VIVA: Virgo Cluster H I Survey

• VIVA is a VLA imaging survey of Virgo-cluster late-type galaxies that defines a high-density benchmark for H I asymmetry studies.
• The survey applies uniform SoFiA data processing and resolution criteria, enabling controlled comparisons with lower-density samples.
• Measured asymmetry metrics in VIVA highlight environmental effects like ram pressure stripping and tidal interactions in dense clusters.

Searching arXiv for papers on the VIVA survey and Virgo-cluster H I studies. VIVA, in extragalactic H I studies, denotes the VLA Imaging of Virgo in Atomic Gas survey, a VLA H I imaging survey of late-type Virgo-cluster galaxies. In analyses of atomic-gas asymmetry, VIVA has a distinctive methodological status: it supplies the high-density cluster benchmark against which lower-density samples such as LVHIS and HALOGAS are compared. In the comparative study of H I asymmetries by Watts et al., VIVA was selected because it provides comparable spectral resolution, sensitivity, physical spatial resolution, and overlap with an optical redshift catalogue for consistent environmental measurements, thereby enabling direct tests of whether H I asymmetry increases with local galaxy density (Reynolds et al., 2020).

1. Survey identity and comparative role

Within the asymmetry study, VIVA is not treated merely as one resolved H I survey among others. Its specific function is to represent the Virgo-cluster / high-density regime, while LVHIS and HALOGAS anchor the low-density regime. This division is central because LVHIS alone mostly samples nearby galaxies in pairs and groups and therefore does not probe truly dense environments. VIVA was included precisely to supply a resolved H I sample in a rich cluster environment under observational conditions close enough to the other surveys for controlled comparison (Reynolds et al., 2020).

A key limitation is equally explicit: VIVA mainly overlaps the high-stellar-mass end of LVHIS. For that reason, HALOGAS was added as a second low-density comparison sample with similar mass coverage to VIVA. This design makes VIVA the environmental anchor of the study rather than a generic reference sample. A plausible implication is that the interpretation of VIVA depends as much on its placement in density–mass parameter space as on the intrinsic properties of the Virgo galaxies themselves.

2. Observational characteristics and sample definition

The comparative analysis uses a subsample of 45 VIVA galaxies with public H I data cubes. These were observed at about 15 arcsec angular resolution, 10 km s$^{-1}$ spectral resolution, and a physical resolution of about 1.5 kpc at the adopted Virgo distance of 16.5 Mpc (Reynolds et al., 2020).

All three surveys in the comparison are processed uniformly with SoFiA source finding, extracting emission above $3.5\sigma$, and producing spectra, source masks, and moment maps. Morphological and kinematic asymmetries are measured only for galaxies resolved by at least 3 beams, following the cited resolution criterion. In the VIVA subsample, 41 galaxies satisfy this resolved-galaxy threshold, which makes VIVA especially important for the spatially resolved component of the analysis.

For stellar masses, the study estimates VIVA values from SDSS photometry using the Taylor et al. relation,

$$\log(M_*/\mathrm{M}_{\odot}) = a + b\,(g-i) - 0.4m + 0.4D_{\mathrm{mod}} + 0.4M_{\mathrm{sol}} - \log(1+z) - 2\log(h/0.7),$$

with $a=-1.197$, $b=1.431$, $m$ the $g$-band apparent magnitude, and all VIVA quantities computed at $16.5$ Mpc. This mass calibration is not ancillary: it is needed to separate density effects from stellar-mass effects in the asymmetry comparisons.

3. Virgo-cluster environment and density calibration

The study quantifies environment with a weighted $N$-th nearest-neighbour estimator,

$\rho_N=\frac{C_N}{\sum_{i=1}^N d_i^3},$

where $3.5\sigma$0 is an empirically calibrated constant and $3.5\sigma$1 are luminosity distances to the $3.5\sigma$2 nearest neighbours. For the adopted fiducial metric,

$3.5\sigma$3

This weighted form is preferred because it responds to the distribution of all ten nearby galaxies rather than only the outermost one (Reynolds et al., 2020).

