1: %\documentclass[manuscript]{aastex}
2: %\documentclass[preprint2]{aastex}
3: \documentclass[preprint,10pt]{aastex}
4: %\documentclass[preprint2,longabstract]{aastex}
5: %\documentclass{emulateapj}
6: %\usepackage{psfig,emulateapj5}
7: %\documentstyle[aaspptwo]{article}
8:
9: \newcommand\sqd{{deg$^{2}$}}
10: \newcommand\etal{{\it et al. }}
11: \newcommand\kms{km~s$^{-1}$}
12: \newcommand\msun{$M_\odot$}
13: \newcommand\mhi{$M_{HI}$}
14: \def\ncat{578}
15: \def\ndet{435}
16: \def\nprior{129}
17: \def\nhvc{14}
18: %\def\nvirgo{195}
19: %\def\nooc{??}
20: \def\be{\begin{equation}}
21: \def\ee{\end{equation}}
22:
23:
24: \shorttitle{Third ALFALFA Catalog}
25: \shortauthors{Kent et al.}
26:
27: \begin{document}
28: %\hskip 3.5in{Version 2.1 \hskip 10pt 04Dec2006}
29: \title{The Arecibo Legacy Fast ALFA Survey: \\
30: VI. Second HI Source Catalog of the Virgo Cluster Region
31: %HI Source Catalog of the Southern Virgo Cluster Region
32: }
33:
34: \author {Brian R. Kent\altaffilmark{1},
35: Riccardo Giovanelli\altaffilmark{1,2}, Martha P. Haynes\altaffilmark{1,2},
36: Ann M. Martin\altaffilmark{1},
37: Am\'elie Saintonge\altaffilmark{3},
38: Sabrina Stierwalt\altaffilmark{1},
39: Thomas J. Balonek\altaffilmark{4},
40: Noah Brosch\altaffilmark{5},
41: Rebecca A. Koopmann\altaffilmark{6}
42: }
43: %\affil{Department of Astronomy and National Astronomy and Ionosphere Center
44: %\altaffilmark{1},\\
45: %Space Sciences Bldg., Cornell University, Ithaca, NY 14853\\
46: %{\it e--mail:} riccardo@astro.cornell.edu, haynes@astro.cornell.edu}
47:
48: \altaffiltext{1}{Center for Radiophysics and Space Research, Space Sciences Building,
49: Cornell University, Ithaca, NY 14853. {\it e--mail:} bkent@astro.cornell.edu, riccardo@astro.cornell.edu,
50: haynes@astro.cornell.edu, amartin@astro.cornell.edu,
51: amelie@astro.cornell.edu, sabrina@astro.cornell.edu}
52:
53: \altaffiltext{2}{National Astronomy and Ionosphere Center, Cornell University,
54: Space Sciences Building,
55: Ithaca, NY 14853. The National Astronomy and Ionosphere Center is operated
56: by Cornell University under a cooperative agreement with the National Science
57: Foundation.}
58:
59:
60:
61: \altaffiltext{3}{Institute for Theoretical Physics, University of Zurich, Winterhurerstrasse 190, CH-8057 Zurich, Switzerland
62: {\it e--mail:} amelie@physik.uzh.ch}
63:
64: \altaffiltext{4}{Dept. of Physics \& Astronomy, Colgate University, Hamilton, NY 13346.
65: {\it e--mail:} tbalonek@mail.colgate.edu}
66:
67: \altaffiltext{5}{The Wise Observatory \& The School of Physics and Astronomy,
68: Raymond \& Beverly Sackler Faculty of Exact Sciences, Tel Aviv University, Israel.
69: {\it e--mail:} noah@wise.tau.ac.il}
70:
71:
72:
73: \altaffiltext{6}{Dept. of Physics \& Astronomy, Union College, Schenectady, NY 12308.
74: {\it e--mail:} koopmanr@union.edu}
75:
76:
77: %\hsize 6.5 truein
78: \begin{abstract}
79:
80: We present the third installment of HI sources extracted from the Arecibo
81: Legacy Fast ALFA extragalactic survey. This dataset
82: continues the work of the Virgo ALFALFA
83: catalog. The catalogs and spectra published
84: here consist of data obtained during the 2005 and 2006 observing sessions
85: of the survey. The catalog consists of \ncat~HI detections within the
86: range $11^h36^m <$ R.A.(J2000) $< 13^h52^m$ and
87: $+08^{\circ} <$ Dec.(J2000) $< +12^{\circ}$, and $cz_{\odot} < 18000$ \kms.
88: The catalog entries are identified with optical counterparts where possible
89: through the examination of digitized optical images. The catalog detections
90: can be classified into three categories: (a) detections
91: of high reliability with S/N $>$ 6.5; (b) high velocity clouds
92: in the Milky Way or its periphery; and (c) signals of lower S/N
93: which coincide spatially with an optical object and known redshift.
94: 75\% of the sources are newly published HI detections. Of particular note is a complex
95: of HI clouds projected between M87 and M49 that do not coincide with any optical
96: counterparts. Candidate
97: objects without optical counterparts are few.
98: The median redshift for this sample is 6500 \kms~ and
99: the $cz$ distribution exhibits the local large scale structure consisting
100: of Virgo and the background void and the A1367-Coma supercluster regime at
101: $cz_{\odot} \sim$7000 \kms. Position corrections for telescope pointing errors are applied
102: to the dataset by comparing ALFALFA continuum centroid with those
103: cataloged in the NRAO VLA Sky Survey. The uncorrected positional accuracy averages
104: 27\arcsec ~(21\arcsec ~median) for all sources with S/N $>$ 6.5 and is of order
105: $\sim$21\arcsec ~(16\arcsec ~median) for signals with S/N $>$ 12.
106: Uncertainties in distances toward the Virgo cluster can affect the calculated
107: HI mass distribution.
108:
109:
110: \end{abstract}
111:
112: \keywords{galaxies: spiral --- galaxies: distances and redshifts ---
113: galaxies: halos --- galaxies: luminosity function, mass function ---
114: galaxies: photometry --- radio lines: galaxies}
115:
116:
117: \section {Introduction}\label{intro}
118:
119: A census of HI in the local Universe
120: will shed light on galaxy evolution and a census
121: of the gas content of galaxies at $z\sim 0$. As such, a complete enumeration
122: tracing the HI at low redshift will yield a homogeneous dataset with
123: quantitative measurements
124: for kinematics, redshifts, and HI masses. An HI catalog provided
125: by a wide-area survey serves as an outstanding complement to existing and future surveys
126: at other wavelengths, including the
127: large area Two Micron All Sky Survey (2MASS; Skrutskie \etal 2006)
128: and the Sloan Digital Sky Survey (SDSS; York \etal 2000). A high sensitivity HI survey
129: sampling a large area of sky completes the picture by adding
130: information about the gas content of galaxies in the local Universe.
131:
132:
133: First generation HI surveys have been limited by sensitivity or areal coverage.
134: The HI Parkes All-Sky Survey (HIPASS; Barnes \etal 2001) has detected 5317 extragalactic sources
135: over 30,000 deg$^{2}$ of sky (Meyer \etal 2004; Wong \etal 2006) using a
136: 13 element receiver system. The survey's
137: successful catalogs provided a wide area of sky coverage. The Arecibo Dual Beam Survey
138: (ADBS; Rosenberg \& Schneider 2002)
139: surveyed a smaller area of sky, but utilized the advantage of the higher
140: sensitivity, smaller beam size, and better
141: positional accuracy of the Arecibo Radio Telescope. The ADBS in particular showed the populations of objects
142: that could be detected with high sensitivity HI surveys that
143: were not subjected to the biases of optical surveys.
144: A ``2nd generation'' extragalactic HI survey can combine high sensitivity with coverage of
145: a wide area and will yield a cosmologically fair sampling of HI
146: sources in the volume of the local Universe.
147: As astronomy pushes into a realm of large scale data mining, optical and IR surveys will be greatly
148: complemented by the catalogs from HI surveys.
149:
150: The Arecibo Legacy Fast ALFA (ALFALFA; Giovanelli \etal 2005a: Paper I) survey, currently underway,
151: utilizes the 7-element Arecibo L-band Feed Array (ALFA) receiver system to conduct a wide area
152: extragalactic HI investigation. Using the high sensitivity of the 305 meter reflector, ALFALFA
153: will survey 7000 deg$^{2}$ of sky at high Galactic latitude up to $cz_{\odot}\sim$18000 \kms.
154: As a second generation HI survey, ALFALFA
155: improves upon previous surveys with larger bandwidth and finer spectral resolution,
156: as well as with higher sensitivity and angular resolution. While the eight-fold
157: increase in sensitivity will assist in the detection of low HI mass galaxies,
158: the wider bandwidth will allow ALFALFA to sample a fair volume of the Universe.
159: Higher angular resolution will largely avoid the source confusion that affected identification of
160: optical counterparts in surveys such as HIPASS.
161: Paper I describes the survey's science goals and objectives. Based on simulations
162: and ALFALFA's initial catalogs (Giovanelli \etal 2005b, 2007; Saintonge \etal 2008),
163: more than 20,000 sources will be detected out to $z \sim 0.06$.
164:
165: Once an HI mass is detectable at an astrophysically significant distance, maximizing
166: the survey area is more advantageous in increasing the detection rate
167: than increasing integration time.
168: The ``fast'' element of ALFALFA aims to survey a large amount of sky quickly, with
169: a 1-Hz sampling rate for scans, and 40 second integration time per pixel.
170: These observing parameters allow ALFALFA to detect
171: low HI mass galaxies down to M$_{HI} \sim 10^{5}$\msun~ in the Local Group and M$_{HI} \sim 2 \times 10^{6}$\msun~
172: at the distance of the Virgo Cluster.
173: As such, a crucial goal of using ALFALFA's
174: low mass sample is a determination of the HI mass function (HIMF) both globally and in a variety of
175: environments. Obtaining a statistically significant sample coupled with accurate
176: distances for nearby galaxies will yield an accurate
177: and robust determination of the faint end slope of the
178: HI mass function, one of the many science goals of ALFALFA and HI astrophysics.
179: Determining the statistics of these low-mass gas rich halos will also shed light on
180: the ``substructure'' problem (Somerville 2002), which predicts that we should observe more satellite objects
181: in the hierarchical structure formation paradigm.
182:
183: ALFALFA is detecting a variety of objects including large disk galaxies and small low mass dwarfs
184: in the Local Universe. Datasets provided by HI surveys like ALFALFA give systemic redshift, integrated
185: line fluxes from spectral profiles, and velocity width measurements for cataloged objects.
186: Approximately 500 galaxies will be resolved nearby. In addition to single galaxies, ALFALFA also provides data
187: for extended tidal features and nearby high--velocity clouds, and gas-rich halos
188: which may exhibit little or no stellar component.
189:
190: The survey's observing strategy has yielded an open shutter time of 97\%. Final
191: data cubes are constructed after two-pass transit observations of targeted areas are completed. ALFALFA's public website is
192: continually updated with schedules, survey information, and links to online catalogs. The survey's
193: first catalogs are being published in periodic installments as target areas and data cubes are completed.
194: The first catalog (Giovanelli \etal~2007)
195: detailed the results of 730 detections from the Northern Virgo region, while the second (Saintonge \etal~2008)
196: yielded 488 detections in a direction opposite of the Virgo cluster. The results presented
197: in this paper extend the release of the ALFALFA survey with \ncat~ detections in the Virgo Cluster region
198: south of first catalog.
199:
200: Virgo is the nearest rich-cluster of galaxies with over 1300 reported members (Binggeli, Sandage \& Tammann 1985; BST).
201: Because of its proximity, high density, and location in the sky visible
202: from Arecibo, Virgo is an important primary target area for ALFALFA.
203: In addition, Virgo is often included in the coverage of both medium and
204: large sky surveys (see ACS Virgo in C{\^o}t{\'e} \etal~2004; BST; HIPASS; SDSS), creating
205: an important synergy with one of ALFALFA's primary target areas. Many late-type spiral galaxies
206: are found in the outlying periphery, which has not yet virialized. As such, many of these gas rich galaxies
207: are still infalling into the cluster, in contrast to the Virgo core spirals, known to be highly
208: deficient in HI (Davies \& Lewis 1973; Chamaraux \etal~1980; Giovanelli \& Haynes 1983; Solanes
209: \etal~2002). The catalog presented in this paper consists of HI sources extracted in Right Ascension
210: between $11^h36^m <$ R.A.(J2000) $< 13^h52^m$ and Declination between $+08^{\circ} <$ Dec.(J2000) $< +12^{\circ}$.
211: This area, covering 132 deg$^{2}$, includes the portion of Virgo between cluster
212: members M87 and M49 and complements the earlier catalog by Giovanelli \etal (2007).
213:
214: The paper organization is as follows. Section 2 describes the observations and data reduction.
215: Section 3 details the catalog contents and signal extraction reliability.
216: Section 4 describes the sample statistics of the dataset.
217: Section 5 deals with the positional accuracy of ALFALFA sources.
218: Section 6 summarizes the results.
219:
220:
221:
222:
223: \section{Observations and Data Reduction}\label{obs}
224:
225: ALFALFA utilizes a fixed azimuth drift mode observation scheme with the 7-element
226: ALFA multi-beam receiver system. Each dual polarization feed has a beam that
227: is $3.3$\arcmin~$\times$~ $3.8$\arcmin~in size. Beam maps
228: of each feed can be found in Paper I. The ALFA array is
229: rotated so that parallel feed tracks
230: of constant J2000 declination are spaced at 1.05\arcmin. Scans are obtained with the telescope
231: parked along the meridian--small corrections in zenith angle are applied
232: between scans such that the beams are tracking along the constant epoch
233: J2000 Declination. Fourteen simultaneous spectra are obtained at a sampling rate of 1 Hz
234: in R.A. and
235: scans are composed of 600 one--second records each sweeping 10 minutes of Right Ascension. The backend correlator setup uses
236: a 100 MHz bandwidth centered at 1385 MHz. Raw scans have 4096 channels, giving
237: a spectral resolution of 24.4 kHz ($\delta v\sim 5.2$~\kms~ at $cz\sim$0).
238: Other survey details about the observing and
239: technical modes can be found in Paper I.
240:
241: Raw data scans are processed offline for each observing session.
242: Initial processing by team members includes flagging of radio interference,
243: bandpass and flux calibration, data quality assessment, and identifying
244: strong continuum sources. Observing runs
245: are scheduled to fill in target area tiles that are
246: composed of $2.4^{\circ} \times 2.4^{\circ}$ areas of sky. Once scheduled target area tiles are completed, the
247: level I data scans are combined into regularly gridded 3-D data cubes, each
248: 2.4 $\times$ 2.4 degrees in spatial extent, and 4~$\times$~1024 channels in velocity space.
249: These data cubes contain spectral data header information, coordinates,
250: both polarizations, continuum data, map weightings, and scan makeup.
251: An automated matched filter algorithm is used for signal extraction (Saintonge 2007).
252: Candidate signals are later examined by eye, and integrated spectral profiles are
253: created. Flux measurements, velocities, and widths are cataloged for publication
254: and noted as possible followup candidates.
255: The resulting data cubes are approximately 4~$\times$~380 MB in size.
256:
257: \subsection{Continuum maps}
258:
259: In addition to the spectral line data, continuum maps are also created
260: using data from the level I data scans, described in Giovanelli \etal (2007).
261: Channels are flagged through the bandpass calibration process and/or
262: the manual flagging process mentioned previously.
263: A background total power continuum value is also computed for all time series records for all
264: records and channels that have not been flagged, excluding point sources; the continuum
265: contribution from these point sources is also stored for the creation of the continuum maps.
266: Sources are detected with an automated
267: peak finding algorithm, which then searches a database of NVSS (Condon \etal~1998) sources
268: and looks for matches, comparing both fluxes and positions. Final positions
269: are fit with a 2D Gaussian and stored along with peak flux measurements.
270: These measurements are used to correct for positions offsets as described in section 5,
271: and continuum source variability studies.
272:
273: \subsection{Data Access}
274:
275: As the ALFALFA collaboration includes many worldwide members,
276: rapid dissemination of the data
277: products is of the utmost importance to the survey's success.
278: The catalog results presented in this paper will be
279: added to the existing ALFALFA archive at
280: {\it http://arecibo.tc.cornell.edu/hiarchive/alfalfa}/.
281: The site provides web services using protocols from the U.S. National
282: Virtual Observatory\footnote{This research has made use of data obtained
283: from or software provided by
284: the US National Virtual Observatory(NVO), which is sponsored by the National
285: Science Foundation.}. The measurements
286: and spectra will join the fast growing archive of HI measurements
287: (Springob \etal~2005; Giovanelli \etal~2007; Saintonge \etal~2008).
288: An ongoing development effort focuses on the
289: long--term public delivery of the 3--D ALFALFA data set through web--based
290: access tools. At this time, delivery of the 3-D data is made possible
291: through the observing team itself, by direct contact to R.G. or M.P.H.
292: A major challenge is data volume: after regridding the 3-D cubes covering the 34 individual
293: ``grids'' from which the current catalog each occupy 50 GB.
294: Allowing access and manipulation of the gridded data publicly will require
295: the development of web services and server applications.
296:
297: In this work, we present a catalog of HI sources extracted
298: from the ALFALFA grids covering a region stretching
299: from $+8^\circ$ to $12^\circ$ in Dec. and from
300: $11^h36^m$ to $13^h52^m$ in R.A. For reference to our database,
301: denomination of the grids from which sources in this catalog were
302: extracted are 1140+09 to 1348+09 and 1140+11 to 1348+11, in
303: steps of 8$^m$ in R.A.
304: The solid angle subtended by this region is $\sim$132 \sqd,
305: which is $\sim1.9$\% of the sky to be ultimately surveyed by
306: ALFALFA. The coverage of the region is complete by the target goals
307: of ALFALFA, i.e. the region has been sampled by two separate passes
308: with the ALFA array in drift mode.
309:
310: \section{Second Virgo Catalog}\label{sourcecat}
311:
312: Raw data scans are processed to Level I (see Paper I for details) datasets upon completion
313: of the observing sessions. The survey
314: employs a two--pass system to yield a well spatially
315: sampled dataset. Once the survey has completed two passes
316: of a designated tile it is converted into regularly
317: gridded data cubes.
318: Base-lining, signal extraction, and source identification are completed and
319: measurements are taken from integrated spectral profiles. These profiles
320: originate from cutouts of the sub-grids, which are then used to create
321: properly weighted source spectra.
322:
323: Signal extraction is initiated in the Fourier domain with an automated matched
324: filter algorithm (Saintonge 2007). Followup by eye confirms these sources and matches
325: them with any possible optical counterpart found in the Sloan Digital
326: Sky Survey (SDSS) or blue band imaging provided by the Digital Sky Survey.
327: In addition, measurements and sources can be compared to NED and
328: listings of existing HI observations. This facilitates
329: quick response for followup with selected sources. Candidate detections are assigned
330: to the following classes:
331:
332: 1. Signals that are extragalactic HI sources above a soft limit of S/N~$\gtrsim$~6.5
333:
334: 2. Signals that are likely galactic or perigalactic in nature - high--velocity clouds.
335:
336: 3. Signals of lower S/N ($4.5 \lesssim$~S/N~$\lesssim 6.5$) but have
337: corroborating evidence based on identification of an
338: optical counterpart with similar redshift measurement at another wavelength.\\
339:
340: Other low S/N candidates are placed in a separate listing for future followup.
341:
342: Table 1 presents the results of the \ncat~ detections from this portion of the survey.
343: The column descriptions are as follows:
344:
345: Col. 1: The catalog and source number
346:
347: Col. 2: Arecibo General Catalog number. This number corresponds to a private database
348: entry maintained by R.G. and M.P.H.
349:
350: Col. 3: Centroid position of the HI source (R.A. and Dec. J2000). The listed position has been corrected
351: for systemic pointing errors. Positional accuracy will be discussed in section 5.
352:
353: Col 4: J2000 position of the most likely optical counterpart of the HI detection. The listed
354: position has been examined by eye using SDSS or the Digital Sky Survey.
355: The centroid accuracy is $\lesssim$2\arcsec. Optical images are examined
356: for counterparts based on spatial proximity, morphology, color, and redshift.
357: If no optical counterpart can be clearly identified, no optical position is recorded in the catalog -- high--velocity
358: cloud detections are included in this category. Comments are provided
359: in column 14 if
360: a preferable source is identified and other possible candidates are present
361: in the field.
362:
363: Col. 5: heliocentric velocity of the HI source in \kms, $cz_{\odot}$, measured
364: as the midpoint
365: between the channels at which the flux density drops to 50\% of
366: each of the two peaks (or of one, if only one is present) at each
367: side of the spectral feature. The error on $cz_\odot$
368: to be adopted is half the error on the width, tabulated in Col. 7.
369:
370: Col. 6: velocity width of the source line profile, $W50$, measured at the 50\%
371: level of each of the two peaks, as described for Col. 5. This value
372: is corrected for instrumental broadening. No corrections due to
373: turbulent motions, disk inclination or cosmological effects are
374: applied. The instrumental correction used is that described by Springob \etal (2005).
375: Therefore, the expression for $W50$ is
376:
377: \be
378: W50=\sqrt{W50_{uncorr}^2-(2\Delta\nu)^2}
379: \ee
380:
381: where $\Delta\nu$ is the channel separation in \kms.
382:
383: Col. 7: estimated error on the velocity width, $\epsilon_w$, in \kms.
384: This error is the sum in quadrature of two components: the first is a
385: statistical error, principally dependent on the S/N
386: ratio of the feature measured; the second is a systematic error
387: associated with the subjective guess with which the observer estimates
388: the spectral boundaries of the feature: maximum and minimum guesses of
389: the spectral extent of the feature are flagged and the
390: ratio of those values is used to estimate systematic errors on the
391: width, the velocity and the flux integral. In the majority of cases,
392: the systematic error is significantly smaller than the statistical
393: error; thus the former is ignored.
394:
395: Col. 8: integrated flux density of the source, $F_c$, in Jy \kms. This
396: is measured on the integrated spectrum, obtained by
397: spatially integrating the source image over a solid angle of at
398: least $7$\arcmin $\times 7$\arcmin ~and dividing by the sum of the survey beam
399: values over the same set of image pixels (Shostak \& Allen 1980).
400: Estimates of integrated fluxes for very extended sources with
401: significant angular asymmetries can be misestimated by our
402: algorithm, which is optimized for measuring sources comparable with
403: or smaller than the survey beam. A special catalog with parameters
404: of extended sources will be produced after completion of the survey.
405:
406: Col. 9: estimated uncertainty of the integrated flux density, in Jy \kms .
407: Uncertainties
408: associated with the quality of the baseline fitting are not included;
409: an analysis of that contribution to the error will be presented
410: elsewhere for the full survey. See the description of Col. 7 for the
411: contribution of a possible systematic measurement error.
412:
413: Col. 10: signal--to--noise ratio S/N of the detection, estimated as
414: \be
415: S/N=\Bigl({1000 F_c\over W50}\Bigr){w^{1/2}_{smo}\over \sigma_{rms}}
416: \ee
417: where $F_c$ is the integrated flux density in Jy \kms, as listed in col. 8,
418: the ratio $1000 F_c/W50$ is the mean flux across the feature in mJy,
419: $w_{smo}$ is either $W50/(2\times 10)$ for $W50<400$ \kms or
420: $400/(2\times 10)=20$ for $W50 \geq 400$ \kms [$w_{smo}$ is a
421: smoothing width expressed as the number of spectral resolution
422: bins of 10 \kms bridging half of the signal width], and $\sigma_{rms}$
423: is the r.m.s noise figure across the spectrum measured in mJy at 10
424: \kms resolution, as tabulated in Col. 11. In a similar analysis, in
425: Giovanelli \etal (~2005b; hereafter Paper II) we adopted a maximum smoothing
426: width $W50/20=10$. See Figure 4 ~and related text
427: below for details. The value of the smoothing width could be
428: interpreted as an indication of the degree to which spectral smoothing
429: aids in the visual detection of broad signals, against broad--band
430: spectral instabilities. The ALFALFA
431: data quality appears to warrant a more optimistic adoption of
432: the smoothing width than previously anticipated.
433:
434: Col. 11: noise figure of the spatially integrated spectral profile, $\sigma_{rms}$,
435: in mJy. The noise figure as tabulated is the r.m.s. as measured over the signal-- and
436: RFI-free portions of the spectrum, after Hanning smoothing to a spectral
437: resolution of 10 \kms.
438:
439: Col. 12: adopted distance in Mpc, $D_{Mpc}$. For objects with $cz_{cmb}> 3000$ \kms,
440: the distance is simply $cz_{cmb}/H_\circ$; $cz_{cmb}$ is the recessional velocity
441: measured in the Cosmic Microwave Background reference frame and $H_\circ$ is
442: the Hubble constant, for which we use a value of 72 \kms Mpc$^{-1}$.
443: For objects of lower $cz_{cmb}$, we use a peculiar velocity model for the
444: local Universe, as described in Paper II. Objects which are thought to be parts
445: of clusters or groups are assigned the $cz_{cmb}$ of the cluster or group.
446: Cluster and group membership are assigned following the method described
447: in Springob \etal (2007). A detailed
448: analysis of group and cluster membership of ALFALFA objects will be presented
449: elsewhere. Note that the Virgo cluster extends over much of the solid angle
450: surveyed. This introduces unavoidable ambiguities in the distance assignment,
451: as the peculiar flow model only corrects for large--scale perturbations in the
452: velocity field and is unable to deal effectively with regions in the
453: immediate vicinity of the cluster and along a section of a cone which contains
454: the cluster, up to $cz\sim 2500$ \kms.
455:
456: Col. 13: logarithm in base 10 of the HI mass, in solar units. This parameter is
457: obtained by using the expression $M_{HI}=2.356\times 10^5 D_{Mpc}^2 F_c$.
458:
459: Col. 14: object code, defined as follows:
460:
461: Code 1 refers to sources
462: of S/N and general qualities that make it a reliable detection.
463: By ``general qualities'' we mean that, in addition to an approximate
464: S/N threshold of 6.5, the signal should exhibit a good match between
465: the two independent polarizations and a spatial extent consistent
466: with expectations given the telescope beam characteristics. Thus, some
467: candidate detections with $S/N>6.5$ have been excluded on grounds
468: of polarization mismatch, spectral vicinity to RFI features or peculiar
469: spatial properties. Likewise, some features of $S/N<6.5$ are included
470: as reliable detections, due to optimal overall characteristics of
471: the feature. The S/N threshold for acceptance of a reliable detection
472: candidate is thus soft. In a preliminary fashion, we estimate that
473: detection candidates with $S/N<6.5$ in Table 1 are reliable, i.e. they
474: will be confirmed in follow--up observations in better than 95\% of
475: cases (Saintonge 2007).
476:
477: Code 2 refers to sources of low S/N ($<$ 6.5), which would
478: ordinarily not be considered
479: reliable detections by the criteria set for code 1. However, those
480: HI candidate sources are matched with optical counterparts with known
481: optical redshifts which match those measured in the HI line. These
482: candidate sources, albeit ``detected'' by our signal finding algorithm,
483: are accepted as likely counterparts only because of the existence of
484: previously available, corroborating optical spectroscopy. We refer to
485: these sources as ``priors''. We include them in our catalog because
486: they are very likely to be real.
487:
488: Code 9 refers to objects assumed to be HVCs; no
489: estimate of their distances is made.
490:
491: Notes flag. An asterisk in this column indicates a comment is included
492: for this source in the text below.
493:
494: Only the first few entries of Table 1 are listed in the printed version of this
495: paper. The full content of Table 1 is accessible through the electronic version
496: of the paper and will be made available also through our public digital
497: archive site.
498:
499: \input tab.tex
500:
501: \subsection{Notes and Unique Detections}
502:
503: In addition to the detections reported by Kent \etal (2007) and Haynes \etal (2007) that exhibit
504: no apparent optical counterparts, a number of other detections exhibit unique properties:
505: ambiguous optical identification in need of optical followup, disturbed HI morphologies,
506: and HI connecting members of a galaxy pair or group. Some have presented
507: challenges in the data reduction process, and special care has been
508: taken to accurately extract the spectra. Other clear detections may or may not
509: be associated with previously uncataloged low surface brightness detections
510: in the vicinity of a given HI centroid.
511: Many of these detections will be described in future publications
512: with the added benefit of followup observations and detailed analysis
513: both at radio and other wavelength regimes (Kent \etal in preparation).\\
514:
515: \noindent Notes associated with the objects listed in Table 1 follow:
516:
517: \input notes.tex
518:
519: \section{Statistics of the Catalog}\label{stats}
520:
521: \ncat~detections are presented for this region: \ndet ~($\sim$76\%)
522: are code ``1'' detections that rate fair to excellent in quality;
523: \nprior ~(23\%) are code ``2'' detections that have known priors
524: associated with them -- optical redshifts corroborate these lower
525: S/N detections; the remaining \nhvc ~detections ~(2\%) are
526: high--velocity clouds. The completeness and reliability
527: of the ALFALFA detection process are discussed by Saintonge (2007).
528: Follow-up observations on lower signal-to-noise candidates from the entire
529: survey are currently underway and will be discussed in future publications.
530:
531: 26\% of the detections presented in this catalog have $cz_{\odot} < 3000$ \kms,
532: a large fraction explained by the fact that this portion of the survey crosses
533: the Virgo region and supergalactic plane where the density of galaxies is higher.
534: The detection rate is 4.3 objects per square degree.
535: Excluding the HVC detections, $\sim$2\% do not have
536: apparent optical counterparts. These ``optically unseen'' HI clouds
537: were reported by Kent \etal (2007). Of particular interest is
538: a cloud complex of five detections located halfway between M87
539: and M49. If not gravitationally bound, the system will disperse
540: in less than a cluster crossing time. The detections
541: and aperture synthesis followup are described by Kent (2007b).
542:
543: The described catalog, when combined with that of strips
544: to the North presented by Giovanelli \etal (2007) presents a wealth
545: of information to complement existing and future Virgo Cluster
546: surveys across the spectrum. The survey's
547: power in redshift detection in the local Universe can
548: be compared with previous HI surveys. ALFALFA improves upon previous
549: HI surveys in numbers
550: and positional accuracy of the first generation
551: HI survey HIPASS.
552: Wong \etal (2006)
553: reported on 40 detections
554: from the HIPASS survey in the same ALFALFA region for this paper's catalog.
555: ALFALFA also complements the {\it GOLDMINE} database (Gavazzi \etal~2003),
556: an optically selected compilation of Virgo galaxies. Long integrations
557: were used on some of the {\it GOLDMINE} detections, as a result
558: some are fainter than the ALFALFA detection limit.
559: 16\% of the detections presented in this catalog give
560: new redshifts, and 60\% are detected in HI for the first time. %BK check these numbers
561:
562: The region of sky presented with this catalog lies south of M87 and
563: is shown in Figure 1. The upper panel displays
564: all detections, while the lower panel shows detections found at
565: $cz_{\odot} < 3000$~\kms. A R.A. versus $cz_{\odot}$ cone diagram is shown in Figure 2.
566: Various distributions of the catalog parameters are shown in Figure 3:
567: (a) shows large scale structures of Virgo and the A1367-Coma supercluster regime in
568: a velocity distribution, as well as the high $cz$ gap near 15,000 \kms resulting from RFI; panels
569: (b), (c), and (d) show the distributions
570: for the velocity width W50, the integrated flux and S/N ratio respectively.
571: The final panel (e) shows the HI mass distribution, with distances computed
572: from velocity flow models (Tonry \etal 2000; Masters \etal 2004). The
573: uncertainty in the distance of galaxies projected on a narrow cone centered on the Virgo Cluster causes detections to
574: have uncertain calculated HI masses. Other distance indicator techniques
575: will need to be employed in order to properly determine an accurate HI mass function.
576: This topic will be discussed in future publications.
577:
578: The top panel of Figure 4 shows the Sp\"{a}nhauer diagram, detailing the HI mass vs. distance for
579: the objects in the catalog. The placement of nearby objects
580: at the Virgo distance by the velocity flow model is clearly seen as the vertical
581: gathering of objects at 16.7 Mpc. Regions affected by RFI are also indicated with
582: vertical dashed lines. The middle panel of Figure 4 shows the S/N ratio versus
583: measured velocity width of the catalog entries, demonstrating the independence of
584: signal extraction on velocity widths. The bottom panel
585: of Figure 4 shows the integrated flux versus measured velocity width. The dashed
586: line indicates the a S/N limit of 6.5, the rough limit of code 1 objects listed
587: in Table 1.
588:
589: \section {Positional Accuracy of Cataloged Sources \label{pos}}
590:
591: Continuum sources are extracted and their positions measured in a ``telescope''
592: coordinate reference frame. This reference frame is determined at the Observatory
593: separately for each receiver system by optimizing a model fit to a set of
594: pointing calibrators. Repeatable residuals from this model translate into small
595: scale pointing errors, which typically have amplitudes on order of 15\arcsec~ for
596: the ALFA array receivers. These errors vary with both azimuth and zenith angle.
597: Because ALFALFA observations are made with ALFALFA at the meridian (constant azimuth),
598: and consist of long runs at constant zenith angle, the ALFALFA sources' positions in
599: the telescope reference frame easily reveal the systematic errors represented by those
600: residuals, when they are compared with the positions of a
601: higher resolution -- and thus higher accuracy -- coordinate reference system.
602: In this section we compare HI source positions obtained in the Arecibo telescope
603: reference system with two such, more accurate position sets: those of optical
604: galaxies and those of VLA radio continuum sources (NVSS; Condon \etal 1998).
605:
606:
607: \subsection{Positional Differences between HI and Optical Sources}
608:
609: The accuracy of positional HI centroids is important in establishing correct optical
610: identifications. The ALFALFA
611: catalogs are presented to the astronomical community with
612: positions as accurate as possible. As this catalog lies just south of the Paper III catalog,
613: the telescope configuration used in obtaining the data is similar. The impact
614: of the telescope pointing is similarly recovered as we describe below.
615:
616: The 3.3\arcmin$\times$3.8\arcmin~ FWHM elliptical beam of the Arecibo telescope
617: affords the ALFALFA
618: survey greater positional accuracy than previous large scale blind HI surveys.
619: This factor in the survey characteristics becomes important
620: when matching HI detections with their optical counterparts,
621: if they exist. Positional accuracy is limited by the beam
622: size, signal--to--noise, and errors in the telescope's pointing. For
623: the region of sky surveyed in this paper, the ALFA receiver system
624: is oriented such that the beam's major axis
625: is in the Declination direction and the telescope's
626: azimuth arm is aligned in the North--South direction.
627:
628: Processed ALFALFA drift scans are combined and re-gridded
629: into a data cube that is spatially sampled at 1\arcmin$\times$1\arcmin~.
630: Gaussian weighting is also applied and the final resolution
631: is 3.8\arcmin$\times$4.3\arcmin.
632: Sources and their associated measured parameters are
633: extracted from these derived data cubes. Source extraction
634: is carried out via an automated matched filter algorithm
635: with visual followup confirmation by a project team member.
636: Once a source is identified, a 2-D map is integrated over the spectral
637: extent of the object, with ellipses fit to the quarter and
638: half--power levels of the measured peak, as well as other preset
639: isophotal levels. As most sources are unresolved by the ALFA beam,
640: the HI source position is taken to be the center of the half--power
641: ellipse. Extended objects that may
642: exhibit warped or asymmetric disks are treated with extra care
643: as the ellipse may not provide the best estimate of the
644: position center.
645:
646: The systematic pointing errors of the telescope can be recovered
647: due to the fixed azimuth telescope observing mode employed by the survey.
648: These errors add to $\sim$15\arcsec~ or more at L--band frequencies with
649: the maximum pointing errors occurring when the telescope is at high (or very low)
650: zenith angles.
651:
652: Positions of optical counterparts of HI sources are obtained by team members through
653: the Digital Sky Survey, from the SDSS database, or through the local
654: Arecibo General Catalog maintained at Cornell.
655: Figure 5 shows the
656: positional differences (HI minus optical) for four different
657: Declination ranges, each binned by 1$^{\circ}$.
658: The points correspond to sources that have optical counterparts --
659: optically inert HI clouds, HVC, and tidal tails are not included.
660: The position differences can be attributed to centroiding statistical
661: errors, optical misidentification, and the previously mentioned
662: telescope pointing offsets. The optical centroid errors are negligible when
663: compared to the other source errors. The telescope pointing errors
664: contribute to the majority of the systematic error budget, as illustrated by
665: offsets in the data distribution from the center of the plots.
666:
667: Signal--to--noise affects the accuracy of
668: the HI position centroid. Figure 6 shows the positional
669: accuracy of the HI positions, after correction for telescope pointing errors, for various S/N regimes.
670: For the higher S/N detections the median centroid error is approximately 15\arcsec.
671:
672:
673:
674: \subsection {Positional Errors as Derived from Radio Continuum Sources}
675:
676:
677: In addition to spectral data cubes, ALFALFA data
678: are used to generate regularly gridded continuum maps.
679: Continuum sources are extracted and their positions determined.
680: These maps are used to determine pointing corrections for the ALFALFA
681: catalog objects by comparing
682: centroid positions with those published by
683: the NRAO VLA Sky Survey (NVSS; Condon \etal 1998).
684: The positions presented in this catalog
685: and in catalogs published on the online ALFALFA data archive are
686: corrected via the method presented here. The pointing
687: corrections utilize data from sixty-four data cubes from the Paper III catalog
688: and the catalog presented in the paper.
689:
690: Continuum sources for position comparison are selected with a priori
691: information from the NVSS.
692: For identification accuracy,
693: the positions of high signal--to--noise objects
694: are used.
695: Therefore, for each region of sky defined by an ALFALFA data cube, an NVSS source
696: subset
697: is identified with a peak intensity $I_{NVSS} > 20$~mJy beam$^{-1}$.
698: The average r.m.s of the NVSS source catalog is 0.48 mJy beam$^{-1}$.
699: For each of the sources in the subset, a 9\arcmin~ box
700: is cutout for the ALFALFA data centered on the NVSS centroid position.
701: A 2-D Gaussian fit is attempted in this window. If the fit converges,
702: then the position from peak of the fit is compared with any NVSS sources
703: in the box. This process consists of comparing the ALFALFA continuum
704: centroid with any possible NVSS positions and flux totals in the 9\arcmin~ box.
705: If the sum of the NVSS fluxes differs from the measured ALFALFA flux
706: by more than 50\%, then the source is not used in the pointing correction.
707: Because of the differences in beam sizes,
708: several NVSS sources may be confused and seen
709: as one source in the ALFALFA beam. Therefore, if
710: flux agreement is achieved between the ALFALFA and NVSS entries,
711: acceptable sources are saved and position differences are computed in Right
712: Ascension and Declination.
713:
714: Using this automated method for cross-identification of NVSS sources yields
715: 1583 objects with centroid positions in the ALFALFA data cubes covering
716: $\sim$240 deg$^{2}$ of sky. The differences between the ALFALFA
717: and NVSS centroids are plotted in Figure 7. These position differences
718: are binned in quarter degree Declination increments.
719: The robust mean position for each Declination bin is plotted in Figure 8,
720: with vertical bars indicating the error on the mean and the horizontal
721: bars indicating the bin width. A third order polynomial is fit through
722: these averaged points, and the resulting coefficients are used to
723: correct the pointing in the catalog. The positions
724: for the HI detections published in this
725: paper reflect these pointing corrections, which remove telescope pointing
726: errors from the HI centroids in our catalogs.
727:
728: \section {Summary}\label{summary}
729:
730: This paper completes another important part of the ALFALFA survey toward the
731: Virgo Cluster region. The catalog presented here completes the Virgo ``core''
732: area centered around M87 and its periphery. The sample of galaxies detected
733: yield a redshift distribution exhibiting large scale structure from Virgo
734: and the Coma Supercluster regime, shown in Figures 2, 3 and 4.
735: The catalog covers 132 \sqd~of sky south of M87, completing another 2\% of the survey.
736: The complementary conclusions to the Paper III catalog are as follows:
737:
738: \begin{itemize}
739: \item
740: The third ALFALFA catalog installment presented here completes the core
741: of the Virgo cluster region initiated with Paper I. Together, Paper I
742: and Paper VI complement the initial anti-Virgo catalog (Saintonge \etal 2008)
743: for future analysis.
744:
745: \item
746: Objects of interest include a cloud complex lying midway between
747: M87 and M49 that do not correlate with any apparent optical counterpart.
748: The complex exhibits a velocity dispersion of $\sim$250~\kms~ between
749: 400 and 640~\kms. The spatial extent of the complex is over 200 kpc
750: if it is at the Virgo distance. An
751: analysis of this system, including aperture synthesis followup,
752: will be detailed in Kent \etal (in preparation) .
753:
754:
755:
756: \item
757: The median $cz$ for the catalog presented here is 6500~\kms; the redshift distribution
758: is analogous to that of the first Virgo catalog (Giovanelli \etal 2007).
759:
760: \item
761: A number of high velocity clouds with velocities between -150 and +200 \kms~are
762: detected. Given the proximity of the region surveyed to the North
763: Galactic Pole, their heliocentric velocities are minimally affected by
764: galactic rotation and represent true infall or outflow motions.
765:
766:
767:
768: \end{itemize}
769:
770: In conclusion, we present the third catalog release of the ALFALFA survey,
771: bringing the total projected sky coverage to $\sim$5\%.
772: This installment of the survey and its associated data products will be incorporated
773: into the more extensive digital HI dataset at
774: {\it http://arecibo.tc.cornell.edu/hiarchive/alfalfa/}.
775:
776: RG and MPH acknowledge the partial support of NAIC as Visiting
777: Scientists during the period of this work. RAK and TB thank NAIC for partial support. RAK, TB, and NB
778: are grateful to NAIC and the Cornell University Astronomy Department
779: for its hospitality to them as sabbatic visitors. This work has been supported by NSF
780: grants AST--0307661, AST--0435697, AST--0347929, AST--0407011, AST--0302049;
781: and by a Brinson Foundation grant. We thank the Director of
782: NAIC, Robert Brown, for stimulating the development of major ALFA surveys,
783: H\' ector Hern\' andez for his attention to the telescope scheduling and the Director,
784: telescope operators and support staff of the Arecibo Observatory for their proactive
785: approach. We thank Tom Shannon for his advice and assistance with hardware,
786: system and network issues at Cornell. We also thank the referee for
787: their useful comments on this work.
788:
789: \vskip 0.3in
790:
791: \newpage
792:
793:
794:
795:
796:
797: %FIGURE 1
798: \begin{figure}[h]
799: %\figurenum{1}
800: \plotone{f1.eps}
801: \caption{Spatial distribution of HI candidate detections
802: listed in Table 1. The upper panel shows all detections
803: presented in this catalog. The lower panel shows
804: all objects with $cz_{\odot}<3000$~\kms~.}
805: \label{skyplot}
806: \end{figure}
807:
808:
809: %FIGURE 2
810: \begin{figure}[h]
811: %\figurenum{1}
812: \epsscale{0.7}
813: \plotone{f2.eps}
814: \caption{Cone diagrams showing right ascension
815: vs. $cz_{\odot}$ for sources listed
816: in Table 1. The top diagram shows all data for this paper's catalog.
817: The bottom diagram shows a close-up view for data at $cz_{\odot}~\leq~$3000~\kms.}
818: \label{coneplot}
819: \end{figure}
820:
821: %FIGURE 3
822: \begin{figure}[h]
823: %\figurenum{2}
824: \epsscale{0.5}
825: \plotone{f3.eps}
826: \caption{Various histograms of the HI detections
827: listed in Table 1. (a) shows the $cz$ distribution in \kms;
828: (b) shows the HI line width W$_{50}$ in \kms;
829: (c) shows the base 10 logarithm of the integrated flux in Jy \kms;
830: (d) shows the base 10 logarithm of the peak S/N ratio;
831: (e) shows the derived HI mass, in base 10 logarithmic units of \msun. The
832: HI masses of some detections may be skewed to higher estimates
833: due to distance uncertainties toward the Virgo cluster region.}
834: \label{histograms}
835: \end{figure}
836:
837: %FIGURE 4
838: \begin{figure}[h]
839: %\figurenum{3}
840: \epsscale{0.8}
841: \plotone{f4.eps}
842: \caption{Statistical properties of the Virgo south catalog.
843: The top panel shows the HI mass vs. distance for all sources
844: of type 1 and 2 presented in Table 1. Note that galaxies
845: assigned to the Virgo cluster appear as the vertical
846: feature at 16.7 Mpc, and the gap at 220 Mpc is due
847: to RFI. The middle panel shows the log$_{10}$ S/N ratio
848: versus velocity width W$_{50}$. The lower envelope
849: is independent of S/N which indicates that
850: no significant bias is present in the detection
851: of sources of large width. The bottom panel
852: shows the integrated flux versus velocity width W$_{50}$,
853: with the dashed line indicating a S/N limit of 6.5.}
854: \label{statistics}
855: \end{figure}
856:
857:
858:
859: %FIGURE 5
860: \begin{figure}[ht!]
861: %\figurenum{4}
862: \epsscale{1.0}
863: \plotone{f5.eps}
864: \caption{Differences between the positions of the HI sources, as indicated by
865: the values embedded in the source names in Col. 1 of Table 1, and the
866: optical counterpart positions as listed in Col. 4 of Table 1. Sources
867: are separated by Declination bins of $1^\circ$: the label ``Dec 8'' identifies
868: sources with Declination between +8$^\circ$ and +9$^\circ$, etc. Average
869: and median offsets, expressed in arcseconds, are inset within each panel.}
870: \label{positions1}
871: \end{figure}
872:
873:
874: %FIGURE 6
875: \begin{figure}[ht!]
876: %\figurenum{5}
877: \epsscale{1.0}
878: \plotone{f6.eps}
879: \caption{Differences between the positions of the HI sources, as indicated by
880: the values in Col. 3 of Table 1, which are corrected for systematic telescope
881: pointing errors, and the
882: optical counterpart positions as listed in Col. 4 of Table 1. Sources
883: are separated by S/N as indicated in each panel. Average and
884: median offsets, expressed in arcseconds, are inset within each panel.
885: }
886: \label{positions2}
887: \end{figure}
888:
889: %FIGURE 7
890: \begin{figure}[ht!]
891: %\figurenum{6}
892: \plotone{f7.eps}
893: \caption{Positional differences between 1568 ALFALFA and NVSS continuum
894: sources (Condon \etal~1998). The offsets in the mean values of $\Delta$R.A. and $\Delta$Decl.
895: are due to the Arecibo telescope pointing errors.}
896: \label{continuumall}
897: \end{figure}
898:
899:
900: %FIGURE 8
901: \begin{figure}[ht!]
902: %\figurenum{7}
903: \plotone{f8.eps}
904: \caption{Positional difference between
905: ALFALFA and NVSS continuum sources (Condon \etal~1998) averaged over 0.25 degree declination bins.
906: Vertical bars indicate the error on the mean for each bin. Horizontal
907: bars indicate the bin width for a given point.
908: }
909: \label{continuumavg}
910: \end{figure}
911:
912:
913:
914: \input bib.tex
915:
916:
917: \vfill
918: \end{document}
919:
920:
921: