Symmetry, Integrability and Geometry: Methods and Applications SIGMA 11 (2015), 032, 14 pages
Three-Phase Freak Waves
Aleksandr O. SMIRNOV, Sergei G. MATVEENKO, Sergei K. SEMENOV
and Elena G. SEMENOVA
St.-Petersburg State University of Aerospace Instrumentation (SUAI),
67 Bolshaya Morskaya Str., St.-Petersburg, 190000, Russia
E-mail: alsmir@guap.ru, MatveiS239@gmail.com, sksemenov@mail.ru, egsemenova@mail.ru
Received December 05, 2014, in final form April 11, 2015; Published online April 21, 2015
http://dx.doi.org/10.3842/SIGMA.2015.032
Abstract. In the article, we describe three-phase finite-gap solutions of the focusing non-
linear Schro¨dinger equation and Kadomtsev–Petviashvili and Hirota equations that exhibit
the behavior of almost-periodic “freak waves”. We also study the dependency of the solution
parameters on the spectral curves.
Key words: nonlinear Schro¨dinger equation; Hirota equation; freak waves; theta function;
reduction; covering; spectral curve
2010 Mathematics Subject Classification: 35Q55; 37C55
1 Introduction
This study was motivated by the intention to demonstrate the behavior of three-phase extreme
waves. Most recent scientific research shows that the simplest and most universal model for
such waves is the focusing nonlinear Schro¨dinger equation (NLS)
ipt + pxx + 2 |p|
2 p = 0, i2 = −1, (1)
Since 1968 the equation (1) has been describing distribution on the surface of the ocean of weakly
nonlinear quasi-monochromatic wave packets with relatively steep fronts [44]. An application of
this equation to the problems of nonlinear optics was known earlier [7]. Since the equation (1) is
a model of first approximation, it appears in simulations of many weakly nonlinear phenomena.
This equation has a wide range of applications ranging from plasma physics [28] to financial
markets [43].
Among the properties of equation (1) there is a modulation instability that leads to the
appearance of the so-called “freak waves” (in hydrodynamics known as “rogue waves”) [2].
These waves represent amplitude peaks localized in space and time. In the last 20 years, first
in hydrodynamics and then in nonlinear optics, these waves have been the object of numerous
theoretical and experimental studies [3]. Such attention to the problem of the “freak waves” is
due to the losses at oil platforms, tankers, container ships and other large vessels caused by the
“rogue waves”.
There are many more precise and more complex models, which give a more exact description
of the “freak waves” [3]. These models can be divided into two classes. In the first class one can
solve them analyticaly while in the second class one can use numerical methods only. Analytical
methods include: inverse scattering transform method; finite-gap integration method; Ba¨cklund
transform method; Darboux transform method; Hirota method.
In the present work, we use a finite-gap integration method. The works of Dubrovin, Novikov,
Marchenko, Lax, McKean, van Moerbeke, Matveev, Its, Krichever [9, 10, 11, 12, 13, 21, 22, 26, 29,
31, 33, 35] give a description of this method (see also the review [32]). However, another method
of constructing finite-gap solutions of integrable nonlinear equations exists [23, 24, 34, 36]. Let
2 A.O. Smirnov, S.G. Matveenko, S.K. Semenov and E.G. Semenova
us remark that first method is based on Baker–Akhiezer function but the second one is based
on some Fay’s identities [14]. In our paper, we use the first method and Its and Kotlyarov’s
classic formulas [18, 20] (see also [6]).
Our goal here is to show the behavior of three-phase algebro-geometric solutions of NLS, KP-I
and Hirota equations. Section 2 of this paper contains the basic notations and classic formulas
for algebro-geometric solutions of integrable nonlinear equations under consideration. Section 3
is devoted to the periodicity of three-phase solutions of NLS, KP-I and Hirota equations. In
Section 4 we consider an example of three-phase algebro-geometric solutions of KP-I and Hirota
