High-Resolution Measurements of the Halos of Four Dark Matter-Dominated Galaxies: Deviations from a Universal Density Profile

TL;DR

This study uses high-resolution, two-dimensional velocity fields (H$\alpha$ and CO) and multiwavelength imaging to derive precise rotation curves for five nearby, low-mass spiral galaxies and to infer their central dark matter density profiles. By subtracting baryonic disks under plausible mass-to-light ratios and testing multiple density profiles (power laws, NFW, pseudo-isothermal, and a newer $\eta$-dependent form), the authors find a wide range of inner slopes from $\alpha_{DM}=0$ to $1.20$ with a mean of $\langle\alpha_{DM}\rangle \approx 0.73$ and a scatter of $\sigma_{\alpha}\approx0.44$, challenging the idea of a universal density profile. One galaxy, NGC 5963, exhibits a cusp consistent with CDM predictions, while the others favor shallower or more varied cores, though none constitute a definitive crisis for CDM given potential baryonic and modeling uncertainties. The paper also detects radial motions in four galaxies and uses them to place lower limits on orbital ellipticity, arguing that halo triaxiality can influence, but not fully account for, the observed diversity in inner halo structure. Overall, the results imply substantial halo-to-halo variation and suggest that improved simulations incorporating baryonic physics and halo geometry are needed to reconcile observations with CDM expectations.

Abstract

We derive rotation curves for four nearby, low-mass spiral galaxies and use them to constrain the shapes of their dark matter density profiles. This analysis is based on high-resolution two-dimensional Halpha velocity fields of NGC 4605, NGC 5949, NGC 5963, and NGC 6689 and CO velocity fields of NGC 4605 and NGC 5963. In combination with our previous study of NGC 2976, the full sample of five galaxies contains density profiles that span the range from alpha_dm = 0 to alpha_dm = 1.20, where alpha_dm is the power law index describing the central density profile. The scatter in alpha_dm from galaxy to galaxy is 0.44, three times as large as in Cold Dark Matter (CDM) simulations, and the mean density profile slope is alpha_dm = 0.73, shallower than that predicted by the simulations. These results call into question the hypothesis that all galaxies share a universal dark matter density profile. We show that one of the galaxies in our sample, NGC 5963, has a cuspy density profile that closely resembles those seen in CDM simulations, demonstrating that while galaxies with the steep central density cusps predicted by CDM do exist, they are in the minority. In spite of these differences between observations and simulations, the relatively cuspy density profiles we find do not suggest that this problem represents a crisis for CDM. Improving the resolution of the simulations and incorporating additional physics may resolve the remaining discrepancies. We also find that four of the galaxies contain detectable radial motions in the plane of the galaxy. We investigate the hypothesis that these motions are caused by a triaxial dark matter halo, and place lower limits on the ellipticity of the orbits in the plane of the disk of 0.043 - 0.175.

Figures (9)

• Figure 1: Three-color images of our four target galaxies. Each image uses the asinh scaling recommended by lupton to retain an optimal combination of color and intensity information. (a) $BVR$ image of NGC 4605 from Lowell Observatory. The southeast of the galaxy is clearly brighter than the remainder. (b) $BVR$ image of NGC 5949 from Lick Observatory. The blacked-out rectangle on the east side of the galaxy is where a large dust grain was not completely removed by flatfielding. We masked out the affected area for the photometric analysis. (c) $BVR$ image of NGC 5963 from Lowell Observatory. (d) $r^{\prime} \hbox{H$\alpha$} i^{\prime}$ image of NGC 6689 from Kitt Peak National Observatory.
• Figure 2: (a) Optical and near-infrared surface brightness profiles of NGC 4605. In each band, the nucleus and exponential disk are visible. The black solid line represents an exponential fit to the $I$-band profile; the maximum deviation from the data is 0.13 magnitudes. (b) Optical and near-infrared surface brightness profiles of NGC 5949. In each band, the nucleus, exponential inner disk, and exponential outer disk are all visible. (c) Optical and near-infrared surface brightness profiles of NGC 5963. (d) Optical and near-infrared surface brightness profiles of NGC 6689. For all galaxies, the $J$ and $K_{\hbox{s}}$ data can be traced further out, but we do not plot the data beyond where the uncertainties reach a factor of two (0.75 mag). The $H$-band profiles of each galaxy have been omitted for clarity; the error bars for $H$ and $K_{\hbox{s}}$ overlap at most radii. The vertical dashed lines in panels b), c), and d) emphasize breakpoints in the profiles. The galaxies are generally well described by exponential disks, except for NGC 5963, which has an unusual and difficult to characterize surface brightness profile. We speculate that this galaxy may have evolved a dense, bulgelike central region through secular processes (a pseudobulge).
• Figure 3: Velocity fields of the four galaxies discussed in this paper. (a) Combined H$\alpha$ (circles) and CO (filled-in) velocity field of NGC 4605. (b) H$\alpha$ velocity field of NGC 5949. (c) Combined H$\alpha$ and CO velocity field of NGC 5963. (d) H$\alpha$ velocity field of NGC 6689.
• Figure 4: (a) Tilted-ring model for NGC 4605. The thick blue curve represents the rotation curve, and the thick red and orange curves show the radial velocity curve and the systemic velocities, respectively. Starting at a radius of 72 the systemic velocities were held fixed at zero. The shaded gray regions surrounding the rotation and radial velocity curves represent the combined systematic and statistical $1\sigma$ uncertainties from the bootstrap and Monte Carlo analysis. Note that for convenience we have plotted the radial velocities as positive; this choice does not indicate whether the radial motions are directed inwards or outwards. The thin black curve is a power law fit to the rotation curve. The residuals from this fit are plotted in the lower panel, and the $1\sigma$ and $2\sigma$ scatter of the data points around the fit is shown by the shaded gray areas. (b) Disk-subtracted rotation curve of NGC 4605. The gray points are the original rotation velocities from panel (a), and the black points are the dark matter halo rotation curve after subtracting the stellar disk ($\hbox{$M_{*}/L_{R}$} = 1.01\hbox{$M_{\odot}/L_{\odot, R}$}$), which is shown as a gray dashed line. The thick green, cyan, and magenta curves show power law, NFW, and pseudo-isothermal fits to the halo rotation curve, respectively. The residuals from these fits are displayed in the lower panel, and the $1\sigma$ and $2\sigma$ scatter of the data points around the power law fit is shown by the shaded gray areas.
• Figure 5: (a) Tilted-ring model for NGC 5949. (b) Disk-subtracted rotation curve of NGC 5949 (for $\hbox{$M_{*}/L_{R}$} = 1.64\hbox{$M_{\odot}/L_{\odot, R}$}$). Symbols, colors, and shading are as in Figure \ref{['rcfig4605']}.