Pericenter advance in general relativity: Comparison of approaches at high post-Newtonian orders

TL;DR

This paper compares different methods for calculating the relativistic pericenter advance of orbits, demonstrating that apparent discrepancies at high post-Newtonian orders are resolved when using invariant quantities and clarifying the meaning of pericenter advance.

Contribution

It clarifies the agreement between approaches at high post-Newtonian orders by expressing variables in terms of invariant quantities and addressing conceptual differences.

Findings

Different methods agree at leading PN order.

Discrepancies at higher PN orders are due to variable definitions.

When properly interpreted, methods are consistent up to 3PN order.

Abstract

The advance of the pericenter of the orbit of a test body around a massive body in general relativity can be calculated in a number of ways. In one method, one studies the geodesic equation in the exact Schwarzschild geometry and finds the angle between pericenters as an integral of a certain radial function between turning points of the orbit. In another method, one describes the orbit using osculating orbit elements, and analyzes the "Lagrange planetary equations" that give the evolution of the elements under the perturbing effects of post-Newtonian (PN) corrections to the motion. After separating the perturbations into periodic and secular effects, one obtains an equation for the secular rate of change of the pericenter angle. While the different methods agree on the leading post-Newtonian contribution to the advance, they do not agree on the higher-order PN corrections. We show that this disagreement is illusory. When the orbital variables in each case are expressed in terms of the invariant energy and angular momentum of the orbit and when account is taken of a subtle difference in the meaning of "pericenter advance" between the two methods, we show to the third post-Newtonian order that the different methods actually agree perfectly.