Partial-rogue waves that come from nowhere but leave with a trace in the Sasa-Satsuma equation Bo Yang1 Jianke Yang2

2025-05-02 0 0 935.25KB 11 页 10玖币
侵权投诉
Partial-rogue waves that come from nowhere but leave with a trace in the
Sasa-Satsuma equation
Bo Yang1, Jianke Yang2
1School of Mathematics and Statistics, Ningbo University, Ningbo 315211, China
2Department of Mathematics and Statistics, University of Vermont, Burlington, VT 05401, U.S.A
Partial-rogue waves, i.e., waves that “come from nowhere but leave with a trace”, are analytically
predicted and numerically confirmed in the Sasa-Satsuma equation. We show that, among a class of
rational solutions in this equation that can be expressed through determinants of 3-reduced Schur
polynomials, partial-rogue waves would arise if these rational solutions are of certain orders, where
the associated generalized Okamoto polynomials have real but not imaginary roots, or imaginary
but not real roots. We further show that, at large negative time, these partial-rogue waves approach
the constant-amplitude background, but at large positive time, they split into several fundamental
rational solitons, whose numbers are determined by the number of real or imaginary roots in the
underlying generalized Okamoto polynomial. Our asymptotic predictions are compared to true
solutions, and excellent agreement is observed.
I. INTRODUCTION
Rogue waves has been a subject of intensive theoreti-
cal and experimental studies in mathematical and phys-
ical communities in the past decade. Hundreds of pa-
pers and several books have been published on it, and
more are still coming. Rogue waves are often defined
as “waves that come from nowhere and leave without
a trace” [1]. For example, they can be localized wave
excitations that arise from the constant-amplitude back-
ground, reach higher amplitude, and then retreat back
to the same background, as time progresses. Almost all
rogue waves that have been theoretically derived or ex-
perimentally observed belong to this category (see [2–6],
among many others).
However, there exists another type of waves that “come
from nowhere but leave with a trace”. Specifically,
these waves also arise from the constant-amplitude back-
ground (thus “come from nowhere”), stay localized, and
reach higher amplitude. Afterwards, instead of retreat-
ing back to the same constant background with no trace,
they evolve into localized waves on the constant back-
ground that persist at large time, thus leaving a trace.
The first report of such peculiar waves seems to be in
[7] for the Davey-Stewartson-II equation, where a two-
dimensional localized wave arose from the constant back-
ground and then split into two localized lumps at large
time (see Fig. 4 of that paper). Later, a similar but
one-dimensional solution was reported in [8] for the Sasa-
Satsuma equation. These peculiar waves resemble rogue
waves in the first half of evolution, but contrast them in
the second half of evolution. Due to these peculiar be-
haviors, let us call them partial-rogue waves. Note that
although two examples of partial-rogue waves can be seen
in [7, 8], there was no explanation for their appearance, as
if they were pure accidents. It was also unclear whether
additional types of partial-rogue waves could be found in
those two systems.
In this paper, we predict partial-rogue waves in the
Sasa-Satsuma equation through large-time asymptotic
analysis on its rational solutions. We show that, among
a class of rational solutions in this equation that can be
expressed through determinants of 3-reduced Schur poly-
nomials, partial-rogue waves arise if and only if these
rational solutions are of certain orders, where the associ-
ated generalized Okamoto polynomials have real but not
imaginary roots, or imaginary but not real roots. We fur-
ther show that, at large negative time, these partial-rogue
waves approach the constant-amplitude background, but
at large positive time, they split into several fundamen-
tal rational solitons, whose numbers are determined by
the number of real or imaginary roots in the underlying
generalized Okamoto polynomial. Our asymptotic pre-
dictions are compared to true solutions, and excellent
agreement is observed.
II. PRELIMINARIES
The Sasa-Satsuma equation was proposed as a higher-
order nonlinear Scodinger equation for optical pulses
that includes some additional physical effects such
as third-order dispersion and self-steepening [9, 10].
Through a variable transformation, this equation can be
written as
ut=uxxx + 6|u|2ux+ 3u(|u|2)x.(1)
Sasa and Satsuma [9] showed that this equation is inte-
grable.
A. A class of rational solutions
Soliton solutions on the zero background in this equa-
tion were derived by Sasa and Satsuma in their original
paper [9]. Later, rational solutions on a constant back-
ground, including rogue waves, were also derived [8, 11–
19]. The solutions that will be the starting point of this
paper are a certain class of rational solutions which, in
the language of Darboux transformation, are associated
arXiv:2210.03414v1 [nlin.SI] 7 Oct 2022
2
with a scattering matrix admitting a triple eigenvalue.
Such solutions have been studied in [8, 14] by Darboux
transformation. However, their solutions are not general
nor explicit for our purpose. For this reason, we will first
present general and explicit expressions for this class of
rational solutions through Schur polynomials.
Before presenting these solutions, we need to specify
the nonzero background. Through variable scalings, we
can normalize the background amplitude to be unity.
Then, this background can be written as
ubg(x, t) = ei[α(x+6t)α3t],(2)
