The nuclear activity of late-type galaxies in the Virgo Cluster

The nuclear activity of late-type galaxies in the Virgo Cluster

Roberto Decarli

Università degli Studi di Milano – BicoccaDipartimento di Fisica

Santiago, September, 6th, 2007

Università degli Studi dell’InsubriaDipartimento di Fisica e Matematica

G. GavazziM. ColpiI. Arosio C. BonfantiL. CorteseA. Boselli

The sample• The Virgo Cluster:

– is near us– is rich– is widely studied

• We select all spiral galaxies in the Virgo Cluster Catalogue (Binggeli et al., 1985, 1993) with mph ≤15.

• Among the complete sample of 237galaxies: • 213 (90%) have a

spectroscopic classification fortheir nuclear activity

• 216 (91%) have NIR imaging• 199 (84%) have r-band imaging

The AGN unified model• The SMBH influence radius is

defined as the radius at which potential energies of the galaxy and of the massive BH are equal:

• The line emission near the SMBH is Doppler broadened because of the gas fast motion (>1000 km/s!) due to the SMBH potential well (Broad Line Region)

• Lines produced outside Rinf are sensitive only of the galactic potential well (Narrow Line Region)

2BH

inf

GMR

Line flux ratios and nuclear activity• Gas emission line fluxes depend on gas

temperature, density, metallicity and on the intensity and energy distribution of the incident radiation field

• Comparing line intensities one can infer information about the ionization source and its strength

VCC 664

Stasinska, 2006

HII-regions

AGNs

Spectroscopic datasets• Fluxes published in Ho, Filippenko & Sargent (1997) – 40 (40)• SDSS spectra – 84 (73)• Nuclear spectra – 29 (13)• Modified drift scan spectra – 22 (4)• Drift scan spectra– 193 (81)• Nuclear activity classification in NED – 41 (2)

February-March, 2005-2006 observing runs in Loiano

+

GOLDMine database (http://goldmine.mib.infn.it)

SDSS fiber

Slit

Nuclear spectra

Drift scan spectra

Modified drift scan spectra

[NII] /H and [OIII] /H

• AGNs have [NII]/H > 0.6

• [OIII]/H > 3 when a strong ionization field is present (Seyfert-like AGN or intensely star-forming region)

• 0.4 < [NII] /H < 0.6 for Transition objects

• The [NII]/Hratio provides a nuclear activity classification without any further information

56 AGNs (23.6%):- 9 Seyfert (3.8%)- 23 Seyfert/LINERs (9.7%)- 24 LINERs (10.1%)

21 Transition objects (8.9%)

121 HII-Regions (51.1%)

15 No Emission Line Galaxies (6.3%)

24 No spectra available (10.1%)

Dynamical mass andH-band luminosity

• Mdyn = dynamical mass up to the optical radius of the galaxy

• Direct measures of Mdyn areobservational time consuming

• For spiral galaxies:

(Gavazzi et al., 1996)

• We use H luminosity as a surrogate of Mdyn. NIR datawere taken from 2MASS and from GOLDMine databases

L6.4

MHLM dyn

Adapted from Gavazzi et al. (1996)

Mass distribution• Only VCC 213 has Mdyn<1010 Mo and hosts

an AGN

• AGN fraction changes from 0% up to 100% when Mdyn changes between ∽1010 Mo and ∽3·1011 Mo

(see also Kauffmann et al., 2003)

• Mdyn > 1010 Mo is a necessary condition for harbouring an AGN

• Mdyn > 3·1011 Mo is also a sufficient condition

The colour-magnitude diagram

AGNs populate the high-mass end of the

same colour-magnitude relation as the one observed for

inactive galaxies!

• The AGN host galaxies are usually found to be redder than non-active galaxies.

• Some authors (e.g., Obric et al., 2006) suggested that the AGN feedback may be responsible of this effect.

- Bulge dominated galaxies have C31>4

- Pure disks have C31∽2.5

- When Mdyn> 1011 Mo, both bulge- and disk-dominated spiral galaxies are found (see Gavazzi Pierini & Boselli, 1996; Scodeggio et al., 2002)

The role of bulges

• From NIR light profile we define:

AGNs are found in all the galaxies having prominent bulges,

but even in bulgeless galaxies!

%25

%7531 r

rC

Bulge-dominated

spiral galaxies

Disk-dominated spiral galaxies

M100

NGC4535

M61

Light cusps

- Sensitive to unresolved nuclei and light cusps

- Nuc increases continuously with Mdyn

),()5.1( %50%25 rrrNuc seeing

AGNs are found mainly at highvalues of Nuc

Are r-band nuclei the due to AGN continuum?

• Nuclear spectra of 4 AGNs in our sample were observed with HST (Spinelli et al., 2006). All of them are well fitted with Bruzual & Charlot libraries of star spectra.

• AGN continuum light is thus negligible with respect to the stellar component.

VCC 1110 - Seyfert VCC 2070 - LINER

VCC 1690 - LINER

VCC 1401 - Seyfert

Environment – I• We divide our sample in two parts, according to

the angular separation of the galaxies from M87:– High density subsample:

114 galaxies, 37 AGNs (32±8 %)– “Low” density subsample:

99 galaxies, 40 AGNs (40±10 %)

• The mass distribution of the subsamples are similar.