For VIVA, densities are computed using the Extended Virgo Cluster Catalogue (EVCC). Because line-of-sight positions within Virgo are uncertain, each VIVA galaxy is assigned a random distance from a Gaussian centred on $3.5\sigma$4 Mpc with $3.5\sigma$5 Mpc, identified as the Virgo virial radius. A completeness correction from mock catalogues is then applied; for VIVA this factor is 1.08.

The resulting mean densities are

$3.5\sigma$6

VIVA therefore occupies the cluster-density regime, roughly $3.5\sigma$7 dex above LVHIS on average. In the architecture of the comparison, this is the quantitative basis for treating VIVA as the dense-environment benchmark.

4. H I asymmetry diagnostics

The paper uses VIVA to test which H I asymmetry measures are most sensitive to environment. For integrated spectra, four standard or semi-standard quantities are defined. The first is the difference between two systemic-velocity definitions,

$3.5\sigma$8

The second is the H I–optical offset,

$3.5\sigma$9

The third is the flux-ratio asymmetry,

$$\log(M_*/\mathrm{M}_{\odot}) = a + b\,(g-i) - 0.4m + 0.4D_{\mathrm{mod}} + 0.4M_{\mathrm{sol}} - \log(1+z) - 2\log(h/0.7),$$0

with inversion when necessary so that $$\log(M_*/\mathrm{M}_{\odot}) = a + b\,(g-i) - 0.4m + 0.4D_{\mathrm{mod}} + 0.4M_{\mathrm{sol}} - \log(1+z) - 2\log(h/0.7),$$1. The fourth is the peak-flux ratio,

$$\log(M_*/\mathrm{M}_{\odot}) = a + b\,(g-i) - 0.4m + 0.4D_{\mathrm{mod}} + 0.4M_{\mathrm{sol}} - \log(1+z) - 2\log(h/0.7),$$2

again inverted if needed so that $$\log(M_*/\mathrm{M}_{\odot}) = a + b\,(g-i) - 0.4m + 0.4D_{\mathrm{mod}} + 0.4M_{\mathrm{sol}} - \log(1+z) - 2\log(h/0.7),$$3.

The most important statistic in the study is the normalised flipped spectrum residual,

$$\log(M_*/\mathrm{M}_{\odot}) = a + b\,(g-i) - 0.4m + 0.4D_{\mathrm{mod}} + 0.4M_{\mathrm{sol}} - \log(1+z) - 2\log(h/0.7),$$4

where the spectrum is mirrored about $$\log(M_*/\mathrm{M}_{\odot}) = a + b\,(g-i) - 0.4m + 0.4D_{\mathrm{mod}} + 0.4M_{\mathrm{sol}} - \log(1+z) - 2\log(h/0.7),$$5. Because the flip is taken around the centre of mass, $$\log(M_*/\mathrm{M}_{\odot}) = a + b\,(g-i) - 0.4m + 0.4D_{\mathrm{mod}} + 0.4M_{\mathrm{sol}} - \log(1+z) - 2\log(h/0.7),$$6 measures profile-shape mismatch rather than merely unequal integrated flux in the two halves. In the comparative analysis, this is the quantity for which VIVA shows the clearest separation from the lower-density samples (Reynolds et al., 2020).

For resolved H I structure, the study defines a moment-0 asymmetry,

$$\log(M_*/\mathrm{M}_{\odot}) = a + b\,(g-i) - 0.4m + 0.4D_{\mathrm{mod}} + 0.4M_{\mathrm{sol}} - \log(1+z) - 2\log(h/0.7),$$7

a Fourier lopsidedness expansion

$$\log(M_*/\mathrm{M}_{\odot}) = a + b\,(g-i) - 0.4m + 0.4D_{\mathrm{mod}} + 0.4M_{\mathrm{sol}} - \log(1+z) - 2\log(h/0.7),$$8

with inner and outer averages

$$\log(M_*/\mathrm{M}_{\odot}) = a + b\,(g-i) - 0.4m + 0.4D_{\mathrm{mod}} + 0.4M_{\mathrm{sol}} - \log(1+z) - 2\log(h/0.7),$$9

and a velocity-field asymmetry

$a=-1.197$0

For $a=-1.197$1, galaxies with $a=-1.197$2 kpc are excluded because random motions can dominate the denominator in very small systems.

