Ill-posedness issues for nonlinear dispersive equations

N. Tzvetkov
2007-02-02
Abstract: These notes are devoted to the notion of well-posedness of the Cauchy problem for nonlinear dispersive equations. We present recent methods for proving ill-posedness type results for dispersive PDE's. The common feature in the analysis is that the proof of such results requires the construction of high frequency approximate solutions on small time intervals (possibly depending on the frequency).
Analysis of PDEs
What problem does this paper attempt to address?
This paper mainly discusses the well-posedness problem of nonlinear dispersive equations. Well-posedness is a key concept in the study of initial value problems for partial differential equations (PDEs), involving the existence, uniqueness, and continuous dependence of solutions. The author points out that for dispersive PDEs such as the KdV equation and the nonlinear Schrödinger equation, determining the well-posedness of an equation in a specific function space requires careful analysis, especially in constructing high-frequency approximate solutions over small time intervals. The paper first defines well-posedness in the classical sense, namely the existence of a unique solution to a given initial value problem and its continuous dependence over a certain time interval. Furthermore, the author introduces a stronger concept of semi-linear well-posedness, which requires that the flow mapping is not only continuous but also uniformly continuous on the set of initial values. The paper discusses the methods of using integral equations (such as the Duhamel formula) contraction principle and compactness argument to prove well-posedness, and demonstrates how to prove well-posedness through specific examples such as the Burgers equation and the KdV equation. The Burgers equation is well-posed in certain function spaces but not semi-linear well-posed, while the KdV equation is semi-linear well-posed. In addition, the well-posedness and semi-linear well-posedness of the nonlinear Schrödinger equation vary under different geometric backgrounds, such as two-dimensional and three-dimensional compact Riemann manifolds. The author also shows how to prove ill-posedness by constructing approximate solutions, for example, by revealing the instability of equation solutions through a change of variables. These results indicate that although certain equations may be technically well-posed, the properties of their solutions may not be sufficient to satisfy the requirements of semi-linear well-posedness. In summary, the paper aims to emphasize the complexity of well-posedness problems for nonlinear dispersive equations, especially when it comes to high-frequency approximate solutions, and how different analysis methods can be used to determine the well-posedness categories of equations.