Constraints on the Coupling between Dark Energy and Dark Matter from CMB data

TL;DR

This work tests a phenomenological DM-DE coupling implemented as $Q = \xi \mathcal{H} \rho_\Lambda$ within a flat cosmology, comparing a DM→DE (MOD1) and a DE→DM (MOD2) scenario against Planck CMB data plus low- and high-redshift probes. Using CAMB/CosmoMC and datasets including Planck TT/lowP, highP, lensing, JLA SN, and BAO/RSD, the authors find that MOD2 better accommodates the $H_0$ and $\sigma_8$ tensions, while MOD1 tends to worsen them due to a larger early DM density and enhanced clustering. Extending the analysis to include a sterile neutrino DM component shows robust constraints on the coupling and negligible qualitative changes, with $H_0$ nudged higher and $\sigma_8$ lower. Overall, a non-zero DE→DM coupling appears to be a viable modification to $\\Lambda$CDM that can coherently link high- and low-redshift cosmological observations.

Abstract

We investigate a phenomenological non-gravitational coupling between dark energy and dark matter, where the interaction in the dark sector is parameterized as an energy transfer either from dark matter to dark energy or the opposite. The models are constrained by a whole host of updated cosmological data: cosmic microwave background temperature anisotropies and polarization, high-redshift supernovae, baryon acoustic oscillations, redshift space distortions and gravitational lensing. Both models are found to be compatible with all cosmological observables, but in the case where dark matter decays into dark energy, the tension with the independent determinations of $H_0$ and $σ_8$, already present for standard cosmology, increases: this model in fact predicts lower $H_0$ and higher $σ_8$, mostly as a consequence of the higher amount of dark matter at early times, leading to a stronger clustering during the evolution. Instead, when dark matter is fed by dark energy, the reconstructed values of $H_0$ and $σ_8$ nicely agree with their local determinations, with a full reconciliation between high- and low-redshift observations. A non-zero coupling between dark energy and dark matter, with an energy flow from the former to the latter, appears therefore to be in better agreement with cosmological data.

Figures (8)

• Figure 1: Dependence of the CMB angular power spectrum $C_l$ on two cosmological parameters: the DM-DE coupling strength $\xi$ (upper panel); the DM energy density today $\Omega_c h^2$ (lower panel). All the other parameters are kept fixed. The black curve is the same in the different panels. The plot shows the degeneracy between $\xi$ and $\Omega_c h^2$.
• Figure 2: Marginalized 1$\sigma$ , 2$\sigma$ and 3$\sigma$ C.L. limits on $\Omega_ch^2$ and $\Omega_\Lambda$ for the "PlanckTT+lowP" dataset and for the three different cosmological models: $\Lambda$CDM, MOD1 and MOD2.
• Figure 3: Marginalized 1$\sigma$ , 2$\sigma$ and 3$\sigma$ C.L. limits for the parameters $\xi$ and $w_\Lambda$ in MOD1, for different datasets. When the error bar is not visible, it coincides with the limit in the prior, as listed in Tab. \ref{['tab:priorscde']}. The red point and lines refer to the MOD1$+\nu_s$ model, discussed in Section \ref{['sec:sterilenuDM']}.
• Figure 4: Marginalized 1$\sigma$, 2$\sigma$ and 3$\sigma$ C.L. limits for the parameters $\xi$ and $w_\Lambda$ in MOD2, for different datasets. When the error bar is not visible, it coincides with the limit in the prior, as listed in Tab. \ref{['tab:priorscde']}. The red point and line refer to the MOD2$+\nu_s$ model, discussed in Section \ref{['sec:sterilenuDM']}.
• Figure 5: Marginalized 1$\sigma$ and 2$\sigma$ C.L. allowed regions in the ($\xi$, $w_\Lambda$) plane in MOD1 (left) and MOD2 (right), for different datasets. The area below the dashed lines ($w_\Lambda^{\rm{eff}}=w_\Lambda+\xi/3=-1$) corresponds to an increasing energy density for DE in the future.