(CUGE + REFORM + ASH, SI base units enforced)
The total action is
with Lagrangian density (all terms in J m⁻³ = kg m⁻¹ s⁻²)
CUGE vacuum response (dimensionless \(n\)):
\(\Phi\) is the positive gravitational potential magnitude \([\Phi]=\rm m^2 s^{-2}\), \(c=299\,792\,458\) m s⁻¹ (exact, invariant), \(G=6.67430\times10^{-11}\) m³ kg⁻¹ s⁻².
Dimensional verification of each term (SI base units):
All consistent; no ε₀c²|∇Φ|² (original error) or unit mixing.
Recovery of Maxwell equations (variation w.r.t. \(\mathbf{A}\)):
U(1) gauge invariance is preserved. The Euler-Lagrange equations yield the macroscopic Maxwell equations in the responsive medium:
with local impedance \(Z_0=\sqrt{\mu(\Phi)/\varepsilon(\Phi)}\) invariant (no reflections). Wave speed \(c_{\rm coord}=c/n(\Phi)\). (Standard derivation, unchanged from variable-medium electrodynamics.)
Recovery of Poisson equation with VSS (variation w.r.t. \(\Phi\)):
The gravitational part of \(\mathcal{L}\) is exactly the standard Newtonian action. Variation yields
The VSS energy density is identified as
so the sourced Poisson equation is
(The EM corrections \(\propto \varepsilon' E^2\), \(\mu' B^2\) are negligible in weak fields, matching all C.O.R.E. documents.) The vacuum strain contributes positively to effective mass density, as required by ZEUS/CUGE.
Recovery of the ray equation (eikonal limit, REFORM v3):
From the EM sector, the wave equation in the geometric-optics limit reduces to Fermat’s principle \(\delta\int n\,ds=0\). Euler-Lagrange variation directly gives the ray equation:
\([n]\) dimensionless, local \(\nabla n\) (m⁻¹), integrated phase dimensionless, \([\ddot{\mathbf{r}}]=\rm m\,s^{-2}\).
Strong-field regularity: Finite MACHO radius \(\varepsilon\) in \(\Phi = GM/\sqrt{r^2+\varepsilon^2}\) keeps \(n\) finite and \(C^\infty\) everywhere; no curvature singularities.
This action is the variational linchpin: it reproduces every equation in the attached documents, supplies the energy-momentum tensor via Noether, closes the framework, and satisfies all style-guide constraints (dimensionless \(n\), SI base units only, local vs. integrated distinction). No free parameters beyond CUGE calibration. All prior versions superseded.