1: %-----------------------------------------------------------------------

2: % Beginning of article.tex

3: %-----------------------------------------------------------------------

4: %

5: % AMS-LaTeX 1.2 sample file for book proceedings, based on amsproc.cls.

6: %

7: % Replace amsproc by the documentclass for the target series, e.g. pspum-l.

8: %

9: \documentclass{conm-p-l}

10: 

11: \usepackage{graphics}

12: 

13: \newtheorem{theorem}{Theorem}[section]

14: \newtheorem{lemma}[theorem]{Lemma}

15: 

16: \theoremstyle{definition}

17: \newtheorem{definition}[theorem]{Definition}

18: \newtheorem{example}[theorem]{Example}

19: \newtheorem{xca}[theorem]{Exercise}

20: 

21: \theoremstyle{remark}

22: \newtheorem{remark}[theorem]{Remark}

23: 

24: \numberwithin{equation}{section}

25: 

26: %    Absolute value notation

27: \newcommand{\abs}[1]{\lvert#1\rvert}

28: 

29: %    Blank box placeholder for figures (to avoid requiring any

30: %    particular graphics capabilities for printing this document).

31: \newcommand{\blankbox}[2]{%

32:   \parbox{\columnwidth}{\centering

33: %    Set fboxsep to 0 so that the actual size of the box will match the

34: %    given measurements more closely.

35:     \setlength{\fboxsep}{0pt}%

36:     \fbox{\raisebox{0pt}[#2]{\hspace{#1}}}%

37:   }%

38: }

39: 

40: \newcommand{\LL}{{\mathcal L}}

41: \newcommand{\tla}{\tilde{\lambda}}

42: \newcommand{\im}{\mathop{\rm Im}\nolimits}

43: \newcommand{\hk}{\hat{k}}

44: \newcommand{\Z}{\ZZ^2/\{0\}}

45: \newcommand{\E}{{\mathcal E}}

46: \newcommand{\HH}{{\mathcal H}}

47: \newcommand{\cS}{{\mathcal S}}

48: \newcommand{\q}{\vec{q}}

49: \newcommand{\vq}{\vec{q}}

50: \newcommand{\p}{\vec{p}}

51: \newcommand{\CC}{{\mathbb C}}

52: \newcommand{\RR}{{\mathbb R}}

53: \newcommand{\ZZ}{{\mathbb Z}}

54: \newcommand{\PP}{{\mathbb P}}

55: \newcommand{\f}{\vec{f}}

56: \newcommand{\F}{{\mathcal F}}

57: \newcommand{\e}{\varepsilon}

58: \newcommand{\vth}{\vartheta}

59: \newcommand{\U}{{\mathcal U}}

60: \newcommand{\N}{{\mathcal N}}

61: \newcommand{\cq}{\tilde{q}}

62: \newcommand{\tDl}{\tilde{\Delta}}

63: \newcommand{\tQ}{\tilde{Q}}

64: \newcommand{\vQ}{\vec{Q}}

65: \newcommand{\A}{{\mathcal A}}

66: \newcommand{\B}{{\mathcal B}}

67: \newcommand{\C}{{\mathcal C}}

68: \renewcommand{\k}{\kappa}

69: \newcommand{\ga}{\gamma}

70: \newcommand{\Ga}{\Gamma}

71: \newcommand{\ve}{\bf{e}}

72: \newcommand{\we}{{\hat{{\bf{e}}}}}

73: \newcommand{\he}{\hat{e}}

74: \newcommand{\vv}{\vec{v}}

75: \newcommand{\hcS}{\hat{\cS}}

76: \newcommand{\tcS}{\tilde{\cS}}

77: \newcommand{\hS}{\hat{S}}

78: \newcommand{\tS}{\tilde{S}}

79: \newcommand{\hD}{\hat{D}}

80: \newcommand{\hJ}{\hat{J}}

81: \newcommand{\tD}{\tilde{D}}

82: \newcommand{\tC}{\tilde{C}}

83: \newcommand{\dl}{\delta}

84: \newcommand{\Dl}{\Delta}

85: \renewcommand{\th}{\theta}

86: \newcommand{\ra}{\rightarrow}

87: \newcommand{\al}{\alpha}

88: \newcommand{\be}{\beta}

89: \newcommand{\sg}{\sigma}

90: \newcommand{\Sg}{\Sigma}

91: \newcommand{\bM}{\bar{M}}

92: \newcommand{\pa}{\partial}

93: \newcommand{\z}{\zeta}

94: \newcommand{\hQ}{\hat{Q}}

95: \newcommand{\hv}{\hat{v}}

96: \newcommand{\hw}{\hat{w}}

97: \newcommand{\hx}{\hat{x}}

98: \newcommand{\bv}{\bar{v}}

99: \newcommand{\bw}{\bar{w}}

100: \newcommand{\La}{\Lambda}

101: \newcommand{\tLa}{\tilde{\Lambda}}

102: \newcommand{\la}{\lambda}

103: \newcommand{\bq}{\bar{q}}

104: \newcommand{\bp}{\bar{p}}

105: \newcommand{\bQ}{\bar{Q}}

106: \newcommand{\bE}{\bar{E}}

107: \newcommand{\nid}{\noindent}

108: \newcommand{\cF}{{\mathcal F}}

109: \newcommand{\rc}{S_\omega}

110: \newcommand{\hrc}{\hat{S}_\omega}

111: \newcommand{\bW}{\bar{W}}

112: \newcommand{\hN}{\hat{N}}

113: \newcommand{\hF}{\hat{F}}

114: \newcommand{\tF}{\tilde{F}}

115: \newcommand{\om}{\omega}

116: \newcommand{\Om}{\Omega}

117: \newcommand{\na}{\nabla}

118: \newcommand{\lag}{\langle}

119: \newcommand{\rag}{\rangle}

120: \newcommand{\tx}{\tilde{x}}

121: \newcommand{\tq}{\tilde{q}}

122: \newcommand{\tE}{\tilde{E}}

123: \newcommand{\cH}{{\mathcal H}}

124: \newcommand{\W}{{\mathcal W}}

125: 

126: 

127: %Commuting Diagram Need

128: \def\maprightu#1{\smash{

129:     \mathop{\longrightarrow}\limits^{#1}}}

130: \def\maprightd#1{\smash{

131:     \mathop{\longrightarrow}\limits_{#1}}}

132: \def\mapdownl#1{

133:     \llap{$\vcenter{\hbox{$\scriptstyle#1$}}$}\Big\downarrow}

134: \def\mapdownr#1{\Big\downarrow

135:     \rlap{$\vcenter{\hbox{$\scriptstyle#1$}}$}}

136: \def\mapupl#1{

137:     \llap{$\vcenter{\hbox{$\scriptstyle#1$}}$}\Big\uparrow}

138: \def\mapupr#1{\Big\uparrow

139:     \rlap{$\vcenter{\hbox{$\scriptstyle#1$}}$}}

140: 

141: 

142: 

143: \begin{document}

144: 

145: \title{Chaos in PDEs and Lax Pairs of Euler Equations}

146: 

147: 

148: %    Information for first author

149: \author{Yanguang (Charles)  Li}

150: %    Address of record for the research reported here

151: \address{Department of Mathematics, University of Missouri, 

152: Columbia, MO 65211}

153: %    Current address

154: \curraddr{}

155: \email{cli@math.missouri.edu}

156: %    \thanks will become a 1st page footnote.

157: \thanks{}

158: 

159: %    Information for second author

160: %\author{Author Two}

161: %\address{Mathematical Research Section, School of Mathematical Sciences,

162: %Australian National University, Canberra ACT 2601, Australia}

163: %\email{two@maths.univ.edu.au}

164: %\thanks{Support information for the second author.}

165: 

166: %    General info

167: \subjclass{Primary 35Q55, 35Q30; Secondary 37L10, 37L50, 35Q99}

168: \date{}

169: 

170: %\dedicatory{This paper is dedicated to our advisors.}

171: 

172: \keywords{Homoclinic orbits, chaos, Lax pairs, Darboux transformations, 

173: Euler equations.}

174: 

175: \begin{abstract}

176: Recently, the author and collaborators have developed a systematic 

177: program for proving the existence of homoclinic orbits in partial 

178: differential equations. Two typical forms of homoclinic orbits thus 

179: obtained are: (1). transversal homoclinic orbits, (2). Silnikov 

180: homoclinic orbits. Around the transversal homoclinic orbits in infinite 

181: dimensional autonomous systems, the author was able to prove the 

182: existence of chaos through a shadowing lemma. Around the Silnikov 

183: homoclinic orbits, the author was able to prove the existence of chaos 

184: through a horseshoe construction. 

185: 

186: Very recently, there has been a breakthrough by the author in finding 

187: Lax pairs for Euler equations of incompressible inviscid fluids. 

188: Further results have been obtained by the author and collaborators. 

189: \end{abstract}

190: 

191: \maketitle

192: 

193: %\section*{}

194: %This is an example of an unnumbered first-level heading.

195: 

196: %\specialsection*{This is a Special Section Head}

197: %This is an example of a special section head%

198: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

199: %\footnote{Here is an example of a footnote. Notice that this footnote

200: %text is running on so that it can stand as an example of how a footnote

201: %with separate paragraphs should be written.

202: %\par

203: %And here is the beginning of the second paragraph.}%

204: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

205: 

206: %\section{This is a numbered first-level section head}

207: %This is an example of a numbered first-level heading.

208: 

209: %\subsection{This is a numbered second-level section head}

210: %This is an example of a numbered second-level heading.

211: 

212: %\subsection*{This is an unnumbered second-level section head}

213: %This is an example of an unnumbered second-level heading.

214: 

215: %\subsubsection{This is a numbered third-level section head}

216: %This is an example of a numbered third-level heading.

217: 

218: %\subsubsection*{This is an unnumbered third-level section head}

219: %This is an example of an unnumbered third-level heading.

220: 

221: \section{Introduction}

222: 

223: Unlike chaos in finite dimensional systems, the area of chaos in partial 

224: differential equations has been a long standing open field.

225: During the past ten years, there have been some remarkable developments 

226: in the area of chaos in partial differential equations by the author 

227: and collaborators. Our main results can be briefly described as follows: 

228: A systematic program is developed for proving the existence of homoclinic 

229: orbits for perturbed soliton equations. The program involves machineries 

230: from integrable theory, dynamical systems, and partial differential equations.

231: The types of homoclinic orbits thus obtained can be either transversal 

232: or of Silnikov. With regard to transversal homoclinic orbits, shadowing 

233: lemmas are utilized or developed to prove the existence of chaos. With 

234: regard to Silnikov homoclinic orbits, Smale horseshoes are constructed 

235: to prove the existence of chaos. The main machineries from integrable theory 

236: are the isospectral theory and Darboux transformations. The main machineries 

237: from dynamical systems are persistence of invariant manifolds, Fenichel 

238: fibers, Melnikov measurement, and other measurements. The main machineries 

239: from partial differential equations are local well-posedness and regularity.

240: 

241: Another exciting development happened to Euler equations of incompressible 

242: inviscid fluids. Lax pairs for both 2D and 3D Euler equations was  

243: found by the author and Steve Childress. A Darboux transformation for the 

244: Lax pair of 2D Euler equation was also found by Artyom Yurov and the author. 

245: The philosophical significance of the existence of Lax pairs for Euler 

246: equations is fundamentally important. If one defines integrability of an 

247: equation by the existence of a Lax pair, then both 2D and 3D Euler equations 

248: are integrable. More importantly, both 2D and 3D Navier-Stokes equations 

249: at high Reynolds numbers are near integrable systems. Such a point of view 

250: changes our old ideology on Euler and Navier-Stokes equations.

251: 

252: The most important application of the theory on chaos in partial 

253: differential equations in theoretical physics will be on the study of 

254: turbulence. Our approach of studying turbulence is different from many 

255: other studies in which one starts with Stokes equation to prove results on

256: Navier-Stokes equations for small Reynolds number. In our studies, we 

257: start with Euler equations and view Navier-Stokes equations at large 

258: Reynolds number as singular perturbations of Euler equations. For this goal, 

259: we chose the 2D Navier-Stokes equations under periodic boundary conditions 

260: to begin a dynamical system study on 2D turbulence. We started with a 

261: simple fixed point of 2D Euler equation, and studied the linearized 2D 

262: Euler equation at the fixed point. A complete spectral theorem is obtained.

263: In particular, we found unstable eigenvalues. Then, naturally we are 

264: interested in the corresponding unstable manifold. Such an unstable manifold 

265: forms the prototype of the attractor of Navier-Stokes equations at large 

266: Reynolds number. And long term turbulence lives around the attractor. 

267: Finding the unstable manifold is not successful yet. Partial success is 

268: accomplished in finding the corresponding unstable manifold for some 

269: Galerkin truncation. Here the unstable manifold is the 2D surface of 

270: an ellipsoid. Stable and unstable manifolds together form a lip-shape 

271: configuration. Whether or not the stable and unstable manifolds for 2D Euler 

272: equation have the same topology as above is not known. In fact, the 

273: existence of stable and unstable manifolds for 2D Euler equation has not been 

274: proved. 

275: 

276: \section{Homoclinic Orbits}

277: 

278: In terms of proving the existence of a homoclinic orbit, 

279: the most common tool is the so-called Melnikov integral method \cite{Mel63} 

280: \cite{Arn64}. This method was subsequently developed by Holmes and Marsden 

281: \cite{GH83}, and most recently by Wiggins \cite{Wig88}. For partial 

282: differential equations, this method was mainly developed by Li et al.

283: \cite{LMSW96} \cite{Li01b} \cite{LM97} \cite{Li02f}.

284: 

285: There are two derivations for the Melnikov integrals. One is the 

286: so-called geometric argument \cite{GH83} \cite{Wig88} \cite{LM97} 

287: \cite{LMSW96} \cite{Li01b}.

288: The other is the so-called Liapunov-Schmitt argument \cite{CHMP80} 

289: \cite{CH82}. The Liapunov-Schmitt argument is a fixed-point type argument 

290: which directly leads to the existence of a homoclinic orbit. The condition 

291: for the existence of a fixed point is the Melnikov integral. 

292: The geometric argument is a signed distance argument which applies 

293: to more general situations than the Liapunov-Schmitt argument. It turns out 

294: that the geometric argument is a much more powerful machinary than the 

295: Liapunov-Schmitt argument. In particular, the geometric argument can handle 

296: geometric singular perturbation problems. I shall also mention an 

297: interesting derivation in \cite{Arn64}.

298: 

299: In establishing the existence of homoclinic orbits in high dimensions, one 

300: often needs other tools besides the Melnikov analysis. For example, when 

301: studying orbits 

302: homoclinic to fixed points created through resonances in 

303: ($n \geq 4$)-dimensional near-integrable systems, one often needs tools 

304: like Fenichel fibers, as presented in previous chapter, to set up geometric 

305: measurements for locating such 

306: homoclinic orbits. Such homoclinic orbits often have a geometric singular 

307: perturbation nature. In such cases, the Liapunov-Schmitt argument can not 

308: be applied. For such works on finite dimensional systems, see for 

309: example \cite{Kov92a} \cite{Kov92b} \cite{LM97}.

310: For such works on infinite dimensional systems, see for example 

311: \cite{LMSW96} \cite{Li01b}.

312: 

313: \subsection{Silnikov Homoclinic Orbits in Nonlinear 

314: Schr\"odinger (NLS) Equation  Under Regular 

315: Perturbations \label{horrnls}}

316: 

317: Consider the regularly perturbed nonlinear Schr\"odinger (NLS) equation 

318: \cite{LMSW96},

319: \begin{equation}

320: iq_t = q_{xx} +2 [ |q|^2 - \om^2] q +i \e [\hat{\pa}^2_xq - \al q +\be ] \ ,

321: \label{rpnls}

322: \end{equation}

323: where $q = q(t,x)$ is a complex-valued function of the two real 

324: variables $t$ and $x$, $t$ represents time, and $x$ represents

325: space. $q(t,x)$ is subject to periodic boundary condition of period 

326: $2 \pi$, and even constraint, i.e., 

327: \[

328: q(t,x + 2 \pi) = q(t,x)\ , \ \ q(t,-x) = q(t,x)\ .

