RESEARCH ARTICLE Cross-Shore Distribution of the Wave-Induced Circulation

10.1029/2021JC018108
Over a Dissipative Beach Under Storm Wave Conditions
Key Points:
Marc Pezerat1 , Xavier Bertin1, Kévin Martins2 , and Laura Lavaud1
• F ield experiment at a dissipative
beach with 6 m Hm0 at breaking and 1
UMR 7266 LIENSs, CNRS/La Rochelle Université, La Rochelle, France, 2UMR 5805 EPOC, CNRS/Université de
undertows reaching 0.25 m/s as far as
Bordeaux, Pessac, France
4 km from the shoreline
• Accurate reproduction of the cross-
shore hydrodynamics using a phase-
averaged 3D circulation model Abstract This study explores the spatial distribution and the driving mechanisms of the wave-induced
• Wave dissipation by breaking locally cross-shore flow within the shoreface and surf zone of a dissipative beach. Unpublished results from a
increases seaward-directed flows by
over 100% compared to the surface
field campaign carried out in early 2021 under storm wave conditions are presented and compared with the
Stokes drift velocity predictions from a state-of-the-art phase-averaged three-dimensional circulation modeling system based on the
vortex force formalism. Under storm wave conditions, the cross-shore flow is dominated by a strong seaward-
Correspondence to:
directed current in the lower part of the water column. The largest current velocities of this return current are
M. Pezerat, located in the surf zone, where the dissipation by depth-induced breaking is most intense, but offshore-directed
[email protected] velocities up to 0.25 m/s are observed as far as 4 km from the shoreline (≃12 m-depth). Numerical experiments
further highlight the key control exerted by non-conservative wave forces and wave-enhanced mixing on the
Citation: cross-shore flow across a transition zone, where depth-induced breaking, whitecapping, and bottom friction all
Pezerat, M., Bertin, X., Martins, K., significantly contribute to the wave energy dissipation. Under storm conditions, this transition zone extended
& Lavaud, L. (2022). Cross-shore almost 6 km offshore and the cross-shore Lagrangian circulation shows a strong seaward-directed jet in the
distribution of the wave-induced
circulation over a dissipative beach lower part of the water column, whose intensity progressively decreases offshore. In contrast, the surf zone
under storm wave conditions. Journal edge appears clearly delimited under fair weather conditions and the seaward-directed current is weakened by a
of Geophysical Research: Oceans, near bottom shoreward-directed current associated with wave bottom streaming in the shoaling region, such that
127, e2021JC018108. https://doi.
org/10.1029/2021JC018108 the clockwise Lagrangian overturning circulation is constrained by an additional anti-clockwise overturning cell
at the surf zone edge.
Received 6 OCT 2021
Accepted 5 MAR 2022 Plain Language Summary As waves propagate toward the shore fluid parcels experience a net
transport in the direction of wave propagation. This onshore mass transport is compensated by a near bed return
Author Contributions: flow, which dynamics remain poorly understood. This study combines measurements from a field campaign
Conceptualization: Marc Pezerat, Xavier carried out in early 2021 in front of a gentle sloping beach and numerical modeling to explore the spatial
Bertin, Kévin Martins
distribution and the driving mechanisms of this wave-induced cross-shore flow. Both observations and model
Data curation: Marc Pezerat, Xavier
Bertin results show that the largest current velocities of this return current are located very close to the shoreline,
Funding acquisition: Xavier Bertin where the wave breaking is the most intense, but values up to 0.25 m/s are observed as far as 4 km from the
Investigation: Marc Pezerat, Xavier
shoreline under storm conditions. Numerical experiments further highlight the key control exerted by the wave
Bertin, Laura Lavaud
Methodology: Marc Pezerat, Xavier forces and the wave-enhanced mixing, which induce very contrasted circulation patterns under fair weather or
Bertin, Kévin Martins storm conditions and strongly constrain the vertical structure of the cross shore flow.
Project Administration: Xavier Bertin
Software: Marc Pezerat, Kévin Martins,
Laura Lavaud
Supervision: Xavier Bertin 1. Introduction
Validation: Marc Pezerat, Xavier Bertin
Writing – original draft: Marc Pezerat The nearshore circulation driven by breaking waves contributes to the cross-shelf transport of material, especially
Writing – review & editing: Xavier in the vicinity of the surf zone, such as the transport of nutrient (e.g., Morgan et al., 2018) or sediment, which
Bertin, Kévin Martins, Laura Lavaud can result in large morphological changes under storm conditions (e.g., Castelle et al., 2015; Coco et al., 2014;
Wright & Short, 1984).

Considering a weak along-shore variability of the topography and a shore normal incidence of waves, the interplay
between waves and currents most notably drives the so-called undertow. Based on the depth-integrated continuity
equation, the undertow commonly designates the time- and depth-averaged Eulerian offshore flow compensating
for the onshore mass transport associated with the Stokes drift. The onshore-directed mass transport is further
enhanced within the surf zone due to contribution from surface wave rollers (e.g., Svendsen, 1984a). The under-
lying dynamics were further investigated both theoretically and experimentally, providing some insights onto
© 2022. American Geophysical Union. the vertical structure of the (Eulerian) cross-shore flow. Within the surf zone, several pioneering studies (e.g.,
All Rights Reserved. Deigaard et al., 1991; Garcez Faria et al., 2000; Haines & Sallenger, 1994; Stive & Wind, 1986; Svendsen, 1984a,

PEZERAT ET AL. 1 of 21
 21699291, 2022, 3, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JC018108 by Hong Kong University of Science and Technology, Wiley Online Library on [21/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Oceans 10.1029/2021JC018108

among many others) proposed theoretical models, which all conceptually rely on the local imbalance between the
depth-uniform barotropic pressure gradient associated with the wave setup and the depth-varying gradient of the
wave radiation stresses. These models adequately predict parabolic velocity profile, whose curvature is a function
of local wave quantities and the vertical eddy viscosity (e.g., Garcez Faria et al., 2000).

In contrast, the wave induced dynamics seaward of the surf zone received much less attention, particularly under
storm waves. Most notably, Lentz et al. (2008) combined long term observations with a one-dimensional vertical
model adapted from Xu and Bowen (1994) to study the vertical structure of the cross-shore flow up to the inner-
shelf. These authors showed that the cross-shore velocity profiles seaward of the surf zone do not resemble the
parabolic profiles observed within the surf zone, but exhibit a maximum near the surface, which is consistent
with a balance between the Coriolis force associated with the offshore flow and the Stokes-Coriolis force, also
referred to as the Hasselmann wave stress (Hasselmann, 1970). As a result, the offshore flow tends to be equal in
magnitude but opposite in direction to the onshore Stokes drift velocity all along the water column, which implies
a nearly depth-uniform zero cross-shore Lagrangian flow seaward of the surf zone.

In recent years, wave-averaged three-dimensional (3D) circulation models have been developed aiming to repre-
sent consistently the effect of short waves on the mean circulation for a wide range of nearshore, coastal and
open-ocean applications. Several theoretical approaches were proposed on the form of the wave-modified prim-
itive equations that would be suitable for such models (e.g., see Bennis et al., 2011). The wave-averaged vortex
force formalism, which separates conservative and non-conservative wave forcing on the 3D quasi-Eulerian mean
circulation, constitutes a theoretically robust framework employed within several of these modeling systems
(e.g., Delpey et al., 2014; Guérin et al., 2018; Kumar et al., 2012; Michaud et al., 2012; Uchiyama et al., 2010;
Zheng et al., 2017). For nearshore applications, non-conservative effects associated with wave energy dissipation
processes through depth-induced breaking, whitecapping and bottom friction are expected to play a crucial role.
While Smith (2006) consistently derived the contribution of these processes to the depth-integrated momentum
equations, no definite theory exists to express these terms for the depth-resolving equations. In particular, it is
assumed that the dissipation of wave energy by breaking acts either like a surface stress on the mean flow or
like a body force, in which case one can thus impose an empirical vertical distribution such that the breaking
contribution applies at appropriate depths near the surface (e.g., Uchiyama et al., 2010). As pointed out by
Rascle (2007), the wave-enhanced vertical mixing associated with the production of turbulence by breaking
waves mostly controls the vertical shear of the horizontal current velocity so that the near-surface distribution of
the momentum source sparsely matters. In this regard, wave-averaged 3D circulation models are usually supple-
mented by a two-equation turbulence closure model, which allows to approximate the wave-enhanced turbulent
kinetic energy (TKE) budget across the water column. There is a consensus in the literature to model the TKE
injection with a flux-type boundary condition at the water surface assuming a power law for the decay of TKE
(Umlauf & Burchard, 2003). Craig and Banner (1994) proposed to express the surface flux of energy injected into
the water column in proportion to the surface wind friction velocity cubed. Following this approach, it is assumed
that the energy flux from the wind to the wavefield very closely matches that transferred from the wavefield to
the water column, which appears especially relevant for the deep ocean where breaking processes (whitecapping)
significantly impacts the atmospheric drag coefficient. For nearshore applications however, observations support
the fact that the surface flux of TKE scales with the energy dissipated through depth-induced breaking (e.g.,
Feddersen & Trowbridge, 2005). It is also interesting to note that the decay of TKE near the surface is particularly
sensitive to the surface mixing length, which remains an empirically parameterized quantity (e.g., see Moghimi
et al., 2016).

Among the above mentioned studies, Kumar et al. (2012), Michaud et al. (2012), Uchiyama et al. (2010) and
Zheng et al. (2017) essentially detailed the implementation of the vortex force formalism within various modeling
framework and further aimed to demonstrate the general applicability of this approach to study surf zone dynam-
ics over commonly used study cases, including applications at Duck, N.C., which serves as a reference bench-
mark. In addition, Kumar et al. (2012) also reproduced the results from Lentz et al. (2008) seaward of the surf
zone using the same data set. In a recent model-based study following Uchiyama et al. (2010), Wang et al. (2020)
further discussed the effect of the bottom wave streaming, which is the stress along the direction of wave prop-
agation that accompanies the wave energy dissipation by bottom friction. Most notably, their results tended to
show that the Lagrangian overturning circulation within the surf zone could be substantially weakened by an
opposite overturning cell arising seaward of the surf zone and extending within it, associated with the bottom

PEZERAT ET AL. 2 of 21
 21699291, 2022, 3, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JC018108 by Hong Kong University of Science and Technology, Wiley Online Library on [21/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Oceans 10.1029/2021JC018108

Figure 1. (a) Location of the study area in the Bay of Biscay, bathymetric map covering the computational domain (the open boundary is symbolized with the red
dotted line) with isobaths reduced to the mean sea level displayed every 10 m (black dash-dotted lines), and position of the Chassiron meteorological station, AWAC,
Acoustic Doppler Current Profilers (ADCP) 600 kHz and inter-tidal area sensors. A cross-shore profile extending from the isobath 25 m to PT5 sensor is also displayed
(brown dashed line). (b) Cross-shore profile from AWAC location to the first sensor deployed within the inter-tidal area. (c) Zoom on inter-tidal area sensors.

wave streaming. Realistic applications of state-of-the-art, fully coupled, wave-averaged 3D circulation models in
the nearshore region remain very scarce (Delpey et al., 2014; Guérin et al., 2018; Michaud et al., 2012), especially
under storm conditions, such that our comprehensive understanding of the wave-induced hydrodynamics remains
somehow limited and the predictive skills of these models uncertain.

This study aims to explore the cross-shore distribution and the driving mechanisms of the wave-induced cross-
shore circulation within the shoreface and the surf zone of a dissipative beach. We present unpublished results
from a field campaign carried out in early 2021 in the central part of the French Atlantic coast under storm
wave conditions, complemented with predictions from the state-of-the-art 3D circulation model SCHISM (Zhang
et al., 2016), fully coupled with the spectral wave model WWM (Roland et al., 2012). The manuscript is organ-
ized as follows. The study area, the field campaign and the processing of in-situ measurements are presented
in Section 2. The parameterization of the modeling system is detailed in Section 3 and its predictive skills are
assessed in Section 4 for the case study considered here. The Section 5 discusses the contrasted wave-induced
circulation patterns and associated driving mechanisms under high and moderate wave energy conditions based
on further numerical experiments. Finally, concluding remarks are provided in Section 6.

2. Study Area and Field Campaign

2.1. Study Area

The study area is located along the South-Western coast of the Oléron Island in the central part of the French
Atlantic coast (see Figure 1a), in front of Saint-Trojan Beach. This beach corresponds to a 8 km-long sandspit
bounded to the South by the Maumusson Inlet and to the North by a rocky shore platform (Lavaud et al., 2020). In
this region, tides are semi-diurnal and range from 1.5 to 5.5 m, which corresponds to a macrotidal regime. Yearly
mean wave conditions along the 30 m isobath are characterized by a significant wave height of 1.6 m, a mean
wave period of 5.9 s and a direction of 285° from the true North (Dodet et al., 2019), but the offshore significant
wave height can exceed 10 m with peak periods over 20 s (Bertin et al., 2015). This area is characterized by a very
gently sloping shoreface (the isobath 20 m being found approximately 10 km offshore) and a non-barred dissipa-
tive beach composed of fine sandy sediments and exposed to an energetic wave climate. Although this stretch of
coast is relatively along-shore uniform, small amplitude inter-tidal bars can develop after the persistence of fair
weather conditions (see Bertin et al., 2008; Guérin et al., 2018, for supplementary studies in this area).

