The ACS Virgo Cluster Survey. VIII. The Nuclei of Early-Type Galaxies
The ACS Virgo Cluster Survey is a Hubble Space Telescope program to obtain high-resolution imaging in widely separated bandpasses (F475W ≈ g and F850LP ≈ z) for 100 early-type members of the Virgo Cluster, spanning a range of 460 in blue luminosity. We use this large, homogenous data set to examine the innermost structure of these galaxies and to characterize the properties of their compact central nuclei. We present a sharp upward revision in the frequency of nucleation in early-type galaxies brighter than M_B ≈ -15 (66% ≾ f_n ≾ 82%) and show that ground-based surveys underestimated the number of nuclei due to surface brightness selection effects, limited sensitivity and poor spatial resolution. We speculate that previously reported claims that nucleated dwarfs are more concentrated toward the center of Virgo than their nonnucleated counterparts may be an artifact of these selection effects. There is no clear evidence from the properties of the nuclei, or from the overall incidence of nucleation, for a change at M_B ~ -17.6, the traditional dividing point between dwarf and giant galaxies. There does, however, appear to be a fundamental transition at M_B ~ -20.5, in the sense that the brighter, "core-Sérsic" galaxies lack resolved (stellar) nuclei. A search for nuclei that may be offset from the photocenters of their host galaxies reveals only five candidates with displacements of more than 0"5, all of which are in dwarf galaxies. In each case, however, the evidence suggests that these "nuclei" are, in fact, globular clusters projected close to the galaxy photocenter. Working from a sample of 51 galaxies with prominent nuclei, we find a median half-light radius of (r_h) = 4.2 pc, with the sizes of individual nuclei ranging from 62 pc down to ≤2 pc (i.e., unresolved in our images) in about a half-dozen cases. Excluding these unresolved objects, the nuclei sizes are found to depend on nuclear luminosity according to the relation r_h α L ^(0.50±0.03). Because the large majority of nuclei are resolved, we can rule out low-level AGNs as an explanation for the central luminosity excess in almost all cases. On average, the nuclei are ≈3.5 mag brighter than a typical globular cluster. Based on their broadband colors, the nuclei appear to have old to intermediate age stellar populations. The colors of the nuclei in galaxies fainter than M_B ≈ -17.6 are tightly correlated with their luminosities, and less so with the luminosities of their host galaxies, suggesting that their chemical enrichment histories were governed by local or internal factors. Comparing the nuclei to the "nuclear clusters" found in late-type spiral galaxies reveals a close match in terms of size, luminosity, and overall frequency. A formation mechanism that is rather insensitive to the detailed properties of the host galaxy properties is required to explain this ubiquity and homogeneity. The mean of the frequency function for the nucleus-to-galaxy luminosity ratio in our nucleated galaxies, (log η) = -2.49 ± 0.09 dex (σ = 0.59 ± 0.10), is indistinguishable from that of the SBH-to-bulge mass ratio, log(M/M_(gal) = -2.61 ± 0.07 dex (σ = 0.45 ± 0.09), calculated in 23 early-type galaxies with detected supermassive black holes (SBHs). We argue that the compact stellar nuclei found in many of our program galaxies are the low-mass counterparts of the SBHs detected in the bright galaxies. If this interpretation is correct, then one should think in terms of "central massive objects"—either SBHs or compact stellar nuclei—that accompany the formation of almost all early-type galaxies and contain a mean fraction ≈0.3% of the total bulge mass. In this view, SBHs would be the dominant formation mode above M_B ≈ -20.5.
Additional Information© 2006 The American Astronomical Society. Received 2005 September 8; accepted 2006 February 27. Based on observations with the NASA/ESA Hubble Space Telescope obtained at the Space Telescope Science Institute, which is operated by the Association of Universities for Research in Astronomy, Inc., under NASA contract NAS5-26555. We thank Peter Stetson for assistance with the construction of the PSFs used in this study. Support for program GO-9401 was provided through a grant from the Space Telescope Science Institute, which is operated by the Association of Universities for Research in Astronomy, Inc., under NASA contract NAS5-26555. P. C. acknowledges additional support provided by NASA LTSA grant NAG5-11714. M. M. acknowledges additional financial support provided by the Sherman M. Fairchild foundation. D.M. is supported by NSF grant AST 02-0631, NASA grant NAG5- 9046, and grant HST-AR-09519.01-A from STScI. M. J. W. acknowledges support through NSF grant AST 02-05960. This research has made use of the NASA/IPAC Extragalactic Database (NED), which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration.
Published - COTapss06.pdf