The Power of Wide Field I: Galaxy Clustering at High Redshift from a Deep, Wide-Area Survey
Several lines of evidence point to a universe that is dominated gravitationally by matter that is unconventional (non-baryonic) and not directly observable, i.e., dark matter. The growth of large-scale structure and clustering of galaxies from a nearly smooth primordial distribution of matter and energy is the direct indication of the influence of dark matter on scales of megaparsecs. The narrow fields of most telescopes capable of deep imaging and spectroscopy have led to investigations of large-scale structure at high redshifts based on narrow pencil beams. The cosmic fluctuations in the density of galaxies in such restricted cones have led to a large dispersion in estimates of the amplitudes of galaxy clustering as a function of cosmic time.
As a preview of the power of the wide-field imaging surveys that are now routinely possible with the CCD Mosaic imagers at KPNO and CTIO, a deep imaging survey was taken with a single-chip imager at the prime focus of the KPNO 4-meter telescope. M. Postman (STScI), T.R. Lauer (NOAO), I. Szapudi (Durham), and W. Oegerle (JHU) report in the October 10th edition of the Astrophysical Journal on a sample of 710,000 galaxies, reliably complete to I = 23 mag. The contiguous 4 x 4 degree region covers some 75 Mpc on a side at z = 1, which allows adequate sampling of structures as large as 20 Mpc, such as are observed locally. The authors find a typical correlation length of ~ 5 Mpc for brighter galaxies, which decreases to some 3 Mpc at z = 0.5. Some 30--50% of the galaxies at I = 23 are at redshifts greater than 1. The galaxy counts require a mild evolution of about a magnitude of intrinsic brightening to fit the results at the faintest magnitudes, which are consistent with the counts in the Hubble Deep Field. The change in clustering strength with brightness suggests that clustering does not evolve as rapidly in the young universe as predicted by simple linear theory.
This survey illustrates the vital importance of sampling over wide enough area. The cosmic fluctuations in galaxy surface density produced by large-scale structure itself are more than a factor of 2. With the large contiguous area of the current survey, Postman et al. were able to get a fair sampling of the peaks and the voids of the galaxy distribution. The result is more accurate by a factor of 3 than that of any previous study. The NOAO Wide-Deep Survey will provide the natural extension of this work, to deeper limits in more colors, covering a comparable area.
The Power of Wide Field II: The Relationship Between Intergalactic Clouds and Groups of Galaxies
Most of the ordinary (baryonic) matter in the early universe is observed to lie in diffuse clouds of predominantly hydrogen gas. As the universe evolves, more and more of the material becomes locked up in galaxies. In the nearby universe, the galaxies are relatively easy to study, while the diffuse clouds are rare. The converse is true in the early universe. Therefore, key questions remain to be answered about this dominant phase of matter. Are the clouds associated with individual galaxies, groups and clusters, or large-scale structure? Conversely, might the clouds be predominantly found in voids between galaxies, particularly in modern times?
To investigate these questions, T. Tripp (Princeton), L. Lu (Caltech), and B. Savage (Wisconsin) obtained high accuracy spectra in the ultraviolet with the Hubble Space Telescope High-Resolution spectrograph. They observed two bright, nearby quasars to search for the signature of diffuse hydrogen clouds in absorption. They detected a sample of 39 such clouds. In addition, they and other colleagues used the WIYN telescope and multi-fiber spectrograph to obtain redshifts of all the brightest galaxies in the one-degree field centered on each quasar.
Tripp et al. found that all galaxies within 600 kiloparsecs of the line of sight of the quasar had an associated absorbing cloud, and that the strength of the absorption drops off systematically with distance from a galaxy. For galaxies with projected distances out to 2 megaparsecs, an associated cloud was not always detected, but when one was, the drop-off of absorption strength with distance from a galaxy continued. Although some clouds may be associated with individual galaxies, the most consistent explanation is that the clouds represent sheaths and filaments of gas associated with the large-scale structure of galaxies. Occasionally, a cloud is observed without a corresponding galaxy. Some wisps may be found in voids, or tongues of gas may extend well beyond the spatial domain of the galaxy structures themselves. Of course, much larger samples of data are required to confirm and flesh out these conclusions.
Two important aspects of this observational program are apparent. There is a powerful synergy between space and groundbased observations---both are required to pursue this problem. The wide field of the WIYN telescope allowed a sampling of galaxy-cloud associations out to distances as great as 2 megaparsecs. Those observations were key in distinguishing the influence of individual galaxies from the range of large-scale structure phenomena.
Extraplanar Dust in Edge-On Spirals
Interstellar dust causes nothing but problems. It reddens starlight. It limits our view, sometimes completely hiding the most fascinating objects. The formation of dust and its distribution in galaxies remains an intriguing question.
In a search for dust structures in nearby edge-on spiral galaxies using the WIYN Observatory at KPNO, C. Howk and B. Savage (U. Wisconsin-Madison; 1999, ApJ, 117, 2077), found extraplanar dust in all galaxies that have diffuse interstellar gas at similar extraplanar distances, |z| > 400 pc (Figure 1). The mechanism for ejection of the dust and gas appears related to vigorous star formation in the underlying disk.
Essentially all of the mechanisms that Howk and Savage consider responsible for the appearance of dust far from the mid-plane of these galaxies---fountains and chimney flows, expulsion by radiation pressure, flows driven by magnetic field instabilities---require the ejection of dust from thin galaxian disks. Whatever the processes, radiation pressure, stellar winds, and supernovae from active star formation in the disk must play a dominant role in the energetics of driving both gas and dust out of the disk. The detection of extraplanar gas is a key element in this conclusion as high star formation rates are common to all galaxies with extraplanar gas, which is easier to detect than dust, and, thus, far more commonly found.
Many of the dust features Howk and Savage observed are comparable in size and mass to giant molecular clouds in the Milky Way. Given this similarity, are stars being formed within these clouds? In related work, they find ionized gas regions at distances of 600--1000 pc from the mid-planes of NGC891 and NGC4013 that must be powered by hot stars. Interestingly enough, ionized gas regions have been found in the extreme outer disks of face-on spiral galaxies (e.g., Ferguson, Wyse & Gallagher, 1996 AJ 112, 2567). Are these two kinds of ionized gas regions, which are found in quite distinct parts of their parent galaxies, related?