PEZERAT ET AL. 3 of 21
 21699291, 2022, 3, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JC018108 by Hong Kong University of Science and Technology, Wiley Online Library on [21/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Oceans 10.1029/2021JC018108

2.2. Field Campaign and Data Processing

The field campaign was carried out between January and February 2021 in two steps. First, two Acoustic
Doppler Current Profilers (ADCP) were mounted on structures anchored in the seabed, at approximately 12.5
and 7.5 m-depth locations below Mean Sea Level (MSL) for a long-term deployment between the January 19 and
the February 26 (see Figure 1). The most offshore instrument is a high resolution ADCP (1 MHz) integrating an
Acoustic Surface Tracker (AST) and is hereafter referred to as the AWAC, whereas the other one is a medium-res-
olution ADCP (600 kHz). Both instruments alternated a “current cycle” and a “wave cycle” each hour. During the
“current cycle,” 10 min-averaged velocity profile measurements were collected along the vertical axis, whereas
during the “wave cycle” velocity measurements within a fixed 2 m-high cell and pressure measurements were
performed at 2 Hz during 20 min. Second, a set of sensors was deployed in the inter-tidal area (Figure 1c) between
January 29th and the 31st during spring tides so as to capture a highly energetic event associated with the storm
Justine. The offshore significant wave height at the Biscay Buoy location (5°W, 45.23°N) reached 10 m, which
corresponds to a return period of the order of 1 year (Nicolae-Lerma et al., 2015). The swell associated with the
storm reached the study area during the night of the 30th from a westward direction, while local winds reached
15 m/s at the storm peak. This set of sensors was deployed along a cross-shore profile and included one 2 MHz
ADCP (with a similar data collection scheme than for the two offshore ADCPs), three pressure transducers with
a 2 Hz sampling frequency (PT) and one Acoustic Doppler Velocimeter deployed 20 cm above the seabed, with a
16 Hz sampling frequency (ADV). The PTs and ADV all performed continuous measurements.

For each sensor, sea-bottom pressure timeseries were split into 20 min-long bursts (consistent with ADCPs “wave
cycle”), corrected for sea level atmospheric pressure using data collected at the nearby meteorological station
of Chassiron (Figure 1a), detrended and converted into a sea-surface elevation signal assuming a hydrostatic
pressure. For the sensors deployed in the inter-tidal area, measurements below a burst-averaged water depth
of 0.5 m were discarded due to the presence of substantial infra-gravity waves, which caused the sensors to be
intermittently dry. Then, pressure attenuation with depth due to non-hydrostatic effects was corrected using the
Transfer Function Method based on the linear wave theory (TFM, e.g., Bishop & Donelan, 1987). This method
requires an upper cutoff frequency to remove high frequency noise that is amplified by the TFM correction, and
to prevent the over-amplification of high-frequency energy levels due to non-linear interactions in intermediate
and shallow-water depths (Mouragues et al., 2019). The cutoff frequency was set to 0.2 Hz for the two offshore
sensors and 0.4 Hz for the sensors in the inter-tidal area. Finally, the sea surface elevation density spectra
𝐴𝐴 𝐴𝐴(𝑓𝑓 )
were computed by means of a Fast Fourier Transform on 10 Hanning-windowed segments with a 50% overlap,
which allows a good compromise between statistical stability (20 degrees of freedom) and frequency resolution
(8.3 mHz). The processing of pressure measurements was further verified at the AWAC location by comparing
the sea-surface elevation spectra with those obtained from the AST measurements. These measurements were
not used directly due to spurious signals during the most energetic events, whose are probably explained by the
presence of air bubbles in the water column induced by the breaking of storm waves (not shown). Wave bulk
parameters (significant wave height,
𝐴𝐴 𝐴𝐴𝑚𝑚0, mean and continuous peak periods,
𝐴𝐴 𝐴𝐴 and 𝐴𝐴𝑝𝑝𝑝𝑝 ) were computed using
𝐴𝐴𝑚𝑚𝑚𝑚
𝐴𝐴 the 𝐴𝐴th moments of the spectra:
𝑓𝑓𝑐𝑐

∫𝑓𝑓min
(1)
𝑚𝑚𝑝𝑝 = 𝑓𝑓 𝑝𝑝 𝐸𝐸(𝑓𝑓 )𝑑𝑑𝑑𝑑

such that:
√
= 4 𝑚𝑚0 𝐻𝐻𝑚𝑚0
√
𝑚𝑚0
𝑇𝑇
(2)
𝑚𝑚02 =
𝑚𝑚2
𝑚𝑚−2 𝑚𝑚1
𝑇𝑇𝑝𝑝𝑝𝑝 =
𝑚𝑚20

For each sensor, an adaptive

𝐴𝐴 𝐴𝐴min value, defined as half the continuous peak frequency computed at the AWAC
location, was used in order to separate the gravity from the infra-gravity bands (e.g., Bertin et al., 2020; Hamm
& Peronnard, 1997).

PEZERAT ET AL. 4 of 21
 21699291, 2022, 3, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JC018108 by Hong Kong University of Science and Technology, Wiley Online Library on [21/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Oceans 10.1029/2021JC018108

For the two offshore ADCPs, 10 min-averaged vertical profiles of current velocities were acquired along bins
spanned respectively every 1 m (AWAC) and 0.5 m (ADCP 600 kHz). The measurements above a distance equal
to the water depth minus half the significant wave height were discarded due to contamination by surface reflec-
tions from the sidelobes of the ADCP acoustic pulses (Appell et al., 1991). Current velocity profile measurements
from the ADCP 2 MHz deployed in the inter-tidal area were discarded because of spurious bin-to-bin velocity
differences. Finally, continuous velocity measurements from the ADV, were split into 30 min-long bursts and
filtered from spikes using the phase-space thresholding method of Goring and Nikora (2002).

The inter-tidal topography was surveyed at low tide during the deployment (29/01/2021) and the recovery
(31/01/2021) of the instruments with PPK GNSS over an area centered on the instrumented transect and extend-
ing 1 km along-shore. The comparison between both datasets showed very limited morphological changes (with
a root mean square difference of 0.10 m along the instrumented profile), a behavior already reported by Guérin
et al. (2018) under similar storm wave conditions. The subtidal bathymetry was surveyed at the location of the
instrumented profile up to a water depth of 11 m below MSL four weeks after the deployment in the inter-tidal
area by means of a Norbit multi-beam echo-sounder. This bathymetric data set was merged with an extensive
single-beam echo-sounder survey carried out in April 2013. Both datasets in the region where they overlap show
a smooth transition in the subtidal zone with changes of the order of 0.4 m.

3. Modeling System
The modeling system used in this study couples the 3D circulation model SCHISM (Zhang et al., 2016) and the
third generation spectral wave model WWM (Roland et al., 2012). This modeling system offers the flexibility
to cover large geographic areas with unstructured grid and very robust numerical schemes for both models.
The 3D wave-induced circulation is modeled through the vortex force formalism, such as presented by Bennis
et al. (2011). Its detailed implementation in SCHISM can be found in Guérin et al. (2018) and is recalled in
Appendix A. In the following, only the parameterization of the relevant part of the model and further improve-
ments since Guérin et al. (2018) are described.

3.1. Parameterization of WWM

3.1.1. Depth-Induced Breaking Parameterization

The wave energy dissipation rate by depth-induced breaking is computed according to the model of Battjes and
Janssen (1978) with an adaptive breaking coefficient (B) as proposed by Pezerat et al. (2021). The local mean
(phase-averaged) rate of energy dissipation per unit
𝐴𝐴 area 𝐴𝐴𝑑𝑑𝑑𝑑 in (W/m 2) reads:
𝐵𝐵
(3)
𝐷𝐷𝑑𝑑𝑑𝑑 = 𝜌𝜌𝜌𝜌𝜌𝜌𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚 𝑄𝑄𝑏𝑏 𝐻𝐻𝑚𝑚2
4

where B = 40 tan β, with tan β the local bottom 𝐴𝐴 slope; 𝐴𝐴mean is a mean frequency, usually computed as the ratio
m1/m0 (see Equation
𝐴𝐴 1); 𝐴𝐴𝑏𝑏 is the local fraction of breaking (and broken) waves 𝐴𝐴 and 𝐴𝐴𝑚𝑚 is the local maximum
possible wave height estimated by means of a parameterized Miche-type breaking criterion. Under the shallow
water assumption,
𝐴𝐴 𝐴𝐴𝑚𝑚 reads:

(4)
𝐻𝐻𝑚𝑚 = 𝛾𝛾𝛾

where
𝐴𝐴 𝐴𝐴 is the breaking index, an adjustable coefficient, usually kept constant at 0.73 following the calibration
performed by Battjes and Stive (1985). However, as pointed out by Pezerat et al. (2021), the introduction of the
adaptive breaking coefficient requires a newly calibrated breaking index. Based on sensitivity tests performed
with the model 𝐴𝐴 on 𝐴𝐴𝑚𝑚0 results considering the entire data set (not shown), a constant value
𝐴𝐴 of 𝐴𝐴 = 0.60 was consid-
ered for this study. It is worth noting that this value might show some site- or wave conditions-specificity, but it
is not the purpose of this study to propose an extensive calibration of the breaking index. Finally, following the
approach of Eldeberky and Battjes (1996), the corresponding source term in WWM is computed by distributing
𝐴𝐴 𝐴𝐴𝑑𝑑𝑑𝑑 over frequencies and directions in proportion to the spectral action density:

∫𝜎𝜎 ∫𝜃𝜃
𝐷𝐷
𝑆𝑆𝑑𝑑𝑑𝑑 = − 𝑑𝑑𝑑𝑑 𝑁𝑁 where 𝐸𝐸𝑡𝑡𝑡𝑡𝑡𝑡 = 𝜌𝜌𝜌𝜌
(5) 𝐸𝐸𝐸𝐸𝐸𝐸 ′ 𝑑𝑑𝑑𝑑 ′
𝐸𝐸𝑡𝑡𝑡𝑡𝑡𝑡

PEZERAT ET AL. 5 of 21
 21699291, 2022, 3, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JC018108 by Hong Kong University of Science and Technology, Wiley Online Library on [21/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Oceans 10.1029/2021JC018108

3.1.2. Other Source Terms

The wind 𝐴𝐴input (𝐴𝐴𝑖𝑖𝑖𝑖) and dissipation by whitecapping
𝐴𝐴 (𝐴𝐴𝑑𝑑𝑑𝑑) are formulated according to the parameterization of
Ardhuin et al. (2010). Non-linear quadruplet interactions
𝐴𝐴 (𝐴𝐴𝑛𝑛𝑛𝑛𝑛4) are taken into account following the approach
of Hasselmann et al. (1985), whereas the non-linear triad interactions 𝐴𝐴 (𝐴𝐴𝑛𝑛𝑛𝑛𝑛3) are parameterized following the
approach of Eldeberky (1997). Finally, the formulation of the energy dissipation by bottom friction 𝐴𝐴 (𝐴𝐴𝑏𝑏𝑏𝑏 ) is
adapted from Tolman (1994). This formulation is based on the model of Madsen et al. (1988) with a parame-
terization of the roughness issued from Grant and Madsen (1982), and was subsequently adapted by Ardhuin
et al. (2003) based on the SHOWEX experiment.

3.2. Parameterization of SCHISM

3.2.1. Non-Conservative Wave Forces
The non-conservative wave forcing due to wave breaking 𝐹𝐹̂𝑏𝑏𝑏𝑏𝑏𝑏𝑏 , 𝐹𝐹̂𝑏𝑏𝑏𝑏𝑏𝑏𝑏 includes the effect of whitecapping and
[ ]
𝐴𝐴
depth-induced breaking further modified due to the action of wave rollers (cf., Appendix A for the description
of the roller model):
( )

∫𝜎𝜎 ∫𝜃𝜃
] 𝛿𝛿𝑧𝑧𝑧𝑧𝑧𝑧 𝐷𝐷
(6)
𝐹𝐹̂𝑏𝑏𝑏𝑏𝑏𝑏𝑏 , 𝐹𝐹̂𝑏𝑏𝑏𝑏𝑏𝑏𝑏 = 𝑘𝑘 ((1 − 𝛼𝛼𝑅𝑅 ) 𝑆𝑆𝑑𝑑𝑑𝑑 + 𝑆𝑆𝑑𝑑𝑑𝑑 ) cos𝜃𝜃 ′ , sin𝜃𝜃 ′ 𝑑𝑑𝑑𝑑 ′ 𝑑𝑑𝑑𝑑 ′ + 𝑅𝑅 𝑘𝑘𝑝𝑝 [cos𝜃𝜃𝑚𝑚 , sin𝜃𝜃𝑚𝑚 ]
[ [ ]
−
𝜌𝜌 𝜎𝜎𝑝𝑝

where the vertical distribution function is given by a delta function

𝐴𝐴 𝐴𝐴𝑧𝑧𝑧𝑧𝑧𝑧 , such that the breaking-induced momen-
tum is injected within the first cell below the free surface (Bennis et al., 2011). The roller model shows bearly any
effect at both offshore locations, while the current intensity slightly increases𝐴𝐴with 𝐴𝐴𝑅𝑅 at the ADV location (not
shown). The model best-fit results were obtained𝐴𝐴with 𝐴𝐴𝑅𝑅 = 0.5.