where αis a free wavenumber parameter, which cannot
be removed since the Sasa-Satsuma equation (1) is not
Galilean-invariant. But αcan be restricted to be posi-
tive, since the Sasa-Satsuma equation is invariant under
the axes reflection of (x, t)(x, t), and negative-
αsolution can be related to positive-αsolution through
this axes reflection.
To present these explicit rational solutions, we also
need to introduce elementary Schur polynomials. These
polynomials Sj(x) with x= (x1, x2, . . .) are defined by
the generating function
X
j=0
Sj(x)j= exp
X
j=1
xjj
.(3)
In addition, we define Sj(x) = 0 when j < 0.
Our expressions for general rational solutions corre-
sponding to a triple eigenvalue in the scattering matrix
of Darboux transformation are given by the following the-
orem.
Theorem 1 When α= 1/2, the
Sasa-Satsuma equation (1) admits bounded
(N1, N2)-th order rational solutions
uN1,N2(x, t) = gN1,N2
fN1,N2
ei[α(x+6t)α3t],(4)
where N1and N2are arbitrary non-negative
integers,
fN1,N2=σ0,0, gN1,N2=σ1,0,(5)
σk,l = det σ[1,1]
k,l σ[1,2]
k,l
σ[2,1]
k,l σ[2,2]
k,l !,(6)
σ[I,J]
k,l =φ(k,l I,J)
3iI, 3jJ1iNI,1jNJ
,(7)
matrix elements in σ[I,J]
k,l are defined by
φ(k,l,I,J)
i,j =
min(i,j)
X
ν=0 p2
1
4p2
0ν
×
Siν(x+
I(k, l) + νs)Sjν(x
J(k, l) + νs),(8)
vectors x±
I(k, l)=(x±
1,I , x±
2,I ,···) are given
by
x+
r,I (k, l) = pr(x+ 6t) + βrt+kθr+lθ
r+ar,I ,(9)
x
r,J (k, l) = pr(x+ 6t) + βrtkθ
rr+ar,J ,(10)
βrand θrare coefficients from expansions
p3(κ) =
X
r=0
βrκr,ln p(κ)+iα
p0+ iα=
X
r=1
θrκr,(11)
the function p(κ) with expansion p(κ) =
P
r=0 prκrand real expansion coefficients pr
is defined by the equation
Q1[p(κ)] = Q1(p0)
3"eκ+ 2eκ/2cos 3
2κ!#,(12)
with
Q1(p)1
piα+1
p+ iα+p, (13)
p0=±3/2, the real vector s= (s1, s2,···)
is defined by the expansion
ln 2p0
p1κp(κ)p0
p(κ) + p0=
X
r=1
srκr,(14)
the asterisk ‘*’ represents complex conjuga-
tion, and
(a1,1,··· , a3N11,1),(a1,2,··· , a3N22,2) (15)
are free real constants.
In these solutions, subscripts of matrix elements in
Eq. (7) jump by 3. Thus, the corresponding determi-
nant in Eq. (6) is called a determinant of 3-reduced Schur
polynomials [20].
Note 1 When we choose p0=3/2, the first few
coefficients of pr,βr,θr, and srare
p1=121/6
2, p2=121/6
2, p3=1
43,(16)
β1=9
8121/6, β2=9
8·35/6
21/3, β3=193
16 ,(17)
θ1=121/6
3+i, θ2=i
121/63+i2, θ3= 0,(18)
s1= 0, s2= 0, s3=1
40.(19)
If we choose p0=3/2, then prand βrwould switch
sign, θrchange to θ
r, and srremain the same.
3
Note 2 If we choose p0=3/2 and keep all inter-
nal parameters (ar,1, ar,2) unchanged, then the resulting
solution ˜u(x, t) would be related to the solution u(x, t)
with p0=3/2 as ˜u(x, t) = u(x, t).
Note 3 Internal parameters a3n,1and a3n,2(n=
1,2,···) do not affect solutions in Theorem 1, for rea-
sons which can be found in [21]. Thus, we will set them
as zero in later text.
The simplest solution of this class — the fundamental
rational soliton, is obtained when we set N1= 0 and
N2= 1 in Theorem 1. In this case, the solution has a
single real parameter a1,2, which can be normalized to
zero through a shift of the xaxis. The resulting solution,
for both p0=±3/2, is
u1(x, t) = ˆu1(x, t)ei[ 1
2(x+6t)1
8t],(20)
where
ˆu1(x, t) = 3ˆx2+ 3iˆx2
3ˆx2+ 1 ,(21)
and
ˆxx+33
4t(22)
is a moving coordinate. The graph of this solution is plot-
ted in Fig. 1. This solution is a rational soliton moving
on the constant-amplitude background (2) with velocity
33/4. Its 3D graph shows a W-shape along the ˆxdirec-
tion and has sometimes been called a W-shaped rational
soliton in the literature [8, 18]. Its height, i.e., max(|u1|),
is 2.
FIG. 1: Graph of the fundamental rational soliton |u1(x, t)|in
Eq. (20). Left: 3D plot. Right: density plot. The horizontal
axes are ˆx=x+ (33/4)t.
B. Generalized Okamoto polynomials
We will show in later text that rational solutions in
Theorem 1 contain partial-rogue waves but are not all
partial-rogue waves. The question of what solutions in
Theorem 1 are partial-rogue waves turns out to be closely
related to root properties of generalized Okamoto poly-
nomials. So, we will introduce these polynomials and
examine their root structures next.
Original Okamoto polynomials arose in Okamoto’s
study of rational solutions to the Painlev´e IV equation
[22]. He showed that a class of such rational solutions can
be expressed as the logarithmic derivative of certain spe-
cial polynomials, which are now called Okamoto polyno-
mials. These original polynomials were later generalized,
and the generalized Okamoto polynomials provide a more
complete set of rational solutions to the Painlev´e IV equa-
tion [20, 23–25]. In addition, determinant expressions for
the original and generalized Okamoto polynomials were
discovered [20, 23–25].
Let pj(z) be Schur polynomials defined by
X
j=0
pj(z)j= exp z+2,(23)
with pj(z)0 for j < 0. Then, generalized Okamoto
polynomials QN1, N2(z), with N1, N2being nonnegative
integers, are defined as
QN1, N2(z) = Wron[p2, p5,··· , p3N11, p1, p4,··· , p3N22],
(24)
or equivalently,
QN1, N2(z) =
p2p1··· p3N1N2
.
.
..
.
..
.
..
.
.
p3N11p3N12··· p2N1N2
p1p0··· p2N1N2
.
.
..
.
..
.
..
.
.
p3N22p3N23··· p2N2N11
,(25)
since p0
j+1(z) = pj(z) from the definition of pj(z) in
Eq. (23), where the prime represents differentiation. The
first few QN1, N2(z) polynomials are
Q1,0(z) = 1
2(z2+ 2),
Q2,0(z) = 1
80(z6+ 10z4+ 20z2+ 40),
Q0,1(z) = z,
Q1,1(z) = 1
2(z2+ 2)
Q2,1(z) = 1
20z(z420),
Q0,2(z) = 1
8(z4+ 4z24),
Q1,2(z) = 1
8(z4+ 4z2+ 4),
Q2,2(z) = 1
80(z6+ 10z420z2+ 40).
Note that our definition of generalized Okamoto polyno-
mials is different from that by Clarkson in Refs. [24, 25].
Denoting the Qm,n(z) polynomial introduced in [24, 25]
摘要:

