CLOUDS TOWARD THE VIRGO CLUSTER PERIPHERY: GAS-RICH OPTICALLY INERT GALAXIES

Galaxies are characterized by their stellar content, morphology, environment, dust, and neutral and molecular gas content. The 21 cm line of neutral hydrogen (H i) continues to play an important role in diagnosing the star formation potential of a galaxy and any past dynamic interactions with a cluster environment or neighboring galaxies. Blind H i surveys can sample the gas-rich population of the local universe and search for low-mass, low-surface brightness systems.

The Arecibo Legacy Fast ALFA (ALFALFA; Giovanelli et al. 2005a) survey is providing such a sample of gas-rich objects. The project utilizes the seven-element Arecibo L-band Feed Array (ALFA) receiver system to conduct a wide area extragalactic H i investigation with the 305 m Arecibo reflector. The survey improves over previous first-generation surveys (HIPASS: Barnes et al. 2001; Meyer et al. 2004; Wong et al. 2006; ADBS: Rosenberg & Schneider 2002) in spectral and spatial resolution, providing 5 km s^-1 resolution with a 3' beam. The 7000 deg² of surveyed sky covers a velocity range out to cz_☉∼ 18,000 km s^-1. This includes interesting areas of the local supercluster such as the Virgo cluster at cz_☉∼ 1100 km s^-1. The first and second ALFALFA Virgo catalogs (Giovanelli et al. 2007; Kent et al. 2008) comprise a complete mass limited data set (M_H i ⩾ 2 × 10⁷ M_☉ at Virgo). Sources identified in Kent et al. (2007, 2009) comprise a sample of optically inert H i detections; these objects are important in understanding the fate of galaxies in a cluster environment and its surrounding periphery. Previous detections that cannot be correlated with optical counterparts have been associated with nearby galaxy groups or nearby spiral galaxies harassed by the cluster environment (Ryder et al. 2001; Minchin et al. 2005; Oosterloo & van Gorkom 2005; Haynes et al. 2007).

Nearby galaxy cluster environments are of great interest to H i studies as both the gravitational potential and intracluster medium (ICM) perturb the gas structure and morphology of galaxies. Ram pressure stripping and tidal encounters at different cluster radii result in spiral galaxy deficiencies in H i of varying degrees. While three-dimensional paths of galaxies through the cluster are often difficult to ascertain, the fingerprint of the cluster–galaxy interaction is well studied through aperture synthesis observations; high-resolution studies show that H i radii are smaller than their optical counterparts and match predictions of theoretical studies (Giovanelli & Haynes 1983).

Aperture synthesis observations have the ability to resolve higher sensitivity single-dish survey detections. Resolved H i observations reveal the truncation of disks, tidal tails, and the disturbed morphology of late-type spiral galaxies (Cayatte et al. 1990, 1994; Chung et al. 2007). All of these properties are indicators of galaxy–galaxy and galaxy–cluster interactions. It has been shown that H i deficiency in late-type galaxies decreases with increasing cluster radius (Giovanelli & Haynes 1985). An important question of galaxy and cluster evolution that remains is: what happens to the stripped gas in the cluster environment? Will it be destroyed by ablation and evaporate into the cluster halo? Is it possible that an isolated cloud can survive and re-initiate star formation?

Here we present H i aperture synthesis observations of two H i clouds in the Virgo Cluster periphery, initially reported in Kent et al. (2007). These clouds, unresolved by Arecibo, are resolved into separate clumps with Very Large Array (VLA)-C observations. In Section 2, we discuss the original Arecibo observations, data reduction, and detections. In Section 3, we describe the follow-up VLA observations, data reduction, and detections. In Section 4, we detail the environment and neighboring galaxies of the H i clouds. In Section 5, we discuss possible cloud origins and compare to other gas-rich optically inert phenomena. Section 6 summarizes the results of the study.

The H i clouds described here were detected by the ongoing ALFALFA survey. The meridian transit observing strategy uses a sky drift mode with a 100 MHz bandwidth and 4096 channels per polarization, centered at 1385 MHz. Each raw scan contains 14 spectra (7 beams × 2 linear polarizations per beam), with a sampling rate of 1 Hz and spectral resolution of 24.4 kHz (δV = 5.1 km s^-1 at the rest frequency of the 21 cm H i line). The system temperatures of the ALFA receivers during the observations were in the range 26 K < T_sys < 30 K, yielding a root mean square (rms) noise of σ_m = 2.5 mJy beam⁻¹ in channels with δV = 5.1 km s^-1. The flagging, calibration, and gridding of the data into cubes are described in detail by Giovanelli et al. (2005a), Kent (2008), and Kent et al. (2009). Table 1 describes the parameters of the Arecibo observations.

Table 1. ALFALFA Observing and Data Cube Parameters

Parameter	Value
Spectral range	25 MHz (−2000 to 3200 km s-1)
Effective integration time	48 s (beam solid angle)−1
Spectral resolution δV	24.4 kHz (5.1 km s-1)
Half-power beam size	33 × 38
rms noise σm for δV = 5.1 km s-1	2.5 mJy beam−1

Download table as:  ASCIITypeset image

The two H i clouds (henceforth Clouds 1 and 2) were detected in the ALFALFA data obtained in the spring 2005 campaign sampling the Virgo cluster and its surrounding periphery. The detections were reported in Kent et al. (2007) as part of an initial collection of gas-rich, optically inert extragalactic objects. A complex of H i clouds situated halfway between M87 and M49 and their VLA observations were examined in Kent et al. (2009). Here we continue this effort with two H i clouds located in the M cloud periphery west of the main A cluster. Upon detection in the ALFALFA survey, both objects were re-observed and confirmed with the single-pixel L-band Arecibo receiver. Integrated spectral profiles for each cloud are depicted in red in Figure 1. The Arecibo observations show both sources as narrow, single-peaked spectral profiles. Clouds 1 and 2 are unresolved by the Arecibo beam and are located 54 (1.5 Mpc in projection) and 42 (1.2 Mpc in projection) from M87, respectively, west toward the direction of the M cloud. The Arecibo detections are unresolved point sources and we cannot deduce any information about the morphology of the sources.

Zoom In Zoom Out Reset image size

Table 2 describes the observed parameters and locations of these two H i detections as measured from the ALFALFA data cubes, derived in the manner described by Giovanelli et al. (2007). The spatial centroid of each cloud is in Column 2. Its accuracy depends on the source strength and is of order ∼30'' for the reported sources. The heliocentric velocity cz_☉, width at 50% of the peak W₅₀, and total flux F_c of the integrated spectral profiles in Figure 1 are in Columns 3–5. The signal-to-noise ratio (S/N) of the detections is in Column 6 and is given by

$$\begin{equation} \mbox{S}/\mbox{N}=\Bigl ({1000 \, F_c \over W_{50}}\Bigr){w^{1/2}_{\rm smo}\over \sigma _{\rm rms}}, \end{equation} \tag{ 1 }$$

where F_c is in Jy km s^-1, W₅₀ is in km s^-1, w_smo is a smoothing width equal to the number of 10  km s^-1 bins bridging half the signal, and σ_rms is the rms noise (in mJy) across the integrated spectrum at 10 km s^-1 resolution. The H i mass M_H i for each cloud is in Column 7 and is computed assuming that the clouds are optically thin:

$$\begin{equation} M_{\rm H\,{\mathsc{i}}}/M_\odot = 2.356 \times 10^5 \, D^{2} \, F_c, \end{equation} \tag{ 2 }$$

where D is in Mpc and F_c is in Jy km s^-1 (Roberts 1975). The uncertainties on M_H i in Table 2 and elsewhere do not include errors in the distance adopted, which is poorly constrained due to the large peculiar velocities of objects near or within the cluster. As described in Giovanelli et al. (2005b) and Springob et al. (2005a), this results in ambiguities for galaxies with cz_☉ < 3000 km s^-1. The peculiar flow model used for the ALFALFA distances published in the catalogs corrects only for large-scale perturbations in the velocity field. The model is not able to deal effectively with regions in the immediate vicinity of Virgo. We adopt the same distance values for consistency with Kent et al. (2007): 16.7 Mpc for Cloud 1 and 34.8 Mpc for Cloud 2. The model used to obtain the distances is based on the work of Tonry et al. (2000) and Masters et al. (2004), using a parametric model and spherical truncated power-law attractor to examine the peculiar motions that arise from a cluster like Virgo.

Both H i cloud centroid positions were observed with the Very Large Array³ in 2006 November. Approximately nine hours of on-source integration were obtained in C configuration for each source. Online Hanning smoothing yielded a channel spacing of 12.2 kHz over a bandpass of 1.5 MHz.

The data from the runs were reduced using the Astronomical Image Processing System (AIPS; Greisen 2003) as described in Kent et al. (2009). Standard flux, phase and bandpass calibration and continuum subtraction routines were applied after flagging. The calibrated data were imaged using various weighting schemes; we analyze the highest sensitivity, naturally weighted cubes with a synthesized beam width of ∼25'' (∼2.0 kpc at the Virgo distance). The data cubes created for the Cloud 1 field are not limited by dynamic range and do not gain image fidelity from self-calibration. For the field with Cloud 2, self-calibration was run using a strong continuum source (NVSS catalog position α= 12^h13^m321, δ= +13°07'204; Condon et al. 1998) of flux density 1.3 Jy, greatly improving the fidelity and phase calibration of the images. As part of the reduction process, each image was smoothed to the resolution of the ALFALFA data cubes to identify emission in channels, as well as their extent in the frequency domain. A summary of the aperture synthesis observing and map parameters is given in Table 3. For clarity, all variables denoting parameters derived from the VLA observations are primed.

Table 3. Aperture Synthesis Observing and Data Cube Parameters

Parameter	Cloud 1	Cloud 2
Pointing center (J2000)	12h08m455, +11° 55' 17''	12h13m418, +12° 53' 51''
Total time on-source	547 minutes	532 minutes
Net bandpass	1.5 MHz (1132–1336 km s-1)	1.5 MHz (2137–2341 km s-1)
Maximum spectral resolution δV'	12.2 kHz (2.6 km s-1)	12.2 kHz (2.6 km s-1)
Synthesized beam/natural weighting	252 × 244 at  567	253 × 244 at  572
σ'm at pointing center, δV' = 2.6 km s-1	1.39 mJy beam−1	1.37 mJy beam−1

Download table as:  ASCIITypeset image

Channel maps of each field are shown in Figures 2 and 3, with solid and dashed contours in the primary beam-corrected maps depicting, respectively, positive and negative multiples of the median rms map noise. All maps are corrected for the attenuation of the primary beam and averaged over 3–4 spectral channels to yield a channel map resolution of δV' = 7.8 or 10.4 km s^-1. The emission from detections in both fields is contiguous over multiple channels in different weighting and imaging deconvolution schemes.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Total intensity (zero moment) and intensity-weighted velocity (first moment) maps of each field are shown in Figures 4 and 5. The zero moment contours are overlaid on Sloan Digital Sky Survey (SDSS) g-band images. For each frequency channel, we blank regions with less than 3σ'_m (defined in Table 3) in the image before combining into the zero moment map. The first moment maps are computed only at locations with column densities of N'_H i ⩾1.5 × 10²⁰ cm⁻² (Cloud 1) and N'_H i ⩾0.7 × 10²⁰ cm⁻² (Cloud 2). Global integrated spectral profiles from all detections in each field, representing the total emission for comparison with the original ALFALFA spectrum, are shown in Figure 1 in blue. The rms error on the computed total emission over the full width at half-maximum range in each individual channel for both spectra ranges from 0.67 to 1.03 mJy and also reflects a 5% uncertainty in calibration.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

3.1. H i Aperture Synthesis Detections

3.2. Field Properties

The properties for the detections in both fields are summarized in Table 4 in the same manner as Kent et al. (2009). The global emission parameters, where measurable, are also listed. Each detected feature in the VLA data cubes was fitted with a centroid ellipse in the same fashion as the Arecibo data. The centroid positions of these fits for each individual cloud detection in the total intensity maps of Fields 1 and 2 (Figures 4 and 5) are given in Column 2. The centroid cz'_☉ of both the individual integrated profiles of Figure 6 is in Column 3, and W'₅₀ of the profiles is in Column 4. The values of W'₅₀ are corrected for instrumental effects by assuming that the unbroadened profile is Gaussian. The integrated flux density F'_cand H i mass M'_H i are in Columns 5 and 8, respectively. The major axis a'_H i measured from the ellipsoidal fits of each individual cloud is in Column 6. We adopt the outermost locations as the edges of the clouds where N'_H i =1.5 × 10²⁰ cm⁻² (Figure 4) and N'_H i =0.7 × 10²⁰ cm⁻² (Figure 5). The position angle PA'_H i at which a'_H i is measured is in Column 7. An estimate of the dynamical mass M'_dyn of each cloud is in Column 9 and is computed via

$$\begin{equation} M_{\rm dyn}^{\prime } = (3.39 \times 10^4)\,a_{\rm H\,{\mathsc{i}}}^\prime D \left(\frac{W_{50}^{\prime }}{2} \right)^2, \end{equation} \tag{ 3 }$$

where a'_H i is the object diameter in arcminutes, W'₅₀ is in km s^-1, and the distance D is in Mpc. We note that M'_dyn has physical meaning only if the clouds are self-gravitating and in dynamical equilibrium, which may or may not be a valid assumption.

Zoom In Zoom Out Reset image size

Figure 7 shows the location of these two H i clouds with respect to the Virgo Cluster, with the boundaries and areas of Binggeli et al. (1993) and plotted against the hot X-ray cluster background detected by ROSAT (Snowden et al. 1995). Both cloud regions are also outside the projected virial radius of the dark matter halo around M87 determined by McLaughlin (1999). The detections lie in the vicinity of the M cloud (Ftaclas et al. 1984) west of M87, where member galaxies are considered to lie behind the main A cluster. These M cloud galaxies have a larger mean velocity (cz_☉ ∼ 2000 km s^-1) than the main cluster (cz_☉ ∼ 1150 km s^-1). However, peculiar velocities due to the large central mass of the cluster yield a velocity dispersion of the projected M cloud region galaxies that ranges from −100 <cz_☉< 2400 km s^-1. The assignment of these H i features to the various areas of Virgo remains ambiguous.

Zoom In Zoom Out Reset image size

Imaging studies by Roberts et al. (2007) analyzed an area extending from the eastern part of the M cloud northward. They arrived at a density of 20–60 low surface brightness (LSB) dwarf galaxies per square degree in the area 15 west of M87. This may have some relation to possible parent galaxies of the Cloud 2 field. Figure 7 shows that the Cloud 1 field is more removed from the main cluster and galaxy density measurement.

No obvious optical features that resemble an extragalactic counterpart can be correlated with any of the cloud components in online imaging databases. The catalogs provided by the NASA/IPAC Extragalactic Database (NED), SDSS (York et al. 2000), Virgo Cluster Catalog (Binggeli et al. 1985), GOLDMine (Gavazzi et al. 2003), and published ALFALFA survey (Giovanelli et al. 2007; Kent et al. 2008) were examined for possible nearby associations to the Clouds 1 and 2 fields. Figure 8 shows 2° × 2° areas of sky surrounding each of the Arecibo detections. Each plot contains open circles from galaxies with published H i or optical redshifts within a given range of the measured Arecibo/H i velocity for the detection; cz_☉⩽ 3000 km s^-1 for the Cloud 1 area and 1400 <cz_☉< 3000 km s^-1 for the Cloud 2 area. Galaxies in those areas with no published redshifts at any wavelength are depicted as small crosses. We next examine a number of nearby galaxies of comparable redshift within the projected vicinity of Clouds 1 and 2.

Zoom In Zoom Out Reset image size

4.1. The Projected Environment of Cloud 1

4.2. The Projected Environment of Cloud 2

Roberts (1988; see references therein) outlined various categories of intergalactic H i clouds: tidal tails, extended H i envelopes, and clouds near groups or within clusters. Starless intergalactic clouds in the field, isolated from other galaxies, have yet to be detected. Recent studies have focused on the search and identification of starless gas-rich halos. As one of the many important science goals, identifying such objects in blind surveys like HIPASS and ALFALFA gives useful information on the formation and evolution of galaxies in a variety of environments. Recently detected optically inert clouds and their respective follow-up studies can be associated with galaxies in nearby clusters (Sancisi et al. 1987; Davies et al. 2004; Minchin et al. 2005; Haynes et al. 2007; Kent et al. 2007), in groups or disturbed galaxies (Schneider et al. 1983; Henning et al. 1993; Ryder et al. 2001), in tidal or harassed tails (Oosterloo & van Gorkom 2005; Giovanelli & Haynes 1989; Salzer et al. 1991), or as a high-velocity cloud or Milky Way/Local Group companion (Kilborn et al. 2000; Giovanelli et al. 2010). However, surveys have not revealed a large population of previously undetected dark matter halos predicted by large-scale simulations (Moore et al. 1999). In the nearby Virgo Cluster, the H i detected in clouds or tidal streams does not make up a significant portion of the H i deficiency in nearby parent spirals; the population of H i clouds does not, by itself, offer a complete solution to the missing satellite or mass problem (Kent et al. 2009; Klypin et al. 1999).

None of the detections discussed here appear to be tidal tails that clearly extend to an obvious parent galaxy. The largest tail extending from a Virgo galaxy is near NGC 4532 at a length of 500 kpc (Koopmann et al. 2008). While late-type spirals and dwarfs are within a projected 500 kpc range of both Clouds 1 and 2, neither has a tail or streamlike morphology leading to another nearby galaxy. The clouds do not belong to a compact group of galaxies, nor are they are part of the main A or B clusters surrounding M87 or M49.

Although the region surrounding the Cloud 1 detection lies at a projected distance of 1.5 Mpc from M87, it has been shown that the spiral galaxy population in the M cloud area is H i deficient. The Cloud 1 detection lies on the boundary of higher H i deficiency (Solanes et al. 2001), whereas the Cloud 2 field lies within it. The intracluster X-ray density in the vicinity of the Virgo M cloud is estimated to be n_icm ∼ 3 × 10⁻⁶ cm⁻³ (computed from Vollmer et al. 2001). The Virgo Cluster ICM temperature maps computed by Shibata et al. (2001) do not cover the region of sky containing the H i clouds. If we entertain the assumption that these clouds came from a spiral disk, then the presence of this gas deficiency means that a ram pressure stripping hypothesis cannot be completely discarded.

We can place upper limits on the optical surface brightness based on models of Bell et al. (2003). As in Kent et al. (2009) we assume a g-band imaging surface brightness limit similar to other SDSS LSB galaxy studies (μ_g∼ 26 mag arcsec⁻²; Kniazev et al. 2004). A feature of source size ∼10'' would have a g-band luminosity of L_g ∼ 10⁶L_☉ and model stellar M/L ratio of M*/L*∼ 1.6. The theoretical upper limits for the stellar to H i mass ratio would range from ∼0.02 to 0.11 for the clouds extracted from the VLA data cubes. Upper limits for the H i mass to stellar luminosity would range from ∼15 to 70. It remains an open issue as to whether or not any optical emission can be positively correlated with these H i detections.

As indicated in Section 4, nine late-type galaxies lie in the vicinity and near redshift range of Cloud 2; only one lies ∼1° northeast of Cloud 1. This makes it difficult to identify a parent galaxy. However, we can hypothetically consider the movement of these clouds through the cluster environment. As both clouds are at higher velocities than the systemic heliocentric cluster velocity (cz_☉,Virgo∼ 1150; km s^-1 Huchra 1988), their line-of-sight velocity with respect to the cluster reference frame is directed away from us. If the clouds were torn from a spiral disk that is in a similar reference frame, with the clouds decelerating, then the parent galaxy would be at a higher systemic velocity than the cluster. The only nearby spiral galaxy of comparable velocity is the aforementioned VCC 58 (IC 769), located one degree northeast of the Cloud 1 detection at a redshift cz_☉ = 2209 km s^-1. VCC 58 also lies one degree southwest of Cloud 2 and stands as a remote, yet possible candidate parent of either cloud.

We have presented new follow-up observations obtained with the VLA that resolve original Arecibo H i detections of extragalactic H i clouds in the Virgo Cluster periphery. The results of these observations are summarized as follows.

• 1.  
  Two H i clouds detected and unresolved with Arecibo using ALFALFA survey data. The H i detections have heliocentric radial velocities of cz_☉= 1230 and 2235 km s^-1. The velocity widths are narrow at 29 and 53 km s^-1. The H i masses of Clouds 1 and 2 are, respectively, 4.3 × 10⁷ and 3.5 × 10⁸M_☉.
• 2.  
  Detections have been made with the VLA in both the Clouds 1 and 2 fields at the same velocities as the Arecibo detections. The data show two and three separated regions of H i emission for the Clouds 1 and 2 fields, respectively. The individual H i masses range from log(M'_H i/M_☉) = 7.1 to 7.8 M_☉. We recover 87% of the flux for the Cloud 1 field and 41% of the flux for the Cloud 2 field. No optical, IR, or UV counterpart can be identified with these H i features using available online imaging databases.
• 3.  
  The galaxy environment is relatively sparse around Cloud 1—one faint object with no redshift information, SDSS J120859.92+115631.2, lies 38 northeast of the Arecibo centroid. The nearest late-type galaxy of comparable Virgo redshift is VCC 58, located one degree to the northeast.
• 4.  
  The Cloud 2 detection lies in a dense galaxy environment showing higher H i deficiency with nine late-type spiral systems of comparable Virgo redshift within a one degree radius. The closest H i detection is VCC 85 at 8'.
• 5.  
  The H i deficient spirals in the M cloud region show that dynamic processes are prevalent even at large distances from the Virgo Cluster center. While there are no larger spirals immediately in the vicinity or at comparable velocity of the H i Clouds, we cannot dismiss a cloud origin hypothesis of ram pressure stripping. Much like previous detections reported in Kent et al. (2007, 2009), it is unlikely that the H i clouds described here are primordial gas structures in dark matter halos. These two clouds are located in the outer parts of the cluster and are in a lower density environment than other H i clouds and tidal tails further toward M87 or M49. Cloud 1 remains unique in its isolation. To date, there are no other gas structures that are both definitively extragalactic and unambiguously not associated with another galaxy outside the Local Group.

We thank David E. Hogg and Morton S. Roberts of the NRAO for their assistance and encouragement of this work. We also thank William Cotton for advice on imaging techniques and the anonymous referee for the careful review.

B.R.K. acknowledges support from a Jansky Fellowship during the completion of this work. This research has made use of the NASA/IPAC Extragalactic Database (NED) which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration. Skyview was developed and maintained under NASA ADP Grant NAS5-32068 under the auspices of the High Energy Astrophysics Science Archive Research Center at the Goddard Space Flight Center Laboratory of NASA.

This research has made use of Sloan Digital Sky Survey (SDSS) data. Funding for the SDSS has been provided by the Alfred P. Sloan Foundation, the Participating Institutions, the National Aeronautics and Space Administration, the National Science Foundation, the U.S. Department of Energy, the Japanese Monbukagakusho, and the Max Planck Society. The SDSS Web site is http://www.sdss.org/. The SDSS is managed by the Astrophysical Research Consortium (ARC) for the Participating Institutions. The Participating Institutions are The University of Chicago, Fermilab, the Institute for Advanced Study, the Japan Participation Group, The Johns Hopkins University, the Korean Scientist Group, Los Alamos National Laboratory, the Max-Planck-Institute for Astronomy (MPIA), the Max-Planck-Institute for Astrophysics (MPA), New Mexico State University, University of Pittsburgh, University of Portsmouth, Princeton University, the United States Naval Observatory, and the University of Washington.