The bottom streaming corresponds to the stress along the direction of wave propagation that accompanies the
dissipation of wave energy by bottom friction within the wave boundary layer (Longuet-Higgins, 1953). The
corresponding body𝐴𝐴 force 𝐹𝐹̂𝑤𝑤𝑤𝑤𝑤𝑤𝑤 , 𝐹𝐹̂𝑤𝑤𝑤𝑤𝑤𝑤𝑤 is formulated by means of an upward decaying vertical distribution of the
[ ]

wave energy dissipation by bottom friction (Uchiyama et al., 2010):

𝑓𝑓 𝑤𝑤𝑤𝑤 (𝑧𝑧)
𝜌𝜌 ∫𝜎𝜎 ∫𝜃𝜃
(7)
𝐹𝐹̂𝑤𝑤𝑤𝑤𝑤𝑤𝑤 , 𝐹𝐹̂𝑤𝑤𝑤𝑤𝑤𝑤𝑤 = −
[ ] [ ]
𝑘𝑘𝑘𝑘𝑏𝑏𝑏𝑏 cos𝜃𝜃 ′ , sin𝜃𝜃 ′ 𝑑𝑑𝑑𝑑 ′ 𝑑𝑑𝑑𝑑 ′

where 𝐴𝐴 𝑤𝑤𝑤𝑤 is a vertical distribution function given by:

𝐴𝐴

1 − tanh(𝑘𝑘𝑤𝑤𝑤𝑤 (𝑑𝑑 + 𝑧𝑧))2

∫−𝑑𝑑 1 − tanh(𝑘𝑘𝑤𝑤𝑤𝑤 (𝑑𝑑 + 𝑧𝑧))2 𝑑𝑑𝑑𝑑
𝑓𝑓 𝑤𝑤𝑤𝑤 (𝑧𝑧) = 𝜂𝜂̄
(8)

where
𝐴𝐴 1∕𝐾𝐾wd = 𝑎𝑎𝑤𝑤𝑤𝑤 𝛿𝛿wbl is a decay length proportional to the wave boundary layer thickness 𝐴𝐴 (𝐴𝐴wbl). Within
SCHISM,
𝐴𝐴 𝐴𝐴wbl is derived from the wave boundary layer model of Madsen (1995) that is used to compute the appar-
ent roughness length for the parameterization of the bottom friction within the circulation model (Section 3.2.3).
The proportionality coefficient
𝐴𝐴 𝐴𝐴wd is taken equal to unity, such that the decay length matches the theoretical wave
boundary layer thickness for monochromatic waves although laboratory measurements of the bottom boundary
layer under random waves suggest a significant increase in the thickness, that 𝐴𝐴 is, 𝐴𝐴𝑤𝑤𝑤𝑤 > 1 (Klopman, 1994).

3.2.2. Wave-Enhanced Turbulence at the Surface

The circulation model is supplemented by a K − ω turbulence closure model retrieved from the Generic Length
Scale (GLS) two-equation turbulence closure model within the framework of the General Ocean Turbulence
Model (GOTM, Burchard et al., 1999; Umlauf et al., 2005). At the water surface, the turbulence closure model
accounts for a TKE injection by breaking waves (K in m 2/s 2) through a flux-type boundary condition assuming a
power law for the vertical decay of K (Umlauf & Burchard, 2003), which reads:

)3
𝑧𝑧0 − 𝑧𝑧′ 2 𝛼𝛼
( 𝑠𝑠
𝜈𝜈
(9) 𝜕𝜕𝜕𝜕
= 𝐹𝐹𝐾𝐾 at 𝑧𝑧 = 𝜂𝜂̄
𝜎𝜎𝐾𝐾 𝜕𝜕𝜕𝜕 𝑧𝑧𝑠𝑠0

PEZERAT ET AL. 6 of 21
 21699291, 2022, 3, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JC018108 by Hong Kong University of Science and Technology, Wiley Online Library on [21/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Oceans 10.1029/2021JC018108

where 𝐴𝐴𝐾𝐾 (in m 3/s 3) is the surface flux of energy injected into the water column, ν is the vertical eddy viscosity,
𝐴𝐴
𝐴𝐴 𝐴𝐴𝐾𝐾 is the turbulent Schmidt number for K, α is the spatial decay rate of TKE in the wave enhanced 𝐴𝐴layer, 𝐴𝐴𝑠𝑠0 is
the surface mixing length and z′ is the distance below the surface at which the flux is imposed. For numerical
reason, z′ is prescribed as half the height of the top cell, such that the boundary condition for K requires a refined
discretization of the vertical grid near the surface. The surface mixing 𝐴𝐴length 𝑧𝑧𝑠𝑠0 controls the depth of penetra-
( )

tion for the injected TKE. There are strong uncertainties over this quantity, which has been either parameterized
as a constant𝐴𝐴(e.g., 𝐴𝐴𝑠𝑠0 = 0.2 m, Feddersen & Trowbridge, 2005) or as a function of the significant wave height:
𝐴𝐴 𝐴𝐴𝑠𝑠0 = 𝛼𝛼𝑤𝑤 𝐻𝐻𝑚𝑚0𝐴𝐴, with 𝐴𝐴𝑤𝑤 = (1) (see Moghimi et al., 2016, for a short review). Following the approach of Feddersen
and Trowbridge (2005), the surface flux of TKE injected at the surface scales with the energy dissipated through
wave-related processes at the surface:
( )

∫𝜎𝜎 ∫𝜃𝜃 𝜌𝜌 ∫𝜎𝜎 ∫𝜃𝜃

𝑐𝑐 𝑐𝑐
(10)
𝐹𝐹𝐾𝐾 = 𝑑𝑑𝑑𝑑 − (1 − 𝛼𝛼𝑅𝑅 ) 𝜎𝜎 ′ 𝑆𝑆𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑 ′ 𝑑𝑑𝑑𝑑 ′ + 𝐷𝐷𝑅𝑅 − 𝑑𝑑𝑑𝑑 𝜎𝜎 ′ 𝑆𝑆𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑 ′ 𝑑𝑑𝑑𝑑 ′
𝜌𝜌

where the coefficients

𝐴𝐴 𝐴𝐴 and 𝐴𝐴𝑑𝑑𝑑𝑑 control the amount of energy injected to the water column. The range of
𝐴𝐴𝑑𝑑𝑑𝑑
value 0.01–0.25 has been proposed in the literature 𝐴𝐴 for 𝐴𝐴𝑑𝑑𝑑𝑑 (e.g., Feddersen, 2012a, 2012b; Feddersen & Trow-
bridge, 2005; Huang et al., 2009), while Paskyabi et al. (2012) suggest 𝐴𝐴 𝐴𝐴𝑑𝑑𝑑𝑑 ≃ 1. The model best-fit results were
obtained𝐴𝐴with 𝐴𝐴𝑠𝑠0 = 𝛼𝛼𝑤𝑤
𝐴𝐴𝐻𝐻𝑚𝑚0, 𝐴𝐴𝑤𝑤
𝐴𝐴 = 1; 𝐴𝐴𝑑𝑑𝑑𝑑 = 0.15
𝐴𝐴 and 𝐴𝐴𝑑𝑑𝑑𝑑 = 1.

3.2.3. Bottom Boundary Condition

The bottom boundary condition imposes a balance between the internal Reynolds stress and the bottom frictional
𝐴𝐴stress: 𝜏𝜏𝑏𝑏𝑏𝑏𝑏 , 𝜏𝜏𝑏𝑏𝑏𝑏𝑏
[ ]

[ ]
𝜕𝜕 [𝑢𝑢𝑢
̂ 𝑢𝑢]
𝑢 𝜏𝜏𝑏𝑏𝑏𝑏𝑏 , 𝜏𝜏𝑏𝑏𝑏𝑏𝑏
(11)
𝜈𝜈 = at 𝑧𝑧 = −𝑑𝑑
𝜕𝜕𝜕𝜕 𝜌𝜌

The law of the wall is then assumed, leading to a logarithmic profile for the velocity within a constant stress
layer that presumably contains the bottom cell while the bottom stress is formulated with a quadratic bottom drag
parameterization. The bottom stress finally reads (e.g., Blumberg & Mellor, 1987):
2
√ ⎛ ⎞
⎜ 𝜅𝜅 ⎟
(12)
[ ] 2 2
𝜏𝜏𝑏𝑏𝑏𝑏𝑏 , 𝜏𝜏𝑏𝑏𝑏𝑏𝑏 = 𝜌𝜌𝜌𝜌𝑑𝑑 𝑢𝑢̂ 𝑏𝑏 + 𝑣𝑣̂ 𝑏𝑏 [𝑢𝑢̂ 𝑏𝑏 , 𝑣𝑣̂ 𝑏𝑏 ] with 𝐶𝐶𝑑𝑑 = ⎜ ( ) ⎟
⎜ ln 𝛿𝛿𝑏𝑏 ⎟
⎝ 𝑧𝑧0 ⎠

where
𝐴𝐴 𝐴𝐴𝑑𝑑 is the friction 𝐴𝐴 factor, [𝑢𝑢̂ 𝑏𝑏 , 𝑣𝑣̂ 𝑏𝑏 ] is the velocity at the top of the bottom computational cell, κ is the von
Kármán's constant, 𝐴𝐴 𝐴𝐴𝑏𝑏 is the thickness of the bottom cell (in m) 𝐴𝐴 and 𝐴𝐴0 is the bottom roughness length (in m).
In the presence of waves, the wave-current interaction theory by Madsen (1995) as modified by Mathisen and
Madsen (1999) is applied to compute an apparent roughness 𝐴𝐴 length 𝐴𝐴𝑎𝑎0, which further replaces
𝐴𝐴 𝐴𝐴0 in the expression
𝐴𝐴 of 𝐴𝐴𝑑𝑑 , and thus allows to account for the enhanced roughness experienced by the current in presence of waves.
This approach takes as input a (physical) bottom roughness length 𝐴𝐴 (𝐴𝐴0) to provide an expression of the wave
boundary layer thickness 𝐴𝐴 and 𝐴𝐴𝑎𝑎0 following a numerical procedure described in H. Zhang et al. (2004). In this
study, 𝐴𝐴0 varies spatially between 0.1 and 5 mm to account for the seabed granulometric variability based on the
𝐴𝐴
sea-bottom nature map provided by the Hydrographic and Oceanographic French Office.

3.3. Model Implementation

The coupling between SCHISM and WWM is made at the source code level. Both models share the same
unstructured grid and domain decomposition. The horizontal spatial resolution ranges from 2 km at the offshore
boundary down to 20 m in the surf zone. The vertical grid for the circulation model is discretized using 25 S-lev-
els stretched near the surface and the bottom. The time step for the circulation model is set to 10 s whereas WWM
is running in implicit mode (Abdolali et al., 2020; Booij et al., 1999). This allows to relax the constraint for the
time step of the wave module, which was set to 300 s. Finally, the spectral space in WWM was discretized in 36
directions covering the entire trigonometric circle and 24 frequencies ranging from 0.03 to 0.4 Hz.

PEZERAT ET AL. 7 of 21
 21699291, 2022, 3, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JC018108 by Hong Kong University of Science and Technology, Wiley Online Library on [21/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Oceans 10.1029/2021JC018108

At the domain offshore boundaries, the tidal forcing was computed considering the 16 main constituents linearly
interpolated from the regional tidal model of Bertin et al. (2012), whereas WWM was forced with timeseries of
energy spectra obtained from a North Atlantic application of the spectral wave model WaveWatch III (WW3,
Tolman, 1991). For both SCHISM and WWM, the atmospheric forcing consisted of MSL pressure and wind
speed at 10 m issued from the meteorological operational model ARPEGE (e.g., Déqué et al., 1994), interpo-
lated onto a 0.1° regular grid. WW3 was forced with wind fields at 10 m originating from the Climate Forecast
System Reanalysis (CFSR, Saha et al., 2011) extracted from a 0.2° regular grid covering the entire North Atlantic
basin. ARPEGE was preferred to CFSR for our local application of the modeling system considering its slightly
improved predictive skills as compared to measurements of wind speed and direction at the nearby meteorologi-
cal station of Chassiron (Figure 1).

