Direction Symmetry of Wave Field Modulation by Tidal Current

Ina Teutsch ,1, 2, ∗ Saulo Mendes ,3, 4, † and Jérôme Kasparian 3, 4, ‡

1
Helmholtz-Zentrum Hereon, Coastal Climate and Regional Sea Level Changes,
Max-Planck-Straße 1, 21502 Geesthacht, Germany
2
Federal Waterways Engineering and Research Institute (BAW), Hamburg, 22559, Germany
3
Group of Applied Physics, University of Geneva,
Rue de l’École de Médecine 20, 1205 Geneva, Switzerland
4
Institute for Environmental Sciences, University of Geneva,
Boulevard Carl-Vogt 66, 1205 Geneva, Switzerland
Theoretical studies on the modulation of unidimensional regular waves over a flat bottom due
to a current typically assign an asymmetry between the effects of opposing/following streams on
arXiv:2407.04723v1 [physics.flu-dyn] 22 Jun 2024

the evolution of major sea variables, such as significant wave height. The significant wave height is
expected to monotonically increase with opposing streams and to decrease with following streams.
To some extent, observations on data sets containing a few thousand of waves or over a continuous
series of about a day confirm this prediction. Here, based on a multi-year dataset, we show that
in very broad-banded seas with high directional spread, the asymptotic behavior of sea variables is
highly non-trivial and does not follow the theoretical predictions, especially at high values of the
ratio between tidal stream and group speed.

I. INTRODUCTION theoretical considerations suggested that the wavelength

should increase when waves propagate in the same direc-
The deterministic study of inhomogeneous waves tion as a tidal stream [1, 26]. Indeed, MacIver et al. [23]
sprung near the middle of the twentieth century [1, 2]. reported from an experimental study that in following
Advanced mathematical techniques arose in the follow- currents waves become longer, while at the same time
ing decades [3–6], in particular with wave-current interac- their height is reduced. They concluded that in waves
tion being assessed through ray theory [7, 8], linear wave with a following current, wave energy is decreased due
theory [9, 10], radiation stress [11, 12], spectral [13] and to the strain rate of the current speed. Swan et al. [22],
perturbative methods [14]. Through these theoretical ad- however, found in an experimental study, supported by
vances, transformations of wave characteristics were mea- a numerical model, that the crest-trough asymmetry of
sured and confirmed in major global currents, such as the waves propagating in a following current increases to-
Agulhas current [15], the Gulf Stream [16, 17] and the wards higher and sharper wave crests. This asymmetry
Kuroshio [18, 19]. These transformations concern both should increase the probability of extreme waves [33–39].
significant wave height and wave energy, as well as spec- Observational studies to date are mostly focused on
tral bandwidth and directional spreading. Additionally, deterministic investigations for specific events or short-
the transformations have been corroborated in laboratory term time series analysis [40–44], even those with larger
experiments [20–24]. The first studies on wave-current data sets [45–48]. Here, we investigate the behavior of
systems focused on tides [1, 25, 26], theoretically treat- the modulation due to the wave-current interaction of
ing them as steady currents. They found that currents long-term series from a statistical point of view, namely
either reduce or expand the wavelength, whose change of average wave properties. We characterize the symme-
is met with a modulation in the wave height [1]. These try in the evolution of fundamental wave properties in
properties of the wave-current system seem to be univer- response to opposing and following currents. We show
sal, featuring similar changes in wave height, length, and that following currents can amplify wave fields in deep
period in estuaries and rivers [27–29]. However, simpli- water in addition to coastal waters [43] and that both
fied quasi-stationary or quasi-homogeneous models based the nonlinearity of the sea state and the speed of the
on the energy balance equations underestimate the effect tidal stream impact the magnitude and asymmetry of
of tidal currents on wave heights [30–32]. the modulation.
The majority of the mentioned studies limit the wave-
current system to waves encountering an opposing cur-
rent, resulting in a shortening and steepening of waves II. DATA AND METHODS
due to energy focusing. Only a few studies considered
the case of waves propagating in the same direction as Wave and current data were recorded between July
the current, and their findings were ambiguous. Initial 2019 and December 2022 at about 200 m distance from
the research platform FINO1 in the southern North Sea,
which is located at 54.015◦ N 6.588◦ E in a water depth
∗ [email protected] of approximately 30 m (figure 1). The currents at this
† [email protected] site result from the tidal cycle, which forms an ellipse
‡ [email protected] with a major axis stretching approximately from west-
 2

