Lisa Horne - Research

The goal of this project is to study the processes of galaxy evolution and star formation by observing the state of baryonic mass in galaxies. This can be accomplished by measuring certain galactic properties such as H I mass, luminosity, and velocity width and determining a relationship between these quantities. This information can provide an understanding of star-formation history in galaxies and the evolution of gas into stars. The project focuses on combining data collected from the Arecibo Legacy Fast ALFA (ALFALFA) H I 21cm radio survey with optical data from the Sloan Digital Sky Survey (SDSS). Optical surveys tend to overlook some gas-rich galaxies such as blue compact dwarf galaxies and low surface brightness galaxies because these systems are too small and too low-contrast to easily be identified by their star light. H I surveys, however, probe the gas in galaxies and can easily identify such objects, but tend to miss elliptical and spherical galaxies which have little gas. Therefore, the combination of the ALFALFA and SDSS data will allow a wider selection of objects to be detected and studied than would be possible with only one survey or the other. This also makes for an interesting census of the state of baryonic mass in different regions of sky. The main goal of this project is to study the optical and HI properties of one region of sky, the Upper Northern Virgo Region.

The Arecibo Legacy Fast ALFA survey (ALFALFA) utilizes Arecibo Observatory, the world’s largest and most powerful single-dish radio telescope. The team employs the Arecibo L-band Feed Array (ALFA) to detect radiation at frequencies between 1225 and 1525 MHz, making it ideal for detecting the atomic hydrogen HI line at 1420.4 MHz. The project aims to explore the local universe between redshifts -1600 km/s and 18,000 km/s and hopes to provide more in-depth information about galactic dynamics and to detect objects with fainter flux limits than has been previously possible with large surveys. The team predicts this study will detect at least 20,000 extragalactic objects upon its completion in several years (Giovanelli et al 2005) .

To achieve these goals, the team uses a simple observation method in which the telescope’s azimuth is fixed (along the local meridian in most cases) and the sky is allowed to “drift” past. This method helps eliminate unnecessary discrepancies calibration, pointing, etc. for each set of data taken. Each strip of sky is observed twice, the observations usually being between 3 to 9 months apart in time (Giovanelli et al 2005). Taking more than one observation helps increase the likelihood that good data will be obtained because problems (Radio Frequency Interference (RFI), disabled equipment, telescope down-time) may arise during any given observation. Multiple observations of the same region also help to identify positions of continuum sources and allow the detection of more objects with low signal-to-noise levels.

Funded mostly by the Alfred P. Sloan Foundation, the Sloan Digital Sky Survey (SDSS) is an ambitious optical survey that aims to map 10,000 deg² (Margon, 1999), over one-fourth, of the entire sky. Information obtained by SDSS is used for a variety of projects including studies of the Milky Way, asteroids, and other Solar System objects, however the stated goals of the survey are to study large scale structures and learn more about the origin of the universe. SDSS actually consists of two separate, but related surveys: a photometric and a spectroscopic survey. The photometric survey measures flux densities in the u, g, r, i, and z bands (Gunn & Weinberg, 1995), which correspond to different wavelengths of visible and near visible light in the electromagnetic spectrum. It also catalogues galaxies, quasars, and stars. The spectrographic survey provides spectra for select galaxies and quasars along with that of some unusual stars. Information from this survey is then used to create 3D maps and accurately measure distances to these objects, thus furthering our understanding of the universe’s extent. The main telescope used for SDSS is the 2.5 meter telescope at Apache Point Observatory in New Mexico. SDSS achieved first light in May of 1998 and released its first complete set of data five years later in May of 2003 (Abazajian et al, 2003). Since then, SDSS has released a new set of data annually. My project utilizes spectrographic information from Data Release 7, the most recent set. Using this information, I will be able to calculate values such as absolute magnitude and luminosity for the galaxies found in both ALFALFA and SDSS and compare them to those values for galaxies found in SDSS only.

Neutral hydrogen gas located throughout the universe emits a 21-cm spectral line, which corresponds to a frequency of 1420.4 MHz, when it transitions from its excited state to its ground state. In the exited energy state, the proton and electron comprising each hydrogen atom spin parallel to each other. When the proton and electron cease spinning parallel to each other and begin to spin anti-parallel, the hydrogen atom releases a photon at and returns to its ground state. (See figure at left.) This photon is detectable in spectra as a bright line at 21cm. Hence the term, "HI Line."

As mentioned previously, the ALFALFA team uses ALFA when observing at Arecibo. ALFA consists of 7 feed horns arranged in a hexagonal pattern, with 6 horns at the 6 vertices and 1 horn in the center. (See photo at right.) Because of this, for each drift observed, a record is created that contains 14 different spectra, one for each polarization of the 7 feed horns. Each of the drifts taken during a single observing session are grouped together. The first steps in processing the raw data are collectively referred to as "Level I Processing." During this phase, a calibration and a bandpass correction are performed. Then each drift’s 14 spectra are examined individually. At this step, referred to as “flagging,” ALFALFA team members flag any interference from terrestrial sources such as airport radar or GPS and get their first glimpse of any celestial HI sources such as galaxies or high-velocity clouds. Any new objects are noted and later cataloged. (See figure below for further detail.)

This is a typical example of what an ALFALFA team member sees when performing Level I Data Processing. (Everything in yellow has been superimposed onto the image to aide in explanation.)

The "Strip" number and the "Pol" number (circled at left) refer to the feed horn and polarization, respectively, from which the displayed data is taken. The blue and white "fuzz" (1 in figure) is a slice through the drift in frequency-position space with the x-axis representing frequency and the y-axis representing a position on the sky. All of the same information collapsed in the spacial direction is displayed in the bottom graph (3 in figure). The middle graph (2 in figure) displays pixel varience over the entire spectrum. Signal from the Milky Way Galaxy is always present at the same frequency and it is important that it is not mistaken for interference. Certain radio frequency interferences like the San Juan International Airport's radar (the left-most interference labeled here) are also present at the same frequency, making it easy for them to be identified and removed from the data. Galaxies and other objects of interest to ALFALFA often appear as 'blobs' of white and are usually present in both polarizations of the spectra from several feed horns. The galaxy circled in the figure is the first galaxy I discovered while working with ALFALFA. After further investigation, we determined that it is an edge-on spiral galaxy.

Once a portion of the sky has been completely observed (two passes have been made), the processed data from the two scans are put together to form “grids” which are in essence velocity cubes, “pictures” of a single position on the sky over a range of velocities. A series of flat-fielding and base-lining processes are applied to the grids before source-detection algorithms are applied. Once an object is positively identified and determined to be real, it is prepared cataloging in the Arecibo General Catalog (AGC) by obtaining its flux, velocity, velocity-width, and other such measurements.

Above: Plot showing right ascensions and velocities of all known ALFALFA galaxies for the entire Northern Virgo Region.
Below: Plot showing right ascensions and velocities for all known SDSS galaxies for the entire Northern Virgo Region.
For both plots, the grouping of galaxies in the point (approximately between 0 and 2000 km/s) is probably associated with the Virgo Cluster and the grouping between approximately 6500 and 8500 km/s is most likely the Great Wall. The Coma Void can also be seen here between approximately 3500 and 6500 km/s where there are very few galaxies present. Right ascension has been divided into four regions on these plots by the dotted diagonal lines. The galaxies that I am working with are shown in the upper two slices of the graph.

A list of 456 galaxies in the SDSS database was compared against a list of 200 galaxies found by the ALFALFA survey between 13° and 16° in declination and between 12h30min and 14h00min in right ascension. (See above two figures.) I am specifically examining the group of galaxies that fall between 13h20min and 14h00min right ascension, dubbed here as the “Upper Northern Virgo” region. My partner is handling the remaining set of galaxies. These galaxies are in the northern part of the Virgo Cluster and were chosen for this study because the region has been observed extensively and much data is available. The region is also very populated, which will enable us to study the effects of galactic interaction on HI mass and other values as well as its effect on detection rates.

The ALFALFA galaxy list was compared to the SDSS galaxy list using a self-written program that matched right ascension, declination, and recessional velocity. If the difference between the SDSS and ALFALFA right ascension values had an absolute value less than 0.025°, the angular separation between the two galaxies was calculated. If the angular separation was less than 0.025° and the difference between the two recessional velocity values was less than 500km/s, then the galaxies from each list were considered to be matches. A list of galaxies with both an ALFALFA and an SDSS record was created as well as two other lists containing the galaxies that were only identified by one survey or the other. The program identified 177 matches, 279 galaxies in SDSS only, and 48 galaxies in ALFALFA only for the entire northern Virgo region. In some cases, two or more SDSS galaxies were chosen as matches for one ALFALFA galaxy. Often the multiple SDSS records were of different regions of the same large galaxy.

Using the coordinates and velocities for the galaxies identified only in SDSS, 9 new ALFALFA galaxies in the Upper Northern Virgo region were found. This was accomplished by looking for visible features in the grids mentioned earlier at the given SDSS coordinates and velocities to obtain flux measurements. Upper flux limits were obtained for the remaining SDSS galaxies that couldn’t be matched using a program entitled “Galflux_up” written by students and a professor at George Mason University.

The collage at left displays 9 of the 14 Upper Northern Virgo galaxies detected by the ALFALFA survey, but not by the SDSS survey. It seems surprising that so many of these galaxies are large, bright spirals. The truth is that these galaxies are too bright and too near to be detected by spectroscopically by SDSS, a well known problem with the survey. Galaxies like AGC 8383 and AGC 8518 are known objects that appear in many other catalogs such as the NASA/IPAC Extragalactic Database (NED). It is possible that some of these galaxies may have recoverable photometric information that would be useful. Some galaxies, like AGC 233668 and AGC 233677, are probably just too small and dim to be detected and to have spectra measured by SDSS.Objects like AGC 232765 are the most interesting finds. No corresponding galaxy appears to be visible in the image. Yet, ALFALFA detects a flux of 1.05 Jy km/s and a signal-to-noise ratio of 13.4. It's possible that this is an example of the elusive dark galaxies, which are theoretically comprised of gas and dust, but few or no stars. In the case of relatively low velocities like AGC 232765 (106 km/s), it is more likely that ALFALFA is detecting a high-velocity cloud, HI within our own galaxy (Giovanelli et al. 2007).

SDSS Only Galaxies:

The collage to the right displays 9 of the 144 Upper Northern Virgo galaxies detected by SDSS, but not by ALFALFA. Most of the galaxies that were detected optically, but not at 21cm tend to be similar in nature. It seems that these galaxies tend to be ellipticals. Elliptical galaxies are composed primarily of older stars, therefore it is not entirely surprising that these galaxies were not detected by ALFALFA, which by design, probes neutral hydrogen. Interestingly, however, some of the galaxies such as SDSSJ133443.5+134507, SDSSJ135812.9+144304, and SDSSJ135024.2+135125 are notably bluer and more disk-like in shape. Such galaxies are expected to be detected by ALFALFA, however there is no indication of any of them in the grids and their upper-limit spectra are relatively uninteresting. Perhaps a third investigation of the grids is warrented, but given that these galaxies are quite small and are located in a crowded velocity region, they may simply have too low of a signal-to-noise value to be detected.

collage - matching galaxies

Matching ALFALFA and SDSS Galaxies:

These 9 galaxies were chosen randomly out of the 177 that were detected in both the SDSS and ALFALFA catalogs. They are presented here mostly to contrast with the galaxies that were only detected in one survey or the other. It is apparent that these galaxies aren't too big, too small, too dim, or too bright. They are "just right." They exist over the entire range of velocities, declinations, and right ascensions in the Upper Northern Virgo region.

Matches recovered from grids

Recovered Matches:

These are the 9 galaxies that were recovered from the grids using the coordinates and velocities for the galaxies identified only in SDSS. Some, but not all, galaxies have already been assigned AGC numbers. Although these galaxies are all rather small, many of them seem to be very blue in color, suggesting that they are rich in hydrogen gas. This is probably why they were recoverable. All of these galaxies, with the exception of SDSSJ133156.9+133101 (4866 km/s), lie within the Great Wall, a highly populated region of sky.

Distances:

At left are a set of histograms displaying the distances of the Upper Northern Virgo galaxy groups from Earth. (Distance = Velocity/70 km/s). There seems to be a clustering of galaxies right around 100 Mpc, probably corresponding to the Great Wall. The clustering of galaxies at approximately 16Mpc usually associated with the Virgo Cluster is not visible here. The Coma Void accounts for the lessened density between approximately 50Mpc and 90Mpc (Fairall, 1998). There is one ALFALFA object, the high velocity cloud from before, and five SDSS objects that fall in the 0Mpc bin. The Hubble relation does not provide an accurrate representation of actual distances for objects with very low velocity values . Other methods will be necessary to determine the actual distances for these galaxies.

Fairall, Anthony. Large Scale Structures in the Universe. Chichester: Praxis. 1998.

Gunn, J. E. & Weinberg, D. H., 1995, in Wide Field Spectroscopy and the Distant Universe, eds. S. Maddox & A.Aragon-Salamanca, (Singapore: World Scientific), 3 (astro-ph/9412080)

Margon, B., 1999, Phil. Trans. Roy. Soc. Lond., 1999, 357, 93. (astro-ph/9805314)

Funding for the SDSS and SDSS-II has been provided by the Alfred P. Sloan Foundation, the Participating Institutions, the National Science Foundation, the U.S. Department of Energy, the National Aeronautics and Space Administration, the Japanese Monbukagakusho, the Max Planck Society, and the Higher Education Funding Council for England. The SDSS Web Site is http://www.sdss.org/.

RESEARCH