1 Introduction
From a historical perspective the mathematical theory of rotational water waves has its roots at the beginning of the 19th century, when Gerstner [1809] constructed a solution in Lagrangian coordinates describing an explicit family of periodic two-dimensional travelling gravity water waves with non-zero vorticity, cf. Gerstner (Reference Gerstner1809). While Gerstner’s solution was found in homogeneous fluid, Dubreil-Jacotin (Reference Dubreil-Jacotin1932) showed that a heterogeneous fluid can also be accommodated; for modern presentations of the Gerstner’s wave solution we refer the reader to Constantin (Reference Constantin2001, Reference Constantin2011a ) and Henry (Reference Henry2008). Nevertheless, substantial rigorous analytical results in the area of rotational water flows appeared relatively recently after the ground-breaking work by Constantin & Strauss (Reference Constantin and Strauss2004) in which the existence of large-amplitude periodic water waves over two-dimensional flows with arbitrary (continuous) vorticity was proved; the same authors extended the result to water flows with discontinuous vorticity, cf. Constantin & Strauss (Reference Constantin and Strauss2011).
The extent of rotationality of water flows is captured by the vorticity function, defined as the curl of the velocity field. The vorticity is essential for describing the interactions of waves with non-uniform currents. Among the most staggering examples of wave–current interactions we quote those at the eastern coast of South Africa (where
$6$
m high sea waves from southwest meeting the Agulhas current lead to many oil tanker wreckages) and those at the Columbia River entrance, one of the most insecure navigational regions in the world, given the doubling of the wave height in just a few hours, cf. Jonsson (Reference Jonsson and Le Méhauté1990). The strength of the interaction is determined – as documented by studies of Peregrine (Reference Peregrine1976), Jonsson (Reference Jonsson and Le Méhauté1990) and Thomas & Klopman (Reference Thomas, Klopman and Hunt1997) – by the vertical structure of the current profile whose description is realized by the vorticity; while zero vorticity models irrotational flows as well as currents which are uniform with depth, the simplest rotational setting is that of linearly sheared currents of constant non-zero vorticity, cf. Lighthill (Reference Lighthill1978).
Although the last decades have witnessed an outburst of studies concerning rotational water flows, the vast majority of works on water flows exhibiting vorticity pertains to two-dimensional flows. That is, while allowing variations in the vertical direction, the flows analysed in the previous mentioned papers present no variation in the direction orthogonal to the direction of propagation. This state of affairs is justified in part by recent results showing that gravity (Constantin Reference Constantin2011b ), capillary (Constantin & Kartashova Reference Constantin and Kartashova2009) or capillary–gravity (Martin Reference Martin2017) wave trains at the surface of water in a flow with constant non-zero vorticity with a flat bed can only occur if the flow is two-dimensional and if the vorticity vector has only one non-vanishing component that points in the horizontal direction orthogonal to the direction of wave propagation. For a similar result concerning solitary waves we refer the reader to the work by Stuhlmeier (Reference Stuhlmeier2012).
A significant result showing the reduced relevance of three-dimensional water flows presenting a constant non-vanishing vorticity vector was obtained by Wahlén (Reference Wahlén2014) who showed that the assumption of a free surface that has a travelling behaviour in both horizontal directions over a flow assumed to be steady results in a flow whose free surface, velocity field and pressure are uniform in the
$y$
-direction, while the second component of the velocity vector is constant.
