Modified cosmology through generalized mass-to-horizon entropy: observational constraints from DESI DR2 BAO data

TL;DR

The paper investigates a cosmological model in which a generalized mass-to-horizon entropy, characterized by a two-parameter set $n$ and $\gamma$, modifies the Friedmann equations via the gravity–thermodynamics correspondence. This framework introduces an effective dark-energy sector without adding an explicit component, recovering standard $\Lambda$CDM in the limit $n=\gamma=1$; the authors derive the modified equations and express the background evolution in terms of the model parameters. They constrain the model using late-time data from Pantheon+ SNIa, cosmic chronometers, and DESI DR2 BAO, with optional SH0ES prior on $H_0$, across four data combinations, and compare to $\Lambda$CDM via the Akaike Information Criterion. The results show near-standard behavior with best-fit $n\approx 0.92$–$0.95$ and $\gamma\approx 1.57$–$1.70$, and while the entropic model can fit comparably or slightly better, AIC mildly favors the cosmological constant due to extra degrees of freedom; the $D_4$ case hints at a possible reduction of the $H_0$ tension, though the absence of CMB data limits definitive conclusions. Overall, the generalized entropic cosmology remains viable and warrants further study in perturbations and multi-probe analyses.

Abstract

A generalized mass-to-horizon entropy has recently been proposed as an extension of the Bekenstein-Hawking area law, derived from a modified mass-horizon relation constructed to ensure consistency with the Clausius equation. Within the gravity-thermodynamics conjecture, this entropy formulation yields modified Friedmann equations, which recover the standard $Λ$CDM cosmology in the appropriate limit of the model's two free parameters. In the present study, we constrain this framework using observations from TypeIa supernovae (SNIa), cosmic chronometers (CC) and baryon acoustic oscillations (BAO, including DESI~DR2), together with the SH0ES distance-ladder prior on $H_0$, across four combinations of data sets. Although the extended entropic scenario yields a slightly better, or statistically comparable, fit to the data, model selection via the Akaike Information Criterion (AIC) mildly favors the cosmological constant as the dark energy candidate. Moreover, the $Λ$CDM limit lies within $\sim 1σ$ of our constraints, indicating no significant deviation from standard cosmology with current data.

Figures (1)

• Figure 1: Likelihood contours for the best-fit parameters of the generalized entropic model. We considered the data sets $\mathbf{D}_{2}$ and $\mathbf{D}_{4}$, respectively.