Dark Matter in Dwarf Galaxies: High Resolution Observations

ABSTRACT We present observations and analysis of rotation curves and dark matter halo density profiles in the central regions of four nearby dwarf galaxies. This observing program has been designed to overcome some of the limitations of other rotation curve studies that rely mostly on longslit spectra. We find that these objects exhibit the full range of central density profiles between constant density and NFW halos. This result suggests that there is a distribution of central density slopes rather than a unique halo density profile.

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We present a spectroscopic deprojection analysis of a sample of ten relaxed galaxy clusters. We use an empirical F-test derived from a set of Markov Chain Monte Carlo simulations to determine if the core plasma in each cluster could contain multiple phases. We derive non-parametric baryon density and temperature profiles, and use these to construct total gravitating mass profiles. We compare these profiles with the standard halo parameterizations. We find central density slopes roughly consistent with the predictions of LCDM: $-1 \lesssim d\log(\rho)/d\log(r) \lesssim -2$. We constrain the core size of each cluster and, using the results of cosmological simulations as a calibrator, place an upper limit of ~0.1 cm^2/g = 0.2 b(GeV/c^2)^{-1} (99% confidence) on the dark matter particle self-interaction cross section.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Self-interacting dark matter has been proposed as a hypothesis to explain the shallow central slopes of the density profiles of dark matter halos in galaxies. In order to be consistent with observational studies at scales of galaxy clusters, the cross section should scale inversely with the velocity of collision. In this paper we consider the mass density profile of the halo of the low surface brightness (LSB) galaxy NGC 5963 to place an upper limit on the dark matter cross section for collisions with velocities $\sim 150$ km s$^{-1}$, i.e. at the low velocity regime. After calibrating against cosmological simulations, we found that the large inferred dark matter concentration and central dark matter density in NGC 5963 are inconsistent with an effective collisional cross section per unit of mass >0.2 cm^2/g. Corrections were applied in order to account for reduction of the core by the adiabatic contraction caused by cooling baryons. Our limits that involve a number of simplifying, but always conservative, assumptions, exclude the last permitted interval for velocity-dependent cross sections to explain the flat density core in LSB galaxies. Implications for the nature of dark matter are also discussed. Comment: 13 pages, 4 figures. Accepted for publication in ApJ
    The Astrophysical Journal 06/2005; · 6.73 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We derive inner dark matter halo density profiles for a sample of 165 low-mass galaxies using rotation curves obtained from high-quality, long-slit optical spectra assuming minimal disks and spherical symmetry. For $\rho(r) \sim r^{-\alpha}$ near the galaxy center we measure median inner slopes ranging from $\alpha_m = 0.22 \pm 0.08$ to $0.28 \pm 0.06$ for various subsamples of the data. This is similar to values found by other authors, and in stark contrast to the intrinsic cusps ($\alpha_{int}\sim1$) predicted by simulations of halo assembly in cold dark matter (CDM) cosmologies. To elucidate the relationship between $\alpha_m$ and $\alpha_{int}$ in our data, we simulate long-slit observations of model galaxies with halo shapes broadly consistent with the CDM paradigm. Simulations with $\alpha_{int}=1/2$ and 1 recover both the observed distribution of $\alpha_m$ and correlations between $\alpha_m$ and primary observational parameters such as distance and disk inclination, whereas those with $\alpha_{int}=5/4$ are marginally consistent with the data. Conversely, the hypothesis that low-mass galaxies have $\alpha_{int}=3/2$ is rejected. While the simulations do not imply that the data favor intrinsic cusps over cores, they demonstrate that the discrepancy between $\alpha_m$ and $\alpha_{int}\sim1$ for our sample does not necessarily imply a genuine conflict between our results and CDM predictions: rather, the apparent cusp/core problem may be reconciled by considering the impact of observing and data processing techniques on rotation curves derived from long-slit spectra. Comment: 17 pages, 14 figures; uses emulateapj. Accepted for publication in AJ. Minor changes made to match AJ proofs
    The Astronomical Journal 02/2005; · 4.97 Impact Factor

Full-text (2 Sources)

Available from

Similar Publications