The initial boundary value problem for free-evolution formulations of General Relativity

TL;DR

This paper develops high-order boundary conditions for free-evolution formulations of general relativity, demonstrating boundary stability and well-posedness in a linear approximation, which is crucial for stable numerical simulations.

Contribution

It introduces high-order derivative boundary conditions for Einstein's equations and proves boundary stability and well-posedness in a linearized setting, advancing numerical relativity methods.

Findings

High-order boundary conditions ensure boundary stability.

Boundary stability demonstrated in the frozen coefficient approximation.

Well-posedness of the linearized IBVP established.

Abstract

We consider the initial boundary value problem for free-evolution formulations of general relativity coupled to a parametrized family of coordinate conditions that includes both the moving puncture and harmonic gauges. We concentrate primarily on boundaries that are geometrically determined by the outermost normal observer to spacelike slices of the foliation. We present high-order-derivative boundary conditions for the gauge, constraint violating and gravitational wave degrees of freedom of the formulation. Second order derivative boundary conditions are presented in terms of the conformal variables used in numerical relativity simulations. Using Kreiss-Agranovich-Metivier theory we demonstrate, in the frozen coefficient approximation, that with sufficiently high order derivative boundary conditions the initial boundary value problem can be rendered boundary stable. The precise number of derivatives required depends on the gauge. For a choice of the gauge condition that renders the system strongly hyperbolic of constant multiplicity, well-posedness of the initial boundary value problem follows in this approximation. Taking into account the theory of pseudo-differential operators, it is expected that the nonlinear problem is also well-posed locally in time.