The census of nuclear activity of late-type galaxies in the Virgo cluster

Abstract

The first spectroscopic census of AGNs associated to late-type galaxies in the Virgo cluster is carried on by observing 213 out of a complete set of 237 galaxies more massive than M_dyn>10^{8.5} solar masses. Among them, 77 are classified as AGNs (including 21 transition objects, 47 LINERs and 9 Seyferts), and comprize 32% of the late-type galaxies in Virgo. Due to spectroscopic incompleteness at most 21 AGNs are missed in the survey, so that the fraction would increase up to 41%. Using corollary Near-IR observations, that enable us to estimate galaxies dynamical masses, it is found that AGNs are hosted exclusively in massive galaxies, i.e. M_dyn\gsim 10^{10} solar masses. Their frequency increases steeply with the dynamical mass from zero at M_dyn\approx10^{9.5} solar masses to virtually 1 at M_dyn>10^{11.5} solar masses. These frequencies are consistent with the ones of low luminosity AGNs found in the general field by the SDSS. Massive galaxies that harbor AGNs commonly show conspicuous r-band star-like nuclear enhancements. Conversely they often, but not necessarily contain massive bulges. Few well known AGNs (e.g. M61, M100, NGC4535) are found in massive Sc galaxies with little or no bulge. The AGN fraction seems to be only marginally sensitive to galaxy environment. We infer the black hole masses using the known scaling relations of quiescent black holes. No black holes lighter than $\sim 10^6$ \msol are found active in our sample. Comment: The paper contains 13 figures and 5 tables; accepted for publication in MNRAS

Full text

Page 1

arXiv:0707.0999v1 [astro-ph] 6 Jul 2007
Mon. Not. R. Astron. Soc. 000, ??–?? (2007)Printed 1 February 2008(MN LATEX style file v2.2)
The census of nuclear activity of late-type galaxies in the
Virgo cluster
R. Decarli,1,2⋆G. Gavazzi,1I. Arosio,1L. Cortese,3A. Boselli,4C. Bonfanti,1M. Colpi1
1Universit` a degli Studi di Milano-Bicocca, Piazza delle scienze 3, 20126 Milano, Italy
2Universit` a degli Studi dell’Insubria, via Valleggio 11, 22100 Como, Italy
3School of Physics and Astronomy, Cardiff University, The Parade, Cardiff CF24 3YB
4Laboratoire d’Astrophysique de Marseille, BP8, Traverse du Siphon, F-13376 Marseille, France
1 February 2008
ABSTRACT
The first spectroscopic census of AGNs associated to late-type galaxies in the Virgo
cluster is carried on by observing 213 out of a complete set of 237 galaxies more
massive than Mdyn> 108.5M⊙. Among them, 77 are classified as AGNs (including
21 transition objects, 47 LINERs and 9 Seyferts), and comprize 32 % of the late-type
galaxies in Virgo. Due to spectroscopic incompleteness at most 21 AGNs are missed
in the survey, so that the fraction would increase up to 41 %. Using corollary Near-
IR observations, that enable us to estimate galaxies dynamical masses, it is found
that AGNs are hosted exclusively in massive galaxies, i.e. Mdyn∼
frequency increases steeply with the dynamical mass from zero at Mdyn≈ 109.5M⊙to
virtually 1 at Mdyn> 1011.5M⊙. These frequencies are consistent with the ones of low
luminosity AGNs found in the general field by the SDSS. Massive galaxies that harbor
AGNs commonly show conspicuous r-band star-like nuclear enhancements. Conversely
they often, but not necessarily contain massive bulges. Few well known AGNs (e.g.
M61, M100, NGC4535) are found in massive Sc galaxies with little or no bulge. The
AGN fraction seems to be only marginally sensitive to galaxy environment. We infer
the black hole masses using the known scaling relations of quiescent black holes. No
black holes lighter than ∼ 106M⊙are found active in our sample.
>1010M⊙ . Their
Key words: galaxies: nuclei – galaxies: clusters: individual (Virgo)
1 INTRODUCTION
Active Galactic Nuclei (AGNs) can be broadly divided in
classes of decreasing luminosity from quasars, Seyfert to
Low-Ionization Nuclear Emission Regions (LINERs) charac-
terized by narrow emission lines of relatively low ionization
and large [NII]/Hα line ratios. A growing body of evidences,
triggered by studies of low luminosity AGNs, from radio to
X-rays (Halderson et al. 2001, Terashima et al. 2002, Filho et
al. 2004, Martini et al. 2006) and in particular in the optical
band, mostly owing to the Sloan Digital Sky Survey (SDSS;
York et al. 2000) is reinforcing the scenario where LINERs
represent the missing link between higher luminosity AGNs
and dormant supermassive black holes. Hao et al. (2005) es-
timate that 4-10 % of all galaxies in the SDSS harbor an
AGN; Kauffmann et al. (2003) estimate that up to 80% of
galaxies of more than 1011M⊙ harbor a supermassive black
hole, either dormant or active.
⋆roberto.decarli@mib.infn.it
The role of the environment in triggering or inhibiting
nuclear activity is still not clear. Dressler et al. (1985) ob-
served an AGN fraction in the field ∼ 5 times higher than in
clusters. Kauffmann et al. (2004) found that the fraction of
powerful AGNs (with high [OIII] luminosity, i.e. dominated
by Seyferts) in the SDSS decreases with increasing galaxy
environment density. They interpret this result in connection
with the fact that strong AGNs are hosted in star-forming
galaxies that avoid dense environments. Popesso & Biviano
(2006) found that the AGN fraction decreases with increas-
ing velocity dispersion of galaxies in groups and clusters,
being higher in dense, low-dispersion groups. On the other
hand, Shen et al. (2007) suggest that the fraction of AGNs in
galaxy groups and in clusters may be the same. Miller et al.
(2003) find that the frequency of low activity AGNs is insen-
sitive to the environment. Furthermore, Boselli & Gavazzi
(2006) showed that gravitational interaction between the
galaxy and the cluster potential as a whole does trigger gas
infall toward the galaxy center, and this may feed nuclear
activity.

Page 2

2Decarli et al.
The Virgo Cluster provides a perfect environment for
studying the relations between the nuclear activity and the
host galaxy, for at least four reasons: 1) it’s near enough to
grant us a highly detailed knowledge of the host galaxy inter-
nal structure and morphology; the nuclear component is also
easily disentangled among inner galaxy regions; 2) a com-
plete survey of a wide range of luminosity can be achieved;
3) many galaxies member of the Virgo Cluster have been
studied individually in almost all the electromagnetic spec-
trum (e.g., Allard et al. 2006; Garca-Burillo, et al., 2005;
Chy˙ zy et al., 2006; Boselli et al. 2006; Yoshida et al. 2004);
4) the environmental effects on the nuclear activity may be
examined through various indicators (i.e. gas deficiency).
Cot´ e et al. (2006), analyzing 100 Elliptical galaxies in
the Virgo cluster surveyed with the ACS on board the HST
by Ferrarese et al. (2006a), find that most (66-82 %) contain
nuclear cusps, mainly in galaxies less massive than 3 · 1010
M⊙ . Wehner & Harris (2006) and Ferrarese et al. (2006b)
found that the masses of these bright cusps follow the same
scale relations as the black hole (BH) masses in giant ellipti-
cals, occupying the same range as the lower-end extrapola-
tion of the MBH – Magnitude correlation. This suggests that
a common mechanism is responsible for the growth of both
nuclear stellar cusps or black holes on these mass scales.
As far as late-type galaxies in the Virgo cluster, instead,
no systematic survey exists providing the census of massive
BHs nor of AGNs of the various species. To fill this gap we
undertook a survey of virtually all spiral galaxies brighter
than 15 mag belonging to the Virgo cluster, probing their ac-
tivity with diagnostic tools that are sensitive to the presence
of low activity AGNs. In spite of the poorer statistics, the
Virgo spirals appear to have AGN properties consistent with
the ones in the SDSS, i.e. their frequency depends critically
on the mass of the hosts galaxies. By exploiting the superior
resolution obtained at the distance of the Virgo cluster we
show that AGNs develop in large mass galaxies that have
significant stellar nuclei, independently from the existence
of conspicuous bulges.
2 SAMPLE
The present analysis is based on a volume limited, com-
plete set of late-type galaxies belonging to the Virgo clus-
ter. The sample is obtained by selecting from the Virgo
Cluster Catalogue (Binggeli et al. 1985, 1993) all cluster
members (−1000 < V el < 3000 km/s) with apparent pho-
tographic magnitude brighter than 15.0 mag and Hubble
type between Sa and Im/BCD. A distance of D=17.0 Mpc
is assumed for all clouds constituting the Virgo cluster, ex-
cept W, M (D=32 Mpc) and B (D=23 Mpc) (see Gavazzi
et al. 1999), equivalent to assuming an Hubble constant of
Ho = 75 km s−1Mpc−1. The resulting sample comprises 237
galaxies brighter than Mpg ≈ −16 mag.
3 DATA
For the aim of the present study, for each of the target
galaxies we collect three sets of observations: Near-IR (H-
band) imaging, optical (Gunn r-band) imaging and inter-
mediate resolution (R ∼ 1000) spectroscopy. These quanti-
Figure 1. Red-channel template spectra near Hα of early-type
galaxies binned in 4 mass intervals and all together (bottom) are
used to estimate the underlying continuum at Hα. The E.W. of
Hα, the number of averaged spectra and the mass interval are
labeled.
Table 1. Completeness.
mpg ? 15.0 Near-IRr-bandSpec.classif.
237 216199 213
ties are partly derived from existing observations available
from the WEB site GOLDMine (Gavazzi et al. 2003), partly
obtained on purpose for the present investigation. Table 1
gives the coverage of the various data sets, stressing the
high degree of completeness of the spectral classification
(213/237=90%), of the optical (199/237=84%) and Near-
IR imaging (216/237=91%). Near-IR imaging was obtained
as described in Gavazzi et al. (2001) (and references therein)
and from 2MASS. H-band total luminosities (LH in L⊙) are
obtained from extrapolation to infinity of light profiles fit-
ted to isophotal ellipses (Gavazzi et al. 2000). For 21 galaxies
without Near-IR data, LH is derived from the photographic
luminosity:
logLH = 0.80 − 0.50(Mpg)(1)
where Mpg = mpg−5logD−25. Equation 1 was obtained by
fitting the H-band luminosity and the Mpg of 440 late-type
galaxies with available data in the GOLDMine database.
The dispersion of this relation is in the order of 0.2 dex.
LH provides us with an estimate of the dynamical mass
within the optical radius of disk galaxies (Gavazzi et al.
1996), according to logMdyn = logLH + 0.66. From the
Near-IR images we also derive the H-band light concentra-
tion index C31, i.e. the ratio of the radii containing 75 and
25 % of the total light. This parameter is sensitive to the
presence of conspicuous central bulges (with C31 in excess