329: \]

330: $\om$ is a positive constant, $\al >0$ and $\be >0$ are constants, 

331: $\hat{\pa}^2_x$ is a bounded Fourier multiplier,

332: \[

333: \hat{\pa}^2_x q = -\sum_{k=1}^{N}k^2 \xi_k \tq_k \cos kx\ ,

334: \]

335: $\xi_k = 1$ when $k \leq N$, $\xi_k = 8k^{-2}$ when $k>N$, for some 

336: fixed large $N$, and $\e > 0$ is the perturbation parameter.

337: The following theorem was proved in \cite{LMSW96}.

338: \begin{theorem}

339: There exists a $\e_0 > 0$, such that for any $\e \in (0, \e_0)$, there 

340: exists a codimension 1 surface in the external parameter space 

341: $(\alpha,\beta, \om) \in \RR^+\times \RR^+\times \RR^+$ where $\om \in 

342: (\frac{1}{2}, 1)$, and $\al \om < \be$. For any 

343: $(\alpha ,\beta, \omega)$ on the codimension 1

344: surface, the regularly perturbed nonlinear Schr\"odinger equation 

345: (\ref{rpnls}) possesses a symmetric pair of Silnikov homoclinic orbits 

346: asymptotic 

347: to a saddle $Q_\epsilon$. The codimension 1 surface has the approximate 

348: representation given by $\al = 1/\k(\om)$, where $\k(\om)$ is plotted 

349: in Figure \ref{kappa}.

350: \label{rhorbit}

351: \end{theorem}

352: \begin{figure}

353: \caption{The graph of $\k(\om)$.}

354: \label{kappa}

355: \end{figure}

356: To prove the theorem, one starts from the invariant plane 

357: \[

358: \Pi=\{ q\mid \ \partial_x q=0 \}.

359: \]

360: On $\Pi$, there is a saddle $Q_\e =\sqrt{I} e^{i\th}$ to which 

361: the symmetric pair of Silnikov homoclinic orbits will be asymptotic to, where

362: \begin{equation}

363: I=\omega^2-\epsilon \frac{1}{2\omega}\sqrt{\beta^2-\alpha^2\omega^2}+\cdots ,

364: \quad \cos \theta  =\frac{\alpha \sqrt{I}}{\beta}, \quad \theta \in 

365: (0,\frac{\pi}{2}).

366: \label{Qec}

367: \end{equation}

368: Its eigenvalues are

369: \begin{equation}

370: \la_n^\pm = -\e [\al +\xi_n n^2]\pm 2 \sqrt{(\frac{n^2}{2} +

371: \om^2-I)(3I -\om^2 -\frac{n^2}{2} )}\ , 

372: \label{Qev}

373: \end{equation}

374: where $n=0,1,2, \cdots $, $\om \in (\frac{1}{2}, 1)$, $\xi_n = 1$ when 

375: $n \leq N$, $\xi_n = 8n^{-2}$ when $n>N$, for some fixed large $N$,

376: and $I$ is given in 

377: (\ref{Qec}). The crucial points to notice are: (1). only $\la_0^+$ and 

378: $\la_1^+$ have positive real parts, $\mbox{Re}\{ \la_0^+\} < \mbox{Re}\{ 

379: \la_1^+\} $; (2). all the other eigenvalues have negative real parts among 

380: which the absolute value of $\mbox{Re}\{ \la_2^+\}=\mbox{Re}\{ \la_2^-\}$ 

381: is the smallest; (3). $|\mbox{Re}\{ \la_2^+\}| < \mbox{Re}\{ \la_0^+\}$. 

382: Actually, items (2) and (3) are the main characteristics of Silnikov 

383: homoclinic orbits. 

384: 

385: The unstable manifold $W^u(Q_\e)$ of $Q_\e$ has a fiber representation 

386: \cite{LMSW96}. The Melnikov measurement measures the 

387: signed distance between $W^u(Q_\e)$ and a center-stable manifold 

388: $W^{cs}_\e$ \cite{LMSW96}. By virtue of a Fenichel Fiber Theorem 

389: \cite{LMSW96}, one can show that, to the leading order in

390: $\e$, the signed distance is given by the Melnikov integral

391: \begin{eqnarray*}

392: M &=& \int^{+\infty}_{-\infty}\int^{2\pi}_0 

393:  [\partial_qF_1(q_0(t))(\hat{\partial}^2_xq_0(t)-\alpha

394: q_0(t)+\beta ) \\

395: & & \quad \quad + \partial_{\bar{q}}F_1(q_0(t))

396: (\hat{\partial}^2_x\overline{q_0(t)}-\alpha 

397: \overline{q_0(t)}+\beta )]dxdt,

398: \end{eqnarray*}

399: where $q_0(t)$ is a homoclinic orbit for the integrable NLS, which 

400: can be generated through Darboux transformations \cite{LM94};

401: and $\partial_qF_1$ and $\partial_{\bar{q}}F_1$ are Melnikov vectors 

402: which can be generated through Floquet discriminant in the isospectral 

403: theory of the integrable NLS \cite{LM94}. The zero of the 

404: signed distance implies the existence of an orbit in $W^u(Q_\e)\cap 

405: W^{cs}_\e$. The stable manifold $W^s(Q_\e)$ of $Q_\e$ is a codimension 

406: 1 submanifold in $W^{cs}_\e$. To locate a homoclinic orbit, one needs 

407: to set up a second measurement measuring the signed distance between 

408: the orbit in $W^u(Q_\e)\cap W^{cs}_\e$ and $W^s(Q_\e)$ inside $W^{cs}_\e$.

409: To set up this signed distance, first one can rather easily track 

410: the (perturbed) orbit by an unperturbed orbit to an $O (\e)$ neighborhood 

411: of $\Pi$, then one needs to prove the size of $W^s(Q_\e)$ to be $O (\e^\nu)$

412: ($\nu <1$) with normal form transform. To the leading order in

413: $\e$, the zero of the second signed distance is given by

414: \[

415: \be \cos \ga = \frac {\al \om (\Dl \ga )} {2 \sin \frac {\Dl \ga }{2}} \ ,

416: \]

417: where $\Dl \ga = -4 \vth_0$ and $\vth_0$ is a phase shift of $q_0(t)$.

418: To the leading order in $\e$, the common zero of the two second signed 

419: distances satisfies $\al = 1/\k(\om)$, where $\k(\om)$ is plotted 

420: in Figure \ref{kappa}. Then the claim of the theorem is proved by virtue 

421: of the implicit function theorem. For rigorous details, see \cite{LMSW96}.

422: 

423: \subsection{Silnikov Homoclinic Orbits in NLS Under Singular 

424: Perturbations \label{horsnls}}

425: 

426: Consider the singularly perturbed nonlinear Schr\"odinger equation 

427: \cite{Li01b},

428: \begin{equation}

429: iq_t = q_{xx} +2 [|q|^2 - \om^2] q +i \e [q_{xx} - \al q +\be ] \ ,

430: \label{spnls}

431: \end{equation}

432: where $q = q(t,x)$ is a complex-valued function of the two real 

433: variables $t$ and $x$, $t$ represents time, and $x$ represents

434: space. $q(t,x)$ is subject to periodic boundary condition of period 

435: $2 \pi$, and even constraint, i.e., 

436: \[

437: q(t,x + 2 \pi) = q(t,x)\ , \ \ q(t,-x) = q(t,x)\ .

438: \]

439: $\om \in (1/2, 1)$ is a positive constant, $\al >0$ and $\be >0$ 

440: are constants, and $\e > 0$ is the perturbation parameter. 

441: The following theorem was proved in \cite{Li01b}.

442: \begin{theorem}

443: There exists a $\e_0 > 0$, such that for any $\e \in (0, \e_0)$, there 

444: exists a codimension 1 surface in the external parameter space 

445: $(\alpha,\beta, \om) \in \RR^+\times  \RR^+\times 

446: \RR^+$ where $\om \in (\frac{1}{2}, 1)/S$, $S$ is a finite subset, and 

447: $\al \om < \be$. For any $(\alpha ,\beta, \omega)$ on the codimension 1

448: surface, the singularly perturbed nonlinear Schr\"odinger equation 

449: (\ref{spnls}) possesses a symmetric pair of Silnikov homoclinic orbits 

450: asymptotic 

451: to a saddle $Q_\epsilon$. The codimension 1 surface has the approximate 

452: representation given by $\al = 1/\k(\om)$, where $\k(\om)$ is plotted 

453: in Figure \ref{kappa}.

454: \label{shorbit}

455: \end{theorem}

456: The proof of the theorem is also completed 

457: through two measurements: the Melnikov measurement and the second measurement.

458: But more powerful machineries are needed \cite{Li01b}.

459: 

460: \subsection{Silnikov Homoclinic Orbits in Vector NLS Under Perturbations}

461: 

462: In recent years, novel results have been obtained on the solutions of 

463: the vector nonlinear Schr\"odinger equations \cite{AOT99} \cite{AOT00} 

464: \cite{YT01}. Abundant ordinary integrable results have been carried through 

465: \cite{WF00} \cite{FSW00}, including linear stability calculations 

466: \cite{FMMW00}. Specifically, the vector nonlinear 

467: Schr\"odinger equations can be written as

468: \begin{eqnarray*}

469: & & ip_t + p_{xx} + \frac{1}{2} (|p|^2 + \chi |q|^2) p = 0 , \\

470: & & iq_t + q_{xx} + \frac{1}{2} (\chi |p|^2 + |q|^2) q = 0 , 

471: \end{eqnarray*}

472: where $p$ and $q$ are complex valued functions of the two real variables 

473: $t$ and $x$, and $\chi$ is a positive constant. These equations describe

474: the evolution of two orthogonal pulse envelopes in birefringent optical 

475: fibers \cite{Men87} \cite{Men89}, with industrial applications in fiber 

476: communication systems \cite{HK95} and all-optical switching devices 

477: \cite{Isl92}. For linearly birefringent fibers \cite{Men87}, $\chi =2/3$.

478: For elliptically birefringent fibers, $\chi$ can take other positive 

479: values \cite{Men89}. When $\chi = 1$, these equations are first 

480: shown to be integrable by S. Manakov \cite{Man74}, and thus called Manakov 

481: equations. When $\chi$ is not 1 or 0, these equations are 

482: non-integrable. Propelled by the industrial applications, extensive 

483: mathematical studies on the vector nonlinear Schr\"odinger equations 

484: have been conducted. Like the scalar nonlinear Schr\"odinger equation, the 

485: vector nonlinear Schr\"odinger equations also possess figure eight 

486: structures in their phase space. Consider the singularly perturbed vector 

487: nonlinear Schr\"odinger equations,

488: \begin{eqnarray}

489: & & ip_t + p_{xx} + \frac{1}{2} [(|p|^2 + |q|^2)-\om^2] p = 

490: i \e [ p_{xx} -\al p - \be ]\ , \label{pnls1}\\

491: & & iq_t + q_{xx} + \frac{1}{2} [(|p|^2 + |q|^2)-\om^2] q = 

492: i \e [ q_{xx} -\al q - \be ]\  , \label{pnls2}

493: \end{eqnarray}

494: where $p(t,x)$ and $q(t,x)$ are subject to periodic boundary condition 

495: of period $2\pi$, and are even in $x$, i.e. 

496: \[

497: p(t,x + 2 \pi) = p(t,x)\ , \ \ p(t,-x) = p(t,x)\ , 

498: \]

499: \[

500: q(t,x + 2 \pi) = q(t,x)\ , \ \ q(t,-x) = q(t,x)\ , 

501: \]

502: $\om \in (1,2)$, $\al > 0$ and $\be$ are real constants, and 

503: $\e > 0$ is the perturbation parameter. We have

504: \begin{theorem}[\cite{Li02d}]

505: There exists a $\e_0 > 0$, such that 

506: for any $\e \in (0, \e_0)$, there exists 

507: a codimension 1 surface in the space of $(\alpha,\beta, \om) \in 

508: \RR^+\times \RR^+\times \RR^+$ where 

509: $\om \in (1, 2)/S$, $S$ is a finite subset, and 

510: $\al \om < \sqrt{2} \be$. For any $(\alpha ,\beta, \omega)$ on the 

511: codimension 1 surface, the singularly perturbed vector nonlinear 

512: Schr\"odinger equations (\ref{pnls1})-(\ref{pnls2}) possesses a 

513: homoclinic orbit asymptotic to a saddle

514: $Q_\epsilon$. This orbit is also the homoclinic orbit 

515: for the singularly perturbed scalar nonlinear Schr\"odinger equation

516: studied in last section, and is the only one asymptotic 

517: to the saddle $Q_\epsilon$ for the singularly perturbed 

518: vector nonlinear Schr\"odinger equations (\ref{pnls1})-(\ref{pnls2}).

519: The codimension 1 surface has the 

520: approximate representation given by $\al = 1/\k(\om)$, where $\k(\om)$ 

521: is plotted in Figure \ref{kappa}.

522: \end{theorem} 

523: 

524: \subsection{Silnikov Homoclinic Orbits in Discrete 

525: NLS Under Perturbations \label{hordnls}}

526: 

527: Consider the following perturbed discrete cubic nonlinear Schr\"odinger 

528: equations \cite{LM97}, 

529: \begin{eqnarray}

530: i\dot{q_n}&=&{1 \over h^2}\bigg[q_{n+1}-2q_n+q_{n-1}\bigg]+|q_n|^2(q_{n+1}+

531:            q_{n-1})-2\om^2 q_n \nonumber \\

532:         & &+i\e \bigg[-\al q_n +{1 \over h^2}(q_{n+1}-2q_n+q_{n-1})

533:            + \be \bigg], \label{PDNLS}

534: \end{eqnarray}

535: \nid

536: where $i=\sqrt{-1}$, $q_n$'s are complex variables,

537: \[

538: q_{n+N}=q_n, \ \ (\mbox{periodic}\  \mbox{condition}); \quad \mbox{and}\ 

539: q_{-n}=q_n, \ \ (\mbox{even}\  \mbox{condition});

540: \]

541: $h={1\over N}$, and 

542: \begin{eqnarray*}

543: & & N\tan{\pi \over N}< \om <N\tan{2\pi \over N},\ \ \mbox{for}\ N>3,\\

544: & & 3\tan{\pi \over 3}< \om < \infty, \ \ \mbox{for}\ N=3. \\

545: & & \e\in[0,\e_1),\ \alpha\ (>0), \ \be\ (>0) \ 

546: \mbox{are}\ \mbox{constants.}

547: \end{eqnarray*}

548: This is a $2(M+1)$ dimensional system, where

549: \[

550: M=N/2,\ \ (N\ \mbox{even}); \quad \mbox{and}\ 

551: M=(N-1)/2, \ \ (N\ \mbox{odd}).

552: \]

553: This system is a finite-difference discretization of the perturbed NLS 

554: (\ref{spnls}). The following theorem was proved in \cite{LM97}.

555: 

556: Denote by $\Sg_N\ (N\geq 7)$ the external parameter space,

557: \begin{eqnarray*}

558: \Sg_N&=&\bigg\{ (\om,\al,\be)\ \bigg | \ \om \in (N\tan{\pi \over N}, 

559:                N\tan{2\pi \over N}),\\ 

560:      & &\al\in (0,\al_0), \be\in (0,\be_0); \\

561:      & &\mbox{where}\ \al_0\ \mbox{and}\ \be_0\ \mbox{are}\ \mbox{any}

562:         \ \mbox{fixed}\ \mbox{positive}\ \mbox{numbers}. \bigg\}

563: \end{eqnarray*}

564: \begin{theorem}

565: For any $N$ ($7\leq N<\infty$), there exists a positive number $\e_0$, 

566: such that for any $\e \in (0,\e_0)$, there exists a 

567: codimension $1$ surface $E_\e$ in $\Sg_N$; for any

568: external parameters ($\om,\al,\be$) on $E_\e$, there exists a 

569: homoclinic orbit asymptotic to a saddle $Q_\e$.

570: The codimension $1$ surface $E_\e$ has the approximate expression

571: $\al=1/\k$, where $\k=\k(\om;N)$

572: is shown in Fig.\ref{nkappa}.

573: \label{dhorbit}

574: \end{theorem}

575: In the cases ($3\leq N \leq 6$), $\k$ is always negative. For 

576: $N \geq 7$, $\k$ 

577: can be positive as shown in Fig.\ref{nkappa}. When $N$ is even and 

578: $\geq 7$, there is in fact a symmetric pair of homoclinic orbits asymptotic to 

579: a fixed point $Q_\e$ at the same values of the external parameters; 

580: since for even $N$, we have the symmetry:

581: If $q_n=f(n,t)$ solves (\ref{PDNLS}),

582: then $q_n=f(n+N/2,t)$ also solves (\ref{PDNLS}). When $N$ is odd

583: and $\geq 7$, the study can not guarantee that two homoclinic orbits 

584: exist at the same value of the external parameters.