A realistic modelling of the water flows occurring in the Earth’s oceans must include not only the vorticity effects but also Coriolis effects arising from the rotation of the Earth. Indeed, in recent years an intense mathematical effort was put into the study of geophysical water waves, as underlined in Constantin & Monismith (Reference Constantin and Monismith2017), Johnson (Reference Johnson2018). This body of work was initiated by Constantin (Reference Constantin2012b
). By a substantial modification of the Gerstner wave, Constantin constructed (Constantin Reference Constantin2012b
) an exact three-dimensional solution of the nonlinear governing equations for geophysical water waves in the
$\unicode[STIX]{x1D6FD}$
-plane approximation near the Equator. This solution described in the Lagrangian framework equatorially trapped waves propagating eastward in a stratified inviscid flow. Further studies presenting explicit exact solutions that accommodate flow reversal and underlying currents in the
$\unicode[STIX]{x1D6FD}$
-plane setting were performed in Constantin (Reference Constantin2013, Reference Constantin2014), Henry (Reference Henry2013, Reference Henry2016, Reference Henry2017), Constantin & Johnson (Reference Constantin and Johnson2015). We would also like to point out the recent emergence of solutions modelling large-scale ocean currents with distinctive and persistence patterns like gyres (Constantin & Johnson Reference Constantin and Johnson2017b
), the Equatorial Undercurrent (Constantin & Johnson Reference Constantin and Johnson2016a
), the Antarctic Circumpolar Current (Constantin & Johnson Reference Constantin and Johnson2016b
; Hsu & Martin Reference Hsu and Martin2017).
We prove here that the only water flows in the
$\unicode[STIX]{x1D6FD}$
-plane approximation that exhibit a constant (not necessarily vanishing) vorticity vector are those with vanishing velocity field and a flat free surface. We will point out below, a situation that comes close to exhibiting constant vorticity vector and which is the forerunner of the upwelling/downwelling, phenomenon giving rise to manifestations of foam streaks on lakes, cf. Langmuir (Reference Langmuir1938), Stommel (Reference Stommel1951). More precisely, the wind blowing on the water surface gives rise to longitudinal surface currents in the direction of the wind. The result of this action is the emergence in water of a series of alternating left and right helical vortices, with their horizontal axes parallel to the wind direction, cf. Langmuir (Reference Langmuir1938), an occurrence that bears the name Langmuir circulation (LC). Hallmarks of the convergence zone between the counter-rotating vortices are lines of bubbles and surface debris aligned with the wind, cf. Moum & Smyth (Reference Moum and Smyth2001). According to Craik & Leibovich (Reference Craik and Leibovich1976), Langmuir circulations can form on flows that initially have constant vorticity. For a comprehensive account on LC we refer the reader to the works by Craik & Leibovich (Reference Craik and Leibovich1976) and Leibovich (Reference Leibovich1977, Reference Leibovich1983).
From a mathematical perspective, constant vorticity flows have the advantage that their velocity field consists of harmonic functions whose properties are well understood. Perturbative methods can be employed thereafter to treat more general flows (Stiassnie & Peregrine Reference Stiassnie and Peregrine1979).
The result we obtain is quite general, in the sense that neither do we require the free surface to have a travelling character nor do we require the underlying flow to be steady. To reinforce our conclusion we point out that the explicit equatorial Kelvin waves described in Constantin (Reference Constantin2012b
, Reference Constantin2013), Henry (Reference Henry2016, Reference Henry2017), Constantin & Johnson (Reference Constantin and Johnson2017a
) have non-constant vorticity. Rather different is the situation of Euler equations in the setting without geophysical effects. They do admit non-trivial solutions exhibiting a constant vorticity vector; either in the form of irrotational flow, cf. Iooss & Plotnikov (Reference Iooss and Plotnikov2009, Reference Iooss and Plotnikov2011) or in the form of a rotational flow (with constant vorticity vector) with a flat free surface, displaying a non-vanishing velocity field that varies only with depth, cf. Remark 2.5. Also the equatorial
$f$
-plane approximation allows for solutions with non-vanishing velocity field and constant (non-zero) vorticity, cf. Remark 2.6.
Summarizing, our analytic investigation shows that assuming the full nonlinear character (of the Euler equations and of their boundary conditions) in the setting of the equatorial
$\unicode[STIX]{x1D6FD}$
-plane approximation is a perspective that cannot accommodate a constant vorticity vector, not even of irrotational type. The latter scenario is a circumstance that allows for thorough investigations in the setting without geophysical effects. Indeed, the constant vorticity of irrotational type allows the utilization of a velocity potential which was essential in proving the existence of doubly periodic capillary–gravity waves, cf. Reeder & Shinbrot (Reference Reeder and Shinbrot1981), the existence of travelling three-dimensional capillary–gravity water waves by Craig & Nicholls (Reference Craig and Nicholls2000), the existence of three-dimensional oblique travelling gravity–capillary water waves by Groves & Haragus (Reference Groves and Haragus2003) or the existence of three-dimensional doubly periodic gravity water waves by Iooss & Plotnikov (Reference Iooss and Plotnikov2009, Reference Iooss and Plotnikov2011). In the context of three-dimensional flows we find highly relevant the theoretical and numerical study of the interaction of surface gravity waves and strongly sheared currents, with shear existing in both horizontal and vertical directions, performed relatively recently by Dong & Kirby (Reference Dong, Kirby, Lynett and Smith2012).
