges is shown to be of secondary importance. The range of validation cases and lack of site-specific calibration show that XBeach-G can be applied to predict storm impacts on pure gravel beaches and barriers with reasonable to high confidence for a range of hydrodynamic forcing conditions and barrier response types.

A comprehensive study of swash-zone hydrodynamics and sediment transport was conducted on a macrotidal beach in Perranporth, United Kingdom. The unique study is the first to simultaneously measure suspended sediment and sheet flow sediment concentrations, water depth, near-bed velocity profiles, and high-resolution swash surface and bed-level changes on a natural beach. Data collected during the study are used to quantify the vertical profile of cross-shore and alongshore velocities and the importance of sheet flow sediment processes in the swash zone. The swash-zone boundary layer for cross-shore velocities is observed to generally occur over at least the lower 0.06 m of the water column. Alongshore velocities are often the same order of magnitude as the cross-shore velocities and are dominant near cross-shore flow reversal. Flows are often logarithmic in profile, but the instantaneous nature of the measurements renders application of the logarithmic model difficult. When valid, the logarithmic model enabled cross-shore shear stress estimates of up to 21.8Nm-2 with maximum alongshore shear stress estimates of 12.3Nm-2, further highlighting the potential importance of alongshore flows. Friction coefficient estimates, assuming a quadratic drag law, showed no statistical difference between onshore- and offshore-directed motion, with typical values of 0.023 +/- 0.013 (mean +/- SD). Sheet flow concentrations exceeded the maximum measured suspended sediment concentrations of approximately 400kgm-3. Sediment loads in the sheet layer are up to 10 times larger than the sediment loads in the lower suspension layer. Simplified sheet flow sediment transport estimates are 3.6 times larger on average than those in the suspension layer. The latter two findings indicate the importance of sheet flow processes in the swash zone that are generally ignored. (C) 2014 American Society of Civil Engineers.

Gravel and mixed sand-gravel beaches are characterised by steep reflective profiles which provide effective forms of wave absorption and therefore coastal defence to many mid-latitude regions, including northwestern Europe and North America. In the UK the combination of energetic wave conditions and large tides creates very dynamic and responsive morphology often dominated by cuspate features. Recent storm-responsive field campaigns at Loe Bar, Cornwall, UK, have captured highly energetic wave conditions (H-s = 2.5-5.8 m) using temporary video camera installations, low tide 3D topographic surveys with real time kinematic GPS, local tide level measurements and inshore directional wave data. Characterised by fine gravel (D-50 = 3 nun) and a steep reflective profile (tan beta = 0.118), the barrier at Loe Bar is exposed to an annual 10% exceedence significant wave height H-s10% of 2.4 in arriving predominantly from the southwest (Atlantic Ocean) and shore-normal to the beach. Under medium-wave conditions (H-s = 2-3 m), contrasting cusp behaviour was recorded with accretion and erosion, principally, through horn growth and decay (bed-level change Delta z = c. 1 in). During more energetic conditions (H-s = 5.8 in), the morphological response is more consistent and the waves drive erosion of the lower profile causing bed-level changes over a tide in excess of 1.5 in. Very rapid recovery to pre-storm bed levels is observed with defined cusp evolution occurring within 12 hours during the falling limb of the storm as incident wave energy decreases. The unique gravel cusp dataset suggests free behavior due to cusp morphodynamic feedback, rather than hydrodynamic forcing, plays an dominant role in cusp evolution.

A 2D laser-scanner was deployed on four different coarse-grained beaches (Chesil, Loe Bar, Hayling Island and Seascale - all in UK) to measure the swash morphodynamics during energetic wave conditions (offshore Hs > 2m). Field observations performed with the laser-scanner showed that different types of coarse-grained beaches present contrasting morphological responses under energetic hydrodynamics. The surf scaling parameter, a proxy of the morphological condition and wave steepness on the swash, showed an inverse relationship with the extreme vertical runup illustrating that runup is enhanced by low steepness swell waves for a given coarse-grained beach\slope; nevertheless additional parameters are needed to explain the entire runup variability.

In this paper we present a process-based numerical model for the prediction of storm hydrodynamics and hydrology on gravel beaches. The model comprises an extension of an existing open-source storm-impact model for sandy coasts (XBeach), through the application of (1) a non-hydrostatic pressure correction term that allows wave-by-wave modelling of the surface elevation and depth-averaged flow, and (2) a groundwater model that allows infiltration and exfiltration through the permeable gravel bed to be simulated, and is referred to as XBeach-G. Although the model contains validated sediment transport relations for sandy environments, transport relations for gravel in the model are currently under development and unvalidated. Consequently, all simulations in this paper are carried out without morphodynamic feedback. Modelled hydrodynamics are validated using data collected during a large-scale physical model experiment and detailed in-situ field data collected at Loe Bar, Cornwall, UK, as well as remote-sensed data collected at four gravel beach locations along the UK coast during the 2012–2013 storm season. Validation results show that the model has good skill in predicting wave transformation (overall SCI 0.14–0.21), run-up levels (SCI <0.12; median error <10%) and initial wave overtopping (85–90% prediction rate at barrier crest), indicating that the model can be applied to estimate potential storm impact on gravel beaches. The inclusion of the non-hydrostatic pressure correction term and groundwater model is shown to significantly improve the prediction and evolution of overtopping events.

Field experiments were conducted on two coastal gravel barriers backed by freshwater lagoons to examine the groundwater dynamics and to investigate the potential for saline intrusion. At Slapton Sands, groundwater, lagoon and ocean water level data were collected over a one year period; at Low Bar, data were collected over a two week period. The groundwater table was highly dynamic at both sites, with the ocean tide and wave event signals propagating to within a few metres of the lagoons.The amplitude and phase lag of the ocean tidal signal as it propagated landwards were used to apply the one-dimensional unsteady groundwater flow equation to estimate the hydraulic conductivity of the barrier aquifers. K is O(0.01) m s(-1) at both field sites, and this was used with the measured hydraulic gradients to estimate the barrier discharge. Net discharge was directed seawards and strongly positively correlated with the lagoon elevation and large wave events. In contrast, discharge was only weakly correlated with ocean tidal range and lagged by 4 days. This is due to strong landward-directed hydraulic gradients during spring tides reducing the lagoon-derived freshwater flux, with peak discharge occurring mid-way between spring and neap tides. The shoreline of the lagoon was decoupled from the groundwater table at both sites. The groundwater elevation was 1-2 m lower, suggesting that seepage from the lagoon to barrier occurs through the base of the lagoon. This is of potential significance to the modelling of coastal gravel barriers. Groundwater conductivity measurements demonstrated that salt water penetrates some distance landwards into the barriers (c. 60 m from spring high tide level). However, the width of the barrier systems (120 and 275 m) and the high water level of the fresh water lagoons, c. 0.75-2 m above spring tide level, inhibit saline intrusion. (C) 2013 Elsevier B.V. All rights reserved.

A process-based non-hydrostatic flow model, which includes the effect of infiltration and exfiltration, but no morphology, is applied to simulate overwash events on gravel barriers. After calibration, the model is shown to produce similar predictions for overwash as the empirical Barrier Inertia Model for parameter combinations within the validity range of the empirical model. When applied to 25 historical storm impacts, the process-based model shows improvement over the empirical model in predicting overwash, The model is applied to study the sensitivity of overwash to input parameters outside the validity range of the empirical Barrier Inertia model. This analysis shows that two parameters currently missing in the Barrier Inertia Model, the depth of the gravel beach toe and the gravel beach slope, greatly affect the threshold criteria for overwash.