585: \begin{figure}

586: \caption{The graph of $\k(\om;N)$.}

587: \label{nkappa}

588: \end{figure}

589: 

590: \subsection{Comments on DSII Under Perturbations}

591: 

592: Consider the perturbed Davey-Stewartson II equations \cite{Li02b},

593: \begin{eqnarray*}

594: iq_t &=& \Upsilon q+ \bigg [2(|q|^2-\omega^2)+ u_y \bigg ]q

595: +i\epsilon f \ , \\

596: & & \ \ \Delta u = -4\partial_y |q|^2 \ ,

597: \end{eqnarray*}

598: where $q$ is a complex-valued function of the three variables ($t,x,y$), 

599: $u$ is a real-valued function of the three variables ($t,x,y$), 

600: $\Upsilon =\partial_{xx}-\partial_{yy}$, $\Delta=\partial_{xx}

601: +\partial_{yy}$, $\omega >0$ is a constant. We also consider periodic 

602: boundary conditions. The perturbation $f$ can be for example a singular 

603: perturbation 

604: \[

605: f = \Dl q - \al q + \be \ ,

606: \]

607: where $\alpha >0$, $\beta >0$ are constants, or a regular perturbation 

608: by mollifying $\Dl$ into a bounded Fourier multiplier

609: \[

610: \hat{\Dl} q = -\sum_{k \in Z^2} \be_k |k|^2 \tilde{q}_k \cos k_1 x 

611: \cos k_2 y \ ,

612: \]

613: in the case of periods ($2\pi , 2\pi$),

614: \[

615: \be_k = 1, \ \ |k| \leq N , \quad \be_k = |k|^{-2},\ \ |k| > N,

616: \]

617: for some large $N$, $|k|^2 = k_1^2 + k_2^2$.

618: 

619: Under both regular and singular 

620: perturbations, the rigorous Melnikov measurement can be established 

621: \cite{Li02b}. It turns out that only local well-posedness is necessary 

622: for rigorously setting up the Melnikov measurement, thanks to the fact 

623: that the unperturbed homoclinic orbits are 

624: classical solutions. The obstacle toward proving the existence of homoclinic 

625: orbits comes from a technical difficulty in solving a linear system 

626: to get the normal form for proving the size estimate of the stable 

627: manifold of a saddle. For details, see \cite{Li02b}.

628: 

629: \subsection{Transversal Homoclinic Orbits in a Periodically Perturbed 

630: Sine-Gordon Equation \label{PPSGE}}

631: 

632: Transversal homoclinic orbits in continuous systems often appear 

633: in two types of systems: (1). periodic systems where the Poincar\'e 

634: period map has a transversal homoclinic orbit, (2). autonomous 

635: systems where the homoclinic orbit is asymptotic to a hyperbolic 

636: limit cycle. 

637: 

638: Consider the periodically perturbed sine-Gordon (SG) equation,

639: \begin{equation}

640: u_{tt}=c^2 u_{xx}+\sin u+\epsilon [-a u_t+u^3 \chi(\| u\|)\cos t],

641: \label{PSG}

642: \end{equation} 

643: where

644: \[

645: \chi(\| u\|)=\left\{ \begin{array}{ll} 1, & \| 

646: u\| \leq M,\\ 0, & \| u\| \geq 2M,\end{array}\right.

647: \]

648: for $M<\| u\| <2M$, $\chi (\| u\|)$ is a smooth bump function, 

649: under odd periodic boundary condition,

650: \[

651: u(x+2\pi ,t)=u(x,t),\quad 

652: u(x,t)=-u(x,t),

653: \]

654: $\frac{1}{4}<c^2<1$, $a>0$, $\epsilon$ is a small 

655: perturbation parameter.

656: \begin{theorem}[\cite{LMSW96}, \cite{SZ00}] There exists an 

657: interval $I\subset \RR^{+}$ such that for any $a\in I$, there exists a

658: transversal homoclinic orbit

659: $u=\xi (x,t)$ asymptotic to $0$ in $H^{1}$.

660: \end{theorem}

661: 

662: \subsection{Transversal Homoclinic Orbits in a Derivative NLS \label{ADNS}}

663: 

664: Consider the derivative nonlinear Schr\"odinger equation,

665: \begin{equation}

666: i q_t = q_{xx} + 2 |q|^2 q +i \e \bigg [ (\frac{9}{16}-|q|^2 )q +\mu 

667: |\hat{\pa}_x q|^2 \bar{q} \bigg ]\ , \label{derNLS}

668: \end{equation}

669: where $q$ is a complex-valued function of two real variables $t$ and $x$,

670: $\e > 0$ is the perturbation parameter, $\mu$ is a real constant, and

671: $\hat{\pa}_x $ is a bounded Fourier multiplier,

672: \[

673: \hat{\pa}_x q = -\sum_{k=1}^K k \tq_k \sin kx\ , \quad 

674: \mbox{for} \ q = \sum_{k=0}^\infty \tq_k \cos kx\ ,

675: \]

676: and some fixed $K$. Periodic boundary condition 

677: and even constraint are imposed,

678: \[

679: q(t,x+2\pi ) = q(t,x)\ , \ \ q(t,-x)=q(t,x) \ . 

680: \] 

681: \begin{theorem}[\cite{Li02a}]

682: There exists a $\e_0 > 0$, such that 

683: for any $\e \in (0, \e_0)$, and $|\mu | > 5.8$,

684: there exist two transversal homoclinic orbits asymptotic to 

685: the limit cycle $q_c = \frac{3}{4} \exp \{ -i [ \frac{9}{8} t + \ga ]\}$.

686: \label{thmdns}

687: \end{theorem}

688: 

689: \section{Existence of Chaos}

690: 

691: The importance of homoclinic orbits with respect to chaotic dynamics was 

692: first realized by Poincar{\'e} \cite{Poi99}. In 1961, Smale constructed 

693: the well-known horseshoe in the neighborhood of a transversal homoclinic 

694: orbit \cite{Sma61} \cite{Sma65} \cite{Sma67}. In particular, 

695: Smale's theorem implies Birkhoff's theorem on the existence of a 

696: sequence of structurely stable periodic orbits in the neighborhood 

697: of a transversal homoclinic orbit \cite{Bir12}. In 1984 and 1988 \cite{Pal84}

698: \cite{Pal88}, Palmer gave a beautiful proof of Smale's theorem using a 

699: shadowing lemma. Later, this proof was generalized to infinite dimensions 

700: by Steinlein and Walther \cite{SW89} \cite{SW90} and Henry \cite{Hen94}. 

701: In 1967, Silnikov proved Smale's theorem for autonomous systems in finite 

702: dimensions using a fixed point argument \cite{Sil67b}. In 1996, Palmer proved 

703: Smale's theorem for autonomous systems in finite dimensions using shadowing 

704: lemma \cite{CKP95} \cite{Pal96}. In 2002, Li proved Smale's theorem for 

705: autonomous systems in infinite dimensions using shadowing lemma \cite{Li02a}.

706: 

707: For nontransversal homoclinic orbits, the most well-known type which leads 

708: to the existence of Smale horseshoes is the so-called Silnikov homoclinic 

709: orbit \cite{Sil65} \cite{Sil67a} \cite{Sil70} \cite{Den89} \cite{Den93}. 

710: Existence of Silnikov homoclinic orbits and new constructions of Smale 

711: horseshoes for concrete nonlinear wave systems have been established 

712: in finite dimensions \cite{LM97} \cite{LW97} and in infinite dimensions 

713: \cite{LMSW96} \cite{Li99a} \cite{Li01b}. 

714: 

715: \subsection{Shift Automorphism}

716: 

717: Let $\W$ be a set which consists of elements of the doubly infinite 

718: sequence form:

719: \[

720: a =(\cdot \cdot \cdot  a_{-2} a_{-1} a_0, a_1 a_2 \cdot \cdot \cdot ),

721: \]

722: where $a_k \in \{ l_1, \cdots, l_m\}$, m labels, and 

723: $k\in \ZZ$. We introduce a topology in $\W$

724: by taking as neighborhood basis of 

725: \[

726: a^* =( \cdot \cdot \cdot a^*_{-2} a^*_{-1} a^*_0, a^*_1 a^*_2 

727: \cdot \cdot \cdot ),

728: \]

729: the set 

730: \[

731: W_j =  \bigg \{ a\in \W \ \bigg | \ a_k=a^*_k\ (|k|<j) \bigg \}

732: \]

733: \nid

734: for $j=1,2,\cdot \cdot \cdot $. This makes $\W$ a topological space.

735: The shift automorphism $\chi$ is defined on $\W$ by

736: \begin{eqnarray*}

737: \chi &:& \W \mapsto \W, \\

738:   & & \forall a \in \W,\ \chi(a) = b,\ \mbox{where}\ b_k=a_{k+1}.

739: \end{eqnarray*}

740: The shift automorphism $\chi$ exhibits {\em{sensitive dependence on 

741: initial conditions}}, which is a hallmark of {\em{chaos}}.

742: 

743: \subsection{NLS Under Regular Perturbations

744: \label{shrnls}}

745: 

746: We continue from section \ref{horrnls} and consider the regularly perturbed 

747: nonlinear Schr\"odinger (NLS) equation (\ref{rpnls}). Starting from the 

748: Homoclinic Orbit Theorem \ref{rhorbit},

749: We have the following theorem on the existence of chaos. 

750: \begin{theorem}[Chaos Theorem, \cite{Li99a}]

751: Under certain generic assumptions for the 

752: perturbed nonlinear Schr\"odinger system (\ref{rpnls}), there exists a 

753: compact Cantor subset 

754: $\La$ of $H^1$ (the Sobolev space), $\La$ consists of points, and is 

755: invariant under a Poincar\'e map $P$.

756: $P$ restricted to $\La$, is topologically conjugate to the shift 

757: automorphism $\chi$ on four symbols $1, 2, -1, -2$. That is, there exists

758: a homeomorphism

759: \[

760: \phi \ : \ \W \mapsto  \La,

761: \]

762: \nid

763: such that the following diagram commutes:

764: \begin{equation} 

765: \begin{array}{ccc}

766: \W &\maprightu{\phi} & \Lambda\\

767: \mapdownl{\chi} & & \mapdownr{P}\\

768: \W & \maprightd{\phi} & \Lambda

769: \end{array} 

770: \nonumber

771: \end{equation}

772: \label{horserhthm}

773: \end{theorem}

774: Although the symmetric pair of Silnikov homoclinic orbits is not structurally 

775: stable, the Smale horseshoes are structurally stable. Thus, the Cantor 

776: sets and the conjugacy to the shift automorphism are structurally stable.

777: 

778: \subsection{NLS Under Singular Perturbations

779: \label{shsnls}}

780: 

781: We continue from section \ref{horsnls} and consider the singularly perturbed 

782: nonlinear Schr\"odinger (NLS) equation (\ref{spnls}). Starting from the 

783: Homoclinic Orbit Theorem \ref{shorbit}, we have

784: \begin{theorem}[Chaos Theorem, \cite{Li02c}]

785: Under certain generic assumptions for the 

786: perturbed nonlinear Schr\"odinger system (\ref{spnls}), there exists a 

787: compact Cantor subset 

788: $\La$ of $H^1$, $\La$ consists of points, and is invariant under a 

789: Poincar\'e map $P$.

790: $P$ restricted to $\La$, is topologically conjugate to the shift 

791: automorphism $\chi$ on four symbols $1, 2, -1, -2$. That is, there exists

792: a homeomorphism

793: \[

794: \phi \ : \ \W \mapsto  \La,

795: \]

796: \nid

797: such that the following diagram commutes:

798: \begin{equation} 

799: \begin{array}{ccc}

800: \W &\maprightu{\phi} & \Lambda\\

801: \mapdownl{\chi} & & \mapdownr{P}\\

802: \W & \maprightd{\phi} & \Lambda

803: \end{array} 

804: \nonumber

805: \end{equation}

806: \label{horseshthm}

807: \end{theorem}

808: 

809: \subsection{Discrete NLS Under Perturbations

810: \label{shdnls}}

811: 

812: We continue from section \ref{hordnls}. Starting from the 

813: Homoclinic Orbit Theorem \ref{dhorbit}, we have the theorem 

814: on the existence of chaos 

815: for the perturbed discrete nonlinear Schr\"odinger equation 

816: (\ref{PDNLS}). 

817: \begin{theorem}[Chaos Theorem, \cite{LW97}]

818: Under certain generic assumptions for the 

819: perturbed discrete nonlinear Schr\"odinger system (\ref{PDNLS}),

820: there exists a compact Cantor subset 

821: $\La$ of $\RR^{2(M+1)}$, $\La$ consists of points, and is invariant under 

822: a Poincar\'e map $P$.

823: $P$ restricted to $\La$, is topologically conjugate to the shift 

824: automorphism $\chi$ on two symbols $0, 1$. That is, there exists

825: a homeomorphism

826: \[

827: \phi \ : \ \W \mapsto  \La,

828: \]

829: \nid

830: such that the following diagram commutes:

831: \begin{equation} 

832: \begin{array}{ccc}

833: \W &\maprightu{\phi} & \Lambda\\

834: \mapdownl{\chi} & & \mapdownr{P}\\

835: \W & \maprightd{\phi} & \Lambda

836: \end{array} 

837: \nonumber

838: \end{equation}

839: \label{horsedhthm}

840: \end{theorem}

841: 

842: \subsection{Numerical Simulation of Chaos}

843: 

844: The finite-difference discretization of both the regularly and the singularly 

845: perturbed nonlinear Schr\"odinger equations (\ref{rpnls}) and (\ref{spnls})

846: leads to the same discrete perturbed nonlinear Schr\"odinger equation 

847: (\ref{PDNLS}).

848:  

849: In the chaotic regime, typical numerical output is shown in 

850: Figure \ref{numo}. Notice that there are two typical profiles at a fixed 

851: time: (1). a breather type profile with its hump located at the center of 

852: the spatial period interval, (2). a breather type profile with its hump 

853: located at the boundary (wing) of the spatial period interval. These two 

854: types of profiles are half spatial period translate of each other.

855: If we label the profiles, with their humps at the center of the spatial 

856: period interval, by ``C''; those profiles, with their humps at the wing

857: of the spatial period interval, by ``W'', then 

858: \begin{equation}

859: ``W\mbox{''}=\sg \circ ``C\mbox{''}, \label{CWJ}

860: \end{equation}

861: where $\sg$ is the symmetry group element representing half spatial period 

862: translate. The time series of the output in Figure \ref{numo} is a 

863: chaotic jumping between ``C'' and ``W'', which we call ``chaotic center-wing 

864: jumping''. We interpret the chaotic center-wing jumping as the 

865: numerical realization of the shift automorphism $\chi$ on 

866: symbols. We can make this more precise in terms of the 

867: phase space geometry. The figure-eight structure of the integrable NLS 

868: projected 

869: onto the plane of the Fourier component $\cos x$ is illustrated in 

870: Figure \ref{pfi}, and labeled by $L_C$ and $L_W$. From 

871: the symmetry, we know that

872: \[

873: L_W = \sg \circ L_C.

874: \]

875: $L_C$ has the spatial-temporal profile realization

876: as in Figure \ref{figlc} with the hump located at the center. $L_W$ 

877: corresponds to the half spatial period translate 

878: of the spatial-temporal profile realization as in Figure \ref{figlc},

879: with the hump located at the boundary (wing).

880: An orbit inside $L_C$, $L_{Cin}$ has a spatial-temporal profile realization as

881: in Figure \ref{figlcin}. The half period translate of 

882: $L_{Cin}$, $L_{Win}$ is 

883: inside $L_W$. An orbit outside $L_C$ and $L_W$, $L_{out}$ has the

884: spatial-temporal profile realization as in Figure \ref{figlout}.

885: $S_l$ and $S_{l,\sg}$ are two phase blocks.

886: From these figures, we can see clearly that 

887: the chaotic center-wing jumping 

888: (Figure \ref{numo}) is the realization of the shift automorphism on 

889: symbols in $S_l \cup S_{l,\sg}$.

