Neutrino and axion hot dark matter bounds after WMAP-7

TL;DR

This work updates cosmological bounds on hot dark matter from neutrinos and hadronic axions using WMAP-7, the SDSS-DR7 halo power spectrum, BAO measurements, and HST $H_0$ priors within an eight-parameter flat, adiabatic framework that allows $\sum m_\nu$ and $m_a$ to vary. Through MCMC analysis, it finds the strongest joint constraints with all data: $\sum m_\nu<0.41$ eV and $m_a<0.72$ eV (95% C.L.), with tighter or looser bounds under fixed $m_a=0$ or fixed $\sum m_\nu=0$. The study explains why CMB data alone constrain $\sum m_\nu$ but not $m_a$: for the same hot-dark-matter fraction, axions must be heavier than neutrinos, and axions are nonrelativistic at recombination, reducing their imprint on the CMB. Consequently, incorporating small-scale structure information, particularly the halo power spectrum, is essential to bound $m_a$; Planck is unlikely to dramatically improve the $m_a$ bound, although it will refine $\sum m_\nu$ limits. Overall, cosmological bounds on hot dark matter complement laboratory searches, with caveats for nonstandard cosmologies that suppress axion production.

Abstract

We update cosmological hot dark matter constraints on neutrinos and hadronic axions. Our most restrictive limits use 7-year data from the Wilkinson Microwave Anisotropy Probe for the cosmic microwave background anisotropies, the halo power spectrum (HPS) from the 7th data release of the Sloan Digital Sky Survey, and the Hubble constant from Hubble Space Telescope observations. We find 95% C.L. upper limits of \sum m_ν<0.44 eV (no axions), m_a<0.91 eV (assuming \sum m_ν=0), and \sum m_ν<0.41 eV and m_a<0.72 eV for two hot dark matter components after marginalising over the respective other mass. CMB data alone yield \sum m_ν<1.19 eV (no axions), while for axions the HPS is crucial for deriving m_a constraints. This difference can be traced to the fact that for a given hot dark matter fraction axions are much more massive than neutrinos.

Figures (4)

• Figure 1: Axion number density $n_a$ as a function of the axion mass $m_a$ assuming the hadronic thermalisation processes (\ref{['eq:axionthermalisation']}) based on the calculations of reference Hannestad:2005df.
• Figure 2: 2D marginal 68% and 95% contours in the $\sum m_\nu$--$m_a$ plane. The blue lines correspond to our results using CMB+HPS, and the red lines using CMB+HPS+HST.
• Figure 3: CMB temperature anisotropy spectra for various mixtures of cold and hot dark matter, where $\omega_{\rm cdm}+\omega_\nu+\omega_a=0.112$ is held constant. Bottom: Differences to standard $\Lambda$CDM; the shaded/grey area indicates cosmic variance. Light blue/dot-dash line: standard $\Lambda$CDM ($\omega_{\rm cdm}=0.112$, $\omega_a=0$, $\omega_\nu=0$). Red/solid line: $\Lambda$CDM+$\nu$ with $\sum m_\nu = 1.2~{\rm eV}$ ($\omega_{\rm cdm}=0.099$, $\omega_\nu=0.013$). Green/dash line: $\Lambda$CDM+$a$ with $m_a = 2.4~{\rm eV}$ ($\omega_{\rm cdm}=0.099$, $\omega_a=0.013$). Dark blue/dotted line: Extreme axion case with $m_a = 10~{\rm eV}$ ($\omega_{\rm cdm}=0.0498$, $\omega_a=0.0622$).
• Figure 4: Same as figure \ref{['fig:cls']}, but for the matter power spectrum.