FIG. 1: (a) Location of the research platform FINO1 in the southern North Sea, close to the Dutch and German
Frisian islands. The green ellipse indicates the tidal cycle at the site, with semi-major and semi-minor axes of
approximately 0.75 m/s and 0.25 m/s, respectively. (b) The research platform FINO1 has been constructed close
to the wind farm Alpha Ventus in the North Sea to observe hydrographical and meteorological parameters, among
others. ©Forschungs- und Entwicklungszentrum Fachhochschule Kiel GmbH

northwest to east-southeast (figure 1). Waves typically Wave parameters Definition

propagated either towards the southeast or towards the Peak wavelength λp Solution to ( T2πp )2 = 2πg
λp
tanh( 2πh
λp
)
2π
east, with peak wave frequencies between 0.11 Hz and Peak wave number kp = λp
0.14 Hz (figure 2). A more detailed description, of the Relative water depth kp h q
2kp h
tidal current in this region, including a tidal chart, can Group speed cg = 21 kgp tanh(kp h) · (1 + sinh(2k p h)
)
e.g. be found in Reynaud and Dalrymple [49]. qm m
Spectral bandwidth ν= 0 2
m2
−1
Wave elevation data were recorded by a surface- √
2
1

following directional Waverider buoy of type MkIII. The Steepness ε= k H

π p s

sampling frequency was 1.28 Hz. The resolution of the

wave buoy is specified as 0.01 m, with an accuracy of less TABLE I: Sea parameter definitions, with gravity g and
than 0.5% of the measured surface elevation relative to spectral moments mn .
the calibrated still water line [51]. The wave buoy de-
livered surface elevation data in samples of 30 minutes
in length, containing 366 waves on average. The sam- rent speeds averaged over a time window of 10 minutes.
ples were quality-controlled according to the procedure These current speeds were further time-averaged over the
described in Teutsch et al. [52] and subsequently used to whole duration of each wave elevation sample. Samples
calculate the directional spectra (figure 2a,b) and the sea with missing wave or current data were excluded from
state parameters in table I. the analysis.
Current velocities and directions were recorded by an The current speed has a smooth mean vertical pro-
acoustic Doppler current profiler (ADCP, Nortek), which file except at the boundaries (figure 3). Close to the
was deployed at the sea bottom in an approximate water bottom and the surface, wave movements, tidal eleva-
depth of 30 m. The current data made available to us by tion and sidelobe interference of the ADCP [54] perturb
the Federal Maritime and Hydrographic Agency (BSH) the measurement. Furthermore, the peak frequency of
were already quality-controlled using the Storm software 0.12 Hz (see figure 2a) corresponds to wavelengths of
provided by the manufacturer [53]. This included correc- more than 100 m, much larger than the water depth of
tions for a possible tilt of the instrument and the removal 30 m. Waves are therefore sensitive to the current in the
of echo spikes, among others. The ADCP delivered cur- entire water column, although their sensitivity slightly
 3

FIG. 2: Mean wave spectra during the period between 2019 and 2022 at the considered site. (a) Unfiltered one-
dimensional spectrum. (b) Mean directional wave spectrum, calculated using the DIrectional WAve SPectra Toolbox
[DIWASP; 50].

FIG. 3: Box plot of approximately 40,000 measured current speed profiles during the considered time period of 3.5
years. Boxes enclose the two central quartiles with the median in red, while the wiskers encompass from the 25th to
75th percentiles and the blue markers indicate outliers of the distributions.