Page 3

AGNs in late-type Virgo galaxies3
Table 2. Spectral available data. ‘Ho’ refers to line fluxes pub-
lished in Ho, Filippenko & Sargent (1997); ‘SDSS’ to SDSS DR5
spectra; ‘Loiano’ to Loiano nuclear spectra; ‘DS’ and ‘mDS’ refer
to Drift-Scan and modified Drift-Scan spectra, as described in the
text; ‘NED’ lists the galaxy nuclear classification available in the
NASA/IPAC Extragalactic Database.
Ho SDSS LoianomDSDS NEDAll
Available
Adopted
40
40
84
73
29
13
22
4
193
81
41
2
409
213
of 4-5, while pure disks have C31 ∼ 2.8 – see Gavazzi et al.,
2000).
As part of the Hα imaging survey of Virgo spirals, com-
plete down to mpg ? 16.5 (Gavazzi et al. 2006), r(Gunn)
images were obtained to estimate the red stellar continua.
Here we use the 199 images available for the present sample
to estimate and quantify the presence of relevant nuclear
enhancements. For each galaxy we construct the parameter
Nuc(r) as the difference of the r-band surface brightness
within 1.5 of the seeing disk and the mean surface bright-
ness between 25 and 50 % of the r-band radius. Galaxies
with Nuc(r) in excess of 1 (mag arcsec−2) have significant
optical nuclei, harboring approximately 109years old stars,
or contributed by the AGN continuum radiation1.
Our analysis is based on the ratios of narrow emission
line fluxes. Different sets of spectroscopic data are used on
this purpose. The first line of Table 2 lists all the avail-
able spectroscopic material, indicating that for many of the
237 sampled galaxies more than one spectroscopic source is
available. The second line gives the 213 independent mea-
surements that were finally adopted after choosing among
various possibilities, according to the following priorities. We
first checked their availability in Ho et al. (1997). If present
we adopt the line ratios, not the final classification given in
that paper. Secondly we measure the line fluxes on avail-
able nuclear spectra. 84 spectra, obtained with apertures
of 3 arcsec, were published in the SDSS in the fifth data re-
lease (Adelman-McCarthy, J., et al. 2007). 73 of these lacked
of a classification in Ho et al. (1997). Other nuclear spectra
were obtained on purpose in February and March 2005, 2006
using the Bologna-Faint-Object-Spectrograph (BFOSC) at-
tached to the Loiano 1.5m telescope. These consist of nuclear
long-slit spectra taken through a 2-arcsec slit, combined with
an intermediate-resolution grism (R ∼ 1000) covering how-
ever only a limited amount of the red-channel (6100 - 8200
˚ A), containing Hα and [NII], but not Hβ nor [OIII]. Out of
the 29 spectra taken at Loiano only 13 were adopted for the
final classification because we often preferred the Ho and
SDSS spectra, covering a broader spectral range.2193 spec-
1The nuclearity parameter was not derived from the Near-IR
data-set because these images are not of sufficient seeing qual-
ity to derive a nuclear parameter, whereas they are adequate for
determining C31
2Note added in proofs: among the 23 unobserved galaxies listed
at the end of Table 5, 7 were observed with the Loiano telescope
in spring 2007, leading to the following results: VCC99, 449, 567,
697, 1442, CGCG14062 (with log(LH/L⊙) < 9.75) have HII-like
nuclei. The brightest VCC362 (log(LH/L⊙) = 10.46) is classified
tra available for galaxies in the present study were taken
from the GOLDMine database. These spectra were taken in
the drift-scan mode (labeled “DS” after Drift-Scan in Table
2), i.e. letting the slit of the spectrograph, parallel to the
galaxy major-axis, slide across the minor-axis (Gavazzi et
al. 2004). Spectra taken in this modality maintain their spa-
tial resolution only along the slit. DS spectra typically cover
the 3500-7000˚ A range with R ∼ 600 − 1000. Since they are
usually employed to study the overall, luminosity-averaged
spectral characteristics of galaxies, the aperture is kept wide
enough to integrate over a large fraction of the slit. Twenty-
two of the DS spectra were also re-extracted with a smaller
(∼ 5 arcsec) aperture. These data are labeled as “modified
DS”. The role of these data is further discussed in section
4.2. Finally, the NASA/IPAC Extragalactic Database re-
ports, without completeness, the nuclear classification of 41
objects in our sample. Only 2 of them had no other spec-
tral data in our study, and were thus classified according to
NED.
In all available spectra (excluding Ho et al. 1997) we
measured the intensity of the [OIII], [NII] and the narrow-
component Hβ and Hα emission lines. Underlying Hβ ab-
sorption is measured separately (as explained in Gavazzi et
al. 2004) and the adopted value of Hβ in emission is cor-
rected accordingly. Underlying Hα is not equally easy to
measure for most emission-line objects. To estimate the un-
derlying Hα and correct for it one could use several ap-
proaches. The most recommended one would be that of
fitting stellar population synthesis models (e.g. Bruzual &
Charlot, 2003) to the observed spectra and measuring Hα
from the models. The drawback of this method is that it
suffers from uncertainties in the extinction correction, un-
less UV and Far-IR data were available, which is not the case
for nuclear regions. We therefore preferred a simpler method
based on the assumption that elliptical-like objects are real-
istic representations of the circumnuclear stellar properties
of galaxies. We therefore assembled template spectra of 51
early-type galaxies available from GOLDMine, binned in 4
mass intervals (each containing tens of spectra). It is ap-
parent from Figure 1 that the amount of absorption at Hα
is independent of the galaxy luminosity. We therefore com-
bined all available early-type spectra into a unique template
spectrum that we subtract (after normalization) from the in-
dividual emission line spectra. On average the E.W. of the
underlying Hα is ∼ 1.7˚ A. If line fluxes fell under the sensi-
tivity limit (∼
<3σ of the rms of the spectral continuum), we
classified the galaxy as No Emission Line (NEL) galaxy.
4ANALYSIS
Table 5 lists the relevant photometric and spectroscopic pa-
rameters derived in this work. The content of the individual
columns is listed at the end of the table.
4.1AGN diagnostics
There are various semi-empirical criteria to separate normal
from AGN galaxies and, within the latter, various levels of
as LINER. The spectroscopic coverage now reaches 93% of the
sample.

