Do WMAP data favor neutrino mass and a coupling between Cold Dark Matter and Dark Energy?

TL;DR

The study investigates whether cosmologies with a self-interacting dark energy scalar field, allowing a CDM–DE coupling $β$ and nonzero neutrino masses $M_ν$, can fit current data as well as or better than the uncoupled case. Using a full MCMC analysis with CosmoMC/CAMB against WMAP5, 2dFGRS, and SNLS data, the authors explore RP and SUGRA tracker potentials and quantify degeneracies between $β$, $M_ν$, and the DE scale $Λ$. They find that permitting $β$ and $M_ν$ relaxes the cosmological neutrino-mass bounds by roughly a factor of two, with best-fit values around $β≃0.07$ and $M_ν≃0.35$ eV, while still allowing substantial coupling within the 95% confidence region. Although the data do not demand a nonzero coupling, this coupled-dark-sector framework remains a viable alternative that could alleviate fine-tuning and coincidence problems and may be testable with future neutrino measurements or tighter cosmological priors.

Abstract

Within the frame of cosmologies where Dark Energy (DE) is a self--interacting scalar field, we allow for a CDM--DE coupling and non--zero neutrino masses, simultaneously. In their 0--0 version, i.e. in the absence of coupling and neutrino mass, these cosmologies provide an excellent fit to WMAP, SNIa and deep galaxy sample spectra, at least as good as \LambdaCDM. When the new degrees of freedom are open, we find that CDM--DE coupling and significant neutrino masses (~0.1eV per νspecies) are at least as likely as the 0--0 option and, in some cases, even statistically favoured. Results are obtained by using a Monte Carlo Markov Chain approach.

Figures (10)

• Figure 1: Transfer functions (left) and angular anisotropy spectrum (right) in cosmologies with/without coupling and with/without 2 massive $\nu$'s (00/CM models). Coupling and mass are selected so to yield an approximate balance. The transfer functions are multiplied by $k^{1.5}$, to help the reader to distinguish different cases. In the lower frame of the $C_l$ plot we also give the spectral differences between 00-- and CM--models, hardly visible in the upper frame. Here dotted lines represent the cosmic variance interval.
• Figure 2: State parameter and its variation in uncoupled RP models. The plot is for $h=0.7$, $\Omega_b=0.046$, $\Omega_c= 0.209$.
• Figure 3: State parameter in models with $w(a)$ given by eq. (\ref{['wa']}) and in a uncoupled SUGRA model with $\Lambda = 0.1\,$GeV. $w_o$ and $w'$ values (in the frame) selected to yield a behavior close to SUGRA, by requiring similar high--$z$ plateau and $w(0)$. Although renouncing to a full coincidence at $z=0$, the fast variablility of $w(a)$ in uncoupled SUGRA cannot be met by any polinomial $w(a)$. The plot is for $h=0.7$, $\Omega_b=0.046$, $\Omega_c= 0.209$.
• Figure 4: State parameter in coupled SUGRA model with $\Lambda = 0.1\,$GeV, $h=0.7$, $\Omega_b=0.046$, $\Omega_c= 0.209$, as in Figure \ref{['SU']}. The uncoupled behavior given there is reported also here.
• Figure 5: State parameter in coupled RP for various $\Lambda$ values, in the case $\beta = 0.1\,$. Here $h=0.7$, $\Omega_b=0.046$, $\Omega_c= 0.209$, as in Figure \ref{['RP']}. The uncoupled behavior at $z=0$ given there is reported also here.