Update on Coupled Dark Energy and the $H_0$ tension
Adrià Gómez-Valent, Valeria Pettorino, Luca Amendola
TL;DR
This work tests a coupled dark energy model with a Peebles-Ratra potential against the Planck 2018 data and an array of late-time probes to reassess constraints on the dark-sector coupling $β$ and potential slope $α$, and to evaluate the impact of SH0ES and H0LICOW on the inferred parameters. The authors implement the model in CLASS, perform Bayesian model comparison with ΛCDM across multiple data combinations, and find data-dependent peaks in $β$ and $α$ that are often diminished by CMB lensing; overall, ΛCDM remains preferred when the full data set is considered. While including SH0ES+H0LICOW modestly shifts the best-fit toward nonzero coupling and a steeper potential, the Bayes factor consistently disfavors the additional CDE parameters, and the inferred Hubble tension is not fully resolved. The results imply that, with current data, the simplest ΛCDM description remains robust, though exploring extended couplings or alternative interaction schemes could still offer marginal improvements for late-time tensions.
Abstract
In this work we provide updated constraints on coupled dark energy (CDE) cosmology with Peebles-Ratra (PR) potential and constant coupling strength $β$. This modified gravity scenario introduces a fifth force between dark matter particles, mediated by a scalar field that plays the role of dark energy. The mass of the dark matter particles does not remain constant, but changes with time as a function of the scalar field. Here we focus on the phenomenological behavior of the model, and assess its ability to describe updated cosmological data sets that include the Planck 2018 cosmic microwave background (CMB) temperature, polarization and lensing, baryon acoustic oscillations, the Pantheon compilation of supernovae of Type Ia, data on $H(z)$ from cosmic chronometers, and redshift-space distortions. We also study which is the impact of the local measurement of $H_0$ from SH0ES and the strong-lensing time delay data from the H0LICOW collaboration on the parameter that controls the strength of the interaction in the dark sector. We find a peak corresponding to a coupling $β> 0$ and to a potential parameter $α> 0$, more or less evident depending on the data set combination. We show separately the impact of each data set and remark that it is especially CMB lensing the one data set that shifts the peak the most towards $Λ$CDM. When a model selection criterion based on the full Bayesian evidence is applied, however, $Λ$CDM is still preferred in all cases, due to the additional parameters introduced in the CDE model.