Page 4

4Decarli et al.
Figure 2. The 2-D diagnostic diagram (the BPT diagram, from Baldwin, Phillips & Terlevich, 1981) used to characterize the spectra
of galaxies in this work based on the [NII]/Hα and [OIII]/Hβ line ratios. Objects for which [OIII]/Hβ is not available either because
the spectra were taken in the red-channel (Loiano spectra) or lines were too weak to be measured, are plotted at log[OIII]/Hβ = -1.2.
Triangles are for drift-scan spectra, squares are for the our nuclear spectra (Loiano+SDSS+modified drift-scan), while circles are from
Ho et al.(1997); empty symbols are for HII-like nuclei, filled symbols for Transition, LINER and Seyfert. Symbol size is proportional to
the dynamical mass.
Table 3. Spectral classification for all the available data. The number of adopted classification separately for each data source is reported
in parenthesis.
Set Ho
7 (7)
0 (0)
17 (17)
6 (6)
10 (10)
0 (0)
30
40
0.75
SDSS
2 (1)
15 (14)
7 (6)
8 (6)
44 (38)
8 (8)
32
84
0.38
Loiano
0 (0)
15 (5)
0 (0)
4 (3)
8 (5)
2 (0)
19
29
0.66
mDS
0 (0)
18 (3)
0 (0)
1 (1)
2 (0)
1 (0)
19
22
0.86
DS
1 (0)
26 (1)
0 (0)
20 (5)
136 (68)
10 (7)
47
193
0.24
NED
12 (1)
3 (0)
11 (1)
2 (0)
13 (0)
0 (0)
28
41
0.68
Adopted
9
23
24
21
121
15
77
213
0.36
Seyfert
Sey/LIN
LINER
LIN/HII
HII
NEL
AGN
All
fAGN
AGN activity. They all rely on 2-D line diagnostic diagrams
involving [OIII]λ5007/Hβ, [NII]λ6584/Hα, [OI]λ6300/Hα and
([SII]λ6717+[SII]λ6731)/Hα (see e.g. Baldwin, Phillips & Ter-
levich, 1981; Veilleux & Osterbrock 1987; Kewley et al. 2001;
Kauffmann et al. 2003). The reader is referred to Stasinska
et al. (2006) for a detailed comparison of these diagnostic
tools. In this paper we adopt the following definitions (see
Figure 2):
1) for all spectra we adopt [NII]/Hα < 0.4 to unam-
biguously identify HII-like nuclei, 0.4 < [NII]/Hα < 0.6 for
transition objects (HII/LIN) and [NII]/Hα > 0.6 for AGNs.
2) if the blue spectrum is available and if [OIII]
and Hβ are detected we split the AGNs among LINERs
([OIII]/Hβ < 3) and Seyfert ([OIII]/Hβ ? 3), otherwise we
classify them as Sey/LIN.
Table 3 summarizes the results of the adopted classifica-

Page 5

AGNs in late-type Virgo galaxies5
λ
[ΝΙΙ]
α
Η
3
2
1
Major Axis (kpc)
Figure 3. Blow-up of the 2-D long-slit Drift-Scan spectrogram of VCC1401 in the region of Hα, showing the inversion of the [NII]/Hα
ratio taking place approximately at 500 pc. The nuclear spectrum in the inner 250 pc from the center of the AGN VCC1401 (bottom-right
panel) and at 2.5 kpc from the center (top-right panel).
Figure 4. The [NII]/Hα (Hα corrected for absorption) ratio in 11 Virgo galaxies with nuclear long-slit spectra taken at Loiano is plotted
as a function of the aperture in which the spectrum was extracted. The inner vertical dashed line corresponds to the seeing. The outer
one gives the seeing if the galaxy was at 5 times the Virgo distance, i.e. at Coma. One galaxy (e.g. VCC 1145) is given twice: as derived
from the Loiano nuclear spectrum (L) and from the Drift-scan mode spectrum (DS), the two being consistent.
tion. Data sources are listed in order of decreasing “weight”
from left to right.
4.2Spatial distribution of the [NII]/Hα ratio
Ionization conditions higher than the normal stellar con-
tribution characterize the AGNs narrow-line region, with a
spatial extent not exceeding some hundred pc (Bennert et al.
2006). A clear-cut example is the giant galaxy VCC 1401,
hosting an AGN (Ho et al. 1997); our 2-dimensional DS
spectrum is shown in Figure 3 (left panel). When integrat-
ing over the inner 250 pc, the [NII] and Hα lines intensity
ratio is typical of AGNs. On the other hand, the line emis-
sion around 2500 pc is typical of normal star-forming regions
(Figure 3, right panels). This example emphasizes that DS
spectra may underestimate the number of AGNs in our sam-
ple, when AGN contribution is strongly contaminated from
outer HII-regions.
It was quite surprising however to find that among the
11 AGNs for which we have long slit nuclear spectra taken

Page 6

6Decarli et al.
Figure 5. The observed distribution of AGNs (shaded) and HII region-like nuclei (thick histogram) as a function of the host galaxy
dynaical mass, given separately for the Ho+Loiano spectra (top-left); for the SDSS spectra (bottom right); for the Drift-Scan spectra
(top-right). The bottom-left panel shows the luminosity distribution of 24 galaxies that were not spectroscopically classified. The dashed
histograms show the expected frequency of AGNs among unclassified and Drift-scan objects, extrapolated from the distribution of AGNs
in the SDSS sample.
at Loiano with sufficient signal to follow the line intensity
at significant distances from the nucleus (see Figure 4), in
only one galaxy (i.e. VCC 73) the high ionization conditions
found in the nucleus drop to HII-like within 1 kpc. In 90% of
the remaining objects they prevail up to radii significantly
exceeding 1 kpc. Some galaxies (e.g. VCC 1110) show an
even increasing [NII]/Hα ratio up to 3 kpc!, in agreement
with Veilleux et al. (2003) who found extended ionization
conditions in NGC1365 (see their Fig 6d). Even in Drift-
scan spectra these galaxies would have been recognized as
AGNs, as it is the case for VCC1145, whose spectra are
available in both the nuclear form (L) and Drift-scan mode
(DS), showing consistency.
It is worth thus estimating the number of possibly
missed AGNs because of the use of DS spectra, as illus-
trated in Figure 5. While the spectra obtained by Ho et al.
(1997) and by us at Loiano were mostly of luminous ob-
jects, DS and SDSS spectra cover all the luminosity range
of the galaxies in our sample. If the DS method would bias
against AGNs, a luminosity dependence would be artificially
injected. If we assume that the correct rate of AGNs as a
function of mass is sampled with the nuclear spectra from
the SDSS, we expect that, among all the 193 DS spectra, 87
should have AGNs signature, that means, the DS data have
a 54 % efficiency in detecting AGNs. Most of these objects
lie in the narrow interval of mass (1010− 1010.5M⊙ ) (see
Figure 5, top-right panel). Among the data with no other
spectral classification, we expect 23 AGNs instead of the 6
observed.
In order to reduce this discrepancy, 1-dimensional spec-
tra are re-extracted from 2-dimensional frames, by integrat-
ing the counts on small apertures (∼ 5 arcsec). These are
labeled “mod-DS” (modified Drift-Scan) in Table 3 and are
obtained in 22 cases out of the 190 available DS spectra. We
re-extracted the spectra only for those galaxies with a sig-
nificant enhancement in nuclear continuum emission or with
clear inversion of the [NII]/Hα ratio in the two-dimensional
spectrum, as in the case of VCC1401. The spectral clas-
sification of VCC524 changed from HII-like to Transition
Objects; VCC596 passed from HII-like to Sey/LIN AGN;
other 5 galaxies changed their classification from Transition
Objects to Sey/LIN subclass. For all the others the classifi-
cation remained unchanged. Owing to the existence of more
suitable spectral material (the nuclear spectra, namely from

Page 7

AGNs in late-type Virgo galaxies7
Figure 6. The dependence of [NII]/Hα on the dynamical mass
separately for HII-like nuclei (empty symbols) and AGNs (filled
symbols).
Ho, SDSS, Loiano) only 81 DS and 4 mod-DS spectra were
adopted in the final classification.
By applying the same argument we expect to find 4
additional AGNs among the 24 galaxies that remain unclas-
sified because the spectroscopic material is unavailable.
In conclusion we predict that once nuclear spectra will
be available for the the whole sample, the number of AGNs
in the Virgo cluster would increase at most by ∼ 21 units,
over the presently known set of 77.
5 RESULTS
5.1 Nuclear activity versus dynamical mass
The data suggest that nuclear activity is very sensitive on
the galaxy dynamical mass (Mdyn). We observe that the
frequency of AGNs increases very steeply with increasing
Mdyn, a trend already apparent in Figure 2 (the symbol size,
set proportional to Mdyn, increases toward the right-bottom
angle of the Figure). Table 4 clearly shows that the majority
of AGNs (89%) resides in galaxies more massive than 1010.5
M⊙ .
Figure 6 shows the dependence of the [NII]/Hα ratio on
the mass of the galaxies, that holds already for the HII like
nuclei. This is a metallicity/luminosity relation well known
since Lequeux et al. (1979); see also Zaritsky (1993) and
Gavazzi et al. (2004). The linear regression for the HII like
nuclei is drawn in Figure 6:
log([NII]/Hα) = 0.44 · logMdyn− 5.20 (2)
It appears that nuclei classified as AGNs have a [NII]/Hα
ratio significantly in excess of the relation found for HII nu-
clei. This provides a side-aspect, yet important when deal-
ing with imaging surveys carried on with Hα filters (that
Table 4. AGN, HII and NEL statistics among 213 galaxies with
available spectroscopic classification.
AGN (77) HII (121) NEL (15)
logM < 10.5
logM ? 10.5
9 (12%)
68 (88%)
84 (69%)
37 (31%)
6 (40%)
9 (60%)
Sa − Sb 41 (53%)
36 (47%)
5 (4%)
116 (96%)
11 (73%)
4 (27%)Sbc − Im − BCD
are broad enough to contain the [NII] lines braketing Hα):
there is a substantial, strongly mass dependent contribution
from [NII] in the circumnuclear regions of galaxies. This re-
minds that significant (mass dependent) corrections to the
Hα E.W. are needed to disentangle the contribution from
[NII]; otherwise, it would produce systematic overestimates
of the Hα luminosity. This has relevant consequences for
those studies that try to quantify the amount of circumnu-
clear star formation as a function of the local galaxy density
using Hα imaging surveys (e.g. Moss & Whittle, 2005) or
studies focused on Hα luminosity function (e.g. Pascual et
al. 2001, Fujita et al. 2003, Ly et al. 2006;)
To further study and quantify the frequency of AGNs
as a function of Mdyn let us consider the four histograms
in Figure 7. The top-left histogram carries the frequency of
AGNs, HII-like+NEL nuclei and the non classified (because
their spectra are unavailable) objects. If we distribute the 24
non-measured ones according to the frequency of the mea-
sured galaxies, we obtain the top-right histogram showing
a clear-cut dichotomy between HII-like and AGN nuclei at
logMdyn∼ 10.5.
Once the AGNs are more finely divided into Transition,
LINERs and Seyfert (bottom-left histogram; Sey/LIN ob-
jects are not plotted) there is a barely significant separation
of these sub-components as a function of mass, because of
the insufficient statistics (see the excellent analysis of SDSS
data by Kewley et al. 2006). Finally the bottom-right his-
togram shows the differential fraction of AGNs emphasizing
that below 1010M⊙ only one late type galaxy harbors an
AGN, while above 1011.5M⊙ all do.
5.2 Nuclear activity versus galaxy morphology
As emphasized in Table 4, there is an obvious tendency for
HII region-like nuclei to avoid early-type spirals and to in-
habit late-type spirals (and vice versa for NEL galaxies). On
the other hand, AGNs are almost as frequent in early and
late-type spirals, i.e. the Hubble type does not appear to be
the driver of the difference.
Figure 8 shows the Near-IR C31 (left) and the optical
“nuclearity” (right) plotted as a function of Mdynfor galax-
ies in our sample, divided in normal (empty) and AGNs
(filled). Albeit a general increase of the C31 with Mdyn, the
two quantities are non-linearly correlated, as pointed out by
Boselli et al. (1997), Gavazzi et al. (2000) and Scodeggio et
al. (2002). There is in fact a number of very large mass
(log(Mdyn/M⊙ ) > 11.5) galaxies with very little bulges
(C31 < 4). AGNs inhabit high mass galaxies, not neces-
sarily high C31 (Bulge-to-disk) ones. Clear-cut examples of
bulge-less AGNs are M61, M100 and NGC4535, three Scs
with log(Mdyn/M⊙ ) ∼ 11.5 and C31 ∼ 2.5.