5. Comparative asymmetry results

Over the full stellar-mass range $a=-1.197$3, VIVA is systematically more asymmetric than the low-density samples in most diagnostics, and most strongly so in $a=-1.197$4. The means are

$a=-1.197$5

with corresponding sample standard deviations $a=-1.197$6, $a=-1.197$7, and $a=-1.197$8. The mean VIVA value is therefore about three times larger than in either low-density sample, and the paper identifies this as the largest separation among all measured asymmetry parameters and the most significant trend with density (Reynolds et al., 2020).

The same pattern persists after controlling for stellar mass. In the restricted range

$a=-1.197$9

the means become

$b=1.431$0

This persistence is the basis for the conclusion that the VIVA offset is driven primarily by environment density, not stellar mass.

Other asymmetry measures also place VIVA above the comparison samples, though less decisively. For the full samples: $b=1.431$1 for VIVA, versus $b=1.431$2 for LVHIS and $b=1.431$3 for HALOGAS;

$b=1.431$4

for VIVA, versus $b=1.431$5 and $b=1.431$6;

$b=1.431$7

for VIVA, versus $b=1.431$8 and $b=1.431$9;

$m$0

for VIVA, versus $m$1 and $m$2;

$m$3

for VIVA, compared with $m$4 for LVHIS and $m$5 for HALOGAS; and

$m$6

for VIVA, versus $m$7 and $m$8.

The statistical language in the paper is intentionally cautious. After controlling for stellar mass, there are signs of statistically significant trends of increasing asymmetry with local density, but outside $m$9 the authors usually describe only hints or tentative trends because the scatter is large and the samples remain modest. VIVA galaxies span a broad $g$0 range and overlap the low-density samples, so the study does not claim that every asymmetry in Virgo must be externally driven; rather, the higher mean and broader upper tail indicate stronger environmental influence.

A related literature comparison uses isolated-galaxy AMIGA thresholds. The fractions with $g$1 and $g$2 are 20% and 13% for VIVA, compared with 8% and 3% for LVHIS and 0% and 0% for HALOGAS. This places VIVA close to other Virgo and Abell 1367 cluster studies and reinforces its interpretation as a genuinely cluster-processed population.

6. Physical interpretation, recommended diagnostics, and survey legacy

The Virgo-cluster setting is central to the physical interpretation. The study argues that the elevated VIVA asymmetries are consistent with dense-environment processes such as ram pressure stripping and tidal encounters acting more frequently or more strongly in clusters. It states explicitly that “ram pressure likely has the greatest effect on the VIVA asymmetries as this sample probes the cluster environment of Virgo,” while also noting that earlier work identified likely tidal-interaction cases among VIVA galaxies (Reynolds et al., 2020). The fact that VIVA exceeds not only LVHIS but also HALOGAS strengthens the interpretation that cluster-specific environmental processing is important.

This comparative role leads directly to methodological recommendations for future H I surveys. For integrated spectra, the preferred measures are

$g$3

Among these, $g$4 is especially strongly motivated by the VIVA comparison because it shows the clearest density dependence. By contrast, $g$5 is regarded as less meaningful because it depends on heterogeneous optical redshifts with uncertainties often larger than the measured offsets, and $g$6 is limited to clean double-horn profiles. For resolved H I galaxies, the recommended diagnostics are

$g$7

VIVA also functions as a proof-of-principle for large future blind H I surveys. The paper argues that WALLABY/ASKAP should detect $g$8 galaxies and sample roughly five orders of magnitude in both local density and stellar mass. This suggests that future datasets will test much more rigorously whether the Virgo-like enhancement of H I asymmetry is universal across dense environments. Because most WALLABY detections will be spatially unresolved, the strong VIVA result for $g$9 makes that statistic especially promising. In that specific sense, VIVA serves not only as a Virgo-cluster survey, but as the empirical benchmark showing that environmentally driven H I asymmetry should be measurable at scale in forthcoming surveys.