equations for different values of parameters.
2 Finite-gap multi-phase solutions of the NLS equation
The nonlinear differential equations that are integrated by methods of the algebraic geome-
try, can be obtained as a compatibility condition of the system of ordinary linear differential
equations with a spectral parameter [6, 15, 16]. In particular, let us consider the following
equations [15, 19, 38]
Yx = UY, Yz = VY, Yt = WY, (2)
where
U = −λ
(
i 0
0 −i
)
+
(
0 iψ
−iφ 0
)
, V = 2λU + V0, W = 4λ2U + 2λW0 + W1,
λ is a spectral parameter. Using these equations and additional relations
(Yx)z = (Yz)x, (Yx)t = (Yt)x
one can easy obtain the so-called equations of zero curvature
Uz −Vx + UV−VU = 0 and Ut −Wx + UW−WU = 0, (3)
which should be valid for all values of spectral parameter λ. Respectively, it follows from
equations (3) that matrixes V0, W0, W1 take the forms
W0 = V0 =
(
−iψφ −ψx
−φx iψφ
)
, W1 =
(
ψxφ− ψφx 2iψ2φ− iψxx
−2iψφ2 + iφxx ψφx − ψxφ
)
,
Also, W = 2λV + W1. Conditions (3) lead to additional system of equations (parities). The
first system is the coupled nonlinear Schro¨dinger equation
iψz + ψxx − 2ψ
2φ = 0,
iφz − φxx + 2ψφ
2 = 0,
(4)
and the second system is the coupled modified Korteweg–de Vries equation
ψt + ψxxx − 6ψφψx = 0,
φt + φxxx − 6ψφφx = 0.
(5)
These two systems of the nonlinear differential equations are closely related to two other
ones. Specifically, differentiating equations (4) with respect to x and substituting them in (5),
one obtains the coupled modified two-dimensional nonlinear Schro¨dinger equation in cone coor-
dinates [27]
iψt + ψxz + 2i(ψφx − φψx)ψ = 0,
iφt − φxz + 2i(φψx − ψφx)φ = 0,
Three-Phase Freak Waves 3
Also, the functions ψ(x, t,−αt) and φ(x, t,−αt) are solutions to the coupled integrable Hirota
equation (α ∈ R)
iψt + ψxx − 2ψ
2φ− iα(ψxxx − 6ψφψx) = 0,
iφt − φxx + 2ψφ
2 − iα(φxxx − 6ψφφx) = 0,
(6)
if ψ(x, z, t) and φ(x, z, t) are solutions of (4) and (5).
Systems of the nonlinear differential equations (4), (5) are the first two integrable systems
from the AKNS hierarchy [15]. One of the features of finite-gap multi-phase solutions of the
integrable nonlinear equations is that fact that in some sense they are the solutions of all
hierarchy. Particulary, our solutions can be used for constructing solutions of generalized non-
linear Schro¨dinger equation [42]. By substituting φ = ±ψ into equation (4) we get a standard
form of the nonlinear Schro¨dinger equation. Particularly, for φ = −ψ equations (4) transform
to (1) [11, 18, 19] and equations (6) transform to the integrable Hirota equation [4, 8, 17, 30]
iψt + ψxx + 2 |ψ|
2 ψ − iα
(
ψxxx + 6 |ψ|
2 ψx
)
= 0. (7)
It is also easy to check that for any ψ and φ, that satisfy both (4) and (5) simultaneously,
the function u(x, z, t) = −2ψφ is a solution of the Kadomtsev–Petviashvili-I equation (KP-I)
3uzz = (4ut + uxxx + 6uux)x. (8)
In case φ = ±ψ this solution is a real function.
Finite-gap solutions of systems (4), (5) are parameterized by the hyperelliptic curve Γ =
{(χ, λ)} of the genus g [15, 38]:
Γ : χ2 =
2g+2∏
j=1
(λ− λj),
The branch points (λ = λj , j = 1, . . . , 2g+ 2) of this curve are the endpoints of the spectral arcs
of continuous spectrum of Dirac operator (2). Infinitely far point of the spectrum corresponds
two different points P±∞ on the curve Γ. In case φ = −ψ the curve Γ has the form
Γ : χ2 =
g+1∏
j=1
(λ− λj)(λ− λj) = λ
2g+2 +
2g+2∑
j=1
χjλ
2g+2−j , =χj = 0, =(λj) 6= 0. (9)
Following a standard procedure of constructing finite-gap solutions [6, 11, 38], for Γ let us
choose a canonical basis of cycles γt = (a1, . . . , ag, b1, . . . , bg) with matrix of intersection indices
C0 =
(
0 I
−I 0
)
.
To satisfy the condition φ = −ψ, it is necessary [6, 11] that this basis of cycles is transformed
according to the rules
τ̂1a = −a, τ̂1b = b +Ka, (10)
where τ1 is anti-holomorphic involution, τ1 : (χ, λ)→ (χ, λ).
Let us also consider normalized holomorphic differentials dUj :
