2
has the advantage that one can in principle calculate the effect on some observable
due to some symmetry breaking terms, which can then be compared with entirely
different observables in different scenarios, for measurements of the same coefficients
controlling the size of the effects. Other formalisms for testing symmetries in gravity
are parametrized directly from the form of a GR observable [23, 24, 25], or are based
on specific models of alternatives to GR [26, 27, 28, 29].
We will consider in this work modifications to the gravity sector that, contrary to
standard GR, break local Lorentz symmetry and diffeomorphism symmetry explicitly or
spontaneously. These spacetime symmetries can be thought of as gauge symmetries for
gravity, and thus GR is a gauge theory of gravity with local Lorentz and diffeomorphism
symmetries as the gauge symmetries, analogous to Standard Model physics based on
gauge groups [30]. The subtle issue of the role of broken spacetime symmetries in the
context of curved spacetime, particularly when assuming asymptotically flat scenarios
or not, has been discussed at length elsewhere [22, 31, 32]. While we do not fully
discuss these concepts and subtleties here, we shall refer to conventions and categories
of transformations in these references as needed.
In the EFT approach taken here, we highlight comparison of short-range (SR)
gravity tests with gravitational wave (GW) observations, thus comparing two tests
“across the universe” for measuring the same quantities describing spacetime-symmetry
breaking for gravity. In fact, we show certain rotational scalar coefficients that can
be measured in GW tests can also be probed in SR tests. Further, there are some
coefficients that cannot be completely disentangled with GW tests alone, but using also
SR gravity tests could accomplish this.
In references [16] and [17] solutions for short-range gravity tests were found, but
these used an approximation of leading order in the coefficients. We show here that
exact, non-perturbative, solutions can reveal where other combinations of coefficients,
not yet disentangled, can show up in experiment. As we are concerned in this paper
with modifications to gravity that do not break the Weak-Equivalence Principle, we do
not discuss WEP violations here. The connection between Lorentz violation and WEP
has been discussed at length elsewhere [33, 34, 35, 36].
Since we examine non-perturbative solutions, the results in this work also touch
on the nature of higher than second order derivatives in the action and how that might
affect gravity. For this latter topic, we do not attempt a comprehensive investigation
of these issues but simply note where results exhibit behavior expected of such models
[37, 38, 39], and how they might be consistent with perturbative approaches.
The paper is organized as follows. In section 2, we review two commonly used EFT
schemes for the description of spacetime symmetry breaking in gravity and we discuss
prior results in short-range gravity signals for Lorentz violation. In section 3, we explore
non-perturbative solutions with a special case model to identify key features. Following
this, we go on to solve the general EFT framework in the static, isotropic coefficient
limit. Features of the solutions are discussed and explained with several plots. We
discuss attempting exact solutions with anistropic coefficients in section 4, and compare