4. Observations and Predictive Skills of the Model

A comparison between measurements and model best-fit results (run Rref) is presented. The overall model
performance for each quantity is assessed with the normalized bias (NB) and the normalized root mean square
error (NRMSE):

𝑋𝑋̂ − 𝑋𝑋
⎧
⎪ × 100 if 𝑋𝑋 ≃ 0
⎪ (max𝑋𝑋 − min𝑋𝑋)
(13)
NB(𝑋𝑋) = ⎨
⎪ 𝑋𝑋̂ − 𝑋𝑋 × 100
⎩ 𝑋𝑋
⎪

⎧ √
)2
𝑋𝑋̂ − 𝑋𝑋
⎪ (

⎪ (max𝑋𝑋 − min𝑋𝑋) × 100 if 𝑋𝑋 ≃ 0

⎪
(14)
NRMSE(𝑋𝑋) = ⎨√
⎪√
√( ̂ )2
⎪√ 𝑋𝑋 − 𝑋𝑋 × 100
𝑋𝑋 2
⎪
⎩

where 𝐴𝐴 and 𝑋𝑋̂ respectively correspond to the measured and modeled quantity and the overbar denotes the aver-
𝐴𝐴
age over the timeseries.

4.1. Water Levels and Short Waves

Phase-averaged surface elevation variations are very well reproduced by the model at the AWAC and ADCP
600 kHz locations (Figures 2a and 2b), with a NRMSE lower than 6% obtained at both locations. The negative
bias observed at the AWAC location (NB = −4.1%) could be attributed to pressure sensor drift, as it increases
in time while such bias is not observed at the nearby ADCP 600 kHz location. Model results on short waves
are compared to observations by means of bulk parameters computed with moments integrated over the same
frequency range as the data. These are very well reproduced by the model at both locations (Figures 2c–2h) with a
NRMSE on𝐴𝐴both 𝐴𝐴𝐴𝐴 𝑚𝑚0 and 𝐴𝐴𝑚𝑚𝑚𝑚 below 7%, only the continuous peak period is slightly underestimated at the storm
peak, resulting in a negative NB of −7.3% at the ADCP 600 kHz location. This problem might be explained by
the fact that spectral wave models allow to represent energy transfers toward higher harmonics by sum self inter-
actions with the LTA formulation of (Eldeberky, 1997) but not by difference interactions. Thus, the model cannot
represent transfers from the gravity band toward the IG band and the subsequent transfers back to the gravity band
by the generation of IG wave higher harmonics (e.g., Bertin et al., 2020), nor direct transfers toward low frequen-
cies in the gravity band (e.g., De Bakker et al., 2015). Both field measurements at the ADCP 600 kHz location
and model results show 𝐴𝐴 that 𝐴𝐴𝑚𝑚0 is tidally modulated when exceeding approximately 2 m at the AWAC location
(Figure 2d). Such modulation also occurs at the storm peak at the AWAC location 𝐴𝐴 (𝐴𝐴𝑚𝑚0 ≃ 6 m , Figure 2c). This
clearly suggests that depth-induced breaking starts being substantial as far as 4 km away from the shoreline
(≃12.5 m-depth) under energetic conditions. In addition 𝐴𝐴 𝐴𝐴𝑚𝑚𝑚𝑚 period is also tidally modulated, highlighting the
significant contribution of the non-linear triad interactions, whose intensity varies with the water-depth. As the
water-depth decreases, more energy is transferred toward the higher harmonics by non-linear triad interactions,

PEZERAT ET AL. 8 of 21
 21699291, 2022, 3, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JC018108 by Hong Kong University of Science and Technology, Wiley Online Library on [21/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Oceans 10.1029/2021JC018108

Figure 2. Measured and simulated phase-averaged free surface elevation 𝐴𝐴 ̄ , significant wave height
(𝜂𝜂) 𝐴𝐴 (𝐴𝐴𝑚𝑚0), mean period
𝐴𝐴 (𝐴𝐴𝑚𝑚𝑚𝑚) and continuous peak period
𝐴𝐴 (𝐴𝐴𝑝𝑝𝑝𝑝 ) timeseries at the AWAC (left panels) and Acoustic Doppler Current Profilers (ADCP)
600 kHz (right panels) locations using the configuration of reference (Rref).

𝐴𝐴 that 𝐴𝐴𝑚𝑚𝑚𝑚 decreases. Within the inter-tidal area most of the differences between model results and measure-
such
ments are observed for water-depths below 1 m (see Figure 3). Water levels and significant wave heights are well
reproduced by the model with a NRMSE computed over all sensors that respectively reaches 10.4% and 10.8%.
𝐴𝐴 The 𝐴𝐴𝑚𝑚𝑚𝑚 period is not shown because a significant amount of energy is transferred toward the IG band, which
can partly go back to the gravity band through the generation of IG wave higher harmonics (Bertin et al., 2020),
resulting in an increase
𝐴𝐴 of 𝐴𝐴𝑚𝑚02, a process that cannot be reproduced by the phase-averaged model.

4.2. Cross-Shore Circulation

The model results for Rref on the cross-shore velocity component show a fairly good agreement with meas-
urements from the AWAC and the ADCP 600 kHz as shown Figure 4 below the lowest sea surface tidal level.
Overall, the NRMSE varies between 13.9% and 20%, while the NB fluctuates between −2.9% and 5.6%. Note,
however, that measurements from the ADCP 600 kHz are more scattered under energetic wave conditions. First,
this might be partly due to measurement artifacts associated with the generation of bubbles in the water column
by depth-induced breaking, which is particularly active at this location (cf. Section 4.1). Second, as vertical
current profile measurements are 10 min-averaged, there may be an aliasing of the signal associated with currents
induced by IG waves, whose period can exceed 300 s at this beach under storm conditions (Bertin et al., 2020). As
averaging currents over a longer period within a thicker cell results in a more accurate measurement of the current
velocity, we arbitrarily discarded current profile velocity measurement that departed by more than 0.15 m/s from
the 20 min-averaged velocity measurement performed during the wave cycle within the fixed 2 m-high cell (see
the red triangles in Figure 4). Both model results and measurements clearly show that the cross-shore velocity
is mostly offshore-directed under energetic wave conditions reaching almost −0.5 m/s at the AWAC location
around the 31/01/2021. The model results for a run performed without waves (Rnowave), which only accounts for
the tides and the atmospheric forcing, show a strong positive bias at both locations (e.g., the NB reaches 27.3%
at the ADCP 600 kHz location), so that the comparison between Rref and Rnowave highlights the significant
contribution of the wave-induced current to the cross-shore flow. Based on this comparison, the contribution of
the wave-induced current at the peak of the storm reaches as much as 0.25 m/s.

PEZERAT ET AL. 9 of 21
 21699291, 2022, 3, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JC018108 by Hong Kong University of Science and Technology, Wiley Online Library on [21/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Oceans 10.1029/2021JC018108

Figure 3. Scatter plot of the measured and simulated water depth (panel a) and significant wave height
𝐴𝐴 (𝐴𝐴𝑚𝑚0, panel b) at the
inner surf zone sensors locations using the configuration of reference (Rref).

Figure 4. Measured and simulated cross-shore velocity component timeseries at the AWAC (left panel) and Acoustic Doppler Current Profilers (ADCP) 600 kHz (right
panel) locations displayed for each vertical bin below the sea surface lowest level. Model results are presented for the run of reference (Rref) and for a run performed
without waves (Rnowave). Red triangles tag the discarded measurements. The vertical position of each bin (denoted b) is measured from the bottom and hab stands for
height above the sea-bed. The time range of the storm event is delimited by the gray background.

PEZERAT ET AL. 10 of 21
 21699291, 2022, 3, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JC018108 by Hong Kong University of Science and Technology, Wiley Online Library on [21/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Oceans 10.1029/2021JC018108

Figure 5. Measured and simulated water depth (panel a), significant wave height
𝐴𝐴 (𝐴𝐴𝑚𝑚0, panel b) and cross-shore velocity
component timeseries
𝐴𝐴 (𝐴𝐴𝐴 , panel c) at the Acoustic Doppler Velocimeter location.

At the ADV location, burst-averaged cross-shore velocity measurements are quite scattered, especially during the
second and third tidal cycle with fluctuations reaching 0.2 m/s. The 30-min averaged current velocities suggest
the presence of Very Low Frequency oscillations (VLF, frequencies below 4 mHz), while IG waves contribution
was presumably filtered. The analysis of the whole data set revealed that waves were mostly normally incident
during the storm event, which is quite common during energetic event at this site (Bertin et al., 2008). As a conse-
quence, mean longshore currents remained weak during the field campaign (ranging from −0.10 to 0.15 m/s) and
were alternatively northward and southward-directed within a very wide surf zone (not shown). Shear instabilities
of mean longshore currents, which require the presence of a strong shear (e.g., associated with highly oblique
large waves breaking over a bar, see Noyes et al., 2004; Oltman-Shay et al., 1989) cannot therefore explain such
VLF motions. Instead, it could be attributed to the breaking of energetic wave groups that has been identified
as a mechanism for the generation of surf zone eddies through the generation of vorticity at the scale of indi-
vidual waves or wave groups (Feddersen, 2014; Long & Özkan-Haller, 2009), which is then transferred to VLF
frequencies and larger spatial scales through non-linear inverse energy cascades (Elgar & Raubenheimer, 2020;
Feddersen, 2014).

These complex dynamical features cannot be reproduced by the present phase-averaged modeling approach,
which undermines the comparison with the field observations (the NRMSE on the cross-shore velocity compo-
nent reaches 24.9%). Both model results and observations qualitatively show that the cross-shore velocity compo-
nent is dominated by a wave-induced seaward-oriented current, whose intensity increases with the significant
wave height (Figures 5b and 5c). As more energy is dissipated by depth-induced breaking when the significant
wave height increases, it further suggests that non-conservative breaking wave force strengthens locally this
seaward-oriented current.

5. Discussion
The results clearly show that the wave-induced circulation plays a crucial role on the cross-shore flow. Circula-
tion patterns show a strong seaward-directed current in the lower part of the water column as far as 4 km from
the shoreline, which cannot be reproduced by solely accounting for the wind and the tidal forcing. This unsteady
cross-shore circulation is quite well reproduced by the model, which shows excellent predictive skills for short
waves with errors on bulk parameters ranging from 4% to 9%. Switching off the wind does not significantly
impact the results on short waves because the local wave growth remains weak (not shown). A new configuration

PEZERAT ET AL. 11 of 21
 21699291, 2022, 3, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JC018108 by Hong Kong University of Science and Technology, Wiley Online Library on [21/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Oceans 10.1029/2021JC018108

Figure 6. Wave energy dissipation rates profiles associated with bottom friction 𝐴𝐴 (𝐴𝐴𝑏𝑏𝑏𝑏 ), whitecapping
𝐴𝐴 (𝐴𝐴𝑑𝑑𝑑𝑑), depth-induced
breaking
𝐴𝐴 (𝐴𝐴𝑑𝑑𝑑𝑑 ) and roller energy dissipation
𝐴𝐴 rates ((1 − 𝛼𝛼𝑅𝑅 )𝐷𝐷𝑑𝑑𝑑𝑑 + 𝐷𝐷𝑅𝑅 ) under high (a) or moderate (b) wave energy conditions.

of the model was thus setup with wind and tidal forcing switched off in order to investigate the driving mech-
anisms of the wave-induced circulation while bypassing the unsteadiness associated with tides and wind. This
configuration was run for two distinct 24 hr-periods associated with storm waves
𝐴𝐴 (𝐴𝐴𝑚𝑚0 = 5.3 m – 30/01/2021) and
moderate wave energy conditions
𝐴𝐴 (𝐴𝐴𝑚𝑚0 = 2.0 m – 04/02/2021).