∮
ak
dUj = δkj , k, j = 1, . . . , g,
4 A.O. Smirnov, S.G. Matveenko, S.K. Semenov and E.G. Semenova
and a matrix of periods B of the curve Γ:
Bkj =
∮
bk
dUj , k, j = 1, . . . , g.
It is well known (see, for example, [5, 11]) that the matrix B is a symmetric matrix with positively
defined imaginary part.
Let us introduce g-dimensional Riemann theta function with characteristics η, ζ ∈ Rg [5, 11,
14]:
Θ
[
ηt; ζt
]
(p|B) =
∑
m∈Zg
exp
{
pii(m + η)tB(m + η) + 2pii(m + η)t(p + ζ)
}
,
Θ
[
0t; 0t
]
(p|B) ≡ Θ(p|B),
where B is a matrix of periods, p ∈ Cg and summation passes over an integer g-dimensional
lattice.
Let us also define on Γ normalized Abelian integrals of the second kind (Ωj(P), j = 1, 2, 3)
and the third kind (ω0(P)) with the following asymptotic at infinitely distant points P±∞:
∮
ak
dΩ1 =
∮
ak
dΩ2 =
∮
ak
dΩ3 =
∮
ak
dω0 = 0, k = 1, . . . , g,
Ω1(P) = ∓i
(
λ−K1 +O
(
λ−1
))
, P → P±∞,
Ω2(P) = ∓i
(
2λ2 −K2 +O
(
λ−1
))
, P → P±∞,
Ω3(P) = ∓i
(
4λ3 −K3 +O
(
λ−1
))
, P → P±∞,
ω0(P) = ∓
(
lnλ− lnK0 +O
(
λ−1
))
, P → P±∞,
χ = ±
(
λg+1 +O (λg)
)
, P → P±∞.
Let us denote the vectors of b-periods of Abelian integrals of the second kind Ω1(P), Ω2(P),
Ω3(P) by 2piiU, 2piiV, 2piiW respectively.
Theorem 1 ([6, 38]). Function
Y (P, x, z, t) =
(
y1(P, x, z, t) y1(τ0P, x, z, t)
y2(P, x, z, t) y2(τ0P, x, z, t)
)
,
where τ0 is hyperelliptic involution, τ0 : (χ, λ)→ (−χ, λ),
y1(P, x, z, t) =
Θ(U(P) + Ux+ Vz + Wt−X)Θ(Z)
Θ(U(P)−X)Θ(Ux+ Vz + Wt+ Z)
× exp{Ω1(P)x+ Ω2(P)z + Ω3(P)t+ iΦ(x, z, t)},
y2(P, x, z, t) = ρ
Θ(U(P) + Ux+ Vz + Wt+ ∆−X)Θ(Z−∆)
Θ(U(P)−X)Θ(Ux+ Vz + Wt+ Z)
× exp{Ω1(P)x+ Ω2(P)z + Ω3(P)t− iΦ(x, z, t) + ω0(P)},
is the eigenfunction of the Dirac operator (2) with functions
ψ(x, z, t) =
2K0
ρ
Θ(Z)Θ(Ux+ Vz + Wt+ Z−∆)
Θ(Z−∆)Θ(Ux+ Vz + Wt+ Z)
exp{2iΦ(x, z, t)},
φ(x, z, t) = 2ρK0
Θ(Z−∆)Θ(Ux+ Vz + Wt+ Z + ∆)
Θ(Z)Θ(Ux+ Vz + Wt+ Z)
exp{−2iΦ(x, z, t)}, (11)
Three-Phase Freak Waves 5
for any z, t and ρ 6= 0. The functions (11) satisfy the equations (4) and (5). Here ∆ is the
vector of holomorphic Abelian integrals, calculated along a path connecting P−∞ and P
+
∞ without
crossing any of the basic cycles,
∆ = U(P+∞)− U(P
−
∞), Φ(x, z, t) = K1x+K2z +K3t,
X = K +
g∑
j=1
U(Pj), Z = U(P+∞)−X,
K is a vector of Riemann constants [5, 11, 14, 25]; Pj, j = 1, . . . , g is a non-special divisor. If
the spectral curve Γ satisfies the condition (9), then the following equalities hold
|ψ|2 = −4K20
Θ(Ux+ Vz + Wt+ Z−∆)Θ(Ux+ Vz + Wt+ Z + ∆)
Θ2(Ux+ Vz + Wt+ Z)
, (12)
=U = =V = =W = =Z = 0, K20 < 0.
It is easy to see that the corresponding solution of KP-I equation (8) has the form
u(x, z, t) = −8K20
Θ(Ux+ Vz + Wt+ Z−∆)Θ(Ux+ Vz + Wt+ Z + ∆)
Θ2(Ux+ Vz + Wt+ Z)
, (13)
and that the square of amplitude of solution of Hirota equation (7) equals
|ψH |
2 (x, t) = −4K20
Θ(Ux+ (V − αW)t+ Z−∆)Θ(Ux+ (V − αW)t+ Z + ∆)
Θ2(Ux+ (V − αW)t+ Z)
.
3 Features of three-phase solutions
In case g = 3, the basis of normalized holomorphic differentials is defined by the formula [6, 11]:
dUk =
(
ck1λ
2 + ck2λ+ ck3
)dλ
χ
,
where
C =
(
At
)−1
, Ajm =
∮
aj
λ3−m
dλ
χ
.
It follows from equation (` is an arbitrary path on Γ)
∫
τ̂ `
dω =
∫
`
τ∗dω,
that
Ajm =
∮
aj
λ3−m
dλ
χ
=
∮
aj
τ∗1
(
λ3−m
dλ
χ
)
=
∮
τ̂1aj
λ3−m
dλ
χ
= −
∮
aj
λ3−m
dλ
χ
= −Ajm.
Therefore A = −A and C = −C. Similarly, with integrals on b-cycles, we obtain
B = −B −K or <B = −
1
2
K.
It follows from bilinear relations of Riemann (see, for example, [5, 6, 11]) that the coordinates
of the vectors U, V, W can be written as
Um = −i
(
dUm
dξ−
∣
∣
∣
∣
ξ−=0
−
dUm
dξ+
∣
∣
∣
∣
ξ+=0
)
, Vm = −2i
(
d2Um
dξ2−
∣
∣
∣
∣
ξ−=0
−
d2Um
dξ2+
∣
∣
∣
∣
ξ+=0
)
,
6 A.O. Smirnov, S.G. Matveenko, S.K. Semenov and E.G. Semenova
Wm = −2i
(
d3Um
dξ3−
∣
∣
∣
∣
ξ−=0
−
d3Um
dξ3+
∣
∣
∣
∣
ξ+=0
)
,
where ξ± = 1/λ are local parameters in the neighborhood of infinitely distant points P±∞.
Calculating the derivatives we obtain the relations
Um = −2icm1, Vm = 2iχ1cm1 − 4icm2, Wm = i
(
4χ2 − 3χ
2
1
)
cm1 + 4iχ1cm2 − 8icm3,
or
(U,V,W) = iC


−2 2χ1 4χ2 − 3χ21
0 −4 4χ1
0 0 −8

 . (14)
It follows from (14) that the vectors U, V, W are real and linearly independent. Therefore,
U, V, W are the basis vectors in R3. Hence, any vector from R3 can be presented in the form
of the linear combinations of these vectors. In particular, for the vectors of the periods of the
three-dimensional theta-functions et1 = (1, 0, 0), e
t
2 = (0, 1, 0), e
t
3 = (0, 0, 1) we can write the
following relations
ek = XkU + ZkV + TkW.
Therefore, three-phase solutions (13) of equation KP-I (8) are the periodic functions in a three-
dimensional space
u(x+ Xk, z + Zk, t+ Tk) = u(x, z, t).
If a three-phase solution of (8) has a form of freak waves, then the maxima of its amplitude
are located in nodes of a three-dimensional lattice with edges (Xk,Zk, Tk). These edges can be
found by an inversion of the matrix (U,V,W):


X1 X2 X3
Z1 Z2 Z3
T1 T2 T3

 = (U,V,W)−1 = i


1/2 χ1/4 χ2/4− χ21/16
0 1/4 χ1/8
0 0 1/8

At.
Therefore, for three-phases solutions of equation KP-I (8) it is possible to describe their behavior
in the following way: after a time interval ∆t = Tk a surface of solution u(x, z) reproduces itself
with a shift on the XOZ plane by the (Xk,Zk) vector.
As the three-phase solution of the equations (1) depends on two coordinates, x and z, and
the third coordinate t is considered to be a parameter, the value of amplitude of this solution
depends on the distance between the nodes of the given three-dimensional lattice and a plane
t = t0. Hence, in contrast to the case of the two-phase solution [39, 40, 41], where the change of
initial phase Z led to trivial shift of the solution on XOZ plane, the amplitude of the three-phase
solution (11) of equations (1) depends on a choice of initial phase Wt0 + Z in a slightly more
complicated fashion.
4 An example of three-phase solution
Let us consider a spectral curve Γ3 = {χ, λ} of genus g = 3:
Γ3 : χ
2 =
(
(λ− λ0)
4 − 2a2(λ− λ0)
2 cos 2ϕ+ a4
)(
(λ− λ0)
4 − 2b2(λ− λ0)
2 cos 2ϕ+ b4
)
,
where 0 < a < b, pi/4 < ϕ < pi/2.
Three-Phase Freak Waves 7
Figure 1. Canonical basis of cycles on Γ3.
Let us choose the basis of cycles on Γ3 as it is shown on Fig. 1.
It is easy to check that the anti-holomorphic involution τ1 transforms the canonical basis of
cycles using the rule (10) with the matrix
K =


