Constraints on Sterile Neutrino Dark Matter

TL;DR

This work analyzes sterile neutrinos as dark matter by comparing non-resonant and resonant production scenarios against X-ray radiative-decay bounds and small-scale structure constraints from the Ly$\alpha$ forest. It shows that the standard non-resonant, zero-lepton-number production is excluded when combining Ly$\alpha$ forest limits with conservative X-ray background bounds, and that entropy-dilution scenarios cannot reopen this window. In contrast, resonant production in the presence of a non-zero lepton number $L>0$ remains allowed, particularly at higher masses $m_s$; the viability of such scenarios depends on the detailed lepton asymmetry and transfer-function effects. Overall, if the X-ray and Ly$\alpha$ constraints are robust, only non-zero $L$ resonant production remains a viable oscillation-based mechanism for sterile neutrino dark matter, with dilution models failing to salvage the standard picture.

Abstract

We present a comprehensive analysis of constraints on the sterile neutrino as a dark matter candidate. The minimal production scenario with a standard thermal history and negligible cosmological lepton number is in conflict with conservative radiative decay constraints from the cosmic X-ray background in combination with stringent small-scale structure limits from the Lyman-alpha forest. We show that entropy release through massive particle decay after production does not alleviate these constraints. We further show that radiative decay constraints from local group dwarf galaxies are subject to large uncertainties in the dark matter density profile of these systems. Within the strongest set of constraints, resonant production of cold sterile neutrino dark matter in non-zero lepton number cosmologies remains allowed.

Figures (3)

• Figure 1: Full parameter space constraints for the sterile neutrino production models, assuming sterile neutrinos constitute the dark matter. Contours labeled with lepton number $L=0$, $L=0.003$, $L=0.01$, $L=0.1$ are production predictions for constant comoving density of $\Omega_s=0.24$ for $L=0$, and $\Omega_s=0.3$ for non-zero $L$Abazajian:2002yz. Constraints from X-ray observations include the diffuse X-ray background (green) Boyarsky:2005us, from XMM-Newton observations of the Coma and Virgo clusters (light blue) Boyarsky:2006zi. The diagonal wide-hatched region is the claimed potential constraint from XMM-Newton observations of the LMC Boyarsky:2006fg. The region at $m_s<0.4\rm\ keV$ is ruled out by a conservative application of the Tremaine-Gunn bound Bode:2000gq. The regions labeled Ly$\alpha$ are those from the amplitude and slope of matter power spectrum inferred from the SDSS Ly$\alpha$ forest [Ly$\alpha$ (1)] Abazajian:2005xn, using high-resolution Ly$\alpha$ data [Ly$\alpha$ (2)] Viel:2005qjAbazajian:2005xn, and that from the high-$z$ SDSS Ly$\alpha$ of SMMT [Ly$\alpha$ (3)] Seljak:2006qw. The grey region to the right of the $L=0$ case is where sterile neutrino dark matter is overproduced. Also shown is the horizontal band of the mass scale consistent with producing a 100 - 300 pc core in the Fornax dwarf galaxy Strigari:2006ue. The parameters consistent with pulsar kick generation are in horizontal hatching Kusenko:1998bkFuller:2003gyKusenko:2004mm.
• Figure 2: The value of the quantity $J [\Delta \Omega(\theta)]$ (see text) for two representative dark matter distributions in the Draco dwarf galaxy. The thickness of both curves corresponds to the range of values in each profile due to the distance uncertainties to Draco. The top curve corresponds to an NFW profile with $\log(\rho_s/M_\odot {\rm kpc}^{-3}) = 7.0$ and $\log (r_s/{\rm kpc}) = 0.85$, while the lower curve depicts a Burkert profile with $\log (\rho_s/M_\odot {\rm kpc}^{-3}) = 9.0$ and $\log( r_s /{\rm kpc})= -0.75$.
• Figure 3: Shown here are the constraints on the massive particle decay dilution model. The diagonally-hatched (blue) region is the lower-mass Ly$\alpha$ limit of Ref. Abazajian:2005xn, while the vertically (red) hatched region the Ly$\alpha$ limit of SMMT. In combination with the conservative XRB limit (green) Boyarsky:2005us, even extreme dilution models of $S=100$ are in conflict with combined constraints. The standard case of no dilution corresponds to $S=1$.