decays with the depth. Hence, we considered the current of the waves at the peak frequency in the directional wave
speed approximately 6 m below the water surface (i.e., spectrum, Θp . The directional spreading of the waves
24.5 m above the bottom, figure 3), as representative of around the peak frequency is calculated as
the whole column. p
σΘp = 2(1 − C1 ), (1)
Furthermore, current turbulence generated in the wa-
ter column by the waves [22] via bottom friction [55] may
p
with C1 = a21 + b21 , in which a1 and b1 denote the first
also feedback on the waves themselves. The mixed layer sine and cosine Fourier coefficients of the directional spec-
depth, in which turbulence may be generated by surface trum, respectively. The significant wave height Hs was
waves, ranges between 2 m and 6 m in our conditions [56, computed as the mean of the highest third of waves in
eq. 7]. Although this feedback likely averages out over each 30-minute sample. To evaluate the effect of cur-
the column in our measurements, we verified that neither rent speed U and current direction on major sea state
wave-induced current turbulence, nor our choice of con- parameters, we selected two sets of samples, for which
sidering the current at 6 m depth influenced the results: the angle between peak wave direction and current di-
repeating the analysis while considering the speed in the rection was either 0 ± 10◦ (henceforth referred to as ”fol-
middle of the water column (15 m depth) or the (vector) lowing current”) or 180 ± 10◦ (henceforth referred to as
averaged speed between 7.5 m and 22.5 m yielded similar ”opposing current”). Within these ranges, sine values
results, as will be discussed in detail in the next section keep below 0.17 and cosine values stay beyond ±0.98, so
(see also figure 6). that the transverse flow is negligible and the axial cur-
Wave direction is defined as the propagation direction rent is virtually unaffected. References without current,
 4

600
500 (a) (b) 600 (c)
500 500
400
400 400
Frequency

300
300 300
200
200 200

100
100 100

0 0 0
-1.0 -0.5 0.0 0.5 1.0 0 1 2 3 4 2 3 4 5 6 7
U (m/s) Tidal Speed Hs (m) Significant Height Tz (s) Zero-Crossing Period
500
(d) (e) 1400 (f)
800
400 1200

600 1000
300
800
400
200 600

400
200
100
200

0 0 0
-0.2 -0.1 0.0 0.1 0.2 2 4 6 8 10 30 40 50 60 70 80
U/cg Normalized Speed kp h Relative Depth Directional Spread (∘ )

800 (g) (h)

400
600

300

400
200

200
100

0 0
0.00 0.05 0.10 0.15 0.30 0.35 0.40 0.45 0.50 0.55
ε steepness  Bandwidth

FIG. 4: Histograms of the frequency of 30-minute wave records as a function of the major dimensional variables as
well as normalized current speed.

hereafter denoted rest conditions, were defined as sam- Figures 4a,d describe the near-symmetrical character of
ples with |U/cg | < 0.02, cg being the group speed of the the tidal speeds, whether in dimensional units or normal-
peak frequency of the wave spectrum. This threshold is ized by the peak group speed. Figures 4b,c provide the
consistent with the residual transverse current at current wave climate of the region, with mild sea characteristics
incidence angles of ±10◦ . for both significant wave height and zero-crossing period.
In total, the quality-controlled data set contained 4686 Most of the data is in intermediate (kp h = 0.3 − 3.1,
samples (2156 with opposing current, 2321 with forward 47% of the dataset) to deep water (kp h > π, 50% of the
current, 209 in rest conditions). Values of U/cg were dataset, or even 79% corresponding to kp h ≥ 2) (figure
binned in intervals of ∼ 1/50, except around zero, where 4e). Adding to the complexity of the sea conditions, the
a single bin covers |U/cg | ≤ 0.02, and for the outermost spectrum is significantly directionally spread (figure 4f).
bins, where data are sparser. Furthermore, the spectrum is very broad (Figure 4h).
The histograms in figure 4 provide descriptive statistics The sea is composed of waves typically of second order
of all selected 30 minute samples along the entire data set. in wave steepness (0.05 ≤ ε ≤ 0.09) [57], albeit smaller
 5

0.12 3.0
(a) (b)
2.5
0.10

Wave Height Hs (m)

2.0
Steepness ε

0.08
1.5
0.06
1.0

0.04
0.5

0.02 0
-0.2 -0.1 0 0.1 0.2 -0.2 -0.1 0 0.1 0.2
U/cg U/cg
120
7 (c) (d)
6 100

λp (m)
kp h

5 80
Relative Depth

Wavelength
4 60

3
40
2
20
-0.2 -0.1 0 0.1 0.2 -0.2 -0.1 0 0.1 0.2
U/cg U/cg

0.50 (e) 65
(f)
60
0.45
σΘ (° )


55
Bandwidth

0.40
50
Spread

0.35 45

0.30 40

-0.2 -0.1 0 0.1 0.2 -0.2 -0.1 0 0.1 0.2

U/cg U/cg

FIG. 5: Average values of the sea state parameters as a function of the relative current speed. Red bands depict
plus or minus one standard deviation on the wave data.