Recently Constantin & Kartashova (Reference Constantin and Kartashova2009), Constantin (Reference Constantin2011b
), Wahlén (Reference Wahlén2014) and Basu & Martin (Reference Basu and Martin2018) showed that constant non-vanishing vorticity below a wave train determines a two-dimensional character of the water flow. Moreover, Martin (Reference Martin2018) showed that constant non-vanishing vorticity in water flows with an arbitrary free surface and governed by the fully nonlinear equations (but without geophysical effects) also implies a two-dimensional motion. In line with the previous conclusion is also the study by Ellingsen (Reference Ellingsen2016) in which solutions to linearized Euler equations that carry a non-vanishing vorticity are presented: while plane wave solutions that travel at an arbitrary angle
$\unicode[STIX]{x1D703}$
with the
$x$
-axis have non-constant vorticity, the situation of waves propagating exactly upstream or downstream, that is
$\sin \unicode[STIX]{x1D703}=0$
, is possible only if the vorticity remains constant.
We conclude by remarking that, in light of the previous discussion, our non-existence result is indicative of the fact that, in the
$\unicode[STIX]{x1D6FD}$
-plane setting, the vorticity steers the fluid motion to a much greater extent than in the case without geophysical effects.
2 The three-dimensional Euler equations in the
$\unicode[STIX]{x1D6FD}$
-plane approximation
In order to formulate the equations governing the motion of the flows we consider we approximate the shape of the Earth by a sphere of radius
$R=6378$
km and consider a rotating framework with the origin at a point on the Earth’s surface, with the
$x$
axis chosen horizontally due east, the
$y$
axis horizontally due north and the
$z$
axis
$z$
axis upward. Denoting with
the velocity field, with
$P(x,y,z,t)$
the pressure within the fluid, and with
$g$
the gravitational acceleration, the equations of motion are, cf. Constantin (Reference Constantin2012b
), the Euler equations in the
$\unicode[STIX]{x1D6FD}$
-plane approximation near the Equator
$$\begin{eqnarray}\left.\begin{array}{@{}c@{}}u_{t}+uu_{x}+vu_{y}+wu_{z}+2\unicode[STIX]{x1D714}w-\unicode[STIX]{x1D6FD}yv=-P_{x},\\ v_{t}+uv_{x}+vv_{y}+wv_{z}+\unicode[STIX]{x1D6FD}yu=-P_{y},\\ w_{t}+uw_{x}+vw_{y}+ww_{z}-2\unicode[STIX]{x1D714}u=-P_{z}-g,\end{array}\right\}\end{eqnarray}$$
(where
$\unicode[STIX]{x1D714}:=7.29^{-5}~\text{rad}~\text{s}^{-1}$
is the rotation speed of the Earth round the polar axis toward east and
$\unicode[STIX]{x1D6FD}:=2\unicode[STIX]{x1D714}/R$
) and the equation of mass conservation
The
$\unicode[STIX]{x1D6FD}$
-plane effect
$\unicode[STIX]{x1D6FD}y$
, present in the first equation and second equation of the system (2.2), appears as a result of linearizing the Coriolis force in the tangent plane approximation, a procedure that – despite the spherical shape of the Earth – is valid due to the moderate spatial scale of the motion: the region occupied by the fluid can be approximated by a tangent plane and the linear term of the Taylor expansion captures the
$\unicode[STIX]{x1D6FD}$
-plane effect (cf. the discussions in Cushman-Roisin & Beckers (Reference Cushman-Roisin and Beckers2011) and Constantin (Reference Constantin2012b
)). In this respect, it is important to note that the
$\unicode[STIX]{x1D6FD}$
-plane equations (2.2) represent a consistent approximation to the governing equations only near the Equator, as emphasized by Dellar (Reference Dellar2011).
