The core density of dark matter halos: a critical challenge to the Lambda-CDM paradigm?

TL;DR

The paper tests whether ΛCDM halos can reproduce the inner mass distribution of disk galaxies by combining high-resolution N-body halos with dynamical constraints from the Milky Way and the TF relation. It demonstrates that halos with $V_{200}\approx 220\,\mathrm{km\,s^{-1}}$ contain roughly $3\times$ more dark mass inside $R_0$ than observations permit, and that even maximal baryon conversion cannot move the TF zero-point into agreement unless $M/L_I$ is unrealistically low or halo concentrations are significantly reduced. The authors argue that resolving this tension would require substantial revisions to the standard ΛCDM model, such as altering the small-scale power spectrum (e.g., a tilt) or invoking exotic DM properties, with potential conflicts for high-redshift galaxy formation and cluster abundances. Overall, the work highlights a significant tension between CDM halo structure and the observed properties of disk galaxies, prompting reevaluation of galaxy formation within the ΛCDM paradigm.

Abstract

We compare the central mass concentration of Cold Dark Matter halos found in cosmological N-body simulations with constraints derived from the Milky Way disk dynamics and from the Tully-Fisher relation. For currently favored values of the cosmological parameters ($Ω_0 \sim 0.3$; $Λ_0=1-Ω_0 \sim 0.7$; $h \sim 0.7$; COBE- and cluster abundance-normalized $σ_8$; Big-Bang nucleosynthesis $Ω_b$), we find that halos with circular velocities comparable to the rotation speed of the Galaxy have typically {\it three times} more dark matter inside the solar circle than inferred from observations of Galactic dynamics. Such high central concentrations of dark matter on the scale of galaxy disks also imply that stellar mass-to-light ratios much lower than expected from population synthesis models must be assumed in order to reproduce the zero-point of the Tully-Fisher relation. Indeed, even under the extreme assumption that {\it all} baryons in a dark halo are turned into stars, disks with conventional $I$-band stellar mass-to-light ratios ($M/L_I \sim 2 \pm 1 (M/L_I)_{\odot}$) are about two magnitudes fainter than observed at a given rotation speed. We examine several modifications to the $Λ$CDM model that may account for these discrepancies and conclude that agreement can only be accomplished at the expense of renouncing other major successes of the model. Reproducing the observed properties of disk galaxies thus appears to demand substantial revision to the currently most successful model of structure formation.

Figures (2)

• Figure 1: Dark mass enclosed within a radius $R_{o}= 8.5$ kpc, the Sun's distance from the center of the Milky Way, versus the circular velocities of s$\Lambda$CDM halos. The shaded region highlights the allowed parameters of the dark halo surrounding the Milky Way, as derived from observations of Galactic dynamics and by assuming that the disk mass cannot exceed the total baryonic content of the halo. The filled circles show the loci of s$\Lambda$CDM halos as determined from high-resolution N-body simulations. The solid line is the circular velocity dependence of the dark mass expected inside $R_{o}$ for halos that follow the density profile proposed by NFW96 and NFW97. The circular velocity dependence of the NFW "concentration" parameter of the simulated halos is well approximated on these scales by $c \approx 20 \, (V_{200}/100 {\rm \, km \, s}^{-1})^{-1/3}$ (dotted line). This is slightly higher than predicted by the approximate formula proposed by NFW97 but consistent with their published results.
• Figure 2: The $I$-band Tully Fisher relation compared with the loci of hypothetical exponential disk galaxies assumed to assemble at the center of three representative s$\Lambda$CDM halos. Dots are a compilation of the data by Giovanelli et al. (1997), Mathewson, Ford & Buchhorn (1992) and Han & Mould (1992). The solid line is the best fit to the data advocated by Giovanelli et al. The hypothetical galaxies have radii consistent with observations and move from left to right along each curve (labeled by the circular velocity of the halo at the virial radius) as the disk mass increases, under the assumption of a constant stellar mass-to-light ratio, $M/L_I=2$ in solar units. The starred symbols correspond to the maximum disk mass, $M_{\rm disk}^{\rm max}$, allowed by the universal baryon fraction of the s$\Lambda$CDM model. Open squares are N-body gasdynamical simulations of the formation of galaxies within these halos. Error bars correspond to two different choices of IMF, as discussed in detail by Steinmetz & Navarro (1998). See that paper for details.