• The observed AGN fractions are consistent.

Environment – II

• We check the [NII]/H dependence on the neutral gas deficiency (Haynes & Giovanelli, 1984).

• Even considering the HI deficiency no clear environmental effect is observed, once the luminosity dependence is taken into account.

The environment does not significantly influence the

nuclear activity

What can we argue about the BH-host galaxy joint evolution?

• At least in spiral galaxies, the nuclear activity is strongly sensitive to the host galaxy mass, while the morphology plays a secondary role.

• The environment is not relevant in nuclear activity.• There is no evidence of a strong feedback mechanism in the

galaxy colours.• Low accretion rates are observed.

• The massive BH may have grown simultaneously with the galaxy, in an anti-hierarchical way:

• The dynamical mass is the driver of both the galaxy and the BH growth, in a top-down scenario

• The scale relations inset during the formation itself of the system

• Galaxies do not require any strong feedback later on, neither strong accretion rates

• The “fragile” morphology of pure disk and bulgeless galaxies is preserved, since the BH is already formed

Bibliography• Decarli, Gavazzi, Arosio, Cortese, Boselli, Bonfanti, Colpi, 2007, arXiv:0707.0999,

accepted for publication in MNRAS• Ferrarese, 2006, in Series in High Energy Physics, Cosmology and Gravitation, `Joint Evolution

of Black Holes and Galaxies', ed. by M. Colpi, V. Gorini, F. Haardt, U. Moschella (New York - London: Taylor & Francis Group), 1

• Gavazzi, Pierini & Boselli, 1996, A&A, 312, 397• Gavazzi, Boselli, Scodeggio, Pierini & Belsole, 1999, MNRAS, 304, 595• Gavazzi, Boselli, Donati, Franzetti & Scodeggio, 2003, A&A, 400, 451• Haynes & Giovanelli, 1984, AJ, 89, 6• Kauffmann et al., 2003, MNRAS, 346, 1055• Kauffmann & Heckman, 2005, RSPTA, 363, 621• Lequeux et al., 1979, A&A, 80, 155• Marconi & Hunt, 2003, ApJ, 589, L21• Obric, et al., 2006, MNRAS, 370, 1677• Scodeggio, Gavazzi, Franzetti, Boselli, Zibetti, Pierini, 2002, A&A, 384, 812• Veilleux, et al., 2003, AJ, 126, 2185

Correcting for line absorptions

• Balmer lines are usually observed both in emission and in absorption

• When possible, both the emission and absorption features are fitted. Otherwise, a correction is applied assuming various templates according to the galaxy luminosity.

Spectroscopic datasets• Fluxes published in Ho, Filippenko & Sargent (1997) – 40 (40)• SDSS spectra – 84 (73)• Nuclear spectra – 29 (13)• Modified drift scan spectra – 22 (4)• Drift scan spectra– 193 (81)• Nuclear activity classification in NED – 41 (2)

February-March, 2005-2006 observing runs in Loiano

+

GOLDMine database (http://goldmine.mib.infn.it)

SDSS fiber

Slit

Nuclear spectra

Drift scan spectra

Modified drift scan spectra

The [NII]/H ratio along the slit

The Narrow Line Region size• For a subsample of AGNs, we extract the [NII]/H ratio at

various apertures, up to some kpc

• Only in one galaxy (VCC 73) the ratio significantly descreases in the first kpc

Veilleux et al., 2003

• Thus, the photoionization of an AGN extends up some kpc!

Mass dependence

• The [NII]/H ratio increases with the dynamical mass of the host galaxy

• A bimodality is observed:– In low-mass host galaxies, the

[NII]/H ratio is a metallicity indicator:

– At higher masses, a further contribution to the [NII]/H ratio is provided by the AGN

2.5M

log43.0Hα

[NII]log

dynM

metallicity

AG

N

Black hole masses

• Following Marconi & Hunt (2003), the MBH of our sample AGNs is estimated from the bulge NIR luminosity:

• Only few tens ‘direct’ supermassive BH mass measures are available, mostly in elliptical galaxies (see Ferrarese, 2006 for a review)

)07.019.8(L10

log)15.025.1(M

log8.10

bulge,H

LM BH

• All the BHs in our sample have MBH>105.8 Mo.

The accretrion rate L/LEdd

BHT

pEdd M

cmGL

4

• The accretion rate is usually identified with the ratio between the AGN bolometric luminosity and the BH Eddington luminosity:

• The AGN bolometric luminosity is roughly proportional to the [OIII] luminosity (Heckman et al., 2004).

• We thus have an order-of-magnitude estimate of L/LEdd for our objects.

Most of our objects have very low accretion rates

Summarizing…1) Mdyn>1010 Mo is a necessary condition to harbour an AGN

2) The AGN fraction steeply rises with the dynamical mass, and it is virtually 1 when Mdyn>3·1011 Mo

3) AGNs are found even in a number of bulgeless galaxies

4) Nuclear stellar cusps are often found in AGN host galaxies

5) The same colour-magnitude relation describes both active and inactive galaxies

6) No significant environmental dependence of the nuclear activity is observed

7) All the AGNs in our sample have MBH>105.8 Mo

8) Very low accretion rates are found in almost all our targets