The water wave problem is completely formulated when we specify the boundary conditions.
and
together with the dynamic boundary condition
which decouples the motion of the motion of the water flow from the motion of the air above it, cf. Constantin (Reference Constantin2011a ).
The grasp into the swirling motion of the water is realized through the vorticity vector
$\unicode[STIX]{x1D6FA}(x,y,z,t)$
, defined as the curl of the velocity field, that is,
Remark 2.1. We will assume that the components of the vorticity vector are constant and, in addition, that
This is a reasonable assumption, since
$\unicode[STIX]{x1D714}\approx 0.73\times 10^{-4}~\text{rad}~\text{s}^{-1}$
, while a typical value for
$\unicode[STIX]{x1D6FA}_{2}$
in the equatorial Pacific is
$25\times 10^{-3}~\text{s}^{-1}$
, cf. Constantin (Reference Constantin2012a
).
Remark 2.2. The choice to work in the geophysical setting, in which the velocity field, the pressure and the free surface obey equations (2.2)–(2.6), clearly implies that
$\unicode[STIX]{x1D6FD}=2\unicode[STIX]{x1D714}/R\neq 0$
. This latter fact will be used in several places within the proof of the main result stated in Theorem 2.4 below.
Lemma 2.3. The velocity components
$u,v,w$
are harmonic functions within the fluid domain.
Proof. We perform the proof for
$u$
. By the definition of the vorticity vector (2.7) and from the equation of mass conservation (2.3) we have
where in the last equality we have used the equation of mass conservation (2.3). Analogously we prove that
$\unicode[STIX]{x0394}v=\unicode[STIX]{x0394}w=0$
.◻
We state now the main result of the paper.
Theorem 2.4. Assuming that the vorticity vector is constant throughout the flow and also (2.8), the only bounded solution of the water wave problem (2.2)–(2.6) is the one with identically vanishing velocity field, flat surface and pressure
where
$a$
is some function of
$t$
.
Proof. Considering the curl of the three equations in (2.2) we find that between the vorticity
$\unicode[STIX]{x1D6FA}$
and the velocity field the relations
hold.
Using now that
$u_{x},u_{y}$
and
$u_{z}$
are also harmonic functions, we obtain from (2.11) that
which can be expanded as
from which we conclude that
Similarly, using (2.12) and (2.13) and the harmonicity of
$v_{x},v_{y},v_{z}$
,
$w_{x},w_{y},w_{z}$
and
$v$
we infer that
and
The expression of
$\unicode[STIX]{x1D6FA}_{2}$
and (2.17) implies that
while the definition of
$\unicode[STIX]{x1D6FA}_{1}$
and (2.18) leads to
Thus, we have from the previous assertions that
that is,
$w_{y}$
is constant within the fluid domain. From the kinematic boundary condition (2.5) we see that
From the constancy of
$w_{y}$
we obtain that
The latter and the definition of
$\unicode[STIX]{x1D6FA}_{1}$
imply that
The next goal is to show that
$w_{x}\equiv 0$
.
