Parametrix for wave equations on a rough background II: construction and control at initial time
Jeremie Szeftel
TL;DR
<3-5 sentence high-level summary>Parametrix for wave equations on a rough background II develops a time-zero parametrix construction for the homogeneous wave equation $\square_{\mathbf{g}}\phi=0$ on a rough Einstein vacuum spacetime and proves $L^2$ boundedness results for the associated Fourier integral operators under minimal regularity of the phase and symbol. The work centers on prescribing the initial-phase functions $u_±$ on the initial slice and solving a boundary data system $(f_+,f_-)$ via $M_+ f_+ + M_-f_- = \phi_0$ and $Q_+(\lambda f_+) - Q_-(\lambda f_-) = i\phi_1$, with quantitative $L^2$ control on $\lambda f_\pm$. Core techniques combine dyadic decompositions in frequency and angle with geometric integrations by parts along tangential directions to foliation level sets, together with an almost-orthogonality framework to bound Fourier integral operators with rough phases and symbols. These results advance the analysis needed for the bounded $L^2$ curvature conjecture by enabling robust control of wave propagation on rough backgrounds and contribute broadly to the theory of $L^2$ bounds for FIOs under limited regularity.
Abstract
This is the second of a sequence of four papers \cite{param1}, \cite{param2}, \cite{param3}, \cite{param4} dedicated to the construction and the control of a parametrix to the homogeneous wave equation $\square_{\bf g} φ=0$, where ${\bf g}$ is a rough metric satisfying the Einstein vacuum equations. Controlling such a parametrix as well as its error term when one only assumes $L^2$ bounds on the curvature tensor ${\bf R}$ of ${\bf g}$ is a major step of the proof of the bounded $L^2$ curvature conjecture proposed in \cite{Kl:2000}, and solved by S. Klainerman, I. Rodnianski and the author in \cite{boundedl2}. On a more general level, this sequence of papers deals with the control of the eikonal equation on a rough background, and with the derivation of $L^2$ bounds for Fourier integral operators on manifolds with rough phases and symbols, and as such is also of independent interest.