fractions of the dataset are also of first and higher or- between bins corresponding to forward and opposing cur-
ders (figure 4g). As discussed in the next section, this rents of the same speed, ±U/cg , for all pairs of bins.
strong nonlinearity will be key to the physical behavior Other variables display a much more symmetric behav-
of the waves. Typically, consideration of the nonlinear ior. The anticipated increase in significant wave height
regime of the wave steepness and its effect on wave-tide due to an opposing current [1, 10] is observed in our data
modulation is overlooked in most studies. (figure 5b), but restricted to |U/cg | ≤ 0.05. Beyond, this
trend is inverted and Hs decreases again. This decrease
might be due to either wave blocking or increasing dissi-
III. RESULTS pation due to high mean steepness, although it seems to
appear earlier than expected by Toffoli et al. [59]. The
Figure 5 displays the dependence of the main wave behavior is comparable to the effect of forward currents:
parameters on the relative current velocity. Negative Hs increases for mild current speeds, unlike the expected
current velocity refers to a current opposing the peak behavior of regular waves [23]. As in the case of an op-
wave direction. Only the mean wave steepness (figure posing stream, the significant wave height decreases for
5a) displays a clear asymmetry between forward and op- U/cg > 0.05, likely due to wave dissipation.
posing currents. In spite of the wide absolute confidence As expected due to the Doppler effect [10], wavelength
ranges, the size of the samples is quite large, typically is downshifted by an opposing current, and upshifted by
with 100,000 entries per bin. As such, pairwise Welch’s a slight following current (figure 5d) for mild current
t-tests [58] show a highly significant difference (p < 10−3 ) speeds. However, over most of the considered range of
 6

0.10
h ∈ [-22.5 m, -7.5 m] 2.0

Wave Height Hs (m)

0.09 h = -5.5 m
Steepness ε

0.08 1.5

0.07
1.0
0.06
(a) (b)
0.05 0.5
-0.2 -0.1 0 0.1 0.2 -0.2 -0.1 0 0.1 0.2
U/cg U/cg
7
(d)
45

λp (m)
6
kp h

5 40
Relative Depth

Wavelength
4 35

3
30
(c)
2
-0.2 -0.1 0 0.1 0.2 -0.2 -0.1 0 0.1 0.2
U/cg U/cg
0.46 60.0

0.44 57.5

σΘ (° )


0.42 55.0
Bandwidth

52.5
0.40
Spread
50.0
0.38
47.5
0.36 (e) (f)
45.0
-0.2 -0.1 0 0.1 0.2 -0.2 -0.1 0 0.1 0.2
U/cg U/cg

FIG. 6: Sensitivity of results to the depth at which the current speed is measured: Average values of the sea
parameters as a function of the relative current speed. The red curves are taken as a function of the normalized speed
measured as an average between the 25th and 75th percentile of the water depth (−22.5 m < h < −7.5 m).

following current (U/cg > 0.03), the wavelength shifts from the dynamic point of view, Ho et al. [43] showed that
back down almost symmetrically to the opposing current. the relevant parameter is the phase speed of the current
Accordingly, this symmetry also appears (although with relative to the wave group speed, so that the interaction
an inverted shape) in the relative water depth kp h (fig- between a following current and a comparatively slow
ure 5c). From the point of view of the classical treatment wave field in the fixed frame appears to be an interaction
of waves interacting with uniform currents, this reversal between an opposing current and waves in the reference
of the Doppler effect is unexpected [10]. For the spectral frame of the tide.
bandwidth and directional spread, both parameters show To rule out artifacts related to the depth at which the
symmetrical and bell-shaped responses to the following tidal stream speed was considered, we compared the pre-
and opposing currents (figures 5e,f). vious analysis to its counterpart performed at the middle
From the classical literature [10], we know that the point of the column (h = -15 m). These returned very
dispersion relation of narrow-banded, unidirectional, and similar results. Note that deeper data typically features
linear waves is affected by the presence of a current: The 10% more samples, due to the reduced perturbations of
dispersion relation ω02 = gk0 becomes (ω ± U k)2 = gk0 . the ADCP signal. Averaging the velocity between 7.5
Bearing in mind that in deep water ω/k = cp = 2cg , and 22.5 m depth also provided similar results, depicted
with cp denoting the phase velocity, one finds the modu- in figure 6.
lation of the wave number to be of the order of k/k0 ≈ The symmetric behavior of the significant wave height
1/[1 ± (U/2cg )]2 ≈ 1 ∓ (U/cg ) + 3/4(U/cg )2 . When in figure 5b with regard to the direction of the tidal cur-
|U/cg | ≪ 0.1, this modulation is expected to be strongly rent U/cg mixes different sea states covering a wide range
asymmetrical, however larger values of U/cg will dimin- of nonlinearity (or, equivalently, the wave steepness ε,
ish such asymmetry, as the quadratic term increases and see figure 4g). Due to higher-order corrections to the
tends to dominate. Furthermore, the wave theory for dispersion relation [60, 61], different effects of current ve-
steeper waves leads to a strengthening of the quadratic locity on wavelength and bandwidth can be expected for
term [60–62], in addition to shear terms [22, 63]. Besides, different levels of nonlinearity. We partitioned the data
 7