890: \begin{figure}

891: \caption{A chaotic solution in the discrete perturbed NLS 

892: system (\ref{PDNLS}).}

893: \label{numo}

894: \end{figure}

895: \begin{figure}

896: \caption{Figure-eight structure and its correspondence with the chaotic

897: center-wing jumping.}

898: \label{pfi}

899: \end{figure}

900: \begin{figure}

901: \caption{Spatial-temporal profile realization of $L_C$ in 

902: Figure \ref{pfi} (coordinates are the same as in Figure \ref{numo}).}

903: \label{figlc}

904: \end{figure}

905: \begin{figure}

906: \caption{Spatial-temporal profile realization of $L_{Cin}$ in 

907: Figure \ref{pfi} (coordinates are the same as in Figure \ref{numo}).}

908: \label{figlcin}

909: \end{figure}

910: \begin{figure}

911: \caption{Spatial-temporal profile realization of $L_{out}$ in 

912: Figure \ref{pfi} (coordinates are the same as in Figure \ref{numo}).}

913: \label{figlout}

914: \end{figure}

915: 

916: \subsection{Shadowing Lemma and Chaos in Finite-D Periodic 

917: Systems \label{Palmer}}

918: 

919: Since its invention \cite{Ano67}, shadowing lemma has been a useful tool

920: for solving many dynamical system problems. Here we only focus upon its 

921: use in proving the existence of chaos in a neighborhood of a transversal 

922: homoclinic orbit. According to the type of the system, the level of 

923: difficulty in proving the existence of chaos with a shadowing lemma

924: is different. 

925: 

926: A finite-dimensional periodic system can be written in the 

927: general form

928: \[

929: \dot{x} = F(x,t)\ ,

930: \]

931: where $x \in \RR^n$, and $F(x,t)$ is periodic in $t$. Let $f$ be the 

932: Poincar\'e period map.

933: \begin{definition}

934: A doubly infinite sequence $\{ y_j \}$ in $\RR^n$ is a $\dl$ pseudo-orbit 

935: of a $C^1$ diffeomorphism $f : \RR^n \mapsto \RR^n$ if for all integers $j$

936: \[

937: |y_{j+1}-f(y_j)| \leq \dl \ .

938: \]

939: An orbit $\{ f^j(x) \}$ is said to $\e$-shadow the $\dl$ pseudo-orbit 

940: $\{ y_j \}$ if for all integers $j$

941: \[

942: |f^j(x)-y_j| \leq \e \ .

943: \]

944: \end{definition}

945: \begin{definition}

946: A compact invariant set $S$ is hyperbolic if there are positive 

947: constants $K$, $\al$ and a projection matrix valued function $P(x)$, 

948: $x \in S$, of constant rank such that for all $x$ in $S$

949: \[

950: P(f(x))Df(x) = Df(x)P(x)\ ,

951: \]

952: \[

953: |Df^j(x)P(x)| \leq Ke^{-\al j}\ ,\quad (j \geq 0 )\ , 

954: \]

955: \[

956: |Df^j(x)(I-P(x))| \leq Ke^{\al j}\ ,\quad (j \leq 0 )\ . 

957: \] 

958: \end{definition}

959: \begin{theorem}[Shadowing Lemma \cite{Pal88}]

960: Let $S$ be a compact hyperbolic set for the $C^1$ diffeomorphism 

961: $f : \RR^n \mapsto \RR^n$. Then given $\e >0$ sufficiently small there 

962: exists $\dl >0$ such that every $\dl$ pseudo-orbit in $S$ has a 

963: unique $\e$-shadowing orbit.

964: \label{shal}

965: \end{theorem}

966: The proof of this theorem by Palmer \cite{Pal88} is overall a 

967: fixed point argument with the help of Green functions for linear maps.

968: 

969: Let $y_0$ be a transversal homoclinic point asymptotic to a saddle $x_0$ 

970: of a $C^1$ diffeomorphism $f : \RR^n \mapsto \RR^n$. Then the set

971: \[

972: S= \{ x_0 \} \cup \{ f^j(y_0): j \in Z\}

973: \]

974: is hyperbolic. Denote by $A_0$ and $A_1$ the two orbit segments of 

975: length $2m+1$

976: \[

977: A_0= \{ x_0,x_0, \cdots, x_0 \}\ , \quad 

978: A_1= \{ f^{-m}(y_0), f^{-m+1}(y_0), \cdots, f^{m-1}(y_0), f^{m}(y_0) \} \ .

979: \]

980: Let 

981: \[

982: a=(\cdots, a_{-1},a_0, a_1, \cdots ) \ , 

983: \]

984: where $a_j \in \{ 0,1\}$, be any doubly infinite binary sequence. Let 

985: $A$ be the doubly infinite sequence of points in $S$, associated with 

986: $a$

987: \[

988: A=\{ \cdots, A_{a_{-1}},A_{a_0}, A_{a_1}, \cdots \} \ .

989: \]

990: When $m$ is sufficiently large, $A$ is a $\dl$ pseudo-orbit in $S$.

991: By the shadowing lemma (Theorem \ref{shal}), there is a unique 

992: $\e$-shadowing orbit that shadows $A$. In this manner, Palmer \cite{Pal88}

993: gave a beautiful proof of Smale's horseshoe theorem.

994: \begin{definition}

995: Denote by $\Sg$ the set of doubly infinite binary sequences

996: \[

997: a=(\cdots, a_{-1},a_0, a_1, \cdots ) \ , 

998: \]

999: where $a_j \in \{ 0,1\}$. We give the set $\{ 0,1\}$ the discrete topology 

1000: and $\Sg$ the product topology. The Bernoulli shift $\chi$ is defined by

1001: \[

1002: [\chi(a)]_j = a_{j+1}\ .

1003: \]

1004: \end{definition}

1005: \begin{theorem}

1006: Let $y_0$ be a transversal homoclinic point asymptotic to a saddle $x_0$ 

1007: of a $C^1$ diffeomorphism $f : \RR^n \mapsto \RR^n$. Then there is a 

1008: homeomorphism $\phi$ of $\Sg$ onto a compact subset of $\RR^n$ which is 

1009: invariant under $f$ and such that when $m$ is sufficiently large

1010: \[

1011: f^{2m+1} \circ \phi = \phi \circ \chi\ ,

1012: \]

1013: that is, the action of $f^{2m+1}$ on $\phi(\Sg)$ is topologically conjugate 

1014: to the action of $\chi$ on $\Sg$.

1015: \end{theorem}

1016: Here one can define $\phi(a)$ to be the point on the shadowing orbit that 

1017: shadows the midpoint of the orbit segment $A_{a_0}$, which is either $x_0$ 

1018: or $y_0$. The topological conjugacy can be easily verified. For details, 

1019: see \cite{Pal88}.

1020: Other references can be found in \cite{Pal84} \cite{Zen95}.

1021: 

1022: \subsection{Shadowing Lemma and Chaos in Infinite-D Periodic Systems}

1023: 

1024: An infinite-dimensional periodic system defined in a Banach space $X$ 

1025: can be written in the general form

1026: \[

1027: \dot{x} = F(x,t)\ ,

1028: \]

1029: where $x \in X$, and $F(x,t)$ is periodic in $t$. Let $f$ be the 

1030: Poincar\'e period map. When $f$ is a $C^1$ map which needs not to be 

1031: invertible, shadowing lemma and symbolic dynamics around a transversal 

1032: homoclinic orbit can both be established \cite{SW89} \cite{SW90} \cite{Hen94}.

1033: Other references can be found in \cite{HL86} \cite{CLP89} \cite{Zen97} 

1034: \cite{Bla86}. There exists also a work on horseshoe construction without

1035: shadowing lemma for 

1036: sinusoidally forced vibrations of buckled beam \cite{HM81}.

1037: 

1038: \subsection{Periodically Perturbed Sine-Gordon (SG) Equation}

1039: 

1040: We continue from section \ref{PPSGE}, and use the notations in  

1041: section \ref{Palmer}. For the periodically perturbed sine-Gordon 

1042: equation (\ref{PSG}), the Poincar\'e period map is a $C^1$ diffeomorphism 

1043: in $H^1$. As a corollary of the result in last section, we have the theorem 

1044: on the existence of chaos.

1045: \begin{theorem} There is an integer $m$ and a homeomorphism $\phi$ 

1046: of $\Sigma$ onto a compact Cantor subset $\Lambda $ of

1047: $H^{1}$. $\Lambda$ is invariant under the Poincar\'e period-$2\pi$ map $P$ of 

1048: the periodically perturbed sine-Gordon equation (\ref{PSG}). The 

1049: action of $P^{2m+1}$ on $\Lambda$ is topologically conjugate to

1050: the action of $\chi$ on

1051: $\Sigma: P^{2m+1} \circ \phi =\phi \circ

1052: \chi$. That is, the following diagram commutes:

1053: \[

1054: \begin{array}{ccc}

1055: \Sg &\maprightu{\phi} & \Lambda\\

1056: \mapdownl{\chi} & & \mapdownr{P^{2m+1}}\\

1057: \Sg & \maprightd{\phi} & \Lambda

1058: \end{array} 

1059: \]

1060: \end{theorem}

1061: 

1062: \subsection{Shadowing Lemma and Chaos in Finite-D Autonomous Systems}

1063: 

1064: A finite-dimensional autonomous system can be written in the 

1065: general form

1066: \[

1067: \dot{x} = F(x)\ ,

1068: \]

1069: where $x \in \RR^n$. In this case, a transversal homoclinic orbit 

1070: can be an orbit asymptotic to a normally hyperbolic limit cycle.

1071: That is, it is an orbit in the intersection of the stable and 

1072: unstable manifolds of a normally hyperbolic limit cycle. Instead of 

1073: the Poincar\'e period map as for periodic system, one may want 

1074: to introduce the so-called Poincar\'e return map which is a map induced by 

1075: the flow on a codimension 1 section which is transversal to the limit 

1076: cycle. Unfortunately, such a map is not even well-defined in the 

1077: neighborhood of the homoclinic orbit. This poses a challenging 

1078: difficulty in extending the arguments as in the case of a Poincar\'e 

1079: period map. In 1996, Palmer \cite{Pal96} completed a proof of a 

1080: shadowing lemma and existence of chaos using Newton's method. It will 

1081: be difficult to extend this method to infinite dimensions, since it 

1082: used heavily differentiations in time. 

1083: Other references can be found in \cite{CKP95} \cite{CKP97} \cite{Sil67b} 

1084: \cite{FS77}.

1085: 

1086: \subsection{Shadowing Lemma and Chaos in Infinite-D Autonomous Systems}

1087: 

1088: An infinite-dimensional autonomous system defined in a Banach space $X$ 

1089: can be written in the general form

1090: \[

1091: \dot{x} = F(x)\ ,

1092: \]

1093: where $x \in X$. In 2002, the author \cite{Li02a} completed a proof of a 

1094: shadowing lemma and existence of chaos using Fenichel's persistence of 

1095: normally hyperbolic invariant manifold idea. The setup is as follows,

1096: \begin{itemize}

1097: \item {\bf Assumption (A1):} There exist a hyperbolic limit cycle

1098: $S$ and a transversal homoclinic orbit $\xi$ 

1099: asymptotic to $S$. As curves, $S$ and $\xi$ are $C^{3}$.

1100: \item {\bf Assumption (A2):} The Fenichel fiber theorem is valid at $S$. 

1101: That is, there exist a family of unstable Fenichel fibers

1102: $\{ {\mathcal F}^{u}(q): \  q \in S\}$ and a family of stable Fenichel 

1103: fibers $\{ {\mathcal F}^{s}(q): \  q\in S\}$. For each fixed $q\in S$,

1104: ${\mathcal F}^{u}(q)$ and ${\mathcal F}^{s}(q)$ are $C^{3}$ submanifolds.

1105: ${\mathcal F}^{u}(q)$ and ${\mathcal F}^{s}(q)$ are $C^{2}$ in $q,

1106: \forall q\in S$. The unions $\bigcup_{q\in S}{\mathcal F}^{u}(q)$ and

1107: $\bigcup_{q\in S}{\mathcal F}^{s}(q)$ are the unstable and stable 

1108: manifolds of $S$. Both families are invariant, i.e.

1109: \[

1110: F^{t}({\mathcal F}^{u}(q))\subset

1111: {\mathcal F}^{u}(F^{t}(q)),  \forall\ t \leq 0, q\in S,

1112: \]

1113: \[

1114: F^{t}({\mathcal F}^{s}(q))\subset {\mathcal F}^{s}(F^{t}(q)), 

1115: \forall\ t \geq 0,  q \in S, 

1116: \]

1117: where $F^{t}$ is the evolution operator. There are positive constants

1118: $\k$ and $\widehat{C}$ such that $\forall q\in S$, $\forall 

1119: q^{-}\in {\mathcal F}^{u}(q)$ and $\forall q^{+}\in

1120: {\mathcal F}^{s}(q)$,

1121: \[

1122: \| F^{t}(q^{-})-F^{t}(q)\| \leq

1123: \widehat{C}e^{\k t}\| q^{-}-q\|,  \forall \ t \leq 0\ ,

1124: \]

1125: \[

1126: \| F^{t}(q^{+})-F^{t}(q)\| \leq \widehat{C}e^{-\k t}\| q^{+}-q\|, 

1127: \forall \ t \geq 0\ .

1128: \]

1129: \item {\bf Assumption (A3):} $F^{t}(q)$ is $C^{0}$ in $t$, for

1130: $t\in (-\infty ,\infty)$, $q\in X$. For any fixed $t\in 

1131: (-\infty ,\infty )$, $F^{t}(q)$ is a $C^{2}$ diffeomorphism on

1132: $X$.

1133: \end{itemize}

1134: \begin{remark}

1135: Notice that we do not assume that as functions of time, $S$ and $\xi$ 

1136: are $C^3$ , and we only assume that as curves, $S$ and $\xi$ 

1137: are $C^3$.

1138: \end{remark}

1139: 

1140: Under the above setup, a shadowing lemma and existence of chaos can be

1141: proved \cite{Li02a}. Another crucial element in the argument is the

1142: establishment of a $\la$-lemma (also called inclination lemma) \cite{Li02a}. 

1143: 

1144: \subsection{A Derivative Nonlinear Schr\"odinger Equation}

1145: 

1146: We continue from section \ref{ADNS}, and consider the derivative nonlinear 

1147: Schr\"odinger equation (\ref{derNLS}). The transversal homoclinic orbit 

1148: given in Theorem \ref{thmdns} is a classical solution. Thus, Assumption (A1) 

1149: is valid. Assumption (A2) follows from the standard arguments 

1150: in \cite{LW97} \cite{LMSW96} \cite{Li01b}. Since the perturbation 

1151: in (\ref{derNLS}) is bounded, Assumption (A3) follows from standard 

1152: arguments. Thus there exists chaos in the derivative nonlinear 

1153: Schr\"odinger equation (\ref{derNLS}) \cite{Li02a}.

1154: 

1155: \section{Lax Pairs of Euler Equations of Inviscid Fluids}

1156: 

1157: The governing equations for the incompressible viscous fluid flow are the 

1158: Navier-Stokes equations. Turbulence occurs in the regime of high Reynolds 

1159: number. By formally setting the Reynolds number equal to infinity, the 

1160: Navier-Stokes equations reduce to the Euler equations of incompressible 

1161: inviscid fluid flow. One may view the Navier-Stokes equations with large 

1162: Reynolds number as a singular perturbation of the Euler equations.

1163: 

1164: Results of T. Kato show that 2D Navier-Stokes equations are globally 

1165: well-posed in $C^0([0, \infty); H^s(R^2)), \ s>2$, and for any 

1166: $0 < T < \infty$, the mild solutions of the 2D Navier-Stokes equations

1167: approach those of the 2D Euler equations in $C^0([0, T]; H^s(R^2))$ 

1168: \cite{Kat86}. 3D Navier-Stokes equations are locally well-posed in 

1169: $C^0([0, \tau]; H^s(R^3)), \ s>5/2$, and the mild solutions of the 3D 

1170: Navier-Stokes equations approach those of the 3D Euler equations in 

1171: $C^0([0, \tau]; H^s(R^3))$, where $\tau$ depends on the norms of the initial 

1172: data and the external force \cite{Kat72} \cite{Kat75}. Extensive studies on 

1173: the inviscid limit have been carried by J. Wu et al. \cite{Wu96} 

1174: \cite{CW96} \cite{Wu98} \cite{BW99}. There is no doubt that mathematical 

1175: study on Navier-Stokes (Euler) equations is one of the most important 

1176: mathematical problems. In fact, Clay Mathematics Institute has posted the 

1177: global well-posedness of 3D Navier-Stokes equations as one of the one 

1178: million dollars problems.