0 1 1
1 0 1
1 1 0

 .
There are also three holomorphic involutions on Γ3:
τ0 : (χ, λ)→ (−χ, λ),
τ2 : (χ, λ)→ (χ, 2λ0 − λ),
τ3 : (χ, λ)→
(
a2b2(λ− λ0)
−4χ, λ0 + ab(λ− λ0)
−1).
As a corollary, the curve Γ3 covers the following two curves:
1) Γ1 = Γ3/τ2 of genus g = 1
Γ1 : χ
2
+ =
(
t2 − 2a2t cos 2ϕ+ a4
)(
t2 − 2b2t cos 2ϕ+ b4
)
,
2) Γ2 = Γ3/(τ0τ2) of genus g = 2
Γ2 : χ
2
− = t
(
t2 − 2a2t cos 2ϕ+ a4
)(
t2 − 2b2t cos 2ϕ+ b4
)
,
where t = (λ− λ0)2, χ+ = χ, χ− = (λ− λ0)χ, and
dt
χ+
=
2(λ− λ0)dλ
χ
,
tdt
χ−
=
2(λ− λ0)2dλ
χ
,
dt
χ−
=
2dλ
χ
.
The curves Γ1 and Γ2 are shown on Figs. 2 and 3, where t1 = b2e2iϕ, t2 = a2e2iϕ.
The coverings generate the following transformations of cycles:


a1
a2
a3

→ S


a1
a21
a22

+ P


b1
b21
b22

 ,


b1
b2
b3

→ Q


a1
a21
a22

+R


b1
b21
b22

 ,
8 A.O. Smirnov, S.G. Matveenko, S.K. Semenov and E.G. Semenova
Figure 2. The curve Γ1. Figure 3. The curve Γ2.
where
S =


−1 1 0
1 0 −1
1 0 1

 , P =


0 −2 0
0 0 2
0 0 −2

 ,
Q =


−1 1 0
0 0 −1
0 0 1

 , R =


0 0 0
1 1 1
1 1 −1

 .
Recall that these matrices should satisfy the relations
StQ = QtS, RtP = P tR, StR−QtP = nI,
where I is identity matrix, n = 2 is the number of covering sheets.
Due to involution τ3, the curve Γ2 covers two elliptic curves Γ± (Figs. 4 and 5):
Γ± : ν
2
± = (s± 2ab)
(
s2 − 2
(
a2 + b2
)
s cos 2ϕ+ a4 + b4 + 2a2b2 cos 4ϕ
)
,
where
s = t+
a2b2
t
, ν± =
t± ab
t2
χ−,
ds
ν±
=
(t∓ ab)dt
χ−
.
The coverings of Γ2 on Γ± generate the following cycles mappings
(
a21
a22
)
→
(
1 1
−1 1
)(
a+
a−
)
,
(
b21
b22
)
→
(
1 1
−1 1
)(
b+
b−
)
,
As a result, we have


a1
a2
a3

→


−1 1 1
1 1 −1
1 −1 1




a1
a+
a−

+


0 −2 −2
0 −2 2
0 2 −2




b1
b+
b−

 , (15)


b1
b2
b3

→


−1 1 1
0 1 −1
0 −1 1




a1
a+
a−

+


0 0 0
1 0 2
1 2 0




b1
b+
b−

 . (16)
Three-Phase Freak Waves 9
Figure 4. The curve Γ+. Figure 5. The curve Γ−.
From (15), (16) and relations
dλ
χ
=
1
4ab
ds
ν−
−
1
4ab
ds
ν+
,
λdλ
χ
=
1
2
dt
χ+
+
λ0
4ab
ds
ν−
−
λ0
4ab
ds
ν+
,
λ2dλ
χ
= λ0
dt
χ+
+
λ20 + ab
4ab
ds
ν−
−
λ20 − ab
4ab
ds
ν+
it follows that
C =


c1 + c3 −2λ0(c1 + c3) (λ20 − ab)c1 + (λ
2
0 + ab)c3
c1 c2 − 2λ0c1 (λ20 − ab)c1 − λ0c2
c3 c2 − 2λ0c3 (λ20 + ab)c3 − λ0c2

 ,
B =


i(b1 + b3) ib1 − 1/2 ib3 − 1/2
ib1 − 1/2 i(b1 + b2) ib2 − 1/2
ib3 − 1/2 ib2 − 1/2 i(b2 + b3)

