MODEL-INDEPENDENT RECONSTRUCTION OF DARK ENERGY EVOLUTION FROM CMB-CALIBRATED EXPANSION HISTORY
DOI:
https://doi.org/10.65009/7z24qd93Keywords:
CMB distance priors; sound horizon calibration; nonparametric reconstruction; Gaussian process regression; spline expansion history; rho_DE(z); w(z) inference,,Abstract
Dark energy inference often inherits a strong prior from a chosen parametric form for w(z),
which can convert limited geometric information into apparently sharp statements about
time variation. A different route starts from the most precise early Universe ruler, calibrates
late time distances against it, and then reconstructs the expansion history with minimal
functional assumptions. Despite a large literature on nonparametric H(z) estimation, a
recurring gap remains: many reconstructions either treat the sound horizon calibration as
fixed, or separate it from the late time inference in a way that understates correlated
uncertainties. The present study develops a model independent reconstruction of dark
energy evolution that is explicitly anchored to CMB calibrated distances and that
propagates the calibration covariance into rho_DE(z) and w(z). The design combines CMB
distance priors from Planck 2018 and ACT DR6 with BAO ratios, Type Ia supernova
distances from Pantheon plus, and cosmic chronometer H(z) estimates. Expansion histories
are reconstructed using two complementary nonparametric families, Gaussian process
regression and constrained cubic splines, and are cross checked with redshift binned E(z)
inference. From the reconstructed E(z) we derive rho_DE(z) and an effective w(z) using
energy conservation, while marginalizing over Omega_m and curvature under controlled
priors. The analysis identifies where apparent departures from a cosmological constant are
driven by calibration, where they are driven by specific data subsets, and which redshift
ranges remain prior dominated. The contribution is an uncertainty disciplined workflow
that yields interpretable w(z) bands, clarifies the role of the CMB ruler, and provides
dataset level diagnostics for future surveys.
References
Aghanim, N., et al. (Planck Collaboration). (2020). Planck 2018 results VI: Cosmological
parameters. Astronomy and Astrophysics, 641, A6.
Louis, T., et al. (ACT Collaboration). (2025). The Atacama Cosmology Telescope: DR6
power spectra, likelihoods and LambdaCDM parameters. arXiv:2503.14452.
Adame, A. G., et al. (DESI Collaboration). (2024). DESI 2024 VI: Cosmological
constraints from the measurements of baryon acoustic oscillations. arXiv:2404.03002.
Scolnic, D., et al. (2022). The Pantheon+ analysis: The full dataset and light-curve release.
arXiv:2112.03863.
Bonilla, A., et al. (2020). Measurements of H0 and reconstruction of the late-time Universe
with Gaussian processes. arXiv:2011.07140.
Johnson, J. P., et al. (2025). Kernel dependence of Gaussian process reconstruction of the
late Universe expansion history. European Physical Journal C, 85, 14732.
Calderon, R., et al. (2024). Reconstructing dark energy using crossing statistics with DESI
BAO. Journal of Cosmology and Astroparticle Physics, 10, 048.
Lodha, K., et al. (2025). DESI constraints on physics-focused aspects of dark energy.
Physical Review D, 111, 023532.
