A geometric approach to Catlin's boundary systems (Q2333214)

From MaRDI portal
scientific article
Language Label Description Also known as
English
A geometric approach to Catlin's boundary systems
scientific article

    Statements

    A geometric approach to Catlin's boundary systems (English)
    0 references
    12 November 2019
    0 references
    Let $M\subset\mathbb{C}^n$ be a smooth real hypersurface and $p\in M$. A fundamental problem in CR-geometry has to do with understanding the singularities of the Cauchy-Riemann structure induced by the ambient complex structure. This paper introduces three new invariants for the study of singularities or points of degeneracy; invariant submodule sheaves of vector fields, invariant ideal sheaves of functions, and invariant quartic tensors. The quartic tensors arise from taking higher derivatives of Levi form, denoted by \[ \tau_p^4: \mathbb{C}T_p\times \mathbb{C}T_p\times K_p^{10}\times \overline{K_p^{10}}\rightarrow \mathbb{C}\otimes \left(T_p/H_p\right) \] with $T$ and $H$ the tangent and complex tangent bundles and the quotient $Q=T/H$ is the normal bundle. Here $K_p^{10}\subset H_p^{10}$ consists of the components of the Levi kernel at $p$. The invariant submodule sheaves are defined as follows. For each integer $1\leq q\leq n-1$, let $\mathcal{T}^{10}$ denote the sheaf of all $(1,0)$ vector fields on $M$ and define $\mathcal{S}^{1 0}(q)\subset \mathcal{T}^{10}$ to be the submodule sheaf consisting of all germs of vector fields on $M$ which satisfy a technical condition termed by the author as being contained in the Levi kernel up to order 1 at every point of Levi rank $\leq q$. Given a local defining function $\rho$ in the neighborhood of a point on $M$ define the contact form $\theta$ by \[ \theta:=i\partial \rho, \] with the standard pairing $\langle \theta, L\rangle:=\theta(L)$, $\theta\in \mathbb{C}\Omega$, $L\in\mathbb{C}T$. More formally, \[ \langle \cdot,\cdot\rangle : \mathbb{C}\Omega^{0}\times \mathbb{C}Q \to \mathbb{C}, \] with the sets on the left denoting the (complexified) bundle of contact forms and the normal bundle. In turn the sheaves $\mathcal{S}^{1 0}(q)$ are used to define invariant ideal sheaves as follows. Under the hypothesis of pseudoconvexity, the ideal sheaf $\ell(q)$ is the sheaf generated by all smooth complex functions $g$, $f$ of the form \[ g=\langle \theta,[L^2,L^1]\rangle,\qquad f=L^3\langle\theta,[L^2,L^1]\rangle. \] Here $\theta\in\Omega^0$ is a contact form, $L^3\in\mathbb{C}T$ an arbitrary vector field, and $L^2, \overline{L^1}\in S^{10}(q)$ arbitrary sections. The invariant ideal sheaves are used to identify the real submanifolds $S\subset M$ through $p$ containing the set of points of Levi rank $q$, a measure of degeneracy for the underlying geometry. These generalize the functions used in Catlin's boundary system. In his earlier approach, Catlin developed his method of boundary systems in conjunction with the alternative measure of Catlin mutlitype. The relationship of the quartic tensor to D'Angelo Finite Type 4 is also discussed. This paper is very nicely written in a clear and self-contained manner.
    0 references
    Catlin multitype
    0 references
    subelliptic estimates
    0 references
    boundary systems
    0 references
    Levi form
    0 references
    pseudoconvexity
    0 references
    real hypersurfaces
    0 references
    invariant tensors
    0 references
    ideal sheaves
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references