Partial-roguewavesthatcomefromnowherebutleavewithatraceintheSasa-SatsumaequationBoYang1,JiankeYang21SchoolofMathematicsandStatistics,NingboUniversity,Ningbo315211,China2DepartmentofMathematicsandStatistics,UniversityofVermont,Burlington,VT05401,U.S.APartial-roguewaves,i.e.,wavesthat\comefromnowhereb...

展开>> 收起<<
Partial-rogue waves that come from nowhere but leave with a trace in the Sasa-Satsuma equation Bo Yang1 Jianke Yang2.pdf

共11页,预览3页

还剩页未读, 继续阅读

声明:本站为文档C2C交易模式,即用户上传的文档直接被用户下载,本站只是中间服务平台,本站所有文档下载所得的收益归上传人(含作者)所有。玖贝云文库仅提供信息存储空间,仅对用户上传内容的表现方式做保护处理,对上载内容本身不做任何修改或编辑。若文档所含内容侵犯了您的版权或隐私,请立即通知玖贝云文库,我们立即给予删除!
分类:图书资源 价格:10玖币 属性:11 页 大小:935.25KB 格式:PDF 时间:2025-05-02

开通VIP享超值会员特权

  • 多端同步记录
  • 高速下载文档
  • 免费文档工具
  • 分享文档赚钱
  • 每日登录抽奖
  • 优质衍生服务
/ 11
客服
关注