Effective gravitational coupling μa
Most tests of modified gravity focus on a single observable. You fit a model to BAO distances, or to supernova luminosities, or to growth-rate measurements — and you report how it compares to ΛCDM. The trouble is that any sufficiently flexible model can fit one dataset at a time. The real question is whether it can fit all of them simultaneously, with the same parameters, without introducing internal contradictions between what geometry says and what growth says.
That is the question addressed in Energy-Flow Cosmology: Unified Analysis of BAO, SN Ia, and RSD with a Derived Effective Gravitational Coupling — a preprint that puts EFC through its first multi-probe consistency test.
The setup
The analysis combines three major classes of cosmological data. Baryon acoustic oscillations (BAO) measure the expansion history through a characteristic distance scale imprinted in the galaxy distribution. Type Ia supernovae (SN Ia) provide an independent distance ladder via calibrated standard candles. And redshift-space distortions (RSD) measure the rate at which cosmic structure is growing, through the observable fσ₈(z).
Figure 1 · Effective gravitational coupling μ(a)
The derived coupling μ(a) = 1 + β·S(a) with β = 0.16. At high redshift, S(a) → 0 and GR is recovered exactly. Enhancement grows as structure forms.
S(a) derived from EFC field equation; γ = 0 (ΛCDM background preserved)
Figure 2 · χ² breakdown by probe
Per-probe χ² for ΛCDM (dark) and EFC (blue). Geometry probes identical; difference from RSD only.
β = 0.16, γ = 0, transition hyperparameters fixed a priori
Figure 3 · Three-probe consistency architecture
EFC preserves ΛCDM background (γ = 0), leaving geometry probes untouched. Only growth is modified through μ(a).
DOI: 10.6084/m9.figshare.31215613
In ΛCDM, these three probes are fitted with a single set of cosmological parameters. The geometry (distances) and the growth (structure formation) are locked together by general relativity. Any modification to gravity must respect this lock — or explain why it breaks.
What EFC does differently
In Energy-Flow Cosmology, the effective gravitational coupling is not a free function bolted onto the Friedmann equations. It is derived from the EFC field equation via an entropy field S(a) that tracks the structural evolution of the universe. The modification takes the form μ(a) = 1 + β·S(a), where μ is the ratio of the effective gravitational constant to Newton's G, and β is a single amplitude parameter.
Crucially, the background expansion is left identical to ΛCDM by construction. The parameter γ, which would modify the Friedmann equation, is set to zero. This means that BAO distances and supernova magnitudes are fitted using exactly the same expansion history as the standard model. The only place EFC differs is in the growth sector — the rate at which density perturbations amplify under the modified coupling.
This is a deliberate design choice. Rather than trying to improve the distance fits (where ΛCDM already works very well), EFC asks whether a derived coupling can produce consistent growth predictions without breaking anything in the geometry sector.
The results
With one free parameter (β = 0.16) and transition hyperparameters fixed a priori, the unified fit yields χ²_total = 51.1 for EFC, compared to 49.4 for ΛCDM — a difference of Δχ² = +1.7. This means ΛCDM is marginally preferred in raw goodness-of-fit, but the difference is well within statistical noise.
The RSD fit is slightly worse under EFC, which is expected: the derived coupling enhances late-time growth in a specific, non-tunable way, and the current RSD data from BOSS DR12 happen to sit close to the ΛCDM prediction at the redshifts probed. There is no tension, no anomaly, and no statistical preference in either direction at this level of data.
What matters is the absence of internal contradiction. The geometry probes (BAO + SN Ia) are unaffected because the background is identical. The growth probe (RSD) is affected by the modified coupling but remains compatible. EFC does not introduce a geometry–growth split — the kind of internal tension that would immediately disqualify a modified gravity model.
Why this matters
The significance of this paper is structural rather than statistical. It establishes that the EFC coupling, when derived from the field equation rather than freely fitted, does not break multi-probe consistency at the current level of observational precision. This is a necessary condition for any viable modification to gravity, and many proposed alternatives fail exactly this test.
It also sets the stage for more discriminating analyses. The derived coupling μ(a) makes specific, testable predictions for how growth should differ from ΛCDM at different redshifts. As data from DESI, Euclid, and the Vera Rubin Observatory come online, the predicted growth enhancement will be tested at higher precision and across a wider redshift range. The unified framework ensures that any future detection — or non-detection — can be interpreted consistently across all probes.
Honest caveats
The analysis uses a ΛCDM background throughout, so it does not test a fully self-consistent EFC cosmology. The entropy field S(a) is specified functionally, not computed from a Boltzmann solver. Neutrino mass degeneracies, which could shift the σ₈ landscape, are not explored. And the transition hyperparameters, while fixed a priori, have not been marginalized over — a more conservative analysis would treat them as nuisance parameters.
These are all directions for future work. The present paper asks a simpler question: does the derived coupling survive first contact with multi-probe data? The answer is yes — not spectacularly, not with any claim of superiority over ΛCDM, but cleanly and without contradiction.
Sometimes that is exactly the result you need.
Preprint available on Figshare: DOI: 10.6084/m9.figshare.31215613
Three Probes, One Coupling: A Unified Test of Energy-Flow Cosmology
About the Author