1179: 

1180: V. Arnold \cite{Arn66} realized that 2D Euler equations are a Hamiltonian 

1181: system. Recently, the author found Lax pair structures for Euler 

1182: equations \cite{Li01a} \cite{LY01} \cite{Li02e} \cite{Li02f} \cite{LS02}.

1183: Understanding the structures of solutions to Euler equations is of 

1184: fundamental interest. Of particular interest is the question on the 

1185: global well-posedness of 3D Navier-Stokes and Euler equations. Our number 

1186: one hope is that the Lax pair structures can be useful in investigating 

1187: the global well-posedness. Our secondary hope is that the Darboux 

1188: transformation \cite{LY01} associated with the Lax pair can generate 

1189: explicit representation of homoclinic structures \cite{Li00a}. 

1190: 

1191: The philosophical significance of the existence of Lax pairs for Euler 

1192: equations is even more important. If one defines integrability of an equation 

1193: by the existence of a Lax pair, then both 2D and 3D Euler equations 

1194: are integrable. More importantly, both 2D and 3D Navier-Stokes equations 

1195: at high Reynolds numbers are singularly perturbed integrable systems. 

1196: Such a point of view changes our old ideology on Euler and Navier-Stokes 

1197: equations.

1198: 

1199: \subsection{A Lax Pair for 2D Euler Equation}

1200: 

1201: The 2D Euler equation can be written in the vorticity form,

1202: \begin{equation}

1203: \pa_t \Om + \{ \Psi, \Om \} = 0 \ ,

1204: \label{euler}

1205: \end{equation}

1206: where the bracket $\{\ ,\ \}$ is defined as

1207: \[

1208: \{ f, g\} = (\pa_x f) (\pa_y g) - (\pa_y f) (\pa_x g) \ ,

1209: \]

1210: $\Om$ is the vorticity, and $\Psi$ is the stream function given by,

1211: \[

1212: u=- \pa_y \Psi \ ,\ \ \ v=\pa_x \Psi \ ,

1213: \]

1214: and the relation between vorticity $\Om$ and stream 

1215: function $\Psi$ is,

1216: \[

1217: \Om =\pa_x v - \pa_y u =\Dl \Psi \ .

1218: \]

1219: \begin{theorem}[Li, \cite{Li01a}]

1220: The Lax pair of the 2D Euler equation (\ref{euler}) is given as

1221: \begin{equation}

1222: \left \{ \begin{array}{l} 

1223: L \varphi = \la \varphi \ ,

1224: \\

1225: \pa_t \varphi + A \varphi = 0 \ ,

1226: \end{array} \right.

1227: \label{laxpair}

1228: \end{equation}

1229: where

1230: \[

1231: L \varphi = \{ \Om, \varphi \}\ , \ \ \ A \varphi = \{ \Psi, \varphi \}\ ,

1232: \]

1233: and $\la$ is a complex constant, and $\varphi$ is a complex-valued function.

1234: \label{2dlp}

1235: \end{theorem}

1236: 

1237: \subsection{A Darboux Transformation for 2D Euler Equation}

1238: 

1239: Consider the Lax pair (\ref{laxpair}) at $\la =0$, i.e.

1240: \begin{eqnarray}

1241: & & \{ \Om, p \} = 0 \ , \label{d1} \\

1242: & & \pa_t p + \{ \Psi, p \} = 0 \ , \label{d2} 

1243: \end{eqnarray}

1244: where we replaced the notation $\varphi$ by $p$.

1245: \begin{theorem}[\cite{LY01}]

1246: Let $f = f(t,x,y)$ be any fixed solution to the system 

1247: (\ref{d1}, \ref{d2}), we define the Gauge transform $G_f$:

1248: \begin{equation}

1249: \tilde{p} = G_f p = \frac {1}{\Om_x}[p_x -(\pa_x \ln f)p]\ ,

1250: \label{gauge}

1251: \end{equation}

1252: and the transforms of the potentials $\Om$ and $\Psi$:

1253: \begin{equation}

1254: \tilde{\Psi} = \Psi + F\ , \ \ \ \tilde{\Om} = \Om + \Dl F \ ,

1255: \label{ptl}

1256: \end{equation}

1257: where $F$ is subject to the constraints

1258: \begin{equation}

1259: \{ \Om, \Dl F \} = 0 \ , \ \ \ \{ \Dl F, F \} = 0\ .

1260: \label{constraint}

1261: \end{equation}

1262: Then $\tilde{p}$ solves the system (\ref{d1}, \ref{d2}) at 

1263: $(\tilde{\Om}, \tilde{\Psi})$. Thus (\ref{gauge}) and 

1264: (\ref{ptl}) form the Darboux transformation for the 2D 

1265: Euler equation (\ref{euler}) and its Lax pair (\ref{d1}, \ref{d2}).

1266: \label{dt}

1267: \end{theorem}

1268: \begin{remark}

1269: For KdV equation and many other soliton equations, the 

1270: Gauge transform is of the form \cite{MS91},

1271: \[

1272: \tilde{p} =  p_x -(\pa_x \ln f)p \ .

1273: \]

1274: In general, Gauge transform does not involve potentials.

1275: For 2D Euler equation, a potential factor $\frac {1}{\Om_x}$

1276: is needed. From (\ref{d1}), one has

1277: \[

1278: \frac{p_x}{\Om_x} = \frac{p_y}{\Om_y} \ .

1279: \]

1280: The Gauge transform (\ref{gauge}) can be rewritten as

1281: \[

1282: \tilde{p} = \frac{p_x}{\Om_x} - \frac{f_x}{\Om_x} \frac{p}{f}

1283: =\frac{p_y}{\Om_y} - \frac{f_y}{\Om_y} \frac{p}{f}\ .

1284: \]

1285: The Lax pair (\ref{d1}, \ref{d2}) has a symmetry, i.e. it is 

1286: invariant under the transform $(t,x,y) \ra (-t,y,x)$. The form 

1287: of the Gauge transform (\ref{gauge}) resulted from the inclusion 

1288: of the potential factor $\frac {1}{\Om_x}$, is consistent with 

1289: this symmetry.

1290: \end{remark}

1291: Our hope is to use the Darboux transformation to generate homoclinic 

1292: structures for 2D Euler equation \cite{Li00a}.

1293: 

1294: 

1295: \subsection{A Lax Pair for Rossby Wave Equation}

1296: 

1297: The Rossby wave equation is

1298: \[

1299: \pa_t \Om + \{ \Psi , \Om \} + \be \pa_x \Psi = 0 \ ,

1300: \]

1301: where $\Om = \Om (t,x,y)$ is the vorticity, 

1302: $\{ \Psi , \Om \} = \Psi_x \Om_y - \Psi_y \Om_x $, 

1303: and $\Psi = \Dl^{-1} \Om$ 

1304: is the stream function. Its Lax pair can be obtained

1305: by formally conducting the transformation, $\Om = \tilde{\Om} +\be y$,

1306: to the 2D Euler equation \cite{Li01a},

1307: \[

1308: \{ \Om , \varphi \} - \be \pa_x \varphi = \la \varphi \ ,

1309: \quad \pa_t \varphi + \{ \Psi , \varphi \} = 0 \ ,

1310: \]

1311: where $\varphi$ is a complex-valued function, and $\la$ is 

1312: a complex parameter. 

1313: 

1314: \subsection{Lax Pairs for 3D Euler Equation}

1315: 

1316: The 3D Euler equation can be written in vorticity form,

1317: \begin{equation}

1318: \pa_t \Om + (u \cdot \na) \Om - (\Om \cdot \na) u = 0 \ ,

1319: \label{3deuler}

1320: \end{equation}

1321: where $u = (u_1, u_2, u_3)$ is the velocity, $\Om = (\Om_1, \Om_2, \Om_3)$

1322: is the vorticity, $\na = (\pa_x, \pa_y, \pa_z)$, 

1323: $\Om = \na \times u$, and $\na \cdot u = 0$. $u$ can be 

1324: represented by $\Om$ for example through Biot-Savart law.

1325: \begin{theorem}

1326: The Lax pair of the 3D Euler equation (\ref{3deuler}) is given as

1327: \begin{equation}

1328: \left \{ \begin{array}{l} 

1329: L \phi = \la \phi \ ,

1330: \\

1331: \pa_t \phi + A \phi = 0 \ ,

1332: \end{array} \right.

1333: \label{alaxpair}

1334: \end{equation}

1335: where

1336: \[

1337: L \phi = (\Om \cdot \na )\phi \ , 

1338: \ \ \ A \varphi = (u \cdot \na )\phi \ , 

1339: \]

1340: $\la$ is a complex constant, and $\phi$ is a complex scalar-valued function.

1341: \end{theorem}

1342: \begin{theorem}[\cite{Chi00}]

1343: Another Lax pair of the 3D Euler equation (\ref{3deuler}) is given as

1344: \begin{equation}

1345: \left \{ \begin{array}{l} 

1346: L \varphi = \la \varphi \ ,

1347: \\

1348: \pa_t \varphi + A \varphi = 0 \ ,

1349: \end{array} \right.

1350: \label{3dlaxpair}

1351: \end{equation}

1352: where

1353: \[

1354: L \varphi = (\Om \cdot \na )\varphi - (\varphi \cdot \na )\Om \ , 

1355: \ \ \ A \varphi = (u \cdot \na )\varphi - (\varphi \cdot \na ) u \ , 

1356: \]

1357: $\la$ is a complex constant, and $\varphi = (\varphi_1, \varphi_2, 

1358: \varphi_3)$ is a complex 3-vector valued function.

1359: \end{theorem}

1360: Our hope is that the infinitely many conservation laws generated by $\la 

1361: \in \CC$ can provide a priori estimates for the global well-posedness of 

1362: 3D Navier-Stokes equations, or better understanding on the global 

1363: well-posedness \cite{LS02}. For more informations on the topics, 

1364: see \cite{LY01} \cite{LS02}.

1365: 

1366: \section{Dynamical System Studies on 2D Euler Equation}

1367: 

1368: \subsection{Linearized 2D Euler Equation at a Fixed Point}

1369: 

1370: To begin an infinite dimensional dynamical system study, we investigate 

1371: the linearized 2D Euler equation at a fixed point \cite{Li00b}

1372: \cite{Li02e} \cite{Li02e}. We consider the 

1373: 2D Euler equation (\ref{euler}) under periodic boundary condition in 

1374: both $x$ and $y$ directions with period $2\pi$. Expanding $\Om$ into 

1375: Fourier series,

1376: \[

1377: \Om =\sum_{k\in \Z} \om_k \ e^{ik\cdot X}\ ,

1378: \]

1379: where $\om_{-k}=\overline{\om_k}\ $, $k=(k_1,k_2)^T$, 

1380: and $X=(x,y)^T$. The 2D Euler equation

1381: can be rewritten as 

1382: \begin{equation}

1383: \dot{\om}_k = \sum_{k=p+q} A(p,q) \ \om_p \om_q \ ,

1384: \label{Keuler}

1385: \end{equation}

1386: where $A(p,q)$ is given by,

1387: \begin{eqnarray}

1388: A(p,q)&=& {1\over 2}[|q|^{-2}-|p|^{-2}](p_1 q_2 -p_2 q_1) \nonumber \\

1389: \label{Af} \\      

1390:       &=& {1\over 2}[|q|^{-2}-|p|^{-2}]\left | \begin{array}{lr} 

1391: p_1 & q_1 \\ p_2 & q_2 \\ \end{array} \right | \ , \nonumber

1392: \end{eqnarray}

1393: where $|q|^2 =q_1^2 +q_2^2$ for $q=(q_1,q_2)^T$, similarly for $p$.

1394: 

1395: Denote $\{ \om_k \}_{k\in \Z}$ by $\om$. We consider the simple fixed point 

1396: $\om^*$ \cite{Li00b}:

1397: \begin{equation}

1398: \om^*_p = \Ga,\ \ \ \om^*_k = 0 ,\ \mbox{if} \ k \neq p \ \mbox{or}\ -p,

1399: \label{fixpt}

1400: \end{equation}

1401: of the 2D Euler equation (\ref{Keuler}), where 

1402: $\Ga$ is an arbitrary complex constant. 

1403: The {\em{linearized two-dimensional Euler equation}} at $\om^*$ is given by,

1404: \begin{equation}

1405: \dot{\om}_k = A(p,k-p)\ \Ga \ \om_{k-p} + A(-p,k+p)\ \bar{\Ga}\ \om_{k+p}\ .

1406: \label{LE}

1407: \end{equation}

1408: \begin{definition}[Classes]

1409: For any $\hk \in \Z$, we define the class $\Sg_{\hk}$ to be the subset of 

1410: $\Z$:

1411: \[

1412: \Sg_{\hk} = \bigg \{ \hk + n p \in \Z \ \bigg | \ n \in \ZZ, \ 

1413: \ p \ \mbox{is specified in (\ref{fixpt})} \bigg \}.

1414: \]

1415: \label{classify}

1416: \end{definition}

1417: \nid

1418: See Fig.\ref{class} for an illustration of the classes. 

1419: According to the classification 

1420: defined in Definition \ref{classify}, the linearized two-dimensional Euler 

1421: equation (\ref{LE}) decouples into infinitely many {\em{invariant subsystems}}:

1422: \begin{eqnarray}

1423: \dot{\omega}_{\hat{k} + np} &=& A(p, \hat{k} + (n-1) p) 

1424:      \ \Gamma \ \omega_{\hat{k} + (n-1) p} \nonumber \\

1425: & & + \ A(-p, \hat{k} + (n+1)p)\ 

1426:      \bar{\Gamma} \ \omega_{\hat{k} +(n+1)p}\ . \label{CLE}

1427: \end{eqnarray}

1428: \begin{figure}[ht]

1429:   \begin{center}

1430:     \leavevmode

1431:       \setlength{\unitlength}{2ex}

1432:   \begin{picture}(36,27.8)(-18,-12)

1433: %%  \begin{picture}(40,310)(-20,-150)

1434: %%%    \linethickness{5pt}

1435:     \thinlines

1436: \multiput(-12,-11.5)(2,0){13}{\line(0,1){23}}

1437: \multiput(-16,-10)(0,2){11}{\line(1,0){32}}

1438: %

1439:     \thicklines

1440: \put(0,-14){\vector(0,1){28}}

1441: \put(-18,0){\vector(1,0){36}}

1442: %

1443: \put(0,15){\makebox(0,0){$k_2$}}

1444: \put(18.5,0){\makebox(0,0)[l]{$k_1$}}

1445: \qbezier(-5.5,0)(-5.275,5.275)(0,5.5)

1446: \qbezier(0,5.5)(5.275,5.275)(5.5,0)

1447: \qbezier(5.5,0)(5.275,-5.275)(0,-5.5)

1448: \qbezier(0,-5.5)(-5.275,-5.275)(-5.5,0)

1449: %

1450:     \thinlines

1451: \put(4,4){\circle*{0.5}}

1452: \put(0,0){\vector(1,1){3.7}}

1453: \put(4.35,4.35){$p$}

1454: %

1455: \put(4,-4){\circle*{0.5}}

1456: %

1457: \put(8,0){\circle*{0.5}}

1458: \put(-8,0){\circle*{0.5}}

1459: \put(-8,-2){\circle*{0.5}}

1460: \put(-12,-4){\circle*{0.5}}

1461: \put(-12,-6){\circle*{0.5}}

1462: \put(-4,2){\circle*{0.5}}

1463: \put(-4,4){\circle*{0.5}}

1464: \put(0,6){\circle*{0.5}}

1465: \put(0,8){\circle*{0.5}}

1466: \put(4,10){\circle*{0.5}}

1467: \put(12,4){\circle*{0.5}}

1468: \put(0,-8){\circle*{0.5}}

1469: %

1470: \put(-4,-12){\line(1,1){17.5}}

1471: \put(-13.5,-7.5){\line(1,1){19.5}}

1472: \put(-13.5,-5.5){\line(1,1){17.5}}

1473: %

1474: \put(-3.6,1.3){$\hat{k}$}

1475: %

1476: \put(-7,12.1){\makebox(0,0)[b]{$(-p_2, p_1)^T$}}

1477: %\put(-7.85,11.75){\vector(1,-2){3.65}}

1478: \put(-6.7,12){\vector(1,-3){2.55}}

1479: %

1480: \put(6.5,13.6){\makebox(0,0)[l]{$\Sg_{\hat{k}}$}}

1481: \put(6.4,13.5){\vector(-2,-3){2.0}}

1482: %

1483: \put(7,-12.1){\makebox(0,0)[t]{$(p_2, -p_1)^T$}}

1484: \put(6.7,-12.25){\vector(-1,3){2.62}}

1485: %

1486: \put(-4.4,-13.6){\makebox(0,0)[r]{$\bar{D}_{|p|}$}}

1487: \put(-4.85,-12.55){\vector(1,3){2.45}}

1488: %

1489: \end{picture}

1490: \end{center}

1491: \caption{An illustration of the classes $\Sg_{\hk}$ and the disk 

1492: $\bar{D}_{|p|}$.}

1493: \label{class}

1494: \end{figure}

1495: Let $\LL_{\hk}$ be the linear operator defined by the right hand side of 

1496: (\ref{CLE}), and $H^s$ be the Sobolev space where $s \geq 0$ is an integer and 

1497: $H^0=\ell_2$. 