(a) ε ∈ [0,0.060]
1.2

Hs (m)
1.0

Wave Height
0.8

0.6

0.4
-0.2 -0.1 0 0.1 0.2
U/cg

(b) ε ∈ [0.061,0.080]
Hs (m)

1.75

1.50
Wave Height

1.25

1.00

0.75
-0.2 -0.1 0 0.1 0.2
U/cg
2.5
(c) ε ∈ [0.081,0.140]
Hs (m)

2.0
Wave Height

1.5

1.0
-0.2 -0.1 0 0.1 0.2
U/cg

FIG. 7: Average values of the significant wave height as a function of the relative current speed with varying levels
of nonlinearity.

into three ranges of wave steepness, in accordance with 0.05. Note that the pattern observed in figure 7a has been
Lé Méhaute’s diagram, considering a similar number of observed previously [30, 47]. Thus, apparent contradic-
waves in each group. tions in the literature regarding the primacy of opposing
For a low nonlinearity (ε ≤ 0.06), the response of the or following tides on the modulation of the significant
significant wave height to tidal stream speeds is slightly wave height might be due to the analysis of different
skewed to the right for |U/cg | < 0.05: a following stream ranges of wave steepness.
boosts the significant wave height more than an opposing Figure 8 superposes the statistics for wavelength and
one (figure 7a). For waves of higher orders in steepness bandwidth, along with significant wave height, for low-
(ε > 0.08) ranging up to breaking, we observed a higher and high-nonlinearity ranges, normalized with regard to
increase of Hs for opposing streams, which is the op- their value in the central bin U/cg = 0. Figure 8a shows
posite of what was observed for linear seas (figure 7c), that the magnitude of the modulation Hs /Hs,0 by an op-
while intermediate nonlinearities feature an intermediate posing current, relative to the no-current case, is higher
behavior, with an approximately symmetrical effect of in nonlinear than in linear seas.
currents on Hs (figure 7b). These skewnesses are highly The wavelength modulation λp /λp,0 is symmetric rela-
significant, the Welch’s t-test between bins with forward tive to U/cg regardless of the wave steepness (figure 8b).
and opposing streams of equal speed displaying p values However, the magnitude of this modulation is larger for
smaller than 10−3 . higher wave steepness. Conversely, the decay of band-
In all cases, the significant wave height was damped by width with increasing normalized tidal current speed is
currents, whether opposing or following, beyond |U/cg | > quite symmetrical for linear seas and asymmetrical for
 8

1.4
(a) ε ∈ [0,0.060]
ε ∈ [0.081,0.140]
1.2

Hs / Hs,0
1.0

0.8

0.6

0.4
-0.2 -0.1 0 0.1 0.2
U/cg
1.4
(b) ε ∈ [0,0.060]
ε ∈ [0.081,0.140]
1.2

1.0
λp / λp,0

0.8

0.6

0.4
-0.2 -0.1 0 0.1 0.2
U/cg

1.00 (c) ε ∈ [0,0.060]

ε ∈ [0.081,0.140]

0.95
 / 0

0.90

0.85

0.80
-0.2 -0.1 0 0.1 0.2
U/cg

FIG. 8: Relative change in the average values of significant wave height, wavelength, and bandwidth as a function
of normalized tidal stream speed over different partitions of the North Sea data.

