Interacting Dark Sectors in light of DESI DR2

TL;DR

This work re-examines interacting dark matter–dark energy (iDMDE) scenarios using DESI DR2 BAO data in combination with Planck 2018 and Pantheon+ SNIa, exploring both background evolution and perturbations. By employing CPL and JBP EoS parametrizations and separating phantom ($w(z)<-1$) from non-phantom regimes to avoid perturbation instabilities, the study finds that phantom iDMDE can alleviate the $S_8$ tension and moderate early-time EoS evolution toward $w\sim -1$ without worsening the $H_0$ tension; however, Bayesian evidence does not decisively favor interacting models over non-interacting ones. In the non-phantom sector, constraints are weaker and $S_8$ remains large with lower precision, indicating less tension relief. Overall, DESI DR2 strengthens the case for phantom iDMDE as a promising route to address cosmological tensions while highlighting the need to explore alternative interaction forms and the continued importance of perturbation analyses with future data.

Abstract

Possible interaction between dark energy and dark matter has previously shown promise in alleviating the clustering tension, without exacerbating the Hubble tension, when Baryon Acoustic Oscillations (BAO) data from the Sloan Digital Sky Survey (SDSS) DR16 is combined with Cosmic Microwave Background (CMB) and Type-Ia Supernovae (SNIa) data sets. With the recent Dark Energy Spectroscopic Instrument (DESI) BAO DR2, there is now a compelling need to re-evaluate this scenario. We combine DESI DR2 with Planck 2018 and Pantheon+ SNIa data sets to constrain interacting dark matter dark energy models, accounting for interaction effects in both the background and perturbation sectors. Our results exhibit similar trends to those observed with SDSS, albeit with improved precision, reinforcing the consistency between the two BAO data sets. In addition to offering a resolution to the $S_8$ tension, in the phantom-limit, the dark energy equation of state exhibits an early-phantom behaviour, aligning with DESI DR2 findings, before transitioning to $w\sim-1$ at lower redshifts, regardless of the DE parametrization. However, the statistical significance of excluding $w=-1$ is reduced compared to their non-interacting counterparts.

Figures (4)

• Figure 1: Constraints on and correlations between $H_0$ and $\sigma_{8,0}$, as well as $w_0$ and $w_a$, are shown for the models considered in this work. The non-interacting cases smoothly incorporate $w = -1$ crossing using the PPF formalism. In contrast, the interacting cases divide the parameter space into phantom and non-phantom regions - only the phantom case is shown here.
• Figure 2: Plots for the evolution of DE EoS for interacting CPL and JBP models. The best-fitting line is shown along with the 1$\sigma$ and 2$\sigma$ confidence intervals in the shaded regions.
• Figure 3: Comparison of constraints obtained for interacting (phantom regime) and non-interacting models considered in this work, using combined Planck 2018 + DESI DR2 BAO + Pantheon+ data sets.
• Figure 4: Comparison of constraints obtained for interacting (non-phantom regime) and non-interacting models considered in this work, using combined Planck 2018 + DESI DR2 BAO + Pantheon+ data sets.