 ,
where
c1 =
1
2(α1 − 2β1)
, c2 =
1
2α2
, c3 =
1
2(α3 − 2β3)
,
ib1 =
α1
2(α1 − 2β1)
, ib2 =
β2
2α2
, ib3 =
α3
2(α3 − 2β3)
,
α1 =
1
2
∮
a+
ds
ν+
, α2 =
1
2
∮
a1
dt
χ+
, α3 =
1
2
∮
a−
ds
ν−
,
β1 =
1
2
∮
b+
ds
ν+
, β2 =
1
2
∮
b1
dt
χ+
, β3 =
1
2
∮
b−
ds
ν−
.
From the structure of the matrix B and from the matrix version of Appel’s theorem [37] it
follows that the Riemann theta function of curve Γ3 equals:
Θ(p|B) = f(p˜1, p˜2, p˜3) = ϑ3(p˜1|h1)ϑ3(p˜2|h2)ϑ3(p˜3|h3) + ϑ4(p˜1|h1)ϑ1(p˜2|h2)ϑ1(p˜3|h3)
+ ϑ1(p˜1|h1)ϑ4(p˜2|h2)ϑ1(p˜3|h3) + ϑ1(p˜1|h1)ϑ1(p˜2|h2)ϑ4(p˜3|h3), (17)
where p˜j = pj + pj+1 − pj+2, pj+3 ≡ pj , hj = exp(−4bj).
10 A.O. Smirnov, S.G. Matveenko, S.K. Semenov and E.G. Semenova
Figure 6. Three-phase solution of KP-I equa-
tion for λ0 = 0, t = 0.
Figure 7. Three-phase solution of KP-I equa-
tion for λ0 = 0, t = 0.1.
Figure 8. Three-phase solution of KP-I equa-
tion for λ0 = 0, t = 0.2.
Figure 9. Three-phase solution of KP-I equa-
tion for λ0 = 0, t = 0.3.
The functions ϑj(p|h) are Jacoby elliptic theta functions [1]:
ϑ1(p|h) = 2
∞∑
m=1
(−1)m−1h(m−1/2)
2
sin[(2m− 1)pip],
ϑ2(p|h) = 2
∞∑
m=1
h(m−1/2)
2
cos[(2m− 1)pip],
ϑ3(p|h) = 1 + 2
∞∑
m=1
hm
2
cos(2mpip),
ϑ4(p|h) = 1 + 2
∞∑
m=1
(−1)mhm
2
cos(2mpip).
Using the reduced form of theta function (17) and values for the vectors of periods, one
obtains the following formula for a squared absolute value of the three-phase solution (12) of
the focusing NLS equation (1)
|ψ|2 = −4K20f(k1x+ κ1t+ δ1, k2z + δ2, k3x+ κ3t+ δ3) (18)
× f(k1x+ κ1t− δ1, k2z − δ2, k3x+ κ3t− δ3)× {f(k1x+ κ1t, k2z, k3x+ κ3t)}
−2,
Three-Phase Freak Waves 11
Figure 10. Three-phase solution of KP-I equa-
tion for λ0 = k2/(4k1), t = 0.
Figure 11. Three-phase solution of KP-I equa-
tion for λ0 = k2/(4k1), t = 0.1.
Figure 12. Three-phase solution of KP-I equa-
tion for λ0 = k2/(4k1), t = 0.2.
Figure 13. Three-phase solution of KP-I equa-
tion for λ0 = k2/(4k1), t = 0.3.
where the function f(p˜1, p˜2, p˜3) is defined by equation (17), and
k1 = −4ic1, k2 = −8ic2, k3 = −4ic3,
κ1 = 4k1
(
3λ20 − ab+
(
a2 + b2
)
cos(2ϕ)
)
, κ3 = 4k3
(
3λ20 + ab+
(
a2 + b2
)
cos(2ϕ)
)
.
(17), (18) imply that for λ0 = 0 the amplitude of the constructed solution of NLS equation (1)
is a periodic function of z, and for λ0 = 0, ϕ = 12 arccos
(
±ab
a2+b2
)
it is a periodic function of z
and t.
Recall that the three-phase solution u(x, z, t) of the KP-I equation (8) and the square of am-
plitude of three-phase solution of Hirota equation (7), |ψH(x, t)|
2, can be constructed from (18)
by using relations u(x, z, t) = 2 |ψ(x, z, t)|2 and |ψH(x, t)|
2 = |ψ(x, t,−αt)|2.
The three-phase solution of KP-I equation for ab = 1,
√
b/a = 1.3, ϕ = 0.3pi at the different
moment of time t and for λ0 = 0 is presented on the Figs. 6–9. One can see the same solution
for λ0 = k2/(4k1) on the Figs. 10–13. It is easy to see all three phases of solution on Figs. 6–13.
Two phases are shortwaves and the third phase is a long-wave envelope. One can see also on
Figs. 6–9 that the solution for λ0 = 0 is periodic in z, and that the long-wave envelope moves
to the right side.
The three-phase solution of Hirota equation for ab = 1,
√
b/a = 1.3, ϕ = 0.3pi, α = 0.1 and
for different values of λ0 is presented in Figs. 14–17. It is easy to see all three phases of solution
only in Fig. 15.
12 A.O. Smirnov, S.G. Matveenko, S.K. Semenov and E.G. Semenova
Figure 14. Amplitude of three-phase solution
of Hirota equation for λ0 = 0.
Figure 15. Amplitude of three-phase solution
of Hirota equation for λ0 = 4.
Figure 16. Amplitude of three-phase solution
of Hirota equation for λ0 = k2/(4k1).