nonlinear seas (figure 8c). In the latter regime, an oppos- trum but increase its directional spread. Moreover, the
ing current hardly affects the bandwidth, while following asymmetry in the modulation of significant wave height
currents reduce it. and the Doppler shift between following and opposing
currents only meet the classical theoretical expectations
at very low normalized current speeds (|U/cg | < 0.03).
IV. CONCLUSION At higher current speeds, the shape of the modulation is
quite symmetrical for all wave parameters except wave
steepness. Whether the maximum modulation appears
We assessed symmetries and asymmetries between the
on the opposing or the following current side depends on
effects of opposing and following currents on wave mod-
the nonlinearity of the sea state and the speed of the tidal
ulation and their dependence on the normalized tidal
stream compared to the wave group velocity.
speed. We showed that following currents will modulate
the wave field as strongly, and sometimes even stronger, Our main findings corroborate and extend the study
as the opposing currents. This strong effect of a fol- of Ho et al. [43] regarding the modulation of wave fields
lowing current is somewhat unexpected by the classical by a following current, by extending them up to deep
wave theory, when linear waves are subject to interaction water conditions and providing multi-year observational
with uniform currents [43]. Furthermore, we showed that data ensuring robust statistics. However, future work
both opposing and following currents narrow the spec- entails reproduction of these findings in the laboratory.
 9

By isolating the effects of current speed, bandwidth and and cooperation partners of the BSH. The sea state por-
directional spread, this could support the elaboration of tal was realized by the RAVE project (Research at alpha
new theoretical understanding of the modification of the ventus), which was funded by the Federal Ministry for
dispersion relation by tidal currents. Economic Affairs and Climate Action on the basis of a
resolution of the German Bundestag.

V. ACKNOWLEDGEMENTS VI. DATA AVAILABILITY STATEMENT

The measurement data were collected and made freely The underlying wave buoy and ACDP data are the
available by the BSH marine environmental monitor- property of and were made available by then Federal Mar-
ing network (MARNET), the RAVE project (www.rave- itime and Hydrographic Agency, Germany. They can be
offshore.de), the FINO project (www.fino-offshore.de) obtained upon request from these organizations.