5.1. Wave Energy Dissipation Patterns

Considering the significant contribution of the wave energy dissipation processes either on the TKE injection
or through the non-conservative wave forces, the energy dissipation rates associated with depth-induced break-
𝐴𝐴 ing (𝐴𝐴𝑑𝑑𝑑𝑑), whitecapping
𝐴𝐴 (𝐴𝐴𝑑𝑑𝑑𝑑) and bottom friction
𝐴𝐴 (𝐴𝐴𝑏𝑏𝑏𝑏 ) are first examined following the approach of Pezerat
et al. (2021), based on the empirical𝐴𝐴ratio 𝐴𝐴 reading:

∫𝜎𝜎 ∫𝜃𝜃
𝐷𝐷𝑑𝑑𝑑𝑑
(15)
𝑅𝑅 = with 𝐷𝐷𝑥𝑥 = 𝜎𝜎 ′ 𝑆𝑆𝑥𝑥 𝑑𝑑𝑑𝑑 ′ 𝑑𝑑𝑑𝑑 ′
(𝐷𝐷𝑑𝑑𝑑𝑑 + 𝐷𝐷𝑏𝑏𝑏𝑏 + 𝐷𝐷𝑑𝑑𝑑𝑑 )

where the subscript

𝐴𝐴 𝐴𝐴 represents either of
𝐴𝐴 the “db, " “ds”
𝐴𝐴 or “𝑏𝑏𝑏𝑏 ” subscripts. As shown in Figure 6 along a
cross-shore profile (see Figure 1), one can distinguish three typical areas. Offshore (zone 𝐴𝐴 I, 𝐴𝐴 < 0.1), wave
energy dissipation is mostly associated with bottom friction. Closer to shore, a transition zone appears (zone II,
0.1 < 𝐴𝐴 < 0.9), within which bottom friction, whitecapping and depth-induced breaking together significantly
𝐴𝐴
contribute to the incident wave energy dissipation, before depth-induced breaking becomes dominant (zone III,
𝐴𝐴 𝐴𝐴 > 0.9). Most notably, the transition zone is much wider under high wave energy conditions (approximately
6 km-long spaning from approximately 15 m- to 5 m-depth) than under moderate wave energy conditions (less
than 1 km-long with water depth comprised between 5 and 7 m). The edge of the third zone is approximately
located at the same distance from the shoreline in both situations (≃5 m-depth). Interestingly, the wave energy
dissipation rates associated with whitecapping and bottom friction are of the same order within the zones I and
II under high wave energy conditions, and even dominate the depth-induced breaking contribution over approx-
imately 5 km up to a water-depth of the order of 15 m (Figure 6a). Although it is commonly understood that the
surf zone is widening under storm waves, it appears more appropriate to introduce this conceptual transition zone,
while the third zone more likely corresponds to the inner surf zone. It is worth noting that the development of this

PEZERAT ET AL. 12 of 21
 21699291, 2022, 3, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JC018108 by Hong Kong University of Science and Technology, Wiley Online Library on [21/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Oceans 10.1029/2021JC018108

Table 1 wide transition zone is also presumably related to the gentle and smoothly
Turbulence Settings for the Runs Rref, Rturb1, Rturb2 and Rturb3 increasing bottom slope that characterizes the study area, while the transi-
tion to a regime dominated by depth-induced breaking is much more abrupt
Mixing length
over barred beach or fringing environments. It would be thus interesting
𝐴𝐴 𝐴𝐴𝑤𝑤 = 1 𝐴𝐴 𝐴𝐴𝑤𝑤=0.5 to perform similar analysis in contrasted environments, with a diversity of
𝐴𝐴 𝐴𝐴𝐾𝐾 𝐴𝐴 𝐴𝐴𝑑𝑑𝑑𝑑 = 1 Rref Rturb2 beach profiles and bottom substrate.
𝐴𝐴 𝐴𝐴𝑑𝑑𝑑𝑑 = 0 Rturb1 Rturb3 The roller model shows a weak effect on the location where energy is dissi-
Note. The mixing length increases𝐴𝐴with 𝐴𝐴𝑤𝑤 𝐴𝐴while 𝐴𝐴𝑑𝑑𝑑𝑑 controls the TKE pated. This could be presumably attributed to the gentle bottom slope char-
injection associated with wave energy dissipation through whitecapping. acterizing the study area, while previous studies highlighted the significant
roller effect over a barred and steeper beach (e.g., Kumar et al., 2012; Reniers
et al., 2004; Uchiyama et al., 2010; Zheng et al., 2017).

5.2. Sensitivity to Vertical Mixing

The results presented above show a substantial wave energy dissipation asso-
ciated with whitecapping and depth-induced breaking as far as 6 km from the
shoreline under energetic conditions, which further suggest that a substantial
Table 2
Relative Difference𝐴𝐴(in %) RD1∕2 (𝑋𝑋) = |𝑋𝑋̂ 1 − 𝑋𝑋̂ 2 |∕|𝑋𝑋̂ 2 | of Modeled amount of TKE could be injected at the surface. Three principal parameters
Turbulent Kinetic Energy
𝐴𝐴 (𝐴𝐴 ), Vertical Eddy Viscosity
𝐴𝐴 (𝜈𝜈) and Cross-Shore control the injection of TKE at the surface𝐴𝐴 (𝐴𝐴𝐴𝐴𝑑𝑑𝑑𝑑 and 𝐴𝐴𝑑𝑑𝑑𝑑) and its vertical decay
Velocity ̂ Vertical Profiles for the Different Configurations of the Model𝐴𝐴 𝑧𝑧0 . If we consider the default value taken 𝐴𝐴 for 𝐴𝐴𝑑𝑑𝑑𝑑 𝐴𝐴 and 𝐴𝐴𝑑𝑑𝑑𝑑, we note that
( 𝑠𝑠 )
𝐴𝐴 (𝑢𝑢)
Within the Three Delimited Area Introduced in Section 5.1, Under High for equivalent energy dissipated through whitecapping and depth-induced
(HE) or Moderate (LE) Wave Energy Conditions
breaking𝐴𝐴(i.e., 𝐴𝐴
𝐴𝐴𝑑𝑑𝑑𝑑 ∼ 𝐴𝐴𝑑𝑑𝑑𝑑), more weight is given to the whitecapping contri-
Rturb1/Rref Rturb3/Rturb2 Rturb2/Rref Rturb3/Rturb1 bution. Previous studies have highlighted the effect of breaking-wave-gener-
HE zone I ated turbulence on the mean circulation within the surf zone, which results in
a reduction of the vertical shear of the horizontal current (e.g., Feddersen &
𝐴𝐴 RD(K) 92 88 30 5
Trowbridge, 2005; Kumar et al., 2012). In intermediate water depths, Pask-
𝐴𝐴 RD(𝜈𝜈) 62 49 42 8
yabi et al. (2012) showed that surface currents associated with Ekman trans-
𝐴𝐴 RD(𝑢𝑢)
̂ 3 6 12 4 port are better reproduced when accounting for whitecapping contribution to
zone II the wave-enhanced mixing, while Lentz et al. (2008) used a crude parameter-
𝐴𝐴 RD(𝐾𝐾) 60 56 27 21 ization based on the wind stress for the vertical eddy viscosity for their study
𝐴𝐴 RD(𝜈𝜈) 25 23 41 28 over the inner continental shelf. Overall, the contribution of the wave-en-
hanced turbulence to the mean circulation under the combined effects of
𝐴𝐴 RD((𝑢𝑢)
̂ 7 4 28 19
depth-induced breaking and whitecapping has thus never been evaluated
zone III
across the shoreface (note that Kumar et al., 2012, gathered both contribu-
𝐴𝐴 RD(K ) 2 2 16 16 tions to compute the surface flux of TKE and𝐴𝐴used 𝐴𝐴 𝑑𝑑𝑑𝑑 = 𝐴𝐴𝑑𝑑𝑑𝑑). In addition,
𝐴𝐴 RD(𝜈𝜈) <1 <1 29 29 a sensitivity analysis of the vertical shear of the mean cross-shore current
𝐴𝐴 RD(𝑢𝑢)
̂ <1 <1 30 30 to the parameterization of the surface mixing length 𝐴𝐴 𝐴𝐴𝑠𝑠0 was performed to
LE zone I supplement the study of Moghimi et al. (2016), who highlighted the sensitiv-
ity of turbulence closure models 𝐴𝐴 to 𝐴𝐴𝑠𝑠0 because of the power law for the decay
𝐴𝐴 RD(K) 66 60 25 12
of the TKE. Three additional runs were thus retained (see Table 1) to assess
𝐴𝐴 RD(𝜈𝜈) 26 18 26 12
the sensitivity of the vertical mixing to the parameterization of the TKE
𝐴𝐴 RD(𝐴𝐴𝐴 ) 2 2 10 7 injection and how it further impacts the cross-shore circulation. The results
zone II are compared along vertical profiles of𝐴𝐴TKE (𝐾𝐾), vertical eddy viscosity 𝐴𝐴 (𝜈𝜈)
𝐴𝐴 RD(K) 30 28 24 21 and cross-shore quasi-Eulerian velocity component 𝐴𝐴 ̂ distributed along the
(𝑢𝑢)
𝐴𝐴 RD(𝜈𝜈) 10 9 29 25 aforementioned cross-shore profile (see Figure 7 and Table 2).
𝐴𝐴 RD(𝐴𝐴𝐴 ) 2 2 32 27 Under storm conditions, whitecapping explains more than two-third of
zone III the TKE injected at the surface in regions I and II. This greatly affects the
𝐴𝐴 RD(K) 4 3 16 16 mixing in the upper third of the water column (see ν almost doubled at some
locations between Rref and Rturb1 in Figure 7). Note, however, that it only
𝐴𝐴 RD(𝜈𝜈) <1 <1 29 29
slightly impacts the cross-shore circulation
𝐴𝐴 (RD (𝑢𝑢)
̂ ⩽ 7%, see the compari-
𝐴𝐴 RD(𝑢𝑢)
̂ <1 <1 30 30
sons Rturb1/Rref and Rturb3/Rturb2 in Table 2), with the largest differences
Note. High RD values indicate a strong sensitivity to turbulence settings, being found in the upper part of the water column. Furthermore, in both high
either to the mixing length or the surface flux of TKE. or moderate wave energy conditions, the model shows a strong sensitivity

PEZERAT ET AL. 13 of 21
 21699291, 2022, 3, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JC018108 by Hong Kong University of Science and Technology, Wiley Online Library on [21/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Oceans 10.1029/2021JC018108

0 K [m²/s²]

5 0.025
0.050
10
z [m]

Rref
15 Rturb1
Rturb2
20 Rturb3 a) I II III

0 [m²/s]

5 0.025
0.050
10
z [m]

15

20
b) I II III

0 û [m/s]

5 0.100
0.200
10
z [m]

15

20
c) I II III
2000 4000 6000 8000 10000 12000
cross-shore distance [m]

Figure 7. Vertical profiles of turbulent kinetic energy (K, panel a), vertical eddy viscosity (ν, panel b) and quasi-Eulerian cross-shore velocity
𝐴𝐴 (𝐴𝐴𝐴 , panel c). The
results are presented for the four configurations of the model (see Table 1) and are extracted under high wave energy conditions. The delimited zones are defined in
Section 5.1.

to the parameterization of the surface mixing length, which strengthens as the injection of TKE at the surface
increases closer to shore (within the zones II and III, see Table 2). With a shorter surface mixing length, the
eddy viscosity at the surface is weaker, such that the cross-shore velocity profiles are more sheared (see Figure 7
under high wave energy conditions, while similar results – not shown – are found under moderate wave energy
conditions). As a result, the orientation of the cross-shore flow near the surface even changes for one profile
within the zone II under high energy condition with a shorter mixing length. The relative difference on the cross-
shore velocity reaches 32% depending on the parameterization 𝐴𝐴 of 𝐴𝐴𝑠𝑠0. Interestingly, comparing Rturb2/Rref and
Rturb3/Rturb1 tends to show that the sensitivity to the parameterization of the surface mixing length is stronger
depending on the contribution of the whitecapping to the TKE injection (see Table 2). Overall, it appears that the
parameterization of the TKE injection impacts the vertical shear of the cross-shore velocity as far as 6 km from
the shoreline under energetic conditions.

5.3. Wave-Induced Circulation Patterns and Forcing Mechanisms

Under moderate wave energy conditions, non-conservative breaking wave forces arise relatively close to shore,
up to 1 km from the shoreline (Figure 8a) and are mostly associated with depth-induced breaking, which domi-
nates the wave energy dissipation within the zones II and III (Figure 6b). The quasi-Eulerian cross-shore flow
therefore shows a strong seaward-directed current reaching 0.3 m/s in the lower part of the water column, whereas
a shoreward-directed current arises near the surface (Figure 8c). The depth-averaged quasi-Eulerian cross-shore
flow (i.e., the undertow) nearly compensates for the depth-averaged cross-shore Stokes drift velocity component

PEZERAT ET AL. 14 of 21
 21699291, 2022, 3, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JC018108 by Hong Kong University of Science and Technology, Wiley Online Library on [21/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Oceans 10.1029/2021JC018108

Figure 8. Cross-shore profiles under moderate wave energy conditions of the wave force cross-shore component, which includes the Stokes-Coriolis term, the vortex
force, the wave-induced pressure term and the non-conservate wave forces (a), the Stokes drift velocity cross-shore component (b), the quasi-Eulerian velocity cross-
shore component (c), and the 2DV Lagrangian circulation streamlines and magnitude (d). For readability, S-level indices are used for representing the wave force profile
(the level 0 corresponds to the bottom and the level 24 to the free surface). Note that the cross-shore distance axis here extends over 2 km.