Figure 17. Amplitude of three-phase solution
of Hirota equation for λ0 = k2/(4k3).
Acknowledgements
Authors thank Professor V.B. Matveev for his support and the discussions that we held over this
paper and quasi-rational solutions of the NLS equation. This research was conducted within the
framework of the State order of the Ministry of Education and Science of Russian Federation,
and partially supported by RFBR (research project 14-01-00589 a).
References
[1] Akhiezer N.I., Elements of the theory of elliptic functions, Translations of Mathematical Monographs, Vol. 79,
Amer. Math. Soc., Providence, RI, 1990.
[2] Akhmediev N.N., Ankiewicz A., Solitons, nonlinear pulses and beams, Chapman & Hall, London, 1997.
[3] Akhmediev N., Pelinovsky E. (Editors), Discussion & debate: rogue waves – towards a unifying concept?,
Eur. Phys. J. Special Topics 185 (2010), 266 pages.
[4] Ankiewicz A., Soto-Crespo J.M., Akhmediev N., Rogue waves and rational solutions of the Hirota equation,
Phys. Rev. E 81 (2010), 046602, 8 pages.
[5] Baker H.F., Abel’s theorem and the allied theory including the theory of theta functions, Cambridge Uni-
versity Press, Cambridge, 1897.
[6] Belokolos E.D., Bobenko A.I., Enol’skii V.Z., Its A.R., Matveev V.B., Algebro-geometric approach to non-
linear integrable equations, Springer Series in Nonlinear Dynamics, Springer-Verlag, Berlin, 1994.
Three-Phase Freak Waves 13
[7] Chiao R.Y., Garmire E., Townes C.H., Self-trapping of optical beams, Phys. Rev. Lett. 13 (1964), 479–482.
[8] Dai C.Q., Zhang J.F., New solitons for the Hirota equation and generalized higher-order nonlinear
Schro¨dinger equation with variable coefficients, J. Phys. A: Math. Gen. 39 (2006), 723–737.
[9] Dubrovin B.A., Inverse problem for periodic finite-zoned potentials in the theory of scattering, Funct. Anal.
Appl. 9 (1975), 61–62.
[10] Dubrovin B.A., Periodic problems for the Korteweg–de Vries equation in the class of finite band potentials,
Funct. Anal. Appl. 9 (1975), 215–223.
[11] Dubrovin B.A., Theta functions and non-linear equations, Russian Math. Surveys 36 (1981), no. 2, 11–92.
[12] Dubrovin B.A., Matveev V.B., Novikov S.P., Nonlinear equations of Korteweg–de Vries type, finite-band
linear operators and Abelian varieties, Russian Math. Surveys 31 (1976), no. 1, 59–146.
[13] Dubrovin B.A., Novikov S.P., A periodicity problem for the Korteweg–de Vries and Sturm–Liouville equa-
tions. Their connection with algebraic geometry, Sov. Math. Dokl. 15 (1974), 1597–1601.
[14] Fay J.D., Theta functions on Riemann surfaces, Lecture Notes in Math., Vol. 352, Springer-Verlag, Berlin –
New York, 1973.
[15] Gesztesy F., Holden H., Soliton equations and their algebro-geometric solutions. Vol. I. (1+ 1)-dimensional
continuous models, Cambridge Studies in Advanced Mathematics, Vol. 79, Cambridge University Press,
Cambridge, 2003.
[16] Gesztesy F., Holden H., Michor J., Teschl G., Soliton equations and their algebro-geometric solutions.
Vol. II. (1 + 1)-dimensional discrete models, Cambridge Studies in Advanced Mathematics, Vol. 114, Cam-
bridge University Press, Cambridge, 2008.
[17] Guo B., Ling L., Liu Q.P., Nonlinear Schro¨dinger equation: generalized Darboux transformation and rogue
wave solutions, Phys. Rev. E 85 (2012), 026607, 9 pages, arXiv:1108.2867.
[18] Its A.R., Inversion of hyperelliptic integrals, and integration of nonlinear differential equations, Vestnik
Leningrad. Univ. (1976), no. 7, 39–46.
[19] Its A.R., “Isomonodromy” solutions of equations of zero curvature, Math. USSR-Izv. 26 (1986), 497–529.
[20] Its A.R., Kotlyarov V.P., On a class of solutions of the nonlinear Schro¨dinger equation, Dokl. Akad. Nauk