[1] P. J. H. Unna, Waves and tidal streams, Nature 149, 219 the shuttle imaging radar-b synthetic aperture radar,
(1942). Journal of Geophysical Research: Oceans 93, 15389
[2] J. Johnson, The refraction of surface waves by cur- (1988).
rents, Transactions American Geophysical Union 28, 867 [16] L. Holthuijsen and H. Tolman, Effects of the gulf
(1947). stream on ocean waves, Journal of Geophysical Research:
[3] F. Ursell, On kelvin’s ship-wave pattern, Journal of Fluid Oceans 96, 12755 (1991).
Mechanics 8, 418 (1960). [17] D. W. Wang, A. K. Liu, C. Y. Peng, and E. A. Meindl,
[4] G. B. Whitham, Mass, momentum and energy flux in Wave-current interaction near the gulf stream during the
water waves, journal of Fluid Mechanics 12, 135 (1962). surface wave dynamics experiment, Journal of Geophys-
[5] G. B. Whitham, A general approach to linear and non- ical Research: Oceans 99, 5065 (1994).
linear dispersive waves using a lagrangian, Journal of [18] P. A. Hwang, Altimeter measurements of wind and wave
Fluid Mechanics 22, 273 (1965). modulation by the kuroshio in the yellow and east china
[6] F. P. Bretherton, C. J. R. Garrett, and M. J. Lighthill, seas, Journal of oceanography 61, 987 (2005).
Wavetrains in inhomogeneous moving media, Proceed- [19] J. Wang, C. Dong, and K. Yu, The influences of the
ings of the Royal Society of London. Series A. Mathe- kuroshio on wave characteristics and wave energy dis-
matical and Physical Sciences 302, 529 (1968). tribution in the east china sea, Deep Sea Research Part
[7] R. S. Arthur, Refraction of shallow water waves: the I 158, 103228 (2020).
combined effect of currents and underwater topogra- [20] G. P. Thomas, Wave-current interactions: an experimen-
phy, Transactions American Geophysical Union 31, 549 tal and numerical study. part 1. linear waves, Journal of
(1950). Fluid Mechanics 110, 457 (1981).
[8] G. B. Whitham, A note on group velocity, Journal of [21] G. P. Thomas, Wave–current interactions: an experimen-
Fluid Mechanics 9, 347 (1960). tal and numerical study. part 2. nonlinear waves, Journal
[9] G. I. Taylor, The action of a surface current used as a of Fluid Mechanics 216, 505 (1990).
breakwater, Proceedings of the Royal Society of London. [22] C. Swan, I. P. Cummins, and R. L. James, An experimen-
Series A. Mathematical and Physical Sciences 231, 466 tal study of two-dimensional surface water waves propa-
(1955). gating on depth-varying currents. part 1. regular waves,
[10] D. H. Peregrine, Interaction of water waves and currents, Journal of Fluid Mechanics 428, 273 (2001).
Advances in applied mechanics 16, 9 (1976). [23] R. MacIver, R. Simons, and G. Thomas, Gravity waves
[11] M. S. Longuet-Higgins and R. W. Stewart, Changes in interacting with a narrow jet-like current, Journal of Geo-
the form of short gravity waves on long waves and tidal physical Research: Oceans 111 (2006).
currents, Journal of Fluid Mechanics 8, 565 (1960). [24] Y. Ma, G. Dong, M. Perlin, X. Ma, G. Wang, and J. Xu,
[12] M. S. Longuet-Higgins and R. W. Stewart, The changes Laboratory observations of wave evolution, modulation
in amplitude of short gravity waves on steady non- and blocking due to spatially varying opposing currents,
uniform currents, Journal of Fluid Mechanics 10, 529 Journal of fluid mechanics 661, 108 (2010).
(1961). [25] H. U. Sverdrup, On wave heights in straits and sounds
[13] N. E. Huang, D. T. Chen, C.-C. Tung, and J. R. Smith, where incoming waves meet a strong, tidal current,
Interactions between steady non-uniform currents and Scripps Inst. Ocean. Wave Report , 4 (1944).
gravity waves with applications for current measure- [26] N. F. Barber, The behaviour of waves on tidal streams,
ments, Journal of Physical Oceanography 2, 420 (1972). Proceedings of the Royal Society of London. Series A.
[14] W. D. McKee, Waves on a shearing current: a uniformly Mathematical and Physical Sciences 198, 81 (1949).
valid asymptotic solution, in Mathematical Proceedings of [27] F. I. González, A case study of wave–current–bathymetry
the Cambridge Philosophical Society, Vol. 75 (1974) pp. interactions at the columbia river entrance, Journal of
295–301. physical oceanography 14, 1065 (1984).
[15] D. E. Irvine and D. G. Tilley, Ocean wave directional [28] L. H. Holthuijsen, N. Booij, and T. H. C. Herbers, A
spectra and wave-current interaction in the agulhas from prediction model for stationary, short-crested waves in
 10

shallow water with ambient currents, Coastal engineering 2820 (2009).