Page 8

8 Decarli et al.
Figure 7. Histogram of the mass distribution of HII-like (continuous), AGNs (shaded) and not spectroscopically classified (dotted)
(top-left panel); same with the “not classified” distributed according to the frequency of the classified (top-right); blow-up of the AGNs
divided in Seyfert (shaded), LINERs (continuous), and Transition (dashed) (bottom-left); fractional differential distribution of AGNs
(bottom-right).
Conversely the right panel in Figure 8 shows that the
presence of optical nuclei in the host galaxy increases with
Mdyn as the fraction of AGNs. An optical nucleus may be
due to the AGN continuum as well as to a steep enhancement
in the concentration of stars. The first one would produce
a power-law continuum. No DS nor SDSS spectrum showed
such a behaviour. Spinelli et al. (2006) published high spa-
tial resolution spectra obtained with HST (0.2 arcsec ≈ 16.4
pc in the Virgo Cloud A) of NGC4450, NGC 4698, M88 and
M90, all having Nuc(r) > 1. Despite of the high resolution,
no noticeable power law continuum is observed in the optical
band. We thus conclude that the role of the AGN contin-
uum is commonly negligible in our sample galaxies. Thus the
galaxy dynamical mass seems to be the driver both for the
formation of star nuclei, and for triggering the AGN. Notice
that the massive, bulge-less Sc M61, M100 and NGC4535
have significant nuclei (Nuc(r) ∼ 1.50).
Obric et al. (2006) suggest that AGN host galaxies are
mainly redder than star forming galaxies. This is interpreted
as an evidence of feedback by the AGN on the host galaxy.
AGNs do reside in red galaxies, as shown in Figure 9, but the
existing and well-known color-magnitude correlation stands
for both active and non-active galaxies. The role of the AGN
in halting star formation in the host galaxy is still debated.
5.3 Environmental dependencies
Kauffmann et al. (2004) find that in the SDSS the fraction
of high luminosity AGNs (those with L[OIII]> 107L⊙), pre-
dominant among type 2 Seyfert galaxies, depends on the
environment, i.e. it decreases with increasing galaxy den-
sity, mimiking the behaviour of emission line galaxies. Con-
versely these authors find, in agreement with Miller et al.
(2003), that the frequency of AGNs in general (dominated
in number by low-luminosity LINERs) does not depend on
the environment.
With a sample like the presently analyzed one, that is
entirely made of galaxies members of a rich cluster, it is
impossible to ensure that the estimate of the frequency of
AGNs is not biased by the density of this particular envi-
ronment. Neither we can perform a detailed study of envi-
ronmental effects for each AGN subclass separately, due to
the poor statistics.
Nevertheless we notice firstly that the overall fraction of
AGNs associated with massive galaxies in the SDSS (> 80

Page 9

AGNs in late-type Virgo galaxies9
Figure 8. The Near-IR concentration index (left panel) and the r-band nuclearity index (right panel) as a function of the dynamical
mass. Empty symbols are for HII region-like nuclei, filled symbols for all kinds of AGN, including transition objects. M61, M100 and
NGC4535 are marked in both diagrams.
Figure 9. The dependence of B-H color on the dynamical mass
of late-type galaxies with HII-like nuclei (empty symbols) and
AGNs (filled symbols).
% for log(Mstars/M⊙ ) > 11, Kauffmann et al. 2003), is
consistent with the one found in Virgo (see Figure 7)3. Such
3We transform Mstars used by the SDSS into Mdyn (within
the optical radius)adopted
log(Mstars/M⊙ ) = 1.41 × log(Mdyn/M⊙ ) − 5.1. This transfor-
mation is the linear correlation between logMdynand logMstars
in the presentpaper,using
a result is in contrast with Popesso & Biviano (2006). If
we adopt an average galaxy velocity dispersion around 600
km/s for each Virgo Cloud (see Gavazzi et al. 1999), cut our
sample at Mdyn ≈ 6.3 · 1010M⊙ (roughly corresponding
to Mpg ≈ −20), and adopt the same limits in classifing
the nuclear activity, our AGN fraction reaches 60%, nearly
a 2 σ deviation from their correlation. We argue that the
reason of such a difference resides in the sample selection.
Unlike Kauffmann et al. (2003), the sample of Popesso &
Biviano has no constraints on line fluxes. Thus, it includes
lots of No Emission Line galaxies, especially among early-
type, gas poor galaxies for which the BPT classification may
be unavailable.
Secondly we remark that, if we divide the Virgo sam-
ple in two subsamples composed of galaxies in and outside
5 degrees (∼ 1.5 Mpc) projected radial distance from M87,
we obtain that the percentage of AGNs varies from 32±8%
“in” to 40 ± 10% “out”, i.e. not significantly. The lack of
a environmental dependence of our sample AGNs is illus-
trated in Figure 10. Similarly by dividing the sample us-
ing an independent probe of the environmental influence,
as represented by the Hydrogen deficiency parameter (see
the parameter definition in Haynes & Giovanelli, 1984 and
Boselli & Gavazzi, 2006), we obtain that the percentage of
AGNs varies from 41 ± 13% among highly deficient objects
(DefHI > 0.5), to 27 ± 7% among the unperturbed objects
(DefHI ? 0.5), i.e. barely significantly (see figure 11). We
thus conclude that the frequency of AGNs is not strongly
environment dependent.
of Bruzual & Charlot synthesis population models fitted to the
SEDs of the sampled spirals

Page 10

10 Decarli et al.
Figure 10. The observed distribution of AGNs (shaded) and
HII region-like+NEL nuclei (thick histogram) as a function of
Mdyn, given separately for galaxies found within (top) or outside
(bottom) 5 degrees projected radial distance from M87.
5.4Indirect estimate of black hole masses and
Eddington ratios
With the current data it is not possible to directly infer the
mass of the black holes that power the AGNs observed in
our sample. We may nevertheless consider the possibility of
extending the known scale relations valid for spheroids to
the bulges and optical nuclei of our late-type galaxies. The
relations used in this section are the MBH – σ (Tremaine et
al., 2002 and references therein) and the MBH – Lbulge(H)
(Marconi & Hunt, 2003). Data on the stellar velocity disper-
sion σ are taken from hypercat4when available for the AGN
sample. The values for σ are estimated using various tech-
niques (mainly the Fourier quotient method). Data on the
bulge luminosity are taken from Gavazzi et al. (2000). The
bulge luminosity is estimated from the deconvolution of the
galaxy light profile in H-band. The light profile is modeled
as a de Vaucouleurs or an exponential curve for the bulge
and an exponential law for the disk component. Since the
AGN continuum flux is found to be negligible in our sample
galaxies, and the fit procedure is performed only out of the
first two seeing radii, we disregarded the AGN contribution
in the fit. Typical values for the Bulge to Total luminosity
ratio (B/T) varies between few % to 30%. A reasonable es-
timate of the bulge luminosity error is around 30%. Twenty
and 64 out of 77 AGNs in our sample have data for these
two estimates to be performed; this subsample includes, for
example, M100 with its optical nucleus and M58 which hosts
a prominent bulge. Figure 12 shows the comparison between
the black hole mass estimates obtained with the two meth-
ods. The slight (0.4 dex) offset between the two estimates
4http://leda.univ-lyon1.fr
Figure 11. [NII]/Hα ratio plotted versus Hydrogen deficiency,
as defined in Haynes & Giovanelli (1984). The AGN fraction in-
creases only slightly from non-deficient to deficient galaxies.
may depend on the heterogeneous data sources adopted for
σ estimates, or on the bulge luminosity underestimate due
to inclination. Since we want to study the black-hole masses
and the Eddington ratios of Virgo AGNs only from a statis-
tical point of view, and the data dispersion is higher, such a
offset is irrelevant for the purposes of our analysis. We no-
tice that the two methods provide consistent values of the
black hole mass for M100 as well as for M58. This result is
surprising given the difference between M100 and M58, the
first hosting a pseudobulge/optical nucleus (Kormendy &
Kennicutt, 2004; see also Allard et al., 2006) while the sec-
ond a massive extended bulge: this coincidence highlights
the close underlying connection between the stellar velocity
dispersion and the nuclear or bulge luminosity in these two
galaxies.
As shown in Figure 12 the black hole masses never ex-
ceed a few 108solar masses, and a number of galaxies hosts
relatively light black holes (MBH∼
holes appear to fill the low mass end of the scaling relation
purely sampled in our Local Universe.
Black hole masses may provide some hints on the Ed-
dington ratios of the AGNs if we have a way to estimate the
bolometric luminosity of the active BH. Since most of our
AGN are LINERs, it is difficult to disentangle the emission
of the AGN from that of the stellar continuum in the op-
tical band. As proposed in Heckman et al. (2004) (see also
Kauffmann & Heckman, 2005), we can estimate the AGN
bolometric luminosity from the [OIII] emission line flux. We
want to stress that this is a rough estimate of bolometric
luminosities, since a good spectrometric calibration is often
hard to achieve, because of the diffuse nature of the source,
and contamination from other ionizing radiation sources is
often present. Combining the data on the [OIII] flux from
Ho and SDSS, and matching them with the subsample of the
<107M⊙ ). These black

Page 11

AGNs in late-type Virgo galaxies 11
Figure 12. Comparison between MBHestimated using the MBH
– σ relation (Tremaine et al., 2002) (x axis) and MBHfrom the
MBH– Lbulge(Marconi & Hunt, 2003) (y axis). Uncertainties in
the relations are accounted for in our error bars; for Lbulgewe
assumed a 30 % error both on the Bulge to Total value and on
total NIR Luminosity.
AGN BH masses, we infer the bolometric to Eddington lu-
minosity ratios of 33 of our AGNs (we could not use Loiano
spectra since they do not include [OIII] line, neither drift
scan spectra since they lack of an absolute spectrophoto-
metric calibration). The mean error of Heckman’s bolomet-
ric correction is 0.38 dex, thus our Eddington ratios mean
error is nearly 0.5 dex. These ratios are found to fall in the
range 10−5−10−2with a mean around 0.001, indicating that
these AGN are accreting at very low rates (see Figure 13).
The only exceptions are NGC4388, with Eddington ratio of
0.06 which reflects the occurrence of ejection in this source
(see Yoshida et al., 2004), and NGC4123, a LIN/HII object,
in which O,B-stars ionization is probably significant.
6 CONCLUSIONS
The present study focused on the nuclear activity of spiral
galaxies in the Virgo cluster, as representative of our Local
Universe. This paper provides the first census of AGNs as-
sociated to the cluster’s spirals, complete down to the 15th
magnitude. This allowed the identification of a number of
low luminosity AGNs even in late type spirals that show a
bright optical stellar nucleus rather than a prominent bulge:
M100 hosts a LINER and offers a clear example.
A first result is that the AGNs, and among them LIN-
ERs, are primarily associated to the spirals with dynamical
mass in excess of M∼
>1010M⊙. This result is in close agree-
ment with Kauffmann et al. (2003) who found that AGNs in
the SDSS are hosted in galaxies (with mean redshift ∼ 0.1)
as massive as ours.
Figure 13. Eddington ratio distribution in a subset of AGNs in
our sample. The Marconi & Hunt relation was applied to estimate
black hole masses and Eddington luminosities.
In particular, we found the fraction of AGNs in late-
type galaxies more massive than ∼ 108.5M⊙ is 77/237 ≈ 32
%. This fraction steeply increases with the galaxy mass: if
we consider only galaxies with Mdyn> 1011.5M⊙ , the AGN
fraction approaches 100 %.
A second result is that the presence of nuclear activity
is insensitive to the morphological type of the host spiral, as
it is equally present in Sa - Sb and later types. Thus, M100 is
not an isolated case. AGNs are often associated to prominent
stellar nuclei. In addition, the nuclear activity displays little
or no environmental density dependence inside the cluster,
neither as a function of the host galaxy position within the
cluster, nor of the neutral gas deficiency. Gas accretion onto
central BHs seems independent of the gaseous content.
As already pointed out, the nuclear activity occurs in
spiral galaxies with dynamical masses above ∼ 1010M⊙ ,
and the fraction of Mdyn in the spheroidal component in
spirals varies between a few % to 30 %. If we apply the
known black hole scaling relations, obeyed in ellipticals, to
the bulges and optical nuclei of our sample (thus assum-
ing self-similarity among early and late type galaxies), this
translates into a lower limit for the black hole mass of ∼ 106
M⊙ . Based on AGN luminosity function evolution, Marconi
et al. (2006) have suggested that also the black holes grow
following the galaxy downsizing, i.e. massive black holes
grow faster and earlier with cosmic time than the lighter
ones. If black holes lighter than 106M⊙ are hosted in spirals
with less prominent nuclei/bulges, we would have expected
them still in their active phase in the galaxies inside our cen-
sus. The lack of AGNs in lower mass spirals may be a hint
that such lighter BHs did not find yet the right conditions
to grow or the environment to form. This dichotomy on the
BH behaviour around 106M⊙ has to be considered valid at
a statistical level.