USSR Ser. A 11 (1976), 965–968.
[21] Its A.R., Matveev V.B., Hill’s operator with finitely many gaps, Funct. Anal. Appl. 9 (1975), 65–66.
[22] Its A.R., Matveev V.B., Schro¨dinger operators with the finite-band spectrum and N -soliton solutions of the
Korteweg–de Vries equation, Theoret. and Math. Phys. 23 (1975), 343–355.
[23] Kalla C., Klein C., New construction of algebro-geometric solutions to the Camassa–Holm equation and
their numerical evaluation, Proc. R. Soc. Lond. Ser. A 468 (2012), 1371–1390, arXiv:1109.5301.
[24] Kalla C., Klein C., On the numerical evaluation of algebro-geometric solutions to integrable equations,
Nonlinearity 25 (2012), 569–596, arXiv:1107.2108.
[25] Krazer A., Lehrbuch der Thetafunktionen, Teubner, Leipzig, 1903.
[26] Krichever I.M., Methods of algebraic geometry in the theory of non-linear equations, Russian Math. Surveys
32 (1977), no. 6, 185–213.
[27] Kundu A., Mukherjee A., Naskar T., Modelling rogue waves through exact dynamical lump soliton controlled
by ocean currents, Proc. R. Soc. Lond. Ser. A 470 (2014), 20130576, 20 pages, arXiv:1204.0916.
[28] Kuznetsov E.A., Solitons in a parametrically unstable plasma, Sov. Phys. Dokl. 22 (1977), 507–508.
[29] Lax P.D., Periodic solutions of the KdV equations, in Nonlinear Wave Motion (Proc. AMS-SIAM Summer
Sem., Clarkson Coll. Tech., Potsdam, N.Y., 1972), Lectures in Appl. Math., Vol. 15, Amer. Math. Soc.,
Providence, R.I., 1974, 85–96.
[30] Li C.Z., He J.S., Darboux transformation and positons of the inhomogeneous Hirota and the Maxwell–Bloch
equation, Sci. China Phys. Mech. Astronomy 57 (2014), 898–907, arXiv:1210.2501.
[31] Marchenko V.A., The periodic Korteweg–de Vries problem, Math. USSR Sb. 24 (1974), 319–344.
[32] Matveev V.B., 30 years of finite-gap integration theory, Philos. Trans. R. Soc. Lond. Ser. A 366 (2008),
837–875.
[33] McKean H.P., van Moerbeke P., The spectrum of Hill’s equation, Invent. Math. 30 (1975), 217–274.
[34] Mumford D., Tata lectures on theta. II. Jacobian theta functions and differential equations, Progress in
Mathematics, Vol. 43, Birkha¨user Boston, Inc., Boston, MA, 1984.
14 A.O. Smirnov, S.G. Matveenko, S.K. Semenov and E.G. Semenova
[35] Novikov S.P., The periodic problem for the Korteweg–de Vries equation, Funct. Anal. Appl. 8 (1974),
236–246.
[36] Previato E., Hyperelliptic quasiperiodic and soliton solutions of the nonlinear Schro¨dinger equation, Duke
Math. J. 52 (1985), 329–377.
[37] Smirnov A.O., A matrix analogue of a theorem of Appell and reductions of multidimensional Riemann
theta-functions, Math. USSR Sb. 61 (1988), 379–388.
[38] Smirnov A.O., Elliptic solutions of the nonlinear Schro¨dinger equation and a modified Korteweg–de Vries
equation, Mat. Sb. 185 (1994), 103–114.
[39] Smirnov A.O., Solution of a nonlinear Schro¨dinger equation in the form of two-phase freak waves, Theoret.
and Math. Phys. 173 (2012), 1403–1416.
[40] Smirnov A.O., Periodic two-phase “rogue waves”, Math. Notes 94 (2013), 897–907.
[41] Smirnov A.O., Semenova E.G., Zinger V., Zinger N., On a periodic solution of the focusing nonlinear
Schro¨dinger equation, arXiv:1407.7974.
[42] Wang L.H., Porsezian K., He J.S., Breather and rogue wave solutions of a generalized nonlinear Schro¨dinger
equation, Phys. Rev. E 87 (2013), 053202, 10 pages, arXiv:1304.8085.
[43] Yan Z., Vector financial rogue waves, Phys. Lett. A 375 (2011), 4274–4279, arXiv:1101.3107.
[44] Zakharov V.E., Stability of periodic waves of finite amplitude on the surface of a deep fluid, J. Appl. Mech.
Tech. Phys. 9 (1968), 190–194.