13, 23 (1989). [46] F. Ardhuin, A. Roland, F. Dumas, A.-C. Bennis,
[29] S. Zippel and J. Thomson, Surface wave breaking over A. Sentchev, P. Forget, J. Wolf, F. Girard, P. Osuna,
sheared currents: Observations from the mouth of the and M. Benoit, Numerical wave modeling in conditions
columbia river, Journal of Geophysical Research: Oceans with strong currents: Dissipation, refraction, and rela-
122, 3311 (2017). tive wind, Journal of Physical Oceanography 42, 2101
[30] C. E. Vincent, The interaction of wind-generated sea (2012).
waves with tidal currents, Journal of Physical Oceanog- [47] J. Gemmrich and C. Garrett, The signature of inertial
raphy 9, 748 (1979). and tidal currents in offshore wave records, Journal of
[31] H. L. Tolman, The influence of unsteady depths and cur- physical oceanography 42, 1051 (2012).
rents of tides on wind-wave propagation in shelf seas, [48] T. Halsne, A. Benetazzo, F. Barbariol, K. H. Chris-
Journal of Physical Oceanography 20, 1166 (1990). tensen, A. Carrasco, and Ø. Breivik, Wave modulation in
[32] H. L. Tolman, Effects of tides and storm surges on north a strong tidal current and its impact on extreme waves,
sea wind waves, Journal of physical oceanography 21, Journal of Physical Oceanography 54, 131 (2024).
766 (1991). [49] J.-Y. Reynaud and R. W. Dalrymple, Shallow-marine
[33] T. Marthinsen, On the statistics of irregular second-order tidal deposits, in Principles of Tidal Sedimentology,
waves, Report No. RMS-11 (1992). edited by R. A. Davis Jr. and R. W. Dalrymple (Springer
[34] N. Mori and N. Kobayashi, Nonlinear distribution of Netherlands, Dordrecht, 2012) pp. 335–369.
neashore free surface and velocity, in Coastal Engineering [50] D. Johnson, Diwasp, a directional wave spectra toolbox
1998 (1998) pp. 189–202. for matlab®: User manual, Centre for Water Research,
[35] N. Mori and T. Yasuda, A weakly non-gaussian model of University of Western Australia. (2002).
wave height distribution random wave train, Ocean Eng. [51] Datawell, Datawell directional waverider mkiii (2023), ac-
29, 1219–1231 (2002). cessed on September 04, 2023.
[36] M. A. Tayfun and M. A. Alkhalidi, Distribution of sea- [52] I. Teutsch, R. Weisse, J. Moeller, and O. Krueger, A
surface elevations in intermediate and shallow water statistical analysis of rogue waves in the southern north
depths, Coastal Eng. 157 (2020). sea, Natural Hazards and Earth System Sciences 20, 2665
[37] S. M. S. Mendes, On the statistics of oceanic rogue waves (2020).
in finite depth: Exceeding probabilities, physical con- [53] Nortek, Storm postprocessing software, NortekSoftware
straints and extreme value theory, UNC Chapel Hill PhD (2022), version 1.17.11.
Thesis. (2020). [54] Nortek, The comprehensive manual for adcps, Nortek
[38] S. Mendes, A. Scotti, M. Brunetti, and J. Kasparian, (2022).
Non-homogeneous model of rogue wave probability evo- [55] J. Wolf and D. Prandle, Some observations of wave–
lution over a shoal, J. Fluid Mech. 939, A25 (2022). current interaction, Coastal Engineering 37, 471 (1999).
[39] S. Mendes and A. Scotti, The rayleigh-haring-tayfun dis- [56] A. V. Babanin, On a wave-induced turbulence and a
tribution of wave heights in deep water, Applied Ocean wave-mixed upper ocean layer, Geophysical Research
Research 113, 102739 (2021). Letters 33, 10.1029/2006gl027308 (2006).
[40] N. Guillou, Modelling effects of tidal currents on waves [57] B. Lé Méhaute, An introduction to hydrodynamics and
at a tidal stream energy site, Renewable Energy 114, 180 water waves, Springer (1976).
(2017). [58] B. L. Welch, The generalization of ‘student’s’ problem
[41] M. J. Lewis, T. Palmer, R. Hashemi, P. Robins, when several different population variances are involved,
A. Saulter, J. Brown, H. Lewis, and S. Neill, Wave-tide Biometrika 34, 28 (1947).
interaction modulates nearshore wave height, Ocean Dy- [59] A. Toffoli, T. Waseda, H. Houtani, T. Kinoshita,
namics 69, 367 (2019). K. Collins, D. Proment, and M. Onorato, Excitation
[42] M. A. Barnes and C. Rautenbach, Toward operational of rogue waves in a variable medium: An experimen-
wave-current interactions over the agulhas current sys- tal study on the interaction of water waves and currents,
tem, Journal of Geophysical Research: Oceans 125, Physical Review E 87, 051201 (2013).
e2020JC016321 (2020). [60] I. Brevik, Remarks on set-down for wave groups and
[43] A. Ho, S. Merrifield, and N. Pizzo, Wave–tide interaction wave-current systems, Coastal Engineering 2, 313 (1978).
for a strongly modulated wave field, Journal of Physical [61] J. H. Pihl, H. Bredmose, and J. Larsen, Shoaling of sixth-
Oceanography 53, 915 (2023). order stokes waves on a current, Ocean engineering 28,
[44] L. Jia, R. Wu, F. Shi, B. Han, and Q. Yang, A numerical 667 (2001).
study of multiscale current effects on waves in the north- [62] I. G. Jonsson, Energy flux and wave action in gravity
ern south china sea, Ocean Modelling , 102342 (2024). waves propagating on a current, Journal of Hydraulic Re-
[45] F. Ardhuin, L. Marié, N. Rascle, P. Forget, and search 16, 223 (1978).
A. Roland, Observation and estimation of lagrangian, [63] C. Swan and R. James, A simple analytical model for
stokes, and eulerian currents induced by wind and waves surface water waves on a depth-varying current, Applied
at the sea surface, Journal of physical oceanography 39, Ocean Research 22, 331 (2000).