Page 12

12 Decarli et al.
It is interesting to notice that ACS imaging surveys on
elliptical galaxies have shown that a sizeable fraction of low
and intermediate luminosity ellipticals contain stellar cusps
at their center with masses in the range expected from the
lower-end extrapolation of the MBH– Magnitude correlation
(Wehner & Harris, 2006; Ferrarese et al. 2006b). Above 1010
M⊙ ellipticals host exclusively massive black holes. Simi-
larly, in our sample of spirals, we have not found at statistical
level an active AGN below such a characteristic dynamical
mass once we correct for the total to bulge ratio. This is
a hint that below a critical mass for the hosting spheroidal
component (either a bulge or a stellar nucleus), lighter black
holes become rarer.
Only a relatively small sample of black holes in spi-
ral galaxies have a direct mass measure. In this paper we
found the occurrence of BHs at the center of spiral galaxies
disregarding of the galaxy morphology, and we provided an
estimate of the black hole mass under the assumption of self-
similarity between the bulges and stellar nuclei of our spirals
with the sample of quiescent black hole hosted in ellipticals.
We do not know if this similarity applies, since either in the
black hole mass as well as in the mass accretion rate there
might be a dependence on the angular momentum of the
host spiral that can affect black hole grow. Further high res-
olution observations focused on black hole mass measures
in late-type galaxies may provide clues on the applicability
of the MBH – bulge relations and the role of the both the
Bulge to Total ratio and the dynamic of the gas and stars
in the host galaxy.
Acknowledgements
We thank F. Haardt and A. Treves for useful discussions.
G.G. thanks A. A. for inspiring hints. This research has
made extensive use of the GOLDMine Database (Gavazzi
et al. 2003). We acknowledge the usage of the HyperLeda
database (Paturel et al. 2003). 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.
REFERENCES
Adelman-McCarthy, J., et al. 2007, submitted to ApJ Sup-
plements
Allard, E. L., Knapen, J. H., Peletier, R. F., Sarzi, M.,
2006, MNRAS, 371, 1087
Baldwin, J. A., Phillips, M. M., Terlevich, R., 1981, PASP,
93, 5
Bennert, N., et al., 2006, A&A, 446, 919
Binggeli, B., Sandage, A., & Tammann, G., 1985, AJ, 90,
1681
Binggeli, B., Popescu, C. & Tammann, G., 1993, AAS, 98,
275
Boselli, A., Tuffs, R. J., Gavazzi, G., Hippelein, H., Pierini,
D., 1997, A&AS, 121, 507
Boselli, A., Boissier, S., Cortese, L., Gil de Paz, A., Seibert,
M., Madore, B. F., Buat, V., Martin, D. C., 2006, ApJ, 651,
811
Boselli, A. & Gavazzi, G., 2006, PASP, 118, 517
Bruzual, G. & Charlot, S. 2003, MNRAS, 344, 1000
Chy˙ zy, K. T., Soida, M., Bomans, D. J., Vollmer, B.,
Balkowski, Ch., Beck, R., Urbanik, M., 2006, A&A, 447,
465
Cote’, P., et al, 2006, ApJS, 165, 57
Dressler, A., Thompson, I. B., Shectman, S. A., 1985, ApJ,
288, 481
Ferrarese, L., et al., 2006a, ApJS, 164, 334
Ferrarese, L., et al., 2006b, ApJ, 644L, 21
Filho, M. E., Fraternali, F., Markoff, S., et al., 2004, A&A,
418, 429
Fujita, S.S., Ajiki, M., Shioya, Y., et al., 2003, ApJ, 586L,
115
Garca-Burillo, S., Combes, F., Schinnerer, E., Boone, F.,
Hunt, L. K., 2005, A&A, 441, 1011
Gavazzi, G., Pierini, D., & Boselli, A., 1996, A&A, 312, 397
Gavazzi, G., Boselli, A., Scodeggio, M., et al., 1999, MN-
RAS, 304, 595
Gavazzi, G., Franzetti, P., Scodeggio, M., et al., 2000,
A&A, 361, 863
Gavazzi, G., Zibetti, S., Boselli, et al., 2001, A&A, 372, 29
Gavazzi, G., Boselli, A., Donati, A., et al., 2003, A&A, 400,
451
Gavazzi, G., Zaccardo, A., Sanvito, G., et al., 2004, A&A,
417, 499
Gavazzi, G., et al., 2006, A&A, 446, 839
Halderson, E. L., Moran, E. C., Filippenko, A. V., Ho, L.
C., 2001, AJ, 122, 637
Hao, L., Strauss, M.A., Tremonti, C.A., et al. 2005, ApJ,
129, 1783
Haynes, M. & Giovanelli, R., 1984, AJ, 89, 758
Heckman, T.M., Kauffmann, G., Brinchmann, J., et al.,
2004, ApJ, 613, 109
Ho, L.C., Filippenko, A.V., & Sargent, W.L.W., 1997,
ApJS, 112, 315
Kauffmann, G., et al. 2003, MNRAS, 346, 1055
Kauffmann, G., White, S.D.M., Heckman, T. M., et al.
2004, MNRAS, 353, 713
Kauffmann, G., Heckman, T.M., 2005, RSPTA, 363, 621
Kewley, L.J., Dopita, M.A., & Southerland, R.S., et al.
2001, ApJ, 556, 121
Kewley, L.J., Groves, B., Kauffmann, G., Heckman, T.,
2006, MNRAS, 372, 961
Kormendy, J., & Kennicutt, R.C., 2004, ARA&A, 42, 603
Lequeux, J., Peimbert, M., Rayo, J. F., Serrano, A., Torres-
Peimbert, S., 1979, A&A, 80, 155
Ly, C., Malkan, M., Kashikawa, N., et al., 2006, AAS, 208,
5302
Marconi, A. & Hunt, L., 2003, ApJ, 589, 21
Marconi, A., Comastri, A., Gilli, R., Hasinger, G., Hunt,
L.K., Maiolino, R., Risaliti, G., Salvati, M., 2006, MmSAI,
77, 742
Martini, P., Kelson, D., Kim, E., et al., 2006, ApJ, 644, 116
Miller, C.J., Nichol, R.C., G´ omez, P.L., Hopkins, A.M.,
Bernardi, M., 2003 ApJ, 597, 142
Moss, C. & Whittle, M., 2005. MNRAS, 357, 1337
Pascual, S., Gallego, J., Arag´ on-Salamanca, A., Zamorano,
J., 2001, A&A, 379, 798
Paturel, G., et al., 2003, A&A, 412, 45
Popesso, P., & Biviano, A., 2006, A&A, 460, L23
Shen, Y., Mulchaey, J.S., Raychaudhury, S., et al., 2007,
ApJ, 654, L115

Page 13

AGNs in late-type Virgo galaxies 13
Spinelli, P.F., Storchi-Bergmann T., Brandt C.H., Calzetti,
D., 2006, ApJ, 166, 498
Stasi´ nska, G., Cid Fernandes, R., et al., 2006, MNRAS,
371, 972
Scodeggio, M., Gavazzi, G., Franzetti, P., et al., 2002,
A&A, 384, 812
Terashima, Y., Iyomoto, N., Ho, L., Ptak, A., 2002, ApJS,
139, 1
Tremaine, S., Gebhardt, K., Bender, R., et al., 2002, ApJ,
574, 740
Veilleux, S., & Osterbrock, D.E., 1987, ApJS, 63, 295
Veilleux, S.; Shopbell, P. L.; Rupke, D. S.; Bland-
Hawthorn, J.& Cecil, G, 2003, AJ, 126, 2185
Wehner, E.H., Harris, W.E., 2006, ApJ, 644L, 17
York, D.G.; Adelman, J.; Anderson, J.E.; et al. 2000, AJ,
120, 1579
Yoshida, M., Ohyama, Y., Iye, M., et al., 2004, AJ, 127, 90
Zaritsky, D., 1993, PASP, 105, 1006

Page 14

14 Decarli et al.
Table 5: spectroscopic results
VCC/CGCGNGC TYPE
mpg
mag
(4)
D
logLH
L⊙
(6)
C31(H)
Nuc(r)
mag arcsec−2
(8)
[NII]/Hα
[OIII]/Hβ
Spec ClassRef
Mpc
(5)(1) (2)(3)(7)(9) (10)(11)(12)
1
10
15
24
25
47
58
66
67
73
87
89
92
97
105
119
120
126
131
135
145
152
157
162
167
170
172
187
199
213
221
222
224
226
227
234
241
267
289
307
315
318
323
324
340
341
358
382
386
393
404
415
459
460
465
483
491
492
497
508
509
522
524
534
552
559
568
570
576
596
613
630
655
656
664
667
675
688
692
699
713
737
739
785
787
792
809
827
836
848
849
851
857
859
865
873
874
905
912
921
938
939
950
957
958
971
-
-
-
-
BCD
BCD
Sa
BCD
Sc
Sa
Sb
Sc
Sc
Sb
Sm
Sc
Sb
Sc
Sd
Sc
Scd
Sd
Sc
S/BCD
Sc
Scd
Sc
Sd
Sb
Sd
BCD
Scd
Sa
S/BCD
Sc
Sa
Scd
Sc
Sdm
Sa
Sd
Sbc
Sc
Sc
Sa
Scd
Sa
BCD
BCD
Sa
Sa
Sc
Sa
Sc
Scd
Sd
BCD
Sa
Sc
Sc
Scd
Sa
Sc
Sc
Sdm
Sa
Sbc
Sa
Sc
Sab
S..
Sab
Sbc
Sc
Sa
Sd
S/BCD
Sb
Sc
Sc
Sa
Sc
Sc
Pec
Sc
S/BCD
Sd
Sa
Scd
Sab
Sc
Sc
Sab
16
Sbc
Sc
Sb
Sc
Sc
Sc
Sc
Sc
Sbc
Sbc
Sc
Sc
Sm
Sc
Sa
Sd
14.78
14.75
14.70
14.95
12.46
14.20
13.17
11.89
13.98
13.35
15.00
12.53
10.92
13.20
13.68
14.76
13.47
14.42
14.34
14.81
12.77
13.48
11.50
14.41
10.97
14.56
14.50
13.91
12.95
14.26
13.43
12.62
14.70
12.53
14.90
12.99
14.60
13.82
14.81
10.43
14.98
14.01
14.91
14.78
14.43
12.70
13.80
12.37
14.47
13.25
15.00
14.82
14.95
11.20
12.62
12.08
12.86
13.76
12.55
10.17
14.98
13.19
12.79
13.59
13.61
12.56
14.91
12.73
13.70
10.11
12.60
13.10
13.21
13.14
13.50
14.24
15.00
13.94
12.93
14.22
14.04
14.94
14.37
12.16
13.69
12.36
14.55
13.76
11.83
14.72
13.27
14.14
11.76
14.61
13.02
12.56
12.99
13.42
12.97
13.14
13.28
12.92
14.49
12.67
12.13
14.28
32
32
32
32
32
32
32
17
32
32
17
32
17
32
32
32
32
17
17
32
17
17
17
17
17
17
32
17
32
17
32
32
17
17
32
32
17
23
32
17
32
32
32
17
32
23
23
32
32
23
17
23
17
17
17
17
17
23
17
17
23
17
23
23
17
17
23
17
23
17
17
17
17
23
17
23
17
23
17
23
23
17
17
17
23
23
17
23
17
23
23
23
17
17
17
17
17
23
17
17
17
23
17
17
17
23
9.27
9.27
-
9.34
10.39
10.09
10.17
10.22
9.86
10.58
8.36
10.65
10.99
10.53
-
-
10.20
-
9.52
9.82
9.98
9.90
10.48
9.23
11.14
9.03
9.30
9.57
10.83
9.21
9.88
10.88
9.19
10.26
9.44
10.68
-
9.72
9.33
10.94
9.70
9.18
10.00
9.09
9.43
10.60
10.03
10.65
10.05
9.85
9.36
-
8.74
10.78
9.88
10.42
9.42
10.21
10.58
10.98
-
9.83
10.27
10.10
9.05
10.22
-
10.32
10.24
11.14
10.35
9.91
9.56
10.35
8.95
9.78
8.54
9.68
9.64
9.71
10.16
8.73
8.65
10.46
9.65
10.66
9.27
10.14
10.54
8.70
9.79
9.80
10.54
9.62
9.69
10.39
9.97
9.66
9.85
9.64
9.74
9.87
8.63
9.91
10.61
9.55
3.36
2.65
-
6.23
3.29
3.78
2.79
3.24
5.55
3.79
3.18
1.89
5.04
2.83
-
-
3.07
-
3.64
3.61
3.73
3.54
2.54
2.44
9.54
2.80
2.80
2.44
4.99
2.94
2.60
6.32
2.89
2.77
3.17
4.09
-
2.50
2.18
3.64
-
2.93
3.49
2.90
2.83
4.43
3.76
3.56
3.46
2.69
3.41
-
3.09
3.52
2.82
3.01
2.73
4.32
3.95
2.41
-
3.00
3.16
3.63
2.54
3.61
-
4.54
2.79
2.48
4.41
3.98
2.50
5.88
2.55
3.74
-
2.03
2.95
4.00
2.88
2.13
2.40
9.36
3.11
3.24
3.24
3.14
4.69
2.86
2.49
3.45
5.64
2.47
4.18
2.95
2.70
3.06
3.14
2.19
2.80
3.25
2.78
3.23
5.15
3.44
0.29
0.33
-
-
0.83
0.69
0.8
0.84
0.4
0.56
0.38
0.81
1.15
0.8
0.64
0.19
0.51
-
0.47
0.58
0.74
0.54
0.7
0.15
1.18
0.61
0.54
0.11
0.78
0.5
0.39
0.62
0.37
0.7
-
-
0.35
0.54
0.34
1.02
-
0.6
0.53
0.63
0.32
0.97
-
0.75
-
0.74
0.43
0.28
0.44
0.97
0.29
0.7
0.08
0.69
0.37
1.58
0.54
-
0.57
0.88
0.71
0.48
0.28
-
0.53
1.53
0.91
0.23
0.48
0.88
0.85
0.57
0.27
0.16
0.76
0.35
0.27
0.37
0.72
1.13
0.44
1.09
0.26
0.43
0.55
0.58
0.59
0.42
1.26
0.28
0.44
0.53
0.77
0.53
0.94
0.92
0.77
0.71
-
0.34
0.82
0.43
0.27
0.25
0.45
0.12
0.32*
0.45
0.38
0.19*
0.12*
2.84
0.09
0.36
1.29
0.43
0.27*
0.08
0.22*
-
0.46
0.32
1.93
0.38
0.35*
0.09
3.27
0.26
0.11
0.24*
-
0.71
0.37
-
0.28
0.74
-
7.69
0.04
0.21*
0.25
0.33*
-
0.16*
2.84
0.05
0.10
2.46
-
0.36
-
0.25*
0.40
-
0.11*
4.19
0.12
0.35
0.13*
1.14
3.49
0.41*
-
-
0.52
-
-
0.46*
0.11*
2.26
0.83
0.66
2.34
0.26*
0.27*
1.05
0.09*
0.16*
0.09
0.10*
0.22
0.21*
0.59*
-
0.17
-
0.21*
2.38
0.21*
0.19*
0.56
0.14*
0.22*
0.24*
1.02
3.58
0.19*
0.35*
0.37
0.29*
0.32
0.42*
0.33*
0.89
0.19*
0.44
1.61
0.11*
0.48
0.71
-
2.37
0.51*
-
-
1.37*
-
-
2.09
0.14
-
0.42
-
1.28
1.14*
-
-
-
-
0.45
-
1.83
-
0.49
4.28
0.88*
-
-
0.48
-
0.62
-
-
-
3.37
-
3.85
0.22*
-
1.68*
-
3.88
2.37
-
-
-
-
-
-
-
2.88*
-
1.94
0.23
2.05*
-
-
0.28*
-
-
0.28
-
-
-
-
-
-
-
-
-
0.17*
-
2.68*
-
2.77
-
0.28
1.31*
-
-
1.11
-
-
-
0.85*
1.80*
11.34
2.05*
0.94*
0.77*
-
-
0.97*
-
-
0.71*
-
0.31*
0.20*
-
-
0.62
-
1.91*
HII
HII
SDSS
SDSS
SDSS
SDSS
Ho
SDSS
SDSS
Ho
DS
LOI
SDSS
SDSS
Ho
SDSS
DS
SDSS
DS
SDSS
SDSS
SDSS
SDSS
SDSS
Ho
SDSS
Ho
SDSS
SDSS
DS
LOI
SDSS
SDSS
Ho
SDSS
SDSS
SDSS
MDS
SDSS
DS
SDSS
Ho
SDSS
DS
SDSS
SDSS
SDSS
SDSS
SDSS
LOI
DS
DS
SDSS
DS
DS
Ho
SDSS
Ho
DS
LOI
SDSS
Ho
DS
SDSS
MDS
DS
DS
DS
DS
SDSS
SDSS
Ho
LOI
DS
DS
LOI
DS
DS
SDSS
DS
SDSS
DS
DS
DS
SDSS
Ho
DS
SDSS
DS
DS
Ho
DS
DS
DS
Ho
SDSS
DS
DS
Ho
DS
LOI
NED
DS
SDSS
DS
SDSS
Ho
DS
LIN/HII
HII
LIN/HII
LIN/HII
HII
HII
HII
Sey/LIN
HII
HII
LIN
LIN/HII
HII
HII
HII
NEL
LIN/HII
HII
LIN
HII
LIN/HII
HII
LIN
HII
HII
HII
Sey/LIN
LIN
HII
Sey
HII
Sey/LIN
NEL
Sey/LIN
HII
HII
HII
LIN/HII
NEL
HII
LIN
HII
HII
Sey/LIN
NEL
HII
NEL
HII
HII
HII
HII
LIN
HII
HII
HII
Sey/LIN
Sey/LIN
LIN
HII
NEL
LIN/HII
NEL
HII
LIN/HII
HII
Sey/LIN
Sey/LIN
LIN
Sey/LIN
HII
HII
Sey/LIN
HII
HII
HII
HII
HII
HII
LIN/HII
HII
HII
Sey
HII
Sey/LIN
HII
HII
Sey
HII
HII
HII
LIN
Sey/LIN
HII
HII
HII
HII
HII
LIN
HII
Sey/LIN
HII
LIN/HII
LIN
HII
4152
4165
-
4178
-
4180
-
4189
4192
4193
-
-
4197
-
-
-
4206
4207
4212
-
4216
-
-
4222
4224
-
4234
4235
-
4237
-
4241
-
-
4252
4254
-
-
4257
-
-
4260
4264
4273
4277
4276
-
-
-
4293
4294
4298
4299
4300
4302
4303
-
4305
4307
4309
-
4312
-
4313
4316
4321
4324
4330
4344
4343
-
-
-
4353
4351
-
4356
-
-
4378
4376
4380
-
-
4388
-
4390
-
4394
-
4396
4402
4405
-
4413
4412
4416
-
-
4420
4419
4423

Page 15

AGNs in late-type Virgo galaxies15
Table 5: spectroscopic results
VCC/CGCGNGCTYPE
mpg
mag
(4)
D
logLH
L⊙
(6)
C31(H)
Nuc(r)
mag arcsec−2
(8)
[NII]/Hα
[OIII]/Hβ
Spec Class Ref
Mpc
(5)(1) (2) (3)(7) (9)(10)(11)(12)
975
979
980
984
1002
1011
1043
1047
1060
1086
1091
1110
1118
1126
1145
1156
1158
1189
1190
1193
1205
1217
1249
1266
1290
1326
1330
1375
1379
1393
1401
1410
1412
1419
1450
1508
1516
1524
1532
1540
1552
1554
1555
1557
1562
1569
1575
1581
1588
1615
1624
1673
1675
1676
1678
1686
1690
1696
1699
1725
1726
1727
1730
1757
1758
1760
1780
1791
1811
1813
1859
1868
1923
1929
1932
1943
1955
1972
1987
1999
2023
2033
2058
2070
13046
14063
14110
15031
15032
15037
15049
15055
41041
41042
43028
43034
43041
43054
43066
43071
43093
69036
71060
71068
71092
100004
- Scd
Sa
Scd
Sa
Sc
Sdm
Sb
Sa
Sm
S..
Sbc
Sab
Sc
Sc
Sb
Scd
Sa
Sc
Sa
Sc
Sc
Sm
Im
Sdm
Sb
Sa
Sa
Sc
Sc
Sc
Sbc
Sm
Sa
18
Sc
Sc
Sbc
Sd
Sc
Sb
Sa
Sm
Sc
Scd
Sc
Scd
Sm
Sm
Scd
Sb
Sc
Sc
Pec
Sc
Sd
Sm
Sab
Sc
Sm
13.58
12.32
14.17
12.82
12.48
14.85
10.91
12.48
15.00
13.66
14.60
10.93
13.31
13.30
11.66
14.13
12.09
13.70
12.22
14.62
13.04
14.59
14.75
14.63
13.09
13.41
13.17
12.00
12.62
14.01
10.27
14.57
12.12
13.64
13.29
12.34
12.73
13.51
14.05
11.32
12.58
12.30
10.51
14.53
11.01
15.00
13.98
14.55
12.81
10.98
13.89
12.08
14.47
11.70
13.70
13.95
10.25
11.81
14.11
14.51
14.54
10.56
12.61
13.60
14.99
12.54
13.70
14.67
12.92
11.51
12.52
13.75
13.14
13.77
13.19
12.19
14.32
12.03
11.14
13.08
13.86
14.65
11.55
11.53
13.50
12.40
12.60
13.30
12.90
13.60
12.90
13.20
13.00
13.10
14.50
13.10
12.30
13.00
14.40
12.50
12.80
13.20
13.30
13.50
11.90
11.30
23
23
17
17
23
17
17
17
17
23
23
17
23
17
17
17
17
17
23
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
9.66
10.38
8.15
10.11
10.22
8.90
10.83
10.34
-
10.08
9.18
10.94
10.08
10.06
10.64
-
10.54
9.30
10.78
9.28
9.73
8.94
8.47
-
9.91
9.78
10.13
9.35
9.84
9.47
11.18
9.02
10.53
9.66
9.44
9.93
9.85
9.49
9.55
10.91
10.22
9.90
10.76
9.12
10.71
8.42
9.31
8.74
10.05
10.91
9.68
10.41
8.96
10.71
8.81
9.32
11.05
10.38
8.83
8.97
8.64
11.11
10.22
9.54
9.03
10.32
9.63
8.95
9.85
10.84
10.10
9.92
9.83
9.46
9.99
10.26
9.36
10.51
10.66
10.04
9.17
8.66
10.48
10.75
10.25
10.78
10.13
10.14
10.36
9.13
10.57
10.04
9.87
9.94
9.52
9.79
9.97
9.43
9.76
10.12
10.27
9.67
9.87
9.76
10.54
10.52
3.03
3.51
2.07
4.68
2.73
2.46
10.21
7.42
-
3.04
3.77
4.33
2.84
3.28
7.17
-
7.45
2.42
4.65
2.61
2.33
2.80
2.73
-
3.88
2.87
4.41
2.78
2.27
3.08
3.73
2.80
5.95
3.44
2.93
3.17
3.12
-
1.69
6.79
3.00
2.92
2.18
2.83
6.96
3.16
2.44
2.77
2.25
4.34
2.77
2.73
2.87
4.27
2.91
2.98
4.37
2.94
2.64
2.92
2.73
4.51
2.68
3.53
3.47
6.11
3.01
2.86
2.71
5.44
3.40
3.02
2.48
3.14
3.08
3.54
4.17
3.06
2.93
6.03
3.09
3.70
2.80
5.78
5.05
2.88
4.03
3.73
8.40
2.12
4.91
2.70
2.27
3.61
2.41
3.25
2.93
2.75
4.59
3.32
1.82
3.34
2.85
3.56
6.28
2.98
0.61
0.83
0.44
-
0.81
0.23
1.13
-
0.37
0.45
0.65
1.4
0.23
0.69
1.34
0.43
-
0.54
0.69
0.18
0.1
-
-
-
0.79
-
1.21
0.72
0.77
0.45
1.22
0.33
-
0.57
0.61
0.83
0.46
0.75
0.45
1.02
-
0.26
1.41
0.34
1.61
0.54
0.41
-
0.49
1.46
0.51
-
0.37
-
0.54
0.45
1.8
0.76
0.57
0.34
-
1.43
0.6
0.63
0.3
0.78
0.43
0.18
0.61
-
0.82
0.42
0.71
0.49
0.15
0.94
1.04
-
1.01
-
0.36
0.38
1.02
1.22
0.97
0.31
0.57
0.5
1.32
0.8
0.9
0.84
0.89
1.3
0.7
0.61
0.64
0.56
0.63
0.51
0.94
0.88
0.16
0.91
-
0.82
0.12*
0.34
0.14*
-
0.43*
-
1.97
-
0.07
-
0.11*
6.74
0.32*
0.37
3.25
-
-
-
0.8
0.25*
0.26*
-
-
-
1.46
-
-
0.27*
0.28*
0.26*
1.42
0.20*
-
0.35
0.24*
0.22*
0.33
0.14*
0.25
0.45*
2.69
0.12*
0.41
0.22
0.50
0.12
0.30*
0.05*
0.34*
22.14
0.31
0.39
0.17*
0.34*
0.10*
0.20*
1.52
0.39*
0.11*
0.14*
-
2.88
1.82
0.34
0.08*
1.10
0.39
-
0.29*
0.67
0.63*
0.68
0.28
0.20*
0.40*
1.18
0.24
0.28
0.28
-
0.15*
0.11*
1.58
4.84
0.89
0.24*
0.37
1.30
1.31*
0.36
0.85
0.32
0.23*
0.54
-
0.24*
0.73
0.11*
1.70
0.41*
0.33
0.79
0.41
0.41
-
1.13
- HII
HII
HII
NEL
LIN/HII
HII
LIN
NEL
HII
LIN/HII
HII
LIN
HII
HII
LIN
HII
NEL
HII
Sey/LIN
HII
HII
NEL
NEL
HII
Sey/LIN
HII
NEL
HII
HII
HII
Sey
HII
NEL
HII
HII
HII
HII
HII
HII
LIN
Sey/LIN
HII
LIN/HII
HII
LIN/HII
HII
HII
HII
Sey
LIN
HII
HII
HII
HII
HII
HII
LIN
HII
HII
HII
HII
Sey
Sey/LIN
HII
HII
Sey
HII
HII
HII
LIN
Sey/LIN
LIN
HII
HII
HII
Sey
HII
HII
HII
NEL
HII
HII
Sey/LIN
Sey
LIN
HII
HII
Sey/LIN
LIN
HII
LIN
HII
HII
LIN/HII
LIN/HII
HII
LIN/HII
HII
Sey/LIN
LIN/HII
HII
Sey/LIN
LIN/HII
LIN/HII
LIN
LIN
DS
Ho
DS
SDSS
DS
DS
Ho
DS
SDSS
DS
DS
Ho
DS
SDSS
Ho
SDSS
DS
DS
SDSS
DS
Ho
DS
DS
SDSS
SDSS
DS
LOI
DS
DS
DS
Ho
DS
LOI
SDSS
DS
DS
SDSS
DS
SDSS
Ho
MDS
Ho
LOI
SDSS
LOI
SDSS
DS
DS
NED
Ho
SDSS
Ho
DS
Ho
DS
DS
Ho
DS
DS
DS
DS
Ho
SDSS
SDSS
DS
SDSS
SDSS
DS
DS
Ho
DS
SDSS
LOI
DS
DS
Ho
SDSS
Ho
Ho
DS
DS
DS
SDSS
Ho
SDSS
DS
LOI
SDSS
Ho
SDSS
SDSS
LOI
DS
Ho
Ho
DS
Ho
DS
SDSS
DS
SDSS
MDS
SDSS
LOI
Ho
Ho
4424
-
4425
4430
-
4438
4440
-
4445
-
4450
4451
-
4457
-
4461
-
4469
4466
4470
-
-
-
4480
4491
4492
-
4498
-
4501
4502
4503
4506
-
4519
4522
4523
-
4527
4531
4532
4535
4533
4536
-
-
-
4540
4548
4544
4567
-
4568
-
-
4569
4571
-
-
-
4579
4580
4584
-
4586
4591
-
4595
4596
4606
4607
4630
4633
4634
4639
4641
4647
4654
4659
-
-
4689
4698
4045
4517
4632
4771
4772
-
4845
4904
4116
4123
4688
4701
4713
4765
4799
4808
4900
4067
4746
4779
4866
4651
0.20
1.71*
-
-
-
-
-
1.60
-
2.28*
-
0.17*
-
-
-
-
-
-
0.71*
0.51*
-
-
-
-
-
-
0.54*
0.28*
0.31*
-
1.05*
-
-
0.42*
1.17*
0.22
0.97*
0.20
-
-
2.00*
-
0.88
-
1.71
0.08*
0.65*
-
-
0.42
-
1.25*
0.14*
2.05*
1.51*
-
-
4.85*
2.14*
1.18*
-
-
0.11
1.31*
8.34
0.20
-
0.60*
-
-
1.28
-
1.05*
0.45*
-
0.31
-
0.20
-
1.91*
1.48*
-
-
1.28
0.85*
-
-
-
0.77
1.69
-
0.51*
0.26
-
0.88*
1.42
2.60*
-
0.68*
0.14
-
0.40
-
-
-
Sm/BCD
Sdm
Sab
Sc
Sa
Sc
Sa
Sb
Sm/BCD
Sc
Sa
Sa
Scd
Sbc
Scd
Sc
Sb
S/BCD
Sc
Sc
Sa
Sc
BCD
Sc
Sa
Sa
Sc
Sc
Sc
Sa
Scd
Sb
Sc
Scd
Sc
Sc
Sc
Sc
Scd
Sd
Sc
Sc
Sb
Sd
Sc
Sa
Sc