Gauge Invariant Treatment of the Energy Carried by a Gravitational Wave

TL;DR

This paper develops a gauge invariant framework for defining and analyzing the energy carried by gravitational waves, ensuring physical quantities remain consistent across different coordinate choices.

Contribution

It constructs a gauge invariant action and energy-momentum tensor for gravitational fluctuations, enabling gauge invariant energy calculations in curved backgrounds.

Findings

Energy and momentum transfer by gravitational waves are gauge invariant.

The framework remains valid even with non-zero asymptotic momentum flux.

Provides a method to analyze gravitational wave energy in arbitrary gauges.

Abstract

Even though the energy carried by a gravitational wave is not itself gauge invariant, the interaction with a gravitational antenna of the gravitational wave which carries that energy is. It therefore has to be possible to make some statements which involve the energy which are in fact gauge invariant, and it is the objective of this paper to provide them. In order to develop a gauge invariant treatment of the issues involved, we construct a specific action for gravitational fluctuations which is gauge invariant to second perturbative order. Then, via variation of this action, we obtain an energy-momentum tensor for perturbative gravitational fluctuations around a general curved background whose covariant conservation condition is also fully gauge invariant to second order. Contraction of this energy-momentum tensor with a Killing vector of the background conveniently allows us to convert this covariant conservation condition into an ordinary conservation condition which is also gauge invariant through second order. Then, via spatial integration we are able to obtain a relation involving the time derivative of the total energy of the fluctuation and its asymptotic spatial momentum flux which is also completely gauge invariant through second order. It is only in making the simplification of setting the asymptotic momentum flux to zero that one would actually lose manifest gauge invariance, with only invariance under those particular gauge transformations which leave the asymptotic momentum flux zero then remaining. However, if one works in an arbitrary gauge where the asymptotic momentum flux is non-zero, the gravitational wave will then deliver both energy and momentum to a gravitational antenna in a completely gauge invariant manner, no matter how badly behaved at infinity the gauge function might be.