1498: \begin{theorem}[\cite{Li00b} \cite{Li02m}]

1499: The eigenvalues of the linear operator $\LL_{\hk}$ in $H^s$ are of 

1500: four types: real pairs ($c, -c$), purely imaginary pairs ($id, -id$), 

1501: quadruples ($\pm c \pm id$), and zero eigenvalues.

1502: \end{theorem}

1503: \begin{definition}[The Disk]

1504: The open disk of radius $|p|$ in $\Z$, denoted by $D_{|p|}$, is defined as

1505: \[

1506:   D_{|p|} = \bigg \{ k \in \Z \ \bigg| 

1507:       \ |k| < |p| \bigg \} \, ,

1508: \]

1509: and the closure of $D_{|p|}$, denoted by

1510: $\bar{D}_{|p|}$, is defined as

1511: \[ 

1512:  \bar{D}_{|p|} = \bigg \{ k \in \Z \ \bigg| 

1513:      \ |k| \leq |p| \bigg \} \, .

1514: \]

1515: \end{definition}

1516: \begin{theorem}[The Spectral Theorem, \cite{Li00b} \cite{Li02m}] We have 

1517: the following claims on the spectrum of the linear operator $\LL_{\hk}$:

1518: \begin{enumerate}

1519: \item If $\Sg_{\hat{k}} \cap \bar{D}_{|p|} = \emptyset$, then the entire

1520: $H^s$ spectrum of the linear operator $\LL_{\hk}$ 

1521: is its continuous spectrum. See Figure \ref{splb}, where

1522: $b= - \frac{1}{2}|\Gamma | |p|^{-2} 

1523: \left|

1524:   \begin{array}{cc}

1525: p_1 & \hat{k}_1 \\

1526: p_2 & \hat{k}_2

1527:   \end{array}

1528: \right| \ .$

1529: That is, both the residual and the point spectra of $\LL_{\hk}$ are empty.

1530: \item If $\Sg_{\hat{k}} \cap \bar{D}_{|p|} \neq \emptyset$, then the entire

1531: essential $H^s$ spectrum of the linear operator $\LL_{\hk}$ is its 

1532: continuous spectrum. 

1533: That is, the residual 

1534: spectrum of $\LL_{\hk}$ is empty. The point 

1535: spectrum of $\LL_{\hk}$ is symmetric with respect to both real and 

1536: imaginary axes. 

1537: See Figure \ref{spla2}.

1538: \end{enumerate}

1539: \label{SST}

1540: \end{theorem}

1541: \begin{figure}[ht]

1542:   \begin{center}

1543:     \leavevmode

1544:       \setlength{\unitlength}{2ex}

1545:   \begin{picture}(36,27.8)(-18,-12)

1546:     \thicklines

1547: \put(0,-14){\vector(0,1){28}}

1548: \put(-18,0){\vector(1,0){36}}

1549: \put(0,15){\makebox(0,0){$\Im \{ \la \}$}}

1550: \put(18.5,0){\makebox(0,0)[l]{$\Re \{ \la \}$}}

1551: \put(0.1,-7){\line(0,1){14}}

1552: \put(.2,-.2){\makebox(0,0)[tl]{$0$}}

1553: \put(-0.2,-7){\line(1,0){0.4}}

1554: \put(-0.2,7){\line(1,0){0.4}}

1555: \put(2.0,-6.4){\makebox(0,0)[t]{$-i2|b|$}}

1556: \put(2.0,7.6){\makebox(0,0)[t]{$i2|b|$}}

1557: \end{picture}

1558:   \end{center}

1559: \caption{The spectrum of $\LL_{\hk}$ in case (1).}

1560: \label{splb}

1561: \end{figure}

1562: \begin{figure}[ht]

1563:   \begin{center}

1564:     \leavevmode

1565:       \setlength{\unitlength}{2ex}

1566:   \begin{picture}(36,27.8)(-18,-12)

1567:     \thicklines

1568: \put(0,-14){\vector(0,1){28}}

1569: \put(-18,0){\vector(1,0){36}}

1570: \put(0,15){\makebox(0,0){$\Im \{ \la \}$}}

1571: \put(18.5,0){\makebox(0,0)[l]{$\Re \{ \la \}$}}

1572: \put(0.1,-7){\line(0,1){14}}

1573: \put(.2,-.2){\makebox(0,0)[tl]{$0$}}

1574: \put(-0.2,-7){\line(1,0){0.4}}

1575: \put(-0.2,7){\line(1,0){0.4}}

1576: \put(2.0,-6.4){\makebox(0,0)[t]{$-i2|b|$}}

1577: \put(2.0,7.6){\makebox(0,0)[t]{$i2|b|$}}

1578: \put(2.4,3.5){\circle*{0.5}}

1579: \put(-2.4,3.5){\circle*{0.5}}

1580: \put(2.4,-3.5){\circle*{0.5}}

1581: \put(-2.4,-3.5){\circle*{0.5}}

1582: \put(5,4){\circle*{0.5}}

1583: \put(-5,4){\circle*{0.5}}

1584: \put(5,-4){\circle*{0.5}}

1585: \put(-5,-4){\circle*{0.5}}

1586: \put(8,6){\circle*{0.5}}

1587: \put(-8,6){\circle*{0.5}}

1588: \put(8,-6){\circle*{0.5}}

1589: \put(-8,-6){\circle*{0.5}}

1590: \end{picture}

1591:   \end{center}

1592: \caption{The spectrum of $\LL_{\hk}$ in case (2).}

1593: \label{spla2}

1594: \end{figure}

1595: \nid

1596: Denote by $L$ the right hand side of (\ref{LE}), i.e. the whole 

1597: linearized 2D Euler operator, the spectral mapping theorem 

1598: holds.

1599: \begin{theorem}[\cite{LLM01}]

1600: $$\sigma(e^{tL})=e^{t\sigma(L)}, t\neq 0.$$

1601: \end{theorem}

1602: \nid

1603: Let $\z$ denote the number of points $q \in

1604: \Z$ that belong to the open disk of radius $|p|$, $D_{|p|}$, 

1605: and such that $q$ is not parallel to $p$. 

1606: \begin{theorem}[\cite{LLM01}]

1607: The number of nonimaginary eigenvalues of $L$ (counting the multiplicities) 

1608: does not exceed $2\z$.

1609: \end{theorem}

1610: \nid

1611: Another interesting discussion upon the discrete spectrum can be found in 

1612: \cite{Fad71}.

1613: 

1614: Since the introduction of continued fractions for calculating 

1615: the eigenvalues of steady fluid flow, by Meshalkin and Sinai \cite{MS61},

1616: this topics had been extensively explored \cite{Yud65} \cite{Liu92a} 

1617: \cite{Liu92b} \cite{Liu93} \cite{Liu94a} \cite{Liu95} \cite{BFY99} 

1618: \cite{Li00b}. Rigorous justification on the continued fraction calculation 

1619: was given in \cite{Li00b} \cite{Liu95}.

1620: As an example, we take $p=(1,1)^T$. When $\Ga \neq 0$, the fixed point 

1621: has $4$ eigenvalues which form a 

1622: quadruple. These four eigenvalues appear in the invariant 

1623: linear subsystem labeled by $\hk = (-3,-2)^T$. One of them is \cite{Li00b}:

1624: \begin{equation}

1625: \tla=2 \lambda / | \Gamma | = 0.24822302478255 \ + \ i \ 0.35172076526520\ .

1626: \label{evun}

1627: \end{equation}

1628: See Figure \ref{figev} for an illustration. The essential spectrum 

1629: (= continuous spectrum) of $\LL_{\hk}$ with $\hk = (-3,-2)^T$ is the segment 

1630: on the imaginary axis shown in Figure \ref{figev}, where 

1631: $b = -\frac{1}{4} \Ga$. The essential spectrum 

1632: (= continuous spectrum) of the linear 2D Euler operator at this fixed point 

1633: is the entire imaginary axis. 

1634: \begin{figure}[ht]

1635:   \begin{center}

1636:     \leavevmode

1637:       \setlength{\unitlength}{2ex}

1638:   \begin{picture}(36,27.8)(-18,-12)

1639:     \thicklines

1640: \put(0,-14){\vector(0,1){28}}

1641: \put(-18,0){\vector(1,0){36}}

1642: \put(0,15){\makebox(0,0){$\Im \{ \la \}$}}

1643: \put(18.5,0){\makebox(0,0)[l]{$\Re \{ \la \}$}}

1644: \put(2.4,3.5){\circle*{0.5}}

1645: \put(-2.4,3.5){\circle*{0.5}}

1646: \put(2.4,-3.5){\circle*{0.5}}

1647: \put(-2.4,-3.5){\circle*{0.5}}  

1648: \put(0.1,-10){\line(0,1){20}}

1649: \put(.2,-.2){\makebox(0,0)[tl]{$0$}}

1650: \put(-0.2,-10){\line(1,0){0.4}}

1651: \put(-0.2,10){\line(1,0){0.4}}

1652: \put(2.0,-9.4){\makebox(0,0)[t]{$-i2|b|$}}

1653: \put(2.0,10.6){\makebox(0,0)[t]{$i2|b|$}}

1654: \end{picture}

1655:   \end{center}

1656: \caption{The spectrum of $\LL_{\hk}$ with $\hk = (-3,-2)^T$, when $p=(1,1)^T$.}

1657: \label{figev}

1658: \end{figure}

1659: A rather well-known open problem is proving the existence of unstable, stable,

1660: and center manifolds. The main difficulty comes from the fact that the 

1661: nonlinear term is non-Lipschitzian.

1662: 

1663: \subsection{Models}

1664: 

1665: To simplify our study, we study only the case when $\om_k$ is real, $\forall 

1666: k \in \Z$, i.e. we only study the cosine transform of the vorticity, 

1667: \[

1668: \Om = \sum_{k \in \Z} \om_k \cos (k \cdot X)\ .

1669: \]

1670: To further simplify our study, we will study a concrete 

1671: dashed-line model based upon the line of fixed points (\ref{fixpt}) with the 

1672: mode $p=(1,1)^T$ parametrized by $\Ga$, and the only unstable invariant 

1673: linear subsystem labeled by $\hk = (-3,-2)^T$. The so-called 

1674: {\em{dashed-line model}} is given by \cite{Li01d},

1675: \begin{eqnarray}

1676: \dot{\omega}_n &=& \epsilon_{n-1} A_{n-1} \omega_p

1677: \omega_{n-1} - \epsilon_{n+1} A_{n+1} \omega_p \omega_{n+1} \ , 

1678: \nonumber \\ 

1679: \label{rdlm} \\

1680: \dot{\omega}_p &=& - \sum_{n \in Z} \epsilon_n \epsilon_{n-1}

1681: A_{n-1,n} \omega_{n-1} \omega_n \, , \nonumber

1682: \end{eqnarray}

1683: where

1684: \begin{equation}

1685:   \omega_n = \omega_{\hat{k}+np} \, , \ \  

1686:   A_n = A(p,\hat{k}+np) \, , \ \  

1687:   A_{m,n} = A(\hat{k}+mp,\hat{k}+np) \, ,  

1688: \label{abbn}

1689: \end{equation}

1690: \[

1691: \epsilon_n = \left\{

1692:     \begin{array}{ll}

1693:       1 \, , & \ \ \hbox{if } n \ne 5j \, , \, \forall j \in Z \ ,  \\

1694: \epsilon \, , & \ \ \hbox{if } n = 5j \, , \, 

1695:       \hbox{ for some }j \in Z \, .

1696:     \end{array} \right.

1697: \]

1698: The model is designed to model the hyperbolic structure of the 2D Euler 

1699: equation, connected to 

1700: the line of fixed points (\ref{fixpt}) with $p=(1,1)^T$. Figure \ref{model}

1701: illustrates the collocation of the modes in this model, which has the 

1702: {\em{``dashed-line''}} nature leading to the name of the model. 

1703: \begin{figure}[ht]

1704:   \begin{center}

1705:     \leavevmode

1706:       \setlength{\unitlength}{2ex}

1707:   \begin{picture}(36,27.8)(-18,-12)

1708:     \thinlines

1709: \multiput(-12,-11.5)(2,0){13}{\line(0,1){23}}

1710: \multiput(-16,-10)(0,2){11}{\line(1,0){32}}

1711:     \thicklines

1712: \put(0,-14){\vector(0,1){28}}

1713: \put(-18,0){\vector(1,0){36}}

1714: \put(0,15){\makebox(0,0){$k_2$}}

1715: \put(18.5,0){\makebox(0,0)[l]{$k_1$}}

1716: \qbezier(-2.75,0)(-2.6375,2.6375)(0,2.75)

1717: \qbezier(0,2.75)(2.6375,2.6375)(2.75,0)

1718: \qbezier(2.75,0)(2.6375,-2.6375)(0,-2.75)

1719: \qbezier(0,-2.75)(-2.6375,-2.6375)(-2.75,0)

1720:     \thinlines

1721: \put(2,2){\circle*{0.5}}

1722: \put(0,0){\vector(1,1){1.85}}

1723: \put(2.275,2.275){$p$}

1724: \put(-12,-10){\circle*{0.5}}

1725: \put(-10,-8){\circle*{0.5}}

1726: \put(-8,-6){\circle*{0.5}}

1727: \put(-6,-4){\circle{0.5}}

1728: \put(-4,-2){\circle*{0.5}}

1729: \put(-2,0){\circle*{0.5}}

1730: \put(0,2){\circle*{0.5}}

1731: \put(2,4){\circle*{0.5}}

1732: \put(4,6){\circle{0.5}}

1733: \put(6,8){\circle*{0.5}}

1734: \put(8,10){\circle*{0.5}}

1735: \put(-14,-12){\line(1,1){24}}

1736: \put(-5.6,-5.4){$\hat{k}$}

1737: \put(-4.4,-13.6){\makebox(0,0)[r]{$\bar{D}_{|p|}$}}

1738: \put(-4.85,-12.55){\vector(1,3){3.4}}

1739: \end{picture}

1740: \end{center}

1741: \caption{The collocation of the modes in the dashed-line model.}

1742: \label{model}

1743: \end{figure}

1744: \begin{figure}

1745: \caption{The unstable and stable manifolds of the fixed point $\om^*$ 

1746: (\ref{fixpt}): (a) and (b) show two ``painted eggs'' and (c) shows a 

1747: ``lip''.}

1748: \label{het}

1749: \end{figure}

1750: When $\e =0$, one can determine the stable and

1751: unstable manifolds of the fixed point $\omega^*$:

1752: %

1753: \begin{equation}

1754: \omega_p =  \Gamma \, , \, \omega_n=0 \quad (n \in \ZZ) \, .

