Optically Unseen HI Detections towards the Virgo Cluster detected in the Arecibo Legacy Fast ALFA Survey
Brian R. Kent, Riccardo Giovanelli, Martha P. Haynes, Amelie Saintonge, Sabrina Stierwalt, Thomas Balonek, Noah Brosch, Barbara Catinella, Rebecca A. Koopmann, Emmanuel Momjian, Kristine Spekkens
aa r X i v : . [ a s t r o - ph ] J u l Optically Unseen HI Detections towards the Virgo Clusterdetected in the Arecibo Legacy Fast ALFA Survey
Brian R. Kent , Riccardo Giovanelli , , Martha P. Haynes , , Am´elie Saintonge , SabrinaStierwalt , Thomas Balonek , , Noah Brosch , Barbara Catinella , Rebecca A.Koopmann , , Emmanuel Momjian , Kristine Spekkens ABSTRACT
We report the discovery by the Arecibo Legacy Fast ALFA (ALFALFA) sur-vey of eight HI features not coincident with stellar counterparts in the VirgoCluster region. All of the HI clouds have cz < − and, if at the Virgodistance, HI masses between 1.9 × M ⊙ and 1.1 × M ⊙ . Four of the eightobjects were reported or hinted at by previous studies and “rediscovered” byALFALFA. While some clouds appear to be associated with optical galaxies intheir vicinity, others show no clear association with a stellar counterpart. Two ofthem are embedded in relatively dense regions of the cluster and are associatedwith M49 and M86; they were previously known. The others are mostly locatedin peripheral regions of the cluster. Especially notable are a concentration of ob-jects towards the so-called M cloud, 3 ◦ to 5 ◦ to the NW of M87, and a complex Center for Radiophysics and Space Research, Space Sciences Building, Cornell University, Ithaca,NY 14853. e–mail: [email protected], [email protected], [email protected],[email protected], [email protected] National Astronomy and Ionosphere Center, Cornell University, Space Sciences Building, Ithaca, NY14853. The National Astronomy and Ionosphere Center is operated by Cornell University under a cooperativeagreement with the National Science Foundation. Dept. of Physics & Astronomy, Colgate University, Hamilton, NY 13346. e–mail: [email protected] The Wise Observatory & The School of Physics and Astronomy, Raymond & Beverly Sackler Faculty ofExact Sciences, Tel Aviv University, Israel. e–mail: [email protected] Arecibo Observatory, National Astronomy and Ionosphere Center, Arecibo, PR 00612. e–mail: [email protected], [email protected] Dept. of Physics & Astronomy, Union College, Schenectady, NY 12308. e–mail: [email protected] National Radio Astronomy Observatory and Dept. of Physics & Astronomy, Rutgers University, 136Frelinghuysen Road, Piscataway, NJ 08854. NRAO is a facility of the National Science Foundation operatedunder cooperative agreement by Associated Universities, Inc. e–mail: [email protected]
Subject headings: galaxies: intergalactic medium — galaxies: halos — individ-ual:Virgo cluster — radio lines: galaxies — galaxies:clusters — galaxies:interactions
1. Introduction
The predictions of the galaxy formation paradigm in the hierarchical scenario requirescorroboration by observational data. Optically selected samples of galaxies do not detect lowluminosity, and hence presumably low mass objects, in the predicted numbers. This has oftenbeen referred to as the “substructure” or “missing satellite” problem (Klypin et al. et al. cz < − . At this timemore than half of the solid angle encompassing the Virgo Cluster has been fully surveyed.ALFALFA can detect ∼ × M ⊙ at the cluster distance, which in this paper will beassumed to be 16.7 Mpc. The Virgo Cluster offers a fertile environment for the possibledetection of gas-rich, optically faint systems.A number of extragalactic HI clouds have been reported in the past (e.g. Schneider etal. et al. et al. etal. et al. et al. et al. et al. et al. in preparation);others are thought to be the result of ram pressure stripping in the cluster environment, e.g.the Oosterloo & van Gorkom feature near NGC 4388. The interactions between sub-groupsin Virgo play an important role in the cluster’s evolution, acting as a preprocessing step ofthe material as the galaxies fall into the cluster.Here we present a catalog of 21 cm line sources detected by ALFALFA in the centralportion of the Virgo Cluster region which have no obvious optical counterparts. Parameters 3 –of the detections and descriptions of their environments are given. Three of the sources werepreviously reported and one hinted at by other studies and “rediscovered” by ALFALFA.
2. Observations and Source Parameters
The sources presented in this paper are part of the ALFALFA catalog and refer tothe region 12 h < R.A.(J2000) < h and +8 ◦ < Dec.(J2000) < +16 ◦ . Complete sourcecatalogs of this region have been reported by Giovanelli et al. (2007) and Kent et al. (2007 inpreparation); they are accessible at http://arecibo.tc.cornell.edu/hiarchive/alfalfa/. Sourcesare extracted from the ALFALFA data set via an automated algorithm (Saintonge 2007),successively inspected by eye, measured and classified according to a code which primarilydepends on signal to noise (S/N). The objects reported in this paper are all classified as bona fide detections (code 1; see Giovanelli et al. & S/N = (cid:16) F c W (cid:17) w / smo σ rms (1)where F c is the integrated flux density in Jy km s − , w smo is either W / (2 ×
10) for W < − or 400 / (2 ×
10) = 20 for W ≥
400 km s − [ w smo is a smoothing width expressed asthe number of spectral resolution bins of 10 km s − bridging half of the signal width], and σ rms is the r.m.s noise figure across the spectrum measured in mJy at 10 km s − resolution.All of them have been confirmed by corroborating observations carried out with the Arecibotelescope or with the Very Large Array (VLA) as discussed below. Details of the ALFALFAobservations can be found in Giovanelli et al. (2005) and Giovanelli et al. (2007).Table 1 contains the observed and derived parameters of the HI clouds. Their velocitiesindicate that an association with the Virgo cluster is possible for most of them, with theexception of the group associated with NGC 4795/4796. The latter is more likely to be inthe background of the Virgo cluster, at a distance of ∼
40 Mpc which is assumed for theclouds in that group.The fields of each of the sources in Table 1 have been inspected in the Sloan Digital Sky 4 –Survey and the DSS2 via Skyview. The contents of Table 1 are as follows.
Col.(1) - Cloud ID number
Col.(2 & 3) - HI source center coordinates (J2000); these positions are typically accurate towithin 30 ′′ or better (see Giovanelli et al. Col.(4) - Heliocentric velocity in km s − Col.(5) - Velocity width measured at half peak power in km s − Col.(6) - Integrated flux in Jy km s − Col.(7) - Signal to noise ratio
Col.(8) - Base 10 logarithm of the HI mass in solar units, assuming HI is optically thin
Col.(9) - Angular distance from M87 in degreesFor three of the sources, we separately list the parameters of several clumps, identifiedwith italic qualifiers. Figure 1 shows locations of the sources within the Virgo cluster region;the grayscale background image shows hard X-ray counts (0.5-2.0 keV) from the ROSATdataset of Snowden et al. (1995), smoothed with a 5 ′ kernel. The approximate boundaries ofthe M and W ′ clouds are indicated by dashed circles (Binggeli, Popescu & Tammann 1993).We note that the fields of the HI clouds often contain one or several small optical objects; thepossibility that one of them may be a small dwarf or low surface brightness galaxy associatedwith the HI source cannot be excluded at this time. We discuss the characteristics of eachHI source below. Cloud 1 .–This object, unresolved by the Arecibo beam of 3.3 ′ × ′ , is near thedetection limit of ALFALFA at the Virgo distance; S/N and spectrum used are those ofthe ALFALFA survey observations. That detection has been confirmed by successive, moresensitive Arecibo observations. This is the object with the lowest HI mass in Table 1. Wenote that the Irr galaxy UGC 7003 lies 30 ′ NW of the HI source at cz = 1286 km s − , the Skyview was developed and maintained under NASA ADP Grant NAS5–32068 under the auspices of theHigh Energy Astrophysics Science Archive Research Center at the Goddard Space Flight Center Laboratoryof NASA. ′ W at cz = 1524 km s − and UGC 7038 lies 32 ′ NW at cz = 889km s − . Cloud 2 .–This object, also unresolved by the Arecibo beam, lies 3.8 ′ away from a smalloptical galaxy for which no optical redshift is known. The optical galaxy AGC 220171 at121035.6+114539, an apparently undisturbed object classified as a BCD, lies 29 ′ to the SEat a redshift of cz = 1296 km s − . A VLA map of the source has been obtained and theresults will be discussed in a forthcoming study. Cloud 3 .–This object, also unresolved by the Arecibo beam, is located in a crowded field- a host of galaxies are in the surrounding periphery at a comparable redshift. The densityof galaxies with known optical or HI redshifts in this region is high and 21 are known with1800 < cz < − within one degree of the HI source. The nearest optical galaxywith similar velocity is AGC 221651 at 121336.4+130201, 8 ′ N of the HI feature at cz = 1932km s − . A VLA map of the source has been obtained and the results will be discussed in aforthcoming study. Cloud 4 .–The HI source is located 3.6 ′ southeast of M86 (=NGC 4406; cz = − − ), the large Virgo S0 galaxy. It was previously reported by Davies et al. (2004)as VIRGOHI4. They also suggested that the source could lie behind M86. HI synthesisimaging by Oosterloo & van Gorkom (2005) showed the HI source to be a plume extendingfrom NGC 4388( cz =2524 km s − ) and possibly resulting from interaction between thatgalaxy and the intracluster gas. Jacoby et al. (2005) indicated that H α filaments detectedcould be associated with this plume. In the same vicinity ALFALFA makes a separatedetection of M86 (Giovanelli et al. ′′ to the south. The features reported by Bregman & Roberts (1990) are associated withthe ALFALFA HI detection of M86. Cloud 5 .–This object was first reported by Sancisi et al. (1987). Later, aperture synthesisobservations were presented by Henning et al. (1993) and McNamara et al. (1994). The HIsource is located 2.6 ′ southeast of M49( cz = 997 km s − ) and it is proposed to be related tothe interaction between M49 and the dwarf irregular UGC 7636( cz = 276 km s − ). Cloud 6 .–Two sources(6 a&b ) are the brightest clumps in a stream found in the vicinityof NGC 4254(M99; cz = 2407 km s − ). The brightest of the two is an object first detectedat Jodrell Bank (Davies et al. et al. et al. c ) lies 18 ′ west of NGC 4254. A more detailed analysis of the ALFALFA evidenceon this object is given in Haynes et al. (2007). Cloud 7 .–A complex of five HI clouds, it projects between M87 and M49, roughly 3 ◦ south of the former. The five clouds, spread in velocity between 480 and 607 km s − , extendover approximately 35 ′ in the plane of the sky, or 200 kpc at the Virgo cluster distance. InHI mass, the clouds range between 0 . × and 2 . × M ⊙ . VLA observations havebeen obtained and clouds 7c and 7d have been clearly detected. Detailed results of both theALFALFA and VLA observations are in preparation by Kent et al.Cloud 8 .–Three clouds comprise an HI complex that surrounds the SB0/a galaxy NGC 4795( cz =2781 km s − ) and its dwarf companion NGC 4796( cz = 2406 km s − ). It was previously notedthat a large flux measurement discrepancy between Arecibo and Effelsberg measurementswas likely due to an offset of the HI from the center of the NGC 4795/4796 pair (Hoffman etal. cz = 2801 and cz = 2856 km s − . The HI massescomputed in Table 1 assume a distance of ∼
40 Mpc. The kinematics of the HI indicate thatall the galaxies in this system are at a comparable redshift and are interacting as a group.
3. Discussion
The characteristics of the eight sources reported here are quite diverse. They can begrouped into three categories: isolated objects, objects in the vicinity of large galaxies, anddisturbed objects that could be remnants of an encounter with a larger system, such asthe collective cluster potential. Except for cloud 8, the projected distances to M87 andheliocentric velocities place the HI sources within the canonical boundaries of the VirgoCluster, as defined by Binggeli, Sandage, & Tammann (1985).Clouds 4 and 5 have been extensively studied in the past, using higher aperture synthesisdata which indicate a clear association with Virgo cluster galaxies M49 and NGC 4388. Theproximity to these galaxies and the cluster environment are important in determining theirproperties. The clouds in the vicinity of NGC 4795 (cloud 8) are most likely the result oftidal stripping in interacting members of a close group, not an uncommon occurrence.Cloud 6 has raised significant attention, as it was proposed to be a rare representativeof a category of massive, yet starless galaxies. The ALFALFA observations clearly show 7 –that rather than an isolated “dark galaxy,” this object is an extended stream connectedto NGC 4254. While no nearby companion with which NGC 4254 may have had a closeencounter is clearly identifiable, the origin of the feature appears most likely to be of a tidalnature, an event of “harassment” of the type graphically illustrated in simulations, e.g. byMoore et al. (1996) and Lake et al. (1998).ALFALFA observations of clouds 1, 2 and 3 offer no clear hints as to their origin. Theyare all unresolved by the Arecibo beam, hence their HI is contained within a ∼
10 kpcdiameter or less. The HI masses of the clouds are relatively low, especially for clouds 1 and2, and the lack of a size determination impedes an estimate of their dynamical masses; upperlimits for the latter on order of 1–6 × M ⊙ can be obtained if one assumes turbulent orrotational motion amplitudes as indicated by the velocity widths. These objects are projectednear or within the boundaries of the M cloud, a loose subclump thought to be behind andfalling into the main cluster around M87, although little coherence in the velocities of thetrio exists to firmly substantiate such association. At any rate, they appear to be relativelyisolated and far removed from the central parts of the cluster so that gravitational, ratherthan hydrodynamical processes involving the intracluster gas, are more likely to be invokedin explaining their nature. While the mean number density of VCC galaxies in the M cloudregion is less than 1.5 × − (arcmin) − (Binggeli, Sandage, & Tammann 1985, Schindler etal. − in extent. With a meanvelocity near 540 km s − , the complex could well be located in the foreground of the clus-ter, although cluster galaxies of similar redshift are found in the vicinity. Most notably,NGC 4424, an SBa at cz = 476 km s − is located about 40 ′ to the west of the complex.ALFALFA and VLA HI maps of that object (Chung et al. et al. ∼
250 kpc Gyr − . Differ-ential motions of this amplitude are consistent with tidal forces within the cluster potentialand suggest the complex may disperse within a cluster crossing time. A more detailed anal-ysis based on ALFALFA and VLA observations will be explored in a future study (Kent etal. , in preparation). 8 –While a detailed statistical study of the ALFALFA catalogs in the Virgo regions awaitscompletion of the survey effort therein, it is interesting to preliminarily note that ALFALFAdoes not detect very large numbers of low HI mass sources in the cluster. This is an indicationthat, at least in the cluster region, the HI mass function faint end slope does not risesufficiently to significantly contribute to solving the “substructure” problem mentioned inour introduction.We gratefully acknowledge Dr. Tom Oosterloo for his critical reading of this paperand providing comments. This research has made use of the NASA/IPAC ExtragalacticDatabase (NED) which is operated by the Jet Propulsion Laboratory, California Institute ofTechnology, under contract with the National Aeronautics and Space Administration. KS isa Jansky Fellow. This work has been supported by NSF grants AST–0307661, AST–0435697,AST–0607007 and the Brinson Foundation. REFERENCES
Binggeli, B., Sandage, A., & Tammann, G. A. 1985, AJ, 90, 1681Binggeli, B., Popescu, C. C., & Tammann, G. A. 1993, A&AS, 98, 275Bregman, J. N., & Roberts, M. S. 1990, ApJ, 362, 468Chengalur, J. N., Giovanelli, R., & Haynes, M. P. 1995, AJ, 109, 2415Chung, A., van Gorkom, J. H., Kenney, J. D. P., & Vollmer, B. 2007, ApJL, 659, L115Cort´es, J. R., Kenney, J. D. P., & Hardy, E. 2006, AJ, 131, 747Davies, J., Minchin, R., Sabatini, S., van Driel, W., Baes, M. et al. et al. et al. et al. et al. et al.
This preprint was prepared with the AAS L A TEX macros v5.2.
10 –Table 1. ALFALFA Optically Unseen Detections
Cloud ID α δ cz ⊙ W FWHM F c S/N log M HI d M J2000 J2000 (km s − ) (km s − ) (Jy km s − ) M ⊙ deg1 a
12 02 44.4 +14 04 56 1121 ± ± ± b
12 08 45.5 +11 55 17 1230 ± ± ± b
12 13 41.8 +12 53 51 2235 ± ± ± c
12 26 19.4 +12 53 30 2246 ± ±
11 2.05 ± d
12 29 54.7 +07 58 12 473 ± ±
10 1.13 ± a e
12 17 55.5 +14 44 45 1984 ± ± ± b e
12 17 49.1 +15 04 52 2200 ± ±
13 0.52 ± c e
12 17 33.8 +14 23 47 2111 ±
10 65 ±
20 0.57 ± a b
12 29 42.8 +09 41 54 524 ± ±
15 1.16 ± b b
12 30 19.4 +09 35 18 603 ± ± ± c b
12 30 25.8 +09 28 01 488 ± ±
11 2.48 ± d b
12 31 19.0 +09 27 49 607 ± ± ± e b
12 31 26.7 +09 18 52 480 ±
10 53 ±
21 0.91 ± a f
12 55 04.3 +08 06 13 2629 ± ± ± b f
12 55 10.2 +08 02 44 2754 ±
14 407 ±
27 2.91 ± c f
12 55 13.7 +08 02 51 2771 ± ± ± a Confirmed by follow-up, high sensitivity observation at Arecibo. b VLA maps obtained, processing underway. c VirgoHI4 (Davies et al. et al. (2005). d Vicinity of M49, Sancisi et al. (1987); synthesis data by Henning et al. (1993). e Clumps in VirgoHI21, WSRT data by Minchin et al. (2005). f NGC 4795/4796 group.
11 –Fig. 1.— Sky distribution of the HI detections (star symbols and corresponding Table 1numbers) presented in this paper. The X-ray peaks of cluster members M87, M86, and M49are labeled for reference. The background grayscale image is a hard X-ray counts image fromROSAT (Snowden et al. ′ Gaussian kernel. The portions of theM and W ′′