Differentiating with respect to
$y$
in (2.13) we get
which, by (2.21), delivers
Differentiating above by
$z$
we get
and
Since
$v_{zz}\equiv 0$
we conclude from (2.28), using the harmonicity of
$v$
, that
$v_{xx}\equiv 0$
. Analogously, since
$w_{yy}\equiv 0$
we conclude from (2.27) that
$w_{xx}\equiv 0$
. We summarize these conclusions as
and
Differentiating with respect to
$z$
in the equation of mass conservation (2.3) we obtain
which, due to (2.27) and (2.24) delivers
Differentiating with respect to
$y$
in (2.3) we obtain
which, by (2.28) and (2.23) becomes
Equation (2.11) gives, after differentiation with respect to
$z$
, the equation
that leads to
where we have used that
$u_{yz}=(v_{z})_{x}=0$
. The latter equality implies that
$u_{zz}(x,y,z)=0$
for all
$(x,y,z)$
with
$y\neq -(\unicode[STIX]{x1D6FA}_{3}/\unicode[STIX]{x1D6FD})$
. From the continuity of
$u_{zz}$
, it follows that
within the fluid domain. From the latter, we infer that
Recall that we already know that
$w_{xy}=(w_{y})_{x}=0$
and
$w_{xx}=0$
. These latter findings say that
$w_{x}$
is constant throughout the fluid domain. Since
$w_{x}=0$
on the bed
$z=-d$
we conclude that
From (2.39) we readily infer that
It is also immediate from (2.39) that
$w_{zx}=0$
throughout the flow. Recalling that
we conclude that
Differentiating now equation (2.12) with respect to
$x$
leads to
Note that from (2.24) we have that
$v_{xz}$
vanishes identically within the flow. This, in conjunction with
$v_{xx}=0$
and formula (2.43), yields
Since
$\unicode[STIX]{x1D6FA}_{2}+2\unicode[STIX]{x1D714}\neq 0$
, we have that
The definition of
$\unicode[STIX]{x1D6FA}_{3}$
and (2.45) also imply that
Since
$u_{zz}=0$
we infer from the harmonicity of
$u$
that also
$u_{xx}=0$
within the flow. The previous facts show that the functions
$u_{x},v_{x},u_{y},v_{y}$
and
$u_{z},v_{z}$
are constant throughout the flow. Therefore, in order for the equations (2.11) and (2.12) to hold throughout the fluid domain, it is necessary that
The previous statement together with (2.24) and (2.40) yield that
$\unicode[STIX]{x1D6FA}_{1}=\unicode[STIX]{x1D6FA}_{2}=0$
. Thus, from (2.11) and (2.12) we infer that
Recalling now that
$w_{x}=w_{y}=0$
within the fluid we notice that (2.13) becomes
We differentiate in (2.49) with respect to
$x$
, taking also into account that
$w_{z}$
is constant throughout the flow we obtain that
Going back to (2.26) we infer that
which immediately implies, via (2.49), that
From (2.51) and from the kinematic condition on the bed (2.5) we conclude that
Moreover, recall that by now we have that
$v_{y}$
and
$w_{z}$
vanish identically throughout the flow. This fact and the equation of mass conservation (2.3) imply that
The previous inferences show that
$u$
is a constant function at any time
$t$
. That is, there exists a function
$t\rightarrow c(t)$
such that for all
$(x,y,z,t)$
(with
$(x,y,z)$
lying in the fluid domain)
Utilizing (2.53) and the third equation in (2.2) we obtain that
Of relevance in proving our assertions is the behaviour of the pressure gradient on the free surface. We note first that since
we have that
and
for all
$(x,y,t)$
. Hence,
and
We will prove next that the free surface
$z=\unicode[STIX]{x1D702}(x,y,t)$
is flat at all times
$t$
. To this end note that the first two equations in (2.2) can now be written as
where we have used that
$v$
and
$w$
vanish throughout the fluid domain.
We claim now that the second equation in (2.62) implies that
$\tilde{c}(t)\neq 0$
for all
$t$
. Indeed, assuming that there is a
$t_{0}$
such that
$\tilde{c}(t_{0})=0$
, this would lead to
$c(t_{0})=0$
. The latter implies, via (2.56), that
$g=0$
, which is an obvious contradiction. Hence,
$\tilde{c}(t)\neq 0$
for all
$t$
, as asserted.
Differentiating now the equations in (2.62) with respect to
$x$
, and taking into account that
$\tilde{c}(t)\neq 0$
, we obtain
which shows that
$\unicode[STIX]{x1D702}_{x}(x,y,t)$
does not depend on
$x$
and
$y$
. Thus, there exist functions
$t\rightarrow p(t)$
and
$(t,y)\rightarrow q(t,y)$
such that
We claim that, in fact,
$p$
is identically zero. Towards proving the claim, let us assume that there is
$t_{0}\in \mathbb{R}$
such that
$f(t_{0})\neq 0$
. We obtain from (2.64) that, for fixed
$y$
, it holds
which is a contradiction with the boundedness of
$\unicode[STIX]{x1D702}$
. This shows that
$p(t)=0$
for all
$t$
. Consequently, (2.64) implies that
$\unicode[STIX]{x1D702}_{x}(x,y,t)=0$
for all
$x,y,t$
. From the first equation in (2.62) we infer that
$c^{\prime }(t)=0$
for all
$t$
. Hence, there is a constant
$c$
such that
$c(t)=c$
for all
$t$
. The system (2.62) can now be rewritten as
The preceding formulas imply that
with
$t\rightarrow f(t)$
being a function that depends solely on
$t$
. Availing now of the boundedness of
$\unicode[STIX]{x1D702}$
we have from (2.67) that
$c=0$
. This readily implies
which renders the kinematic boundary condition as
Hence, using (2.67), we have that
$f^{\prime }(t)=0$
for all
$t$
. That is, the free surface is flat and constant at each time
$t$
.◻
We have thus proved that the only water flow that satisfies the governing equations and their boundary conditions (corresponding to the
$\unicode[STIX]{x1D6FD}$
-plane approximation near the Equator) is the stationary flow with
$u=v=w=0$
within the fluid domain and having a flat free surface. Very different is the case without geophysical effects. We formulate this in the next remark.
Remark 2.5. The tuple
$(u,v,w,P,\unicode[STIX]{x1D702})$
representing the velocity field, the pressure and the free surface, given by
(where
$a,b,\unicode[STIX]{x1D702}_{0},\unicode[STIX]{x1D6FA}_{1},\unicode[STIX]{x1D6FA}_{2}$
are constants and
$P_{atm}$
is the constant atmospheric pressure) satisfies (within the fluid domain bounded below by the flat bed
$z=-d$
and above by the flat surface
$z=\unicode[STIX]{x1D702}_{0}$
) the Euler equations
$$\begin{eqnarray}\left.\begin{array}{@{}c@{}}u_{t}+uu_{x}+vu_{y}+wu_{z}=-P_{x},\\ v_{t}+uv_{x}+vv_{y}+wv_{z}=-P_{y},\\ w_{t}+uw_{x}+vw_{y}+ww_{z}=-P_{z}-g,\end{array}\right\}\end{eqnarray}$$
the equation of mass conservation
the dynamic boundary condition
and the kinematic boundary conditions
Moreover, the vorticity of the flow (2.70) equals
$(\unicode[STIX]{x1D6FA}_{1},\unicode[STIX]{x1D6FA}_{2},0)$
.
Interestingly enough, the governing equations of the equatorial
$f$
-plane approximation also admit solutions with non-vanishing velocity field and constant non-vanishing vorticity vector. This is the content of the following remark.
Remark 2.6. Let
$\tilde{a},\tilde{b},\tilde{\unicode[STIX]{x1D702}}_{0},\widetilde{\unicode[STIX]{x1D6FA}}_{1},\widetilde{\unicode[STIX]{x1D6FA}}_{2}$
be constants and let
$P_{atm}$
denote the constant atmospheric pressure. Then, the tuple
$(u,v,w,P,\unicode[STIX]{x1D702})$
representing the velocity field, the pressure and the free surface, given by
$$\begin{eqnarray}\left.\begin{array}{@{}c@{}}u(x,y,z,t)=\widetilde{\unicode[STIX]{x1D6FA}}_{2}z+\tilde{a},\quad v(x,y,z,t)=-\widetilde{\unicode[STIX]{x1D6FA}}_{1}z+\tilde{b},\quad w(x,y,z,t)=0,\\ P(x,y,z,t)=(2\unicode[STIX]{x1D714}\tilde{a}-g)(z-\tilde{\unicode[STIX]{x1D702}}_{0})+\unicode[STIX]{x1D714}\widetilde{\unicode[STIX]{x1D6FA}}_{2}(z^{2}-\tilde{\unicode[STIX]{x1D702}}_{0}^{2})+P_{atm},\\ \unicode[STIX]{x1D702}(x,y,t)=\tilde{\unicode[STIX]{x1D702}}_{0},\end{array}\right\}\end{eqnarray}$$
satisfies the Euler equations in the
$f$
-plane approximation
$$\begin{eqnarray}\left.\begin{array}{@{}c@{}}u_{t}+uu_{x}+vu_{y}+wu_{z}+2\unicode[STIX]{x1D714}w=-P_{x},\\ v_{t}+uv_{x}+vv_{y}+wv_{z}=-P_{y},\\ w_{t}+uw_{x}+vw_{y}+ww_{z}-2\unicode[STIX]{x1D714}u=-P_{z}-g,\end{array}\right\}\end{eqnarray}$$
as well as the equation of mass conservation
in the domain bounded below by the bed
$z=-d$
and above by the flat surface
$z=\tilde{\unicode[STIX]{x1D702}}_{0}$
. Clearly, the dynamic boundary condition (2.6) and the kinematic boundary conditions (2.4) and (2.5) are also satisfied. Moreover, the vorticity of the flow (2.75) is given by the (constant) vector
$(\widetilde{\unicode[STIX]{x1D6FA}}_{1},\widetilde{\unicode[STIX]{x1D6FA}}_{2},0)$
.
Remark 2.7. The question of existence of non-trivial solutions to the water waves equations incorporating the full Coriolis term and exhibiting a constant vorticity vector remains open. That is, denoting with
$\unicode[STIX]{x1D719}$
the latitude and replacing the system (2.2) with
$$\begin{eqnarray}\left.\begin{array}{@{}c@{}}u_{t}+uu_{x}+vu_{y}+wu_{z}+2\unicode[STIX]{x1D714}w\cos \unicode[STIX]{x1D719}-2\unicode[STIX]{x1D6FA}v\sin \unicode[STIX]{x1D719}=-P_{x},\\ v_{t}+uv_{x}+vv_{y}+wv_{z}+2\unicode[STIX]{x1D6FA}u\sin \unicode[STIX]{x1D719}=-P_{y},\\ w_{t}+uw_{x}+vw_{y}+ww_{z}-2\unicode[STIX]{x1D714}u\cos \unicode[STIX]{x1D719}=-P_{z}-g,\end{array}\right\}\end{eqnarray}$$
keeping (2.3)–(2.6), we see that, instead of (2.11)–(2.13), the velocity field and the vorticity vector satisfy the relations
$$\begin{eqnarray}\left.\begin{array}{@{}c@{}}\unicode[STIX]{x1D6FA}_{1}u_{x}+(\unicode[STIX]{x1D6FA}_{2}+2\unicode[STIX]{x1D714}\cos \unicode[STIX]{x1D719})u_{y}+(\unicode[STIX]{x1D6FA}_{3}+2\unicode[STIX]{x1D714}\sin \unicode[STIX]{x1D719})u_{z}=0,\\ \unicode[STIX]{x1D6FA}_{1}v_{x}+(\unicode[STIX]{x1D6FA}_{2}+2\unicode[STIX]{x1D714}\cos \unicode[STIX]{x1D719})v_{y}+(\unicode[STIX]{x1D6FA}_{3}+2\unicode[STIX]{x1D714}\sin \unicode[STIX]{x1D719})v_{z}=0,\\ \unicode[STIX]{x1D6FA}_{1}w_{x}+(\unicode[STIX]{x1D6FA}_{2}+2\unicode[STIX]{x1D714}\cos \unicode[STIX]{x1D719})w_{y}+(\unicode[STIX]{x1D6FA}_{3}+2\unicode[STIX]{x1D714}\sin \unicode[STIX]{x1D719})w_{z}=0.\end{array}\right\}\end{eqnarray}$$
Unlike in the beginning of the proof of Theorem 2.4 we cannot conclude anymore that
$yu_{z}$
is a harmonic function. Thus, all the advantages emerging from the latter fact are lost.
Acknowledgements
The support of the Science Foundation Ireland (SFI) under research grant 13/CDA/2117 is gratefully acknowledged. The author would like to thank the editor (Professor J. Kirby) and the referees for many useful suggestions and remarks. The author also thanks Professor R. S. Johnson and Professor S. Leibovich for helpful discussions.