1755: \label{nfixpt}

1756: \end{equation}

1757: %

1758: Using the polar coordinates:

1759: %

1760: \begin{displaymath}

1761:   \omega_1 = r \cos \theta \, , \, 

1762:   \omega_4 = r \sin \theta \, ; \, 

1763:   \omega_2 = \rho \cos \vartheta \, , \, 

1764:   \omega_3 = \rho \sin \vartheta \, ;

1765: \end{displaymath}

1766: %

1767: we have the following explicit expressions for the stable and

1768: unstable manifolds of the fixed point $\om^*$ and its

1769: negative $-\om^*$ represented through {\em{perverted 

1770: heteroclinic orbits}}:

1771: %

1772: \begin{eqnarray}

1773: \omega_p &=& \Gamma \ \tanh \tau \, , \nonumber \\

1774: r &=& \sqrt{ \frac{A_2}{A_2-A_1}}\, \ \Gamma \ \mbox{sech}\ \tau \, , \nonumber  \\[1ex]

1775: \theta &=& - \ \frac{A_2}{2\k} \ \mbox{ln}\ \cosh \tau + \theta_0 \, , 

1776: \label{exus}\\[1ex]

1777: \rho &=& \sqrt{\frac{-A_1}{A_2}} \  r \, ,\nonumber  \\[1ex]

1778: \theta + \vartheta &=& \left\{

1779:   \begin{array}{ll}

1780:     - \arcsin \left[ \frac{1}{2} \sqrt{\frac{A_2}{-A_1}}\,  \right] 

1781:         \ , & (\k>0) \ , \\[2ex]

1782: \pi + \arcsin  \left[ \frac{1}{2} \sqrt{\frac{A_2}{-A_1}} \right]

1783: \, , & (\k<0) \, ,

1784:   \end{array} \right.  \nonumber 

1785: \end{eqnarray}

1786: %

1787: where $\tau = \k \Gamma t 

1788: + \tau_0$, $(\tau_0, \theta_0)$ are the two parameters

1789: parametrizing the two-dimensional stable (unstable) manifold,

1790: and

1791: %

1792: \begin{displaymath}

1793:   \k = \sqrt{-A_1 A_2} \cos (\theta + \vartheta) 

1794:    = \pm \sqrt{-A_1 A_2} \sqrt{1+ \frac{A_2}{4A_1}} \ .

1795: \end{displaymath}

1796: %

1797: The two auxilliary variables $\om_0$ and $\om_5$ have the

1798: expressions:

1799: %

1800: \begin{eqnarray*}

1801:   \omega_0 &=& \frac{\alpha \beta}{1+ \beta^2} \ \mbox{sech}\ \tau \left\{ 

1802:             \sin [ \beta \ \mbox{ln}\ \cosh \tau +  \theta_0]

1803:             - \frac{1}{\beta} \cos

1804:             [ \beta \ \mbox{ln}\ \cosh \tau + \theta_0 ] \right\} \,

1805:    , \\

1806: %

1807:   \omega_5 &=& \frac{\alpha \beta}{1+ \beta^2} \ \mbox{sech}\ \tau

1808:   \left\{ \cos [ \beta \ \mbox{ln}\ \cosh \tau + \theta_0 ]

1809:     + \frac{1}{\beta} \sin [ \beta \ \mbox{ln}\ \cosh \tau + \theta_0]

1810:     \right\} \, ,

1811: \end{eqnarray*}

1812: %

1813: where

1814: %

1815: \begin{displaymath}

1816:   \alpha = -A_1 \Gamma \k^{-1} \sqrt{\frac{A_2}{A_2-A_1}} \ , \ \ 

1817:   \beta = - \frac{A_2}{2\k} \ .

1818: \end{displaymath}

1819: The explicit expression (\ref{exus}) of the perverted heteroclinic orbits

1820: shows that the unstable manifold $W^u(\om^*)$ 

1821: of the fixed 

1822: point $\om^*$ is the same as the stable 

1823: manifold $W^s(\om^*)$ of the negative $-\om^*$ of 

1824: $\om^*$, and the 

1825: stable manifold of the fixed point $\om^*$ 

1826: is the same as the unstable manifold of the negative $-\om^*$ 

1827: of $\om^*$. Both $W^u(\om^*)$ and $W^s(\om^*)$ have 

1828: the shapes of ``painted eggs'' (Figure \ref{het}). $W^u(\om^*)$ and 

1829: $W^s(\om^*)$ together form the ``lip'' (Figure \ref{het}) which is a 

1830: higher dimensional generalization of the heteroclinic connection on plane.

1831: 

1832: As the first step toward understanding the degeneracy v.s. nondegeneracy 

1833: of the hyperbolic structure of 2D Euler 

1834: equation, we are interested in the $\e$-homotopy deformation of such 

1835: hyperbolic structure for the dashed-line model. We have calculated 

1836: the Melnikov functions to study the ``breaking'' or ``persistence'' 

1837: of such hyperbolic structure when $\e$ is small. It turns out that both the 

1838: first and the second order Melnikov functions are identically zero 

1839: \cite{Li01d}. 

1840: 

1841: \section{Conclusion}

1842: 

1843: Results on chaos in partial differential equations are reported. 

1844: Results on Lax pairs of Euler equations of incompressible inviscid 

1845: fluids are also reported. Preliminary results on dynamical system 

1846: studies of 2D Euler equation are summarized. 

1847: 

1848: 

1849: \begin{thebibliography}{99}

1850: 

1851: 

1852: \bibitem{AOT99}

1853: M.~J. Ablowitz, Y.~Ohta, and A.~D. Trubatch.

1854: \newblock On {D}iscretizations of the {V}ector {N}onlinear {S}chr{\"{o}}dinger

1855:   {E}quation.

1856: \newblock {\em Phys. Lett. A}, 253, no.5-6:287--304, 1999.

1857: 

1858: \bibitem{AOT00}

1859: M.~J. Ablowitz, Y.~Ohta, and A.~D. Trubatch.

1860: \newblock On {I}ntegrability and {C}haos in {D}iscrete {S}ystems.

1861: \newblock {\em Chaos Solitons Fractals}, 11, 1-3:159--169, 2000.

1862: 

1863: 

1864: \bibitem{Ano67}

1865: D.~V. Anosov.

1866: \newblock Geodesic {F}lows on {C}ompact {R}iemannian {M}anifolds of {N}egative

1867:   {C}urvature.

1868: \newblock {\em Proc. Steklov Inst. Math.}, 90, 1967.

1869: \bibitem{Arn64}

1870: V.~I. Arnold.

1871: \newblock Instability of {D}ynamical {S}ystems with {M}any {D}egrees of

1872:   {F}reedom.

1873: \newblock {\em Sov. Math. Dokl.}, 5 No. 3:581--585, 1964.

1874: 

1875: \bibitem{Arn66}

1876: V.~I. Arnold.

1877: \newblock Sur la {G}eometrie {D}ifferentielle des {G}roupes de {L}ie de

1878:   {D}imension {I}nfinie et ses {A}pplications a {L}'hydrodynamique des

1879:   {F}luides {P}arfaits.

1880: \newblock {\em Ann. Inst. Fourier, Grenoble}, 16,1:319--361, 1966.

1881: 

1882: \bibitem{BFY99}

1883: L.~Belenkaya, S.~Friedlander, and V.~Yudovich.

1884: \newblock The {U}nstable {S}pectrum of {O}scillating {S}hear {F}lows.

1885: \newblock {\em SIAM J. Appl. Math.}, 59, no.5:1701--1715, 1999.

1886: 

1887: \bibitem{Bir12}

1888: G.~D. Birkhoff.

1889: \newblock Some {T}heorems on the {M}otion of {D}ynamical {S}ystems.

1890: \newblock {\em Bull. Soc. Math. France}, 40:305--323, 1912.

1891: 

1892: \bibitem{Bla86}

1893: C.~M. Blazquez.

1894: \newblock Transverse {H}omoclinic {O}rbits in {P}eriodically {P}erturbed

1895:   {P}arabolic {E}quations.

1896: \newblock {\em Nonlinear Analysis, Theory, Methods and Applications}, 10,

1897:   no.11:1277--1291, 1986.

1898: 

1899: \bibitem{BW99}

1900: J.~Bona and J.~Wu.

1901: \newblock Zero {D}issipation {L}imit for {N}onlinear {W}aves.

1902: \newblock {\em Preprint}, 1999.

1903: 

1904: \bibitem{Chi00}

1905: S.~Childress.

1906: \newblock A {L}ax pair of 3{D} {E}uler equation.

1907: \newblock {\em personal communication}, 2000.

1908: 

1909: 

1910: \bibitem{CH82} S.-N. Chow and J.K. Hale, {\it Methods of Bifurcation 

1911: Theory}, Springer-Verlag, New York-Berlin, 1982.

1912: \bibitem{CHMP80} S.-N. Chow and J. K. Hale and J. Mallet-Paret,

1913: An example of bifurcation to homoclinic orbits, {\it J. Differential 

1914: Equations} {\bf 37, no.3} (1980), 351.

1915: \bibitem{CLP89}

1916: S.~N. Chow, X.~B. Lin, and K.~J. Palmer.

1917: \newblock A {S}hadowing {L}emma with {A}pplications to {S}emilinear {P}arabolic

1918:   {E}quations.

1919: \newblock {\em SIAM J. Math. Anal.}, 20, no.3:547--557, 1989.

1920: 

1921: \bibitem{CW96}

1922: P.~Constantin and J.~Wu.

1923: \newblock The {I}nviscid {L}imit for {N}on-smooth {V}orticity.

1924: \newblock {\em Indiana Univ. Math. J.}, 45, no.1:67--81, 1996.

1925: 

1926: 

1927: \bibitem{CKP95}

1928: B.~Coomes, H.~Kocak, and K.~Palmer.

1929: \newblock A {S}hadowing {T}heorem for {O}rdinary {D}ifferential {E}quations.

1930: \newblock {\em Z. Angew. Math. Phys.}, 46, no.1:85--106, 1995.

1931: 

1932: 

1933: 

1934: \bibitem{CKP97}

1935: B.~Coomes, H.~Kocak, and K.~J. Palmer.

1936: \newblock Long {P}eriodic {S}hadowing.

1937: \newblock {\em Numerical Algorithms}, 14:55--78, 1997.

1938: 

1939: 

1940: 

1941: 

1942: 

1943: 

1944: \bibitem{Den89}

1945: B.~Deng.

1946: \newblock Exponential {E}xpansion with {S}ilnikov's {S}addle-{F}ocus.

1947: \newblock {\em J. Differential Equations}, 82, no.1:156--173, 1989.

1948: 

1949: \bibitem{Den93}

1950: B.~Deng.

1951: \newblock On {S}ilnikov's {H}omoclinic-{S}addle-{F}ocus {T}heorem.

1952: \newblock {\em J. Differential Equations}, 102, no.2:305--329, 1993.

1953: 

1954: \bibitem{Fad71}

1955: L.~D. Faddeev.

1956: \newblock On the {S}tability {T}heory for {S}tationary {P}lane-{P}arallel

1957:   {F}lows of {I}deal {F}luid.

1958: \newblock {\em Kraevye Zadachi Mat. Fiziki (Zapiski Nauchnykh Seminarov LOMI,

1959:   v.21), Moscow, Leningrad: Nauka}, 5:164--172, 1971.

1960: 

1961: 

1962: \bibitem{FMMW00}

1963: M.~G. Forest, D.~W. McLaughlin, D.~J. Muraki, and O.~C. Wright.

1964: \newblock Nonfocusing {I}nstabilities in {C}oupled, {I}ntegrable {N}onlinear

1965:   {S}chr{\"{o}}dinger pdes.

1966: \newblock {\em J. Nonlinear Sci.}, 10, no.3:291--331, 2000.

1967: 

1968: \bibitem{FSW00}

1969: M.~G. Forest, S.~P. Sheu, and O.~C. Wright.

1970: \newblock On the {C}onstruction of {O}rbits {H}omoclinic to {P}lane {W}aves in

1971:   {I}ntegrable {C}oupled {N}onlinear {S}chr{\"{o}}dinger {S}ystem.

1972: \newblock {\em Phys. Lett. A}, 266, no.1:24--33, 2000.

1973: 

1974: \bibitem{FS77}

1975: J.~E. Franke and J.~F. Selgrade.

1976: \newblock Hyperbolicity and {C}hain {R}ecurrence.

1977: \newblock {\em J. Diff. Eq.}, 26:27--36, 1977.

1978: 

1979: \bibitem{GH83} J. Guckenheimer and P. J. Holmes, {\it Nonlinear 

1980: Oscillations, Dynamical Systems, and Bifurcations of Vector Fields},

1981: Springer-Verlag, Applied Mathematical Sciences, vol.42, 1983.

1982: 

1983: \bibitem{HL86}

1984: J.~K. Hale and X.~B. Lin.

1985: \newblock Symbolic {D}ynamics and {N}onlinear {S}emiflows.

1986: \newblock {\em Ann. Mat. Pura Appl.}, 144:229--259, 1986.

1987: 

1988: 

1989: \bibitem{HK95}

1990: A.~Hasegawa and Y.~Kodama.

1991: \newblock {\em Solitons in {O}ptical {C}ommunications}.

1992: \newblock Academic Press, San Diego, 1995.

1993: 

1994: \bibitem{Hen94}

1995: D.~Henry.

1996: \newblock Exponential {D}ichotomies, the {S}hadowing {L}emma and {H}omoclinic

1997:   {O}rbits in {B}anach {S}paces.

1998: \newblock {\em Resenhas}, 1, no.4:381--401, 1994.

1999: 

2000: \bibitem{HM81}

2001: P.~J. Holmes and J.~E. Marsden.

2002: \newblock A {P}artial {D}ifferential {E}quation with {I}nfinitely {M}any

2003:   {P}eriodic {O}rbits: {C}haotic {O}scillations of a {F}orced {B}eam.

2004: \newblock {\em Arch. Rat. Mech. Anal.}, 76:135--166, 1981.

2005: 

2006: 

2007: \bibitem{Isl92}

2008: M.~N. Islam.

2009: \newblock {\em Ultrafast {F}iber {S}witching {D}evices and {S}ystems}.

2010: \newblock Cambridge University Press, New York, 1992.

2011: 

2012: \bibitem{Kat72}

2013: T.~Kato.

2014: \newblock Nonstationary {F}lows of {V}iscous and {I}deal {F}luids in ${R}^3$.

2015: \newblock {\em Journal of Functional Analysis}, 9:296--305, 1972.

2016: 

2017: \bibitem{Kat75}

2018: T.~Kato.

2019: \newblock Quasi-{L}inear {E}quations of {E}volution, with {A}pplications to

2020:   {P}artial {D}ifferential {E}quations.

2021: \newblock {\em Lecture Notes in Mathematics}, 448:25--70, 1975.

2022: 

2023: \bibitem{Kat86}

2024: T.~Kato.

2025: \newblock Remarks on the {E}uler and {N}avier-{S}tokes {E}quations in ${R}^2$.

2026: \newblock {\em Proc. Symp. Pure Math., Part 2}, 45:1--7, 1986.

2027: 

2028: 

2029: \bibitem{Kov92a} G. Kovacic, Dissipative dynamics of orbits homoclinic 

2030: to a resonance band, {\it Phys Lett A} {\bf 167} (1992), 143.

2031: \bibitem{Kov92b} G. Kovacic, Hamiltonian dynamics of orbits homoclinic 

2032: to a resonance band, {\it Phys Lett A} {\bf 167} (1992), 137.

2033: \bibitem{LLM01} Y. Latushkin and Y. Li and M. Stanislavova, The spectrum 

2034: of a linearized 2D Euler operator, {\it Submitted} (2001).

2035: 

2036: \bibitem{Li99a} Y. Li, Smale horseshoes and symbolic dynamics 

2037:           in perturbed nonlinear Schr{\"{o}}dinger equations, {\it 

2038: Journal of Nonlinear Sciences} {\bf 9, no.4} (1999), 363.

2039: \bibitem{Li00a} Y. Li, B{\"{a}}cklund-Darboux transformations 

2040:          and Melnikov analysis for Davey-Stewartson II equations, {\it

2041: Journal of Nonlinear Sciences} {\bf 10, no.1} (2000), 103.

2042: \bibitem{Li00b} Y. Li, On 2D Euler equations. I. On the energy-Casimir 

2043:            stabilities and the spectra for a linearized two 

2044:            dimensional Euler equation, {\it Journal of Mathematical 

2045: Physics} {\bf 41, no.2} (2000), 728.

2046: \bibitem{Li01a} Y. Li, A Lax pair for the 2D Euler equation, {\it 

2047: Journal of Mathematical Physics} {\bf 42, no.8} (2001), 3552.

2048: \bibitem{Li01d} Y. Li, On 2D Euler equations: part II. Lax pairs 

2049: and homoclinic strucutres, {\it Communications on Applied Nonlinear Analysis,

2050: to appear, available at:

2051:   http://xxx.lanl.gov/abs/math.AP/0010200, or

2052:   http://www.math.missouri.edu/\~{}cli} (2002). 

2053: \bibitem{Li01b} Y. Li, Persistent homoclinic orbits for nonlinear 

2054: Schr{\"{o}}dinger equation under singular perturbation, {\it Submitted,

2055: available at:

2056:   http://xxx.lanl.gov/abs/math.AP/0106194, or

2057:   http://www.math.missouri.edu/\~{}cli} 

2058: (2001).

2059: 

2060: \bibitem{Li02a}

2061: Y.~Li.

2062: \newblock Chaos and {S}hadowing {L}emma for {A}utonomous {S}ystems of

2063:   {I}nfinite {D}imensions.

2064: \newblock {\em Submitted, available at:

2065:   http://xxx.lanl.gov/abs/nlin/0203024, or

2066:   http://www.math.missouri.edu/\~{}cli}, 2002.

2067: 

2068: \bibitem{Li02b}

2069: Y.~Li.

2070: \newblock Melnikov {A}nalysis for {S}ingularly {P}erturbed {D}{S}{I}{I}

2071:   {E}quation.

2072: \newblock {\em Submitted, available at:

2073:   http://xxx.lanl.gov/abs/math.AP/0206272, or

2074:   http://www.math.missouri.edu/\~{}cli}, 2002.

2075: 

2076: \bibitem{Li02c}

2077: Y.~Li.

2078: \newblock Existence of {C}haos for a {S}ingularly {P}erturbed {N}{L}{S}

2079:   {E}quation.

2080: \newblock {\em Submitted, available at:

2081:   http://xxx.lanl.gov/abs/math.AP/0206270, or

2082:   http://www.math.missouri.edu/\~{}cli}, 2002.

2083: 

2084: 

2085: \bibitem{Li02d}

2086: Y.~Li.

2087: \newblock Singularly Perturbed Vector and Scalar Nonlinear {S}chroedinger

2088:   Equations with Persistent Homoclinic Orbits.

2089: \newblock {\em Studies in Applied Mathematics}, 109:19--38, 2002.

2090: 

2091: \bibitem{Li02e}

2092: Y.~Li.

2093: \newblock Integrable Structures for 2{D} {E}uler Equations of Incompressible

2094:   Inviscid Fluids.

2095: \newblock {\em Proc. Institute Of Mathematics of NAS of Ukraine}, 43, part

2096:   1:332--338, 2002.

2097: 

2098: 

2099: \bibitem{Li02f} Y. Li, Chaos in Partial Differential Equations, {\it 

2100: Contemporary Mathematics, to appear, available at:

2101:   http://xxx.lanl.gov/abs/math.AP/0205114, or

2102:   http://www.math.missouri.edu/\~{}cli}, 2002.

2103: 

2104: \bibitem{Li02m}

2105: Y.~Li.

2106: \newblock On 2{D} {E}uler {E}quations: {I}{I}{I}. {A} {L}ine {M}odel.

2107: \newblock {\em available at:

2108:   http://xxx.lanl.gov/abs/math.AP/0206278, or

2109:   http://www.math.missouri.edu/\~{}cli}, 2002.

2110: 

2111: \bibitem{LMSW96} Y. Li et al., Persistent homoclinic orbits for a 

2112: perturbed nonlinear Schr{\"{o}}dinger equation, 

2113: {\it Comm. Pure Appl. Math.} {\bf XLIX} (1996), 1175.

2114: \bibitem{LM94} Y. Li and D. W. McLaughlin, Morse and Melnikov functions 

2115: for NLS pdes, {\it Comm. Math. Phys.} {\bf 162} (1994), 175.

2116: \bibitem{LM97} Y. Li and D. W. McLaughlin, Homoclinic orbits and chaos 

2117: in discretized perturbed NLS system, part I. homoclinic orbits,

2118: {\it Journal of Nonlinear Sciences} {\bf 7} (1997), 211.

2119: 

2120: \bibitem{LS02}

2121: Y.~Li and R.~Shvidkoy.

2122: \newblock Isospectral {T}heory of {E}uler {E}quations.

2123: \newblock {\em Submitted, available at:

2124:   http://xxx.lanl.gov/abs/math.AP/0203125 , or

2125:   http://www.math.missouri.edu/\~{}cli}, 2002.

2126: 

2127: \bibitem{LW97}

2128: Y.~Li and S.~Wiggins.

2129: \newblock Homoclinic {O}rbits and {C}haos in {P}erturbed {D}iscrete {N}{L}{S}

2130:   {S}ystem. {P}art {II} {S}ymbolic {D}ynamics.

2131: \newblock {\em Journal of Nonlinear Sciences}, 7:315--370, 1997.

2132: 

2133: 

2134: \bibitem{LY01} Y. Li and A. Yurov, Lax pairs and Darboux transformations 

2135: for Euler equations, {\it Submitted , available at: 

2136: http://xxx.lanl.gov/abs/math.AP/0101214, or 

2137: http://www.math.missouri.edu/\~{}cli} (2001).

2138: 

2139: \bibitem{Liu92a}

2140: V.~X. Liu.

2141: \newblock An {E}xample of {I}nstability for the {N}avier-{S}tokes {E}quations

2142:   on the 2-{D}imensional {T}orus.

2143: \newblock {\em Comm. Partial Differential Equations}, 17, no.11-12:1995--2012,

2144:   1992.

2145: 

2146: \bibitem{Liu92b}

2147: V.~X. Liu.

2148: \newblock Instability for the {N}avier-{S}tokes {E}quations on the

2149:   2-{D}imensional {T}orus and a {L}ower {B}ound for the {H}ausdorff {D}imension

2150:   of {T}heir {G}lobal {A}ttractors.

2151: \newblock {\em Comm. Math. Phys.}, 147, no.2:217--230, 1992.

2152: 

2153: \bibitem{Liu93}

2154: V.~X. Liu.

2155: \newblock A {S}harp {L}ower {B}ound for the {H}ausdorff {D}imension of the

2156:   {G}lobal {A}ttractors of the 2{D} {N}avier-{S}tokes {E}quations.

2157: \newblock {\em Comm. Math. Phys.}, 158, no.2:327--339, 1993.

2158: 

2159: \bibitem{Liu94a}

2160: V.~X. Liu.

2161: \newblock Remarks on the {N}avier-{S}tokes {E}quations on the {T}wo- and

2162:   {T}hree-{D}imensional {T}orus.

2163: \newblock {\em Comm. Partial Differential Equations}, 19, no.5-6:873--900,

2164:   1994.

2165: 

2166: \bibitem{Liu95}

2167: V.~X. Liu.

2168: \newblock On {U}nstable and {N}eutral {S}pectra of {I}ncompressible {I}nviscid

2169:   and {V}iscid {F}luids on the 2{D} {T}orus.

2170: \newblock {\em Quart. Appl. Math.}, 53, no.3:465--486, 1995.

2171: 

2172: \bibitem{Man74}

2173: S.~V. Manakov.

2174: \newblock On the {T}heory of {T}wo-{D}imensional {S}tationary {S}elf-{F}ocusing

2175:   of {E}lectromagnetic {W}aves.

2176: \newblock {\em Soviet Physics JETP}, 38, no.2:248--253, 1974.

2177: 

2178: \bibitem{MS91}

2179: V.~B. Matveev and M.~A. Salle.

2180: \newblock {\em Darboux {T}ransformations and {S}olitons}, volume~5.

2181: \newblock Springer Series in Nonlinear Dynamics, 1991.

2182: 

2183: \bibitem{Mel63} V. K. Melnikov, On the stability of the center for time 

2184: periodic perturbations, {\it Trans. Moscow Math. Soc.} {\bf 12} (1963), 1.

2185: 

2186: \bibitem{Men87}

2187: C.~R. Menyuk.

2188: \newblock Nonlinear {P}ulse {P}ropagation in {B}irefringent {O}ptical {F}ibers.

2189: \newblock {\em IEEE J. Quantum Electron}, 23, no.2:174--176, 1987.

2190: 

2191: \bibitem{Men89}

2192: C.~R. Menyuk.

2193: \newblock Pulse {P}ropagation in an {E}lliptically {B}irefringent {K}err

2194:   {M}edium.

2195: \newblock {\em IEEE J. Quantum Electron}, 25, no.12:2674--2682, 1989.

2196: 

2197: \bibitem{MS61}

2198: L.~D. Meshalkin and Ia.~G. Sinai.

2199: \newblock Investigation of the {S}tability of a {S}tationary {S}olution of a

2200:   {S}ystem of {E}quations for the {P}lane {M}ovement of an {I}ncompressible

2201:   {V}iscous {L}iquid.

2202: \newblock {\em J. Appl. Math. Mech. (PMM)}, 25:1140--1143, 1961.

2203: 

2204: 

2205: \bibitem{Pal84}

2206: K.~Palmer.

2207: \newblock Exponential {D}ichotomies and {T}ransversal {H}omoclinic {P}oints.

2208: \newblock {\em J. Differential Equations}, 55, no.2:225--256, 1984.

2209: 

2210: \bibitem{Pal88}

2211: K.~Palmer.

2212: \newblock Exponential {D}ichotomies, the {S}hadowing {L}emma and {T}ransversal

2213:   {H}omoclinic {P}oints.

2214: \newblock {\em Dynamics Reported}, 1:265--306, 1988.

2215: 

2216: \bibitem{Pal96}

2217: K.~Palmer.

2218: \newblock Shadowing and {S}ilnikov {C}haos.

2219: \newblock {\em Nonlinear Anal.}, 27, no.9:1075--1093, 1996.

2220: 

2221: \bibitem{Poi99}

2222: H.~Poincar{\'{e}}.

2223: \newblock {\em Les {M}ethodes {N}ouvelles de la {M}ecanique {C}eleste, {V}ols.

2224:   1-3. {E}nglish {T}ranslation: {N}ew {M}ethods of {C}elestial {M}echanics.

2225:   {V}ols. 1-3, edited by {D}aniel {L}. {G}oroff, {A}merican {I}nstitute of

2226:   {P}hysics, {N}ew {Y}ork, 1993}.

2227: \newblock Gauthier-Villars, Paris, 1899.

2228: 

2229: \bibitem{SZ00}

2230: J.~Shatah and C.~Zeng.

2231: \newblock Homoclinic {O}rbits for the {P}erturbed {S}ine-{G}ordon {E}quation.

2232: \newblock {\em Comm. Pure Appl. Math.}, 53, no.3:283--299, 2000.

2233: 

2234: \bibitem{Sil65}

2235: L.~P. Silnikov.

2236: \newblock A {C}ase of the {E}xistence of a {C}ountable {N}umber of {P}eriodic

2237:   {M}otions.

2238: \newblock {\em Soviet Math. Doklady}, 6:163--166, 1965.

2239: 

2240: \bibitem{Sil67a}

2241: L.~P. Silnikov.

2242: \newblock The {E}xistence of a {D}enumerable {S}et of {P}eriodic {M}otions in

2243:   {F}our-dimensional {S}pace in an {E}xtended {N}eighborhood of a

2244:   {S}addle-{F}ocus.

2245: \newblock {\em Soviet Math. Doklady}, 8:54--58, 1967.

2246: 

2247: \bibitem{Sil67b}

2248: L.~P. Silnikov.

2249: \newblock On a {P}oincare-{B}irkoff {P}roblem.

2250: \newblock {\em Math. USSR Sb.}, 3:353--371, 1967.

2251: 

2252: \bibitem{Sil70}

2253: L.~P. Silnikov.

2254: \newblock A {C}ontribution to the {P}roblem of the {S}tructure of an {E}xtended

2255:   {N}eighborhood of a {R}ough {E}quilibrium {S}tate of {S}addle-focus {T}ype.

2256: \newblock {\em Math. USSR Sb.}, 10:91--102, 1970.

2257: 

2258: \bibitem{Sma61}

2259: S.~Smale.

2260: \newblock A {S}tructurally {S}table {D}ifferentiable {H}omeomorphism with an

2261:   {I}nfinite {N}umber of {P}eriodic {P}oints.

2262: \newblock {\em Qualitative Methods in the Theory of Non-linear Vibrations

2263:   (Proc. Internat. Sympos. Non-linear Vibrations)}, II:365--366, 1961.

2264: 

2265: \bibitem{Sma65}

2266: S.~Smale.

2267: \newblock Diffeomorphisms with {M}any {P}eriodic {P}oints.

2268: \newblock {\em Differential and Combinatorial Topology (A Symposium in Honor of

2269:   Marston Morse), Princeton Univ. Press, Princeton, N.J.}, pages 63--80, 1965.

2270: 

2271: \bibitem{Sma67}

2272: S.~Smale.

2273: \newblock Differentiable {D}ynamical {S}ystems.

2274: \newblock {\em Bull. Amer. Math. Soc.}, 73:747--817, 1967.

2275: 

2276: \bibitem{SW89}

2277: H.~Steinlein and H.~Walther.

2278: \newblock Hyperbolic {S}ets and {S}hadowing for {N}oninvertible {M}aps.

2279: \newblock {\em Advanced Topics in the Theory of Dynamical Systems, Academic

2280:   Press, Boston, MA}, pages 219--234, 1989.

2281: 

2282: \bibitem{SW90}

2283: H.~Steinlein and H.~Walther.

2284: \newblock Hyperbolic {S}ets, {T}ransversal {H}omoclinic {T}rajectories, and

2285:   {S}ymbolic {D}ynamics for $c^1$-{M}aps in {B}anach {S}paces.

2286: \newblock {\em J. Dynamics Differential Equations}, 2, no.3:325--365, 1990.

2287: 

2288: \bibitem{Wig88} S. Wiggins, {\it Global Bifurcations and Chaos: Analytical 

2289: Methods}, Springer-Verlag, New York, 1988.

2290: \bibitem{WF00}

2291: O.~C. Wright and M.~G. Forest.

2292: \newblock On the {B}{\"{a}}cklund-{G}auge {T}ransformation and {H}omoclinic

2293:   {O}rbits of a {C}oupled {N}onlinear {S}chr{\"{o}}dinger {S}ystem.

2294: \newblock {\em Phys. D}, 141, no.1-2:104--116, 2000.

2295: 

2296: \bibitem{Wu96}

2297: J.~Wu.

2298: \newblock The {I}nviscid {L}imits for {I}ndividual and {S}tatistical

2299:   {S}olutions of the {N}avier-{S}tokes {E}quations.

2300: \newblock {\em Ph.D. Thesis, Chicago University}, 1996.

2301: 

2302: \bibitem{Wu98}

2303: J.~Wu.

2304: \newblock The {I}nviscid {L}imit of the {C}omplex {G}inzburg-{L}andau

2305:   {E}quation.

2306: \newblock {\em J. Diff. Equations}, 142, No.2:413--433, 1998.

2307: 

2308: \bibitem{YT01}

2309: J.~Yang and Y.~Tan.

2310: \newblock Fractal {D}ependence of {V}ector-{S}oliton {C}ollisions in

2311:   {B}irefringent {F}ibers.

2312: \newblock {\em Phys. Lett. A}, 280:129--138, 2001.

2313: 

2314: \bibitem{Yud65}

2315: V.~I. Yudovich.

2316: \newblock Example of the {G}eneration of a {S}econdary {S}tationary or

2317:   {P}eriodic {F}low {W}hen {T}here is {L}oss of {S}tability of the {L}aminar

2318:   {F}low of a {V}iscous {I}ncompressible {F}luid.

2319: \newblock {\em J. Appl. Math. Mech. (PMM)}, 29:527--544, 1965.

2320: 

2321: 

2322: 

2323: \bibitem{Zen95}

2324: W.~Zeng.

2325: \newblock Exponential {D}ichotomies and {T}ransversal {H}omoclinic {O}rbits in

2326:   {D}egenerate {C}ases.

2327: \newblock {\em J. Dyn. Diff. Eq.}, 7, no.4:521--548, 1995.

2328: 

2329: \bibitem{Zen97}

2330: W.~Zeng.

2331: \newblock Transversality of {H}omoclinic {O}rbits and {E}xponential

2332:   {D}ichotomies for {P}arabolic {E}quations.

2333: \newblock {\em J. Math. Anal. Appl.}, 216:466--480, 1997.

2334: 

2335: 

2336: \end{thebibliography}

2337: 

2338: 

2339: 

2340: 

2341: \end{document}

2342: 

2343: %-----------------------------------------------------------------------

2344: % End of article.tex

2345: %-----------------------------------------------------------------------

2346: