Widefield Arecibo Virgo Extragalactic Survey: II. Characterizing the HI properties and environment of the WAVES South region

Abstract: Context. Galaxy clusters are extreme environments where interactions with the hot intracluster medium drive rapid galaxy evolution. These processes can result in the formation of optically dark gas clouds, as previously observed in Virgo and other clusters. Aims. We investigate the distribution and properties of neutral hydrogen (HI) in two large adjoining regions of the Virgo cluster to understand how the cluster environment influences galaxy transformation. Specifically, we examine the gas content of both star-forming and quiescent populations and search for evidence of gas-loss driven evolution. Methods. We cataloged the 21cm HI Widefield Arecibo Virgo Extragalactic Survey (WAVES) South data using visual and automatic source extraction methods. By combining these results with an optically selected sample, we compared the HI properties of WAVES South with the previously studied VC1 region. To probe gas reservoirs below the nominal detection limit, we performed a stacking analysis of radio spectra across the WAVES South, VC1 and VC2 footprints. Results. We detected 56 HI sources with a median root mean square (rms) noise of 0.8 mJy, including 50 galaxies, two gas clouds (one being optically dark), and the ALFALFA Virgo 7 complex. Our results reveal a significantly lower detection fraction in WAVES South compared to the VC1 region. Stacking showed no new HI detection at a 0.080 mJy rms with a maximum of 157 stacked objects from WAVES South, VC1, and VC2. Conclusions. The lower HI detection fraction suggests that WAVES South is a more dynamically relaxed and evolved environment than the VC1 region. The presence of residual HI in a small subset of early-type galaxies supports a model of dwarf irregular to dwarf elliptical transformation via environmental stripping. Finally, we note a possible evolutionary link between optically dark clouds and recently discovered "blue blobs."

• The paper analyzed H I properties in a 20 deg2 area of the Virgo cluster, discovering 47 individual galaxies, 3 unresolved pairs, and 6 unique sources for a total of 56 sources.
• WAVES South exhibits a more dynamically evolved galaxy population compared to VC1, with lower H I detection fractions and a lack of optically dark clouds, indicating a gas-poor environment.
• Stacking analysis reveals no hidden H I reservoirs below the detection limit, suggesting an environmental critical density for neutral gas survival near $10^7.0 ⚹{\[M\|unsorted\\odot\]{rc v 2.992}$ exists, pointing to effective gas removal mechanisms.

Overview

This paper presents the complete catalog and analysis of the WAVES South region, a $20~\rm{deg^2}$ Arecibo 21-cm survey of the Virgo cluster centered on the X-ray filament connecting M87 and M49. Conducted with the ALFA receiver in drift-scan mode between 2017 January and 2018 March, the survey achieved a median rms noise of $0.8~\rm{mJy\,beam^{-1}}$ at a velocity resolution of $10~\rm{km\,s^{-1}}$ after Hanning smoothing. The authors cataloged 56 H I sources via two independent visual searches (using FRELLED) supplemented by automated extraction with SoFiA, and constructed a comparison sample of 78 spectroscopically confirmed cluster galaxies without detectable H I. Because the observational setup is essentially identical to that of the AGES VC1 region studied by Taylor et al. (2012), the paper's central analytical strategy is a direct comparison between the two footprints, which share a boundary and probe different substructures of the same cluster.

Source catalog and completeness

The final catalog combines detections confirmed by both examiners with single-examiner sources passing an integrated signal-to-noise threshold of ${S/N}_{\rm{int}} > 6.5$, plus a handful of marginal sources retained because of plausible optical counterparts. Automated SoFiA extraction at a threshold iteratively raised to 3.85 recovered all 56 visual detections but produced 211 candidates overall; 20 unconfirmed candidates are excluded pending follow-up observations — a limitation the authors state explicitly.

The detected population comprises 47 individual galaxies (44 late-type, 3 early-type), three unresolved galaxy pairs, the ALFALFA Virgo 7 complex (counted as four sources), one star-forming cloud, and one optically dark cloud candidate (WCS 54). H I masses span $8.1\times10^6$ to $1.7\times10^9~\rm{M_\odot}$. Nine H I-detected galaxies (~16% of the sample) are absent from the Virgo Cluster Catalog, including two low-surface-brightness systems (WCS 10 and WCS 23) that remain absent from all dedicated Virgo optical surveys — demonstrating the continued capacity of deep H I surveys to uncover optically missed members.

Relative to ALFALFA, whose median rms is $2.4~\rm{mJy}$, WAVES South recovers all 48 known ALFALFA sources plus eight new detections, typically of mass $\sim10^7~\rm{M_\odot}$. Notably, the detection gain is only ~17%, despite a factor-of-three sensitivity improvement, whereas VC1 showed a 45% gain for a factor-of-four improvement. The authors argue this discrepancy is not an extraction artifact but reflects an intrinsic absence of fainter sources below $\sim10^7~\rm{M_\odot}$ in this environment — a claim substantiated by their stacking analysis.

Evidence for a dynamically relaxed region

Three independent lines of evidence support the conclusion that WAVES South is more dynamically evolved than VC1:

• Velocity distributions: In VC1, H I-detected and non-detected populations have statistically distinct velocity profiles (significant at the 99.9% level), consistent with recent infall of gas-rich systems. In WAVES South, both populations are indistinguishable, implying dynamical mixing.
• Detection fraction: The H I detection fraction among VCC galaxies is ~14% in WAVES South versus ~22% in VC1, widening to 16% versus 28% when all galaxies are included. The disparity persists when restricted to the main cluster body at 17 Mpc (11% vs. 17%), despite WAVES South containing a higher proportion of 17-Mpc objects (~71% of its VCC sample).
• Dark cloud census: VC1 hosts eight isolated optically dark clouds; WAVES South contains only one dark candidate, WCS 54, which is not isolated but connected by an H I bridge to the galaxy WCS 40 (VCC 952) at a projected separation of ~66 kpc.

The deficiency analysis complicates this picture: median H I deficiency is actually slightly lower in WAVES South (0.48) than in VC1 (0.70), differing at 99.3% confidence. However, deficiency does not date gas loss, and the sample-size-limited significance tempers any strong inference. Within WAVES South itself, deficiency declines steadily with projected distance from the M87–M49 filament spine (confirmed at 99.7% confidence by a Student's t-test), consistent with ram-pressure stripping scaling with intracluster medium density.

Stacking analysis

The stacking methodology is carefully validated: inverse-variance weighting consistently outperforms unweighted averaging, and measured stacked rms tracks the theoretical prediction from the weighted combination of individual noise values, providing a reliable performance diagnostic. No new H I emission was detected in any subsample. The deepest stack — 157 objects across WAVES South, VC1, and VC2 — reached an rms of $0.080~\rm{mJy}$, corresponding to a $0.8~\rm{mJy\,beam^{-1}}$0 H I mass upper limit of $0.8~\rm{mJy\,beam^{-1}}$1, roughly an order of magnitude below the nominal detection limit and marginally deeper than the best VC1 stack ($0.8~\rm{mJy\,beam^{-1}}$2).

The persistent non-detection carries a strong implication: once a galaxy's neutral gas falls below single-dish detection thresholds, no hidden reservoir survives. Combined with the sharp drop in detections below $0.8~\rm{mJy\,beam^{-1}}$3, this suggests the Virgo environment imposes an effective critical density below which H I cannot persist — contrasting sharply with field environments where stacking routinely recovers signals. The comparison with Coma (H I mass function cutoff near $0.8~\rm{mJy\,beam^{-1}}$4) and Fornax (cutoff near $0.8~\rm{mJy\,beam^{-1}}$5) suggests the cutoff mass scales with ICM harshness and galaxy density, though the mechanisms remain unresolved.

Gas-loss driven evolution of dwarfs

The two H I-detected early-type dwarfs in WAVES South, WCS 47 (VCC 21) and WCS 51 (VCC 651), have gas fractions of 0.16 and 0.62 respectively — systematically lower than the late-type population and than the gas-rich dEs reported in VC1. This supports the Boselli et al. (2008) scenario in which dwarf irregulars are stripped and transform into gas-poor dwarf ellipticals, with these objects caught in the final phase of transition. Conversely, the paper identifies a group of compact, blue, H I-non-detected late-type dwarfs (SDSS/WISEA objects) with stellar masses of $0.8~\rm{mJy\,beam^{-1}}$6–$0.8~\rm{mJy\,beam^{-1}}$7 closely resembling recently stripped systems found by MeerKAT in Fornax — candidates for very recent ram-pressure stripping that have not yet faded in color or structure.

Dark clouds, blue blobs, and the missing gas budget

WCS 54, the sole dark cloud candidate, differs fundamentally from the isolated VC1 clouds: its H I bridge to WCS 40, together with compressed contours and a secondary tail on WCS 40 and a marginal tail on WCS 52 (itself highly deficient, $0.8~\rm{mJy\,beam^{-1}}$8), indicates formation through ongoing ram-pressure stripping. It may represent an early stage in the production of isolated dark clouds or merely a transient knot destined to disperse into the ICM; high-resolution observations are required to distinguish these cases.

The paper also documents a striking spatial association between "blue blobs" (BBs) and dark clouds: BC17 and BC25 in VC1 are optical counterparts to the dark clouds AGESVC1 266 and AGESVC1 274. Combined with the similar isolation of both populations, this motivates a tentative evolutionary sequence in which stripped gas persists in a dark phase before triggering late-stage star formation as a BB. The authors appropriately caution that this rests on very small number statistics. Notably, WAVES South hosts as many BBs as VC1 despite being more relaxed and hosting fewer dark clouds, suggesting BB numbers are not simply proportional to recent stripping events. Quantitatively, the total H I deficit implied by galaxy deficiencies is $0.8~\rm{mJy\,beam^{-1}}$9, of which all detected extragalactic gas accounts for less than 7% — compatible with most stripped gas evaporating into the hot ICM.

The paper also refutes two previously claimed features: the extended H I tail on NGC 4424 and the KW Cloud near NGC 4451, neither of which is recovered in data deeper than the original KAT-7 detections, strongly indicating artifacts.

Limitations and open questions

Several caveats bear directly on the results. The 33% higher rms relative to VC1, attributed to Hurricane Maria damage and proximity to M87, partially degrades sensitivity, although the identical H I mass distributions suggest the impact on the detected population is minor. Distance assignments rely on the Gavazzi et al. substructure scheme rather than direct distance indicators, introducing systematic uncertainty into all derived masses. The velocity cut at $10~\rm{km\,s^{-1}}$0 excludes up to 13% of the cluster population. Twenty SoFiA candidates remain unconfirmed, and the BB/dark-cloud evolutionary link awaits AGES-depth coverage over a much larger area, which the authors propose to pursue with FAST and the ViCTORIA project. Whether WCS 54 contains ordered motions — and hence is long-lived — remains undetermined, as does the survival mechanism of long-lived dark clouds in an environment where far more massive galaxies are efficiently stripped.

Conclusion

This work delivers a complete, dual-validated H I census of a $10~\rm{km\,s^{-1}}$1 Virgo field and demonstrates, through comparison with the adjacent VC1 region, that WAVES South hosts a more dynamically mixed and gas-poor galaxy population. The stacking results establish that no significant hidden H I reservoir exists below the nominal detection limit anywhere in the surveyed cluster regions, pointing to an environmental critical density for neutral gas survival near $10~\rm{km\,s^{-1}}$2. The identification of a stripping-connected dark cloud and the spatial coincidence of blue blobs with dark clouds offer a concrete, testable framework for the lifecycle of stripped gas, while the residual H I in transitioning early-type dwarfs provides direct observational support for environmentally driven morphological transformation.