On post-Newtonian orbits and the Galactic-center stars

TL;DR

This paper develops a Hamiltonian formulation and a symplectic integrator to accurately compute post-Newtonian relativistic effects on stars near the Galactic center, aiding the detection of general relativity in strong gravitational fields.

Contribution

It introduces a novel Hamiltonian-based method and an adaptive symplectic integrator for precise numerical calculation of relativistic orbital effects near the Galactic center.

Findings

Relativistic effects cause measurable astrometric and kinematic shifts for stars like S2.

The integrator enables fast, accurate simulations of post-Newtonian orbits.

Relativistic perturbations have distinctive time-dependent signatures that can be distinguished from Newtonian influences.

Abstract

Stars near the Galactic center reach a few percent of light speed during pericenter passage, which makes post-Newtonian effects potentially detectable. We formulate the orbit equations in Hamiltonian form such that the $O(v^2/c^2)$ and $O(v^3/c^3)$ post-Newtonian effects of the Kerr metric appear as a simple generalization of the Kepler problem. A related perturbative Hamiltonian applies to photon paths. We then derive a symplectic integrator with adaptive time-steps, for fast and accurate numerical calculation of post-Newtonian effects. Using this integrator, we explore relativistic effects. Taking the star S2 as an example, we find that general relativity would contribute tenths of mas in astrometry and tens of $\rm km s^{-1}$ in kinematics. (For eventual comparison with observations, redshift and time-delay contributions from the gravitational field on light paths will need to be calculated, but we do attempt these in the present paper.) The contribution from stars, gas, and dark matter in the Galactic center region is still poorly constrained observationally, but current models suggest that the resulting Newtonian perturbation on the orbits could plausibly be of the same order as the relativistic effects for stars with semi-major axes $\gtrsim 0.01$ pc (or 250 mas). Nevertheless, the known and distinctive {\it time dependence} of the relativistic perturbations may make it possible to disentangle and extract both effects from observations.