(the divergence of the alongshore velocity component is not exactly zero), while the depth-varying cross-shore
flow is locally strengthened by the effect of the non-conservative breaking wave forces within the zones II and
III, resulting in a 100% stronger seaward-directed current in the lower part of the water column than the surface
Stokes drift velocity (Figures 8b and 8c and 10b). In order to better evaluate the contribution of breaking, an addi-
tional run was performed with the non-conservative breaking wave forces uniformly distributed over the vertical
(RFbr). The resulting quasi-Eulerian cross-shore flow is less sheared near the surface within the zones II and III,
yielding a weaker seaward-directed current in the lower part of the water column than that obtained with a near
surface momentum source by up to a factor three (Figure 10b). Within zone I, the quasi-Eulerian cross-shore
flow is also mostly seaward-directed, with a maximum intensity in the upper part of the water column, while a
weak shoreward-directed current associated with wave streaming arises near the bottom and extends inside the
zone II (Figures 8a and 8c and 10b). As a result, a strong clockwise Lagrangian overturning circulation develops
within the zones II and III (the magnitude of the current locally reaches 0.3 m/s), while a weaker anti-clockwise
overturning cell arises at the seaward edge of the zone II, which constrains the offshore flow (see the upward
deflection of the streamlines of the clockwise cell within the zone II in Figure 8d). A similar circulation patern in
the vicinity of the surf zone was found by Wang et al. (2020). The magnitude of the Lagrangian circulation then
decreases relatively rapidly within zone I (Figure 8d). Interestingly, the cross-shore quasi-Eulerian flow does not
exactly compensate for the cross-shore Stokes drift velocity component, which would have been expected assum-
ing a balance between the Coriolis force associated with the quasi-Eulerian flow and the Stokes-Coriolis force, as
pointed out by Lentz et al. (2008). Presumably, this could be attributed to the shape of the coastline, which is not
alongshore-uniform at the scale of several kilometers, the distance where the instruments were located.

Under high wave energy conditions, non-conservative breaking wave forces arise at the surface as far as 6 km
from the shoreline and strengthen shoreward (Figure 9a), associated with a significant wave energy dissipa-
tion occurring through whitecapping and depth-induced breaking within the zones II and III (Figure 6a). The
quasi-Eulerian cross-shore flow shows a strong seaward-directed current of the order of 0.2 m/s up to 4 km from
the shoreline, whose intensity progressively decreases offshore reaching 0.1 m/s as far as 10 km from the shore-
line (Figure 9c). Similar to moderate wave energy conditions, this current is strengthened locally by the effect of
the non-conservative breaking wave forces within the zones II and III (Figure 10a). Interestingly, no near-bottom

PEZERAT ET AL. 15 of 21
 21699291, 2022, 3, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JC018108 by Hong Kong University of Science and Technology, Wiley Online Library on [21/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Oceans 10.1029/2021JC018108

Figure 9. Same as Figure 8 under high wave energy conditions. Note that the cross-shore distance axis here extends over 12 km.

shoreward-directed current arises as the intensity of the bottom streaming decreases beyond 20 m-depth, while
closer to shore the cross-shore flow in the lower part of the water column is dominated by the strong seaward-di-
rected current. The Lagrangian circulation therefore shows a wide clockwise overturning cell extending over
8 km, which generates a seaward-oriented jet in the lower part of the water column (Figure 9d) that contrasts with
the circulation patern obtained under moderate wave energy conditions.

6. Conclusions and Perspectives

In this study, we presented a data set comprising water levels, wave parameters and currents collected under
fair weather and storm conditions issued from a field campaign carried out in early 2021 within the shoreface
and the surf zone of the dissipative beach of Saint-Trojan (France). These results were complemented with the
predictions from a state-of-the-art wave-averaged 3D circulation modeling system coupling the circulation model
SCHISM (Zhang et al., 2016) with the spectral wave model WWM (Roland et al., 2012) in order to study the
wave-induced cross-shore circulation in the nearshore area up to the surf zone. Both measurements and model
results clearly showed that the cross-shore flow is dominated by a strong seaward-directed current in the lower
part of the water column, which under storm wave conditions can reach 0.2 m/s as far as 4 km from the shoreline.
The model was further employed to study the driving mechanisms of the wave-induced cross-shore circulation. It
was shown that a wide (≃6 km) transition zone appears under high wave energy conditions where depth-induced
breaking, whitecapping and bottom friction all contribute significantly to the wave energy dissipation, whereas
the surf zone edge appears clearly delimited under moderate wave energy conditions. Once wave energy dissi-
pation through breaking processes starts being substantial, sensitivity tests performed with the model tended to
show that the parameterization of the wave-enhanced mixing could substantially impact the vertical shear of the
cross-shore velocity close to the surface, while non-conservative breaking wave forces strengthen locally the
seaward-directed current in the lower part of the water column. The wave-induced cross-shore circulation thus
presents very contrasting patterns. Under moderate wave energy conditions, the dominance of the wave energy
dissipation by bottom friction seaward of the surf zone allows the development of a near-bottom onshore flow
associated with bottom streaming of the order of a few centimeters per second. Thus, the surf zone clockwise
Lagrangian overturning circulation weakens, while an additional anti-clockwise overturning cell arises seaward
of the surf zone and extends within it. In contrast, under high energy conditions the Lagrangian circulation shows
a seaward-directed jet in the lower part of the water column, whose intensity progressively decreases offshore.

PEZERAT ET AL. 16 of 21
 21699291, 2022, 3, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JC018108 by Hong Kong University of Science and Technology, Wiley Online Library on [21/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Oceans 10.1029/2021JC018108

Figure 10. Vertical profiles of the Stokes drift velocity cross-shore component
𝐴𝐴 (𝑢𝑢𝑆𝑆 ) and the quasi-Eulerian cross-shore
velocity
𝐴𝐴 ̂ under high (a) or moderate (b) wave energy conditions for the baseline run (Rref) and for a run performed with
(𝑢𝑢)
the non-conservative breaking wave forces uniformly distributed (RFbr). The delimited zones are defined in Section 5.1.

Among different implications, this study opens perspectives for sediment transport modeling using a similar 3D
framework in order to produce realistic morphological evolutions across the shoreface.

Appendix A: Vortex Force Framework in the Modeling System SCHISM

In the vortex force framework the continuity and momentum equations read:

𝜕𝜕 𝜕𝜕𝜕 𝜕𝜕 𝜕𝜕𝜕 𝜕𝜕 𝑤𝑤̂

(A1) + + =0
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
( ( ))
𝐷𝐷𝐷𝐷𝐷 1 𝜕𝜕𝜕𝜕𝐻𝐻 𝜕𝜕 𝜕𝜕 𝜕𝜕𝜕 𝜕𝜕 𝜕𝜕𝜕 𝜕𝜕 𝜕𝜕𝜕 𝜕𝜕 𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
( )
(A2) − 𝑓𝑓 𝑓𝑓𝑓 = − + 𝜈𝜈 + 𝑣𝑣𝑆𝑆 𝑓𝑓𝐶𝐶 + − − 𝑤𝑤𝑆𝑆 − + 𝐹𝐹̂𝑥𝑥
𝐷𝐷𝐷𝐷 𝜌𝜌 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
( ( ))
𝐷𝐷𝐷𝐷𝐷 1 𝜕𝜕𝜕𝜕𝐻𝐻 𝜕𝜕 𝜕𝜕 𝜕𝜕𝜕 𝜕𝜕 𝜕𝜕𝜕 𝜕𝜕 𝜕𝜕𝜕 𝜕𝜕 𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
( )
(A3) + 𝑓𝑓 𝑓𝑓𝑓 = − + 𝜈𝜈 − 𝑢𝑢𝑆𝑆 𝑓𝑓𝐶𝐶 + − − 𝑤𝑤𝑆𝑆 − + 𝐹𝐹̂𝑦𝑦
𝐷𝐷𝐷𝐷 𝜌𝜌 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕

𝐴𝐴where 𝑢𝑢𝑢 𝑢 𝑤𝑤̂ is the quasi-Eulerian velocity, equal to the mean Lagrangian velocity [𝑢𝑢𝑢 𝑢𝑢𝑢 𝑢𝑢] minus the Stokes
[ ]
̂ 𝑢𝑢𝑢 𝐴𝐴
drift velocity
𝐴𝐴 [𝑢𝑢𝑆𝑆 , 𝑣𝑣𝑆𝑆 , 𝑤𝑤𝑆𝑆 ]. In Equations A2 and 𝐴𝐴 A3, 𝐴𝐴𝐶𝐶 is the Coriolis parameter,
𝐴𝐴 𝐴𝐴 is the water density,
𝐴𝐴 𝐴𝐴 is the
gravitational acceleration, 𝐴𝐴 𝐴𝐴 is the vertical eddy viscosity,
𝐴𝐴 𝐴𝐴𝐻𝐻 is the hydrostatic pressure,
𝐴𝐴 𝐴𝐴 is the wave-induced
mean pressure 𝐴𝐴 and 𝐹𝐹̂𝑥𝑥 , 𝐹𝐹̂𝑦𝑦 gathers the non-conservative wave forces.
[ ]

The three components of the Stokes drift velocities, the wave-induced pressure term and the non-conservative
wave forces are all computed from local variables issued from WWM, which simulates the generation, propaga-
tion and transformation of short waves by solving the Wave Action Equation (e.g., Komen et al., 1994):
̇
( )
𝜕𝜕𝜕𝜕 𝜕𝜕 𝜕𝜕 𝜕𝜕 (𝜎𝜎𝜎𝜎)
̇ 𝜕𝜕 𝜃𝜃𝜃𝜃
(A4) + ((𝐶𝐶𝑔𝑔𝑔𝑔𝑔 + 𝑢𝑢)
̃ 𝑁𝑁) + ((𝐶𝐶𝑔𝑔𝑔𝑔𝑔 + 𝑣𝑣)
̃ 𝑁𝑁) + + = 𝑆𝑆𝑡𝑡𝑡𝑡𝑡𝑡
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕

PEZERAT ET AL. 17 of 21
 21699291, 2022, 3, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JC018108 by Hong Kong University of Science and Technology, Wiley Online Library on [21/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Oceans 10.1029/2021JC018108

where
𝐴𝐴 𝐴𝐴 is the relative wave frequency, which is related to the wavenumber 𝐴𝐴 𝐴𝐴 by the linear dispersion relation;
𝐴𝐴 𝐴𝐴 corresponds to the wave direction;
𝐴𝐴 𝐴𝐴 = 𝜌𝜌𝜌𝜌𝜌𝜌∕𝜎𝜎 is the wave action density spectrum,𝐴𝐴with 𝐴𝐴 the sea surface
elevation density spectrum; 𝐶𝐶𝑔𝑔𝑔𝑔𝑔 , 𝐶𝐶𝑔𝑔𝑔𝑔𝑔 is the intrisic group velocity; 𝑢 is the advective current velocity, here
[ ]
𝐴𝐴 𝐴𝐴 [𝑢𝑢𝑢
̃ 𝑢𝑢]
equal to the depth-averaged horizontal quasi-Eulerian velocity. Finally, 𝐴𝐴 𝐴𝐴tot is a sum of source terms that account
for the energy input due to wind, non linear wave-wave interactions, and energy dissipation due to whitecapping,
depth-induced breaking and bottom friction (cf., Section 3.1).

The horizontal Stokes drift velocity vector is given by:

∫𝜎𝜎 ∫𝜃𝜃
cosh(2𝑘𝑘(𝑧𝑧 + 𝑑𝑑)) [ ]
(A5)
[𝑢𝑢𝑆𝑆 , 𝑣𝑣𝑆𝑆 ] = 𝜎𝜎 ′ 𝑘𝑘𝑘𝑘 2
cos𝜃𝜃 ′ , sin𝜃𝜃 ′ 𝑑𝑑𝑑𝑑 ′ 𝑑𝑑𝑑𝑑 ′
sinh (𝑘𝑘𝑘)

where 𝐴 = 𝑑𝑑 + 𝜂𝜂̄ is the (local) phase-averaged water depth𝐴𝐴(with 𝐴𝐴 , the still water depth
𝐴𝐴 𝐴𝐴 and 𝐴𝐴𝐴 , the phase-averaged
free surface elevation). The vertical Stokes drift component is given by the horizontal divergence 𝐴𝐴 of [𝑢𝑢𝑆𝑆 , 𝑣𝑣𝑆𝑆 ] as
the full Stokes drift flow is non divergent at the lowest order (Ardhuin et al., 2008). The wave-induced pressure
term reads:

∫𝜎𝜎 ∫𝜃𝜃 sinh(2𝑘𝑘𝑘)

𝑔𝑔𝑔𝑔𝑔𝑔
(A6)
𝐽𝐽 = 𝑑𝑑𝑑𝑑 ′ 𝑑𝑑𝑑𝑑 ′

Finally, the formulation of the non-conservative wave forces are detailed in Section 3.2.1.
The wave model is supplemented by a roller model that helps representing the inertia of depth-induced breaking
processes by slightly advecting the location where energy is actually dissipated toward the shoreline (Svend-
sen, 1984b). As compared to the implementation detailed in Guérin et al. (2018), the roller model solves a balance
equation for the roller kinetic energy
𝐴𝐴 𝐴𝐴𝑅𝑅 (e.g., Reniers et al., 2004), slightly modified to account for the modifi-
cation of the wave phase velocity by the mean current within the advection term, such that:

𝜕𝜕𝜕𝜕𝑅𝑅 𝜕𝜕 𝜕𝜕
(A7) + (2𝐸𝐸𝑅𝑅 (𝑐𝑐𝑥𝑥 + 𝑢𝑢))
̃ + (2𝐸𝐸𝑅𝑅 (𝑐𝑐𝑦𝑦 + 𝑣𝑣))
̃ = 𝛼𝛼𝑅𝑅 𝐷𝐷𝑑𝑑𝑑𝑑 − 𝐷𝐷𝑅𝑅
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕

where
𝐴𝐴 [𝐴𝐴𝑥𝑥 , 𝑐𝑐𝑦𝑦 ] is the wave phase velocity computed by means of the short wave (continuous) peak wavenum-
𝐴𝐴 ber (𝑘𝑘𝑝𝑝 ) and mean direction 𝐴𝐴 𝐴𝐴(𝜃𝜃𝑚𝑚 ); 𝐴𝐴𝑑𝑑𝑑𝑑 is the bulk wave energy dissipation rate by depth-induced breaking;
𝐴𝐴 𝐴𝐴𝑅𝑅
is the percentage of wave energy dissipation by depth-induced breaking transferred to the rollers (Tajima &
Madsen, 2006) 𝐴𝐴 and 𝐴𝐴𝑅𝑅 is the roller energy dissipation rate, which reads:

2𝑔𝑔sin𝛽𝛽𝑅𝑅 𝐸𝐸𝑅𝑅
𝐷𝐷𝑅𝑅 = √
(A8)
𝑐𝑐𝑥𝑥2 + 𝑐𝑐𝑦𝑦2

where
𝐴𝐴 sin 𝐴𝐴𝑅𝑅 = 0.1 is the roller angle (Nairn et al., 1991; Reniers et al., 2004). Surface rollers contribute to the
total mass flux in proportion to the roller energy. Although this transport primarily occurs near the surface, above
trough level, there is no consensus on its vertical distribution. We here follow the choice to impose an homoge-
neous vertical distribution. This contribution is accounted for through an additional term in the horizontal Stokes
drift velocity vector, which reads:

2𝐸𝐸
𝑢𝑢𝑆𝑆𝑆𝑆𝑆 , 𝑣𝑣𝑆𝑆𝑆𝑆𝑆 = √ 𝑅𝑅
[ ]
[cos𝜃𝜃𝑚𝑚 , sin𝜃𝜃𝑚𝑚 ] ∀ 𝑧𝑧 ∈ [−𝑑𝑑𝑑 𝑑𝑑]
𝑑
(A9)
𝜌𝜌𝜌 𝑐𝑐𝑥𝑥2 + 𝑐𝑐𝑦𝑦2

Note, however,𝐴𝐴 that 𝐴𝐴𝑆𝑆 is still computed

𝐴𝐴with 𝐴𝐴𝑆𝑆 , 𝑣𝑣𝑆𝑆 from Equation A5 as it is assumed that rollers do not contrib-
ute to the vertical transport.

Data Availability Statement

The processed field data presented in this study are available through a Zenodo repository (Pezerat et al., 2022).
The instructions to download and install the model used in this study can be accessed freely at https://github.
com/schism-dev/schism. Figures were made with python module Matplotlib version 3.1.1, available under the
Matplotlib licenses at https://matplotlib.org/.

PEZERAT ET AL. 18 of 21
 21699291, 2022, 3, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JC018108 by Hong Kong University of Science and Technology, Wiley Online Library on [21/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Oceans 10.1029/2021JC018108

Acknowledgments References
M. Pezerat is supported by a PhD fellow-
ship from CDA La Rochelle and from the Abdolali, A., Roland, A., Van Der Westhuysen, A., Meixner, J., Chawla, A., Hesser, T. J., et al. (2020). Large-scale hurricane modeling using
FEDER project DURALIT. K. Martins domain decomposition parallelization and implicit scheme implemented in wavewatch iii wave model. Coastal Engineering, 157, 103656.
greatly acknowledges the financial https://doi.org/10.1016/j.coastaleng.2020.103656
support from the University of Bordeaux, Appell, G. F., Bass, P., & Metcalf, M. A. (1991). Acoustic doppler current profiler performance in near surface and bottom boundaries. IEEE
through an International Postdoctoral Journal of Oceanic Engineering, 16(4), 390–396. https://doi.org/10.1109/48.90903
Grant (Idex, nb. 1024R-5030). L. Lavaud Ardhuin, F., O’Reilly, W. C., Herbers, T. H. C., & Jessen, P. F. (2003). Swell transformation across the continental shelf. Part 1: Attenuation and
is supported by a PhD fellowship from directional broadening. Journal of Physical Oceanography, 33, 1921–1939. https://doi.org/10.1175/1520-0485(2003)033<1921:statcs>2.0
the Region Nouvelle-Aquitaine and the .co;2
UNIMA engineering consulting company. Ardhuin, F., Rascle, N., & Belibassakis, K. A. (2008). Explicit wave-averaged primitive equations using a generalized lagrangian mean. Ocean
The authors appreciate the administra- Modelling, 20(1), 35–60. https://doi.org/10.1016/j.ocemod.2007.07.001
tive support of the DDTM to carry out Ardhuin, F., Rogers, E., Babanin, A. V., Filipot, J.-F., Magne, R., Roland, A., et al. (2010). Semiempirical dissipation source functions
long term deployment of sensors. The for ocean waves. Part I: Definition, calibration, and validation. Journal of Physical Oceanography, 40(9), 1917–1941. https://doi.
topo-bathymetric data of the studied area org/10.1175/2010jpo4324.1
were acquired in the scope of the National Battjes, J. A., & Janssen, J. (1978). Energy loss and set-up due to breaking of random waves. Coastal Engineering, 569–587. https://doi.
Observation System DYNALIT (https:// org/10.1061/9780872621909.034
www.dynalit.fr), part of the research Battjes, J. A., & Stive, M. (1985). Calibration and verification of a dissipation model for random breaking waves. Journal of Geophysical
infrastructure ILICO. The Observatoire Research, 90(C5), 9159–9167. https://doi.org/10.1029/jc090ic05p09159
de la Côte Aquitaine (OCA) partly funded Bennis, A.-C., Ardhuin, F., & Dumas, F. (2011). On the coupling of wave and three-dimensional circulation models: Choice of theoretical
the field campaign presented in this framework, practical implementation and adiabatic tests. Ocean Modelling, 40(3–4), 260–272. https://doi.org/10.1016/j.ocemod.2011.09.003
study. Model development carried out in Bertin, X., Bruneau, N., Breilh, J.-F., Fortunato, A. B., & Karpytchev, M. (2012). Importance of wave age and resonance in storm surges: The
this study is a contribution to the project case xynthia, Bay of Biscay. Ocean Modelling, 42, 16–30. https://doi.org/10.1016/j.ocemod.2011.11.001
"Segundo Torrao", funded by the Fonda- Bertin, X., Castelle, B., Chaumillon, E., Butel, R., & Quique, R. (2008). Longshore transport estimation and inter-annual variability at a
tion de France and Fondation Edouard et high-energy dissipative beach: St. Trojan beach, SW Oléron Island, France. Continental Shelf Research, 28(10–11), 1316–1332. https://doi.
Geneviève Buffard through the program org/10.1016/j.csr.2008.03.005
"Nouveaux commanditaires Science". N. Bertin, X., Li, K., Roland, A., & Bidlot, J.-R. (2015). The contribution of short-waves in storm surges: Two case studies in the bay of Biscay.
Lachaussée built the structures to anchor Continental Shelf Research, 96, 1–15. https://doi.org/10.1016/j.csr.2015.01.005
the sensors, whereas T. Coulombier and Bertin, X., Martins, K., de Bakker, A., Chataigner, T., Guérin, T., Coulombier, T., & de Viron, O. (2020). Energy transfers and reflection of
D. Dausse provided valuable assistance infragravity waves at a dissipative beach under storm waves. Journal of Geophysical Research: Oceans, 125(5), e2019JC015714. https://doi.
during the field campaign. Lastly, the org/10.1029/2019jc015714
authors want to thank O. de Viron for his Bishop, C. T., & Donelan, M. A. (1987). Measuring waves with pressure transducers. Coastal Engineering, 11(4), 309–328. https://doi.
enlightening insights on data post-pro- org/10.1016/0378-3839(87)90031-7
cessing, and the two anonymous review- Blumberg, A. F., & Mellor, G. L. (1987). A description of a three-dimensional coastal ocean circulation model. Three-dimensional coastal ocean
ers for their constructive comments. models, 4, 1–16. https://doi.org/10.1029/co004p0001
Booij, N., Ris, R. C., & Holthuijsen, L. H. (1999). A third-generation wave model for coastal regions: 1. Model description and validation. Journal
of geophysical research, 104(C4), 7649–7666. https://doi.org/10.1029/98jc02622
Burchard, H., Bolding, K., & Villarreal, M. (1999). Gotm, a general ocean turbulence model: Theory, implementation and test cases (Tech. Rep.
No. EUR18745). European Commisssion.
Castelle, B., Marieu, V., Bujan, S., Splinter, K. D., Robinet, A., Sénéchal, N., & Ferreira, S. (2015). Impact of the winter 2013–2014 series of
severe western Europe storms on a double-barred sandy coast: Beach and dune erosion and megacusp embayments. Geomorphology, 238,
135–148. https://doi.org/10.1016/j.geomorph.2015.03.006
Coco, G., Senechal, N., Rejas, A., Bryan, K. R., Capo, S., Parisot, J., et al. (2014). Beach response to a sequence of extreme storms. Geomorphol-
ogy, 204, 493–501. https://doi.org/10.1016/j.geomorph.2013.08.028
Craig, P. D., & Banner, M. L. (1994). Modeling wave-enhanced turbulence in the ocean surface layer. Journal of Physical Oceanography, 24(12),
2546–2559. https://doi.org/10.1175/1520-0485(1994)024<2546:mwetit>2.0.co;2
De Bakker, A., Herbers, T., Smit, P., Tissier, M., & Ruessink, B. (2015). Nonlinear infragravity–wave interactions on a gently sloping laboratory
beach. Journal of Physical Oceanography, 45(2), 589–605. https://doi.org/10.1175/jpo-d-14-0186.1
Deigaard, R., Justesen, P., & Fredsøe, J. (1991). Modelling of undertow by a one-equation turbulence model. Coastal Engineering, 15(5–6),
431–458. https://doi.org/10.1016/0378-3839(91)90022-9
Delpey, M., Ardhuin, F., Otheguy, P., & Jouon, A. (2014). Effects of waves on coastal water dispersion in a small estuarine bay. Journal of
Geophysical Research: Oceans, 119(1), 70–86. https://doi.org/10.1002/2013jc009466
Déqué, M., Dreveton, C., Braun, A., & Cariolle, D. (1994). The arpege/ifs atmosphere model: A contribution to the French community climate
modelling. Climate Dynamics, 10(4), 249–266. https://doi.org/10.1007/bf00208992
Dodet, G., Bertin, X., Bouchette, F., Gravelle, M., Testut, L., & Wöppelmann, G. (2019). Characterization of sea-level variations along the metro-
politan coasts of France: Waves, tides, storm surges and long-term changes. Journal of Coastal Research, 88, 10–24. https://doi.org/10.2112/
si88-003.1
Eldeberky, Y. (1997). Nonlinear transformation of wave spectra in the nearshore zone. Oceanographic Literature Review, 4(44), 297.
Eldeberky, Y., & Battjes, J. A. (1996). Spectral modeling of wave breaking: Application to boussinesq equations. Journal of Geophysical
Research, 101(C1), 1253–1264. https://doi.org/10.1029/95jc03219
Elgar, S., & Raubenheimer, B. (2020). Field evidence of inverse energy cascades in the surfzone. Journal of Physical Oceanography, 50(8),
2315–2321. https://doi.org/10.1175/jpo-d-19-0327.1
Feddersen, F. (2012a). Observations of the surf-zone turbulent dissipation rate. Journal of Physical Oceanography, 42(3), 386–399. https://doi.
org/10.1175/jpo-d-11-082.1
Feddersen, F. (2012b). Scaling surf zone turbulence. Geophysical Research Letters, 39(18). https://doi.org/10.1029/2012gl052970
Feddersen, F. (2014). The generation of surfzone eddies in a strong alongshore current. Journal of Physical Oceanography, 44(2), 600–617.
https://doi.org/10.1175/jpo-d-13-051.1
Feddersen, F., & Trowbridge, J. (2005). The effect of wave breaking on surf-zone turbulence and alongshore currents: A modeling study. Journal
of Physical Oceanography, 35(11), 2187–2203. https://doi.org/10.1175/jpo2800.1
Garcez Faria, A., Thornton, E., Lippmann, T., & Stanton, T. (2000). Undertow over a barred beach. Journal of Geophysical Research, 105(C7),
16999–17010. https://doi.org/10.1029/2000jc900084

PEZERAT ET AL. 19 of 21
 21699291, 2022, 3, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JC018108 by Hong Kong University of Science and Technology, Wiley Online Library on [21/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Oceans 10.1029/2021JC018108

Goring, D. G., & Nikora, V. I. (2002). Despiking acoustic doppler velocimeter data. Journal of Hydraulic Engineering, 128(1), 117–126. https://
doi.org/10.1061/(asce)0733-9429(2002)128:1(117)
Grant, W. D., & Madsen, O. S. (1982). Movable bed roughness in unsteady oscillatory flow. Journal of Geophysical Reasearch, 87, 469–481.
https://doi.org/10.1029/jc087ic01p00469
Guérin, T., Bertin, X., Coulombier, T., & de Bakker, A. (2018). Impacts of wave-induced circulation in the surf zone on wave setup. Ocean
Modelling, 123, 86–97. https://doi.org/10.1016/j.ocemod.2018.01.006
Haines, J. W., & Sallenger, A. H., Jr. (1994). Vertical structure of mean cross-shore currents across a barred surf zone. Journal of Geophysical
Research, 99(C7), 14223–14242. https://doi.org/10.1029/94jc00427
Hamm, L., & Peronnard, C. (1997). Wave parameters in the nearshore: A clarification. Coastal Engineering, 32(2–3), 119–135. https://doi.
org/10.1016/s0378-3839(97)81746-2
Hasselmann, K. (1970). Wave-driven inertial oscillations. Geophysical and Astrophysical Fluid Dynamics, 1(3–4), 463–502. https://doi.
org/10.1080/03091927009365783
Hasselmann, S., Hasselmann, K., Allender, J., & Barnett, T. (1985). Computations and parameterizations of the nonlinear energy transfer in a
gravity-wave specturm. Part II: Parameterizations of the nonlinear energy transfer for application in wave models. Journal of Physical Ocean-
ography, 15(11), 1378–1391. https://doi.org/10.1175/1520-0485(1985)015<1378:capotn>2.0.co;2
Huang, Z.-C., Hsiao, S.-C., Hwung, H.-H., & Chang, K.-A. (2009). Turbulence and energy dissipations of surf-zone spilling breakers. Coastal
Engineering, 56(7), 733–746. https://doi.org/10.1016/j.coastaleng.2009.02.003
Klopman, G. (1994). Vertical structure of the flow dur to waves and currents – Laser-doppler flow measurements for waves following or opposing
a current (Techical Report No., H840–30). Deltares (WL).
Komen, G. J., Cavaleri, L., Donelan, M., Hasselmann, K., Hasselmann, S., & Janssen, P. A. E. M. (1994). Dynamics and modelling of ocean
waves. Cambridge, U.K: Cambridge University Press. https://doi.org/10.1017/CBO9780511628955
Kumar, N., Voulgaris, G., Warner, J. C., & Olabarrieta, M. (2012). Implementation of the vortex force formalism in the coupled ocean-atmos-
phere-wave-sediment transport (coawst) modeling system for inner shelf and surf zone applications. Ocean Modelling, 47, 65–95. https://doi.
org/10.1016/j.ocemod.2012.01.003
Lavaud, L., Pezerat, M., Coulombier, T., Bertin, X., & Martins, K. (2020). Hydrodynamics on a rocky shore under moderate-energy wave condi-
tions. Journal of Coastal Research, 95, 1473–1479. https://doi.org/10.2112/si95-284.1
Lentz, S. J., Fewings, M., Howd, P., Fredericks, J., & Hathaway, K. (2008). Observations and a model of undertow over the inner continental shelf.
Journal of Physical Oceanography, 38(11), 2341–2357. https://doi.org/10.1175/2008jpo3986.1
Long, J. W., & Özkan-Haller, H. T. (2009). Low-frequency characteristics of wave group–forced vortices. Journal of Geophysical Research,
114(C8). https://doi.org/10.1029/2008jc004894
Longuet-Higgins, M. (1953). Mass transport in water waves. Philosophical Transactions of the Royal Society, 245, 535–581.
Madsen, O. S. (1995). Spectral wave-current bottom boundary layer flows. Coastal Engineering, 1994, 384–398. https://doi.
org/10.1061/9780784400890.030
Madsen, O. S., Poon, Y. K., & Graber, H. C. (1988). Spectral wave attenuation by bottom friction: Theory. In 21st international conference on
coastal engineering (pp. 492–504). Malaga: ASCE. https://doi.org/10.9753/icce.v21.34
Mathisen, P. P., & Madsen, O. S. (1999). Waves and currents over a fixed rippled bed: 3. Bottom and apparent roughness for spectral waves and
currents. Journal of Geophysical Research, 104(C8), 18447–18461. https://doi.org/10.1029/1999jc900114
Michaud, H., Marsaleix, P., Leredde, Y., Estournel, C., Bourrin, F., Lyard, F., et al. (2012). Three-dimensional modelling of wave-induced current
from the surf zone to the inner shelf. Ocean Science, 8(4), 657–681. https://doi.org/10.5194/os-8-657-2012
Moghimi, S., Thomson, J., Özkan-Haller, T., Umlauf, L., & Zippel, S. (2016). On the modeling of wave-enhanced turbulence nearshore. Ocean
Modelling, 103, 118–132. https://doi.org/10.1016/j.ocemod.2015.11.004
Morgan, S. G., Shanks, A. L., MacMahan, J. H., Reniers, A. J., & Feddersen, F. (2018). Planktonic subsidies to surf-zone and intertidal commu-
nities. Annual Review of Marine Science, 10, 345–369. https://doi.org/10.1146/annurev-marine-010816-060514
Mouragues, A., Bonneton, P., Lannes, D., Castelle, B., & Marieu, V. (2019). Field data-based evaluation of methods for recovering surface wave
elevation from pressure measurements. Coastal Engineering, 150, 147–159. https://doi.org/10.1016/j.coastaleng.2019.04.006
Nairn, R. B., Roelvink, J., & Southgate, H. N. (1991). Transition zone width and implications for modelling surfzone hydrodynamics. Coastal
Engineering, 1990, 68–81. https://doi.org/10.1061/9780872627765.007
Nicolae-Lerma, A., Bulteau, T., Lecacheux, S., & Idier, D. (2015). Spatial variability of extreme wave height along the atlantic and channel
French coast. Ocean Engineering, 97, 175–185. https://doi.org/10.1016/j.oceaneng.2015.01.015
Noyes, T. J., Guza, R., Elgar, S., & Herbers, T. (2004). Field observations of shear waves in the surf zone. Journal of Geophysical Research,
109(C1), C01031. https://doi.org/10.1029/2002jc001761
Oltman-Shay, J., Howd, P., & Birkemeier, W. (1989). Shear instabilities of the mean longshore current: 2. Field observations. Journal of Geophys-
ical Research, 94(C12), 18031–18042. https://doi.org/10.1029/jc094ic12p18031
Paskyabi, M. B., Fer, I., & Jenkins, A. D. (2012). Surface gravity wave effects on the upper ocean boundary layer: Modification of a one-dimen-
sional vertical mixing model. Continental Shelf Research, 38, 63–78. https://doi.org/10.1016/j.csr.2012.03.002
Pezerat, M., Bertin, X., Martins, K., & Lavaud, L. (2022). Cross-shore distribution of the wave-induced circulation over a dissipative beach under
storm wave conditions: The dataset. https://doi.org/10.5281/zenodo.5878857
Pezerat, M., Bertin, X., Martins, K., Mengual, B., & Hamm, L. (2021). Simulating storm waves in the nearshore area using spectral model:
Current issues and a pragmatic solution. Ocean Modelling, 158, 101737. https://doi.org/10.1016/j.ocemod.2020.101737
Rascle, N. (2007). Impact of waves on the ocean circulation (Unpublished doctoral dissertation). Université de Bretagne Occidentale.
Reniers, A. J., Roelvink, J., & Thornton, E. (2004). Morphodynamic modeling of an embayed beach under wave group forcing. Journal of
Geophysical Research, 109(C1), C01030. https://doi.org/10.1029/2002jc001586
Roland, A., Zhang, Y. J., Wang, H. V., Meng, Y., Teng, Y.-C., Maderich, V., & Zanke, U. (2012). A fully coupled 3d wave-current interaction
model on unstructured grids. Journal of Geophysical Research, 117(C11), C00J33. https://doi.org/10.1029/2012jc007952
Saha, S., Moorthi, S., Wu, X., Wang, J., Nadiga, S., Tripp, P., & Becker, E. (2011). Ncep climate forecast system version 2 (cfsv2) selected hourly
time-series products. Boulder CO: Research Data Archive at the National Center for Atmospheric Research, Computational and Information
Systems Laboratory. https://doi.org/10.5065/D6N877VB
Smith, J. A. (2006). Wave–current interactions in finite depth. Journal of Physical Oceanography, 36(7), 1403–1419. https://doi.org/10.1175/
jpo2911.1
Stive, M., & Wind, H. (1986). Cross-shore mean flow in the surf zone. Coastal Engineering, 10(4), 325–340. https://doi.
org/10.1016/0378-3839(86)90019-0
Svendsen, I. A. (1984a). Mass flux and undertow in a surf zone. Coastal Engineering, 8(4), 347–365. https://doi.org/10.1016/0378-3839(84)90030-9

PEZERAT ET AL. 20 of 21
 21699291, 2022, 3, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2021JC018108 by Hong Kong University of Science and Technology, Wiley Online Library on [21/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Oceans 10.1029/2021JC018108

Svendsen, I. A. (1984b). Wave heights and set-up in a surf zone. Coastal Engineering, 8(4), 303–329. https://doi.org/10.1016/0378-3839(84)90028-0
Tajima, Y., & Madsen, O. S. (2006). Modeling near-shore waves, surface rollers, and undertow velocity profiles. Journal of Waterway, Port,
Coastal, and Ocean Engineering, 132(6), 429–438. https://doi.org/10.1061/(asce)0733-950x(2006)132:6(429)
Tolman, H. L. (1991). A third-generation model for wind waves on slowly varying, unsteady, and inhomogeneous depths and currents. Journal of
Physical Oceanography, 21(6), 782–797. https://doi.org/10.1175/1520-0485(1991)021<0782:atgmfw>2.0.co;2
Tolman, H. L. (1994). Wind waves and movable-bed bottom friction. Journal of Physical Oceanography, 24, 994–1009. https://doi.
org/10.1175/1520-0485(1994)024<0994:wwambb>2.0.co;2
Uchiyama, Y., McWilliams, J. C., & Shchepetkin, A. F. (2010). Wave–current interaction in an oceanic circulation model with a vortex-force
formalism: Application to the surf zone. Ocean Modelling, 34(1–2), 16–35. https://doi.org/10.1016/j.ocemod.2010.04.002
Umlauf, L., & Burchard, H. (2003). A generic length-scale equation for geophysical turbulence models. Journal of Marine Research, 61(2),
235–265. https://doi.org/10.1357/002224003322005087
Umlauf, L., Burchard, H., & Bolding, K. (2005). The general ocean turbulence model (gotm) scientific documentation: Version 3.2 (Technical
Report No. 63). Warnemunde, Germany: Leibnitz Institute for Baltic Sea Research.
Wang, P., McWilliams, J. C., Uchiyama, Y., Chekroun, M. D., & Yi, D. L. (2020). Effects of wave streaming and wave variations on nearshore
wave-driven circulation. Journal of Physical Oceanography, 50(10), 3025–3041. https://doi.org/10.1175/jpo-d-19-0304.1
Wright, L. D., & Short, A. D. (1984). Morphodynamic variability of surf zones and beaches: A synthesis. Marine Geology, 56(1–4), 93–118.
https://doi.org/10.1016/0025-3227(84)90008-2
Xu, Z., & Bowen, A. (1994). Wave-and wind-driven flow in water of finite depth. Journal of Physical Oceanography, 24(9), 1850–1866. https://
doi.org/10.1175/1520-0485(1994)024<1850:wawdfi>2.0.co;2
Zhang, H., Madsen, O. S., Sannasiraj, S., & Chan, E. S. (2004). Hydrodynamic model with wave–current interaction in coastal regions. Estuarine,
Coastal and Shelf Science, 61(2), 317–324. https://doi.org/10.1016/j.ecss.2004.06.002
Zhang, Y. J., Ye, F., Stanev, E. V., & Grashorn, S. (2016). Seamless cross-scale modeling with schism. Ocean Modelling, 102, 64–81. https://doi.
org/10.1016/j.ocemod.2016.05.002
Zheng, P., Li, M., van der Zanden, J., Wolf, J., Chen, X., Wang, C., & Wang, C. (2017). A 3D unstructured grid nearshore hydrodynamic model
based on the vortex force formalism. Ocean Modelling, 116, 48–69. https://doi.org/10.1016/j.ocemod.2017.06.003

PEZERAT ET AL. 21 of 21