ACTIVE GALACTIC NUCLEI

1
ACTIVE GALACTIC NUCLEI

• Optical spectral classification and Luminosity
  Function

2
Introduction and some caveats

• Sy1/QSO
• QSO/quasar
• NLRG/QSO2/Sy2
• RL QSO/RQ QSO
• Point-like/extended
• Active galactic nuclei (AGN) are a class of
  galaxies where a significant fraction of the
  energy output, emerging from their centers, is
  not produced by the normal galaxy components
  stars, dust and interstellar gas. This energy can
  be emitted across the whole electromagnetic
  spectrum, from radio waves to gamma rays

Reducing the AGN zoo as much as possible !
3
The UV/optical/NIR spectrum

• Power-law emitted by an highly compact
  non-thermal source (power law)
• Big Blue Bump this component possibly comes
  from the BH accretion disk (black body)

4
The UV/optical AGN spectrum

• Power-law
• Big Blue Bump
• Small Blue Bump FeIIBalmer Continuum
• Broad emission lines FWHM gt 1500 Km/s
• Narrow emission lines FWHM lt 900 Km/s
• The emission lines characterize the AGN spectra
  they are produced in two separate regions, a
  subarcsec Broad Line Region (BLR) close to the
  central engine, and a more extended Narrow Line
  Region (NLR).
• Galaxy Starlight usually overwhelmed by the
  AGN non-thermal continuum and emerging in the red
  part of the spectrum

5
The UV/optical emission line spectrum
6
What we learn from theAGN UV/optical spectra

• Redshift
• Classification (optical) Type 1/Type 2
• AGN Unification evidences
• Physical state of the emitting-line gas
• AGN/StB separation Diagnostic Diagrams
• Unconventional AGNs

7
Redshift

• Discovery of the true nature of quasar
• Cosmological Distance
• Absolute physical quantities (L,M,size)
• Only the optical spectrum (sometimes UV, near-IR)
  gives z
• Easy way to measure z

8
In 1963 Schmidt identifies highly redshifted
Balmer lines in 3C273s spectrum
z 0.158 (vr 47500 km/s)
9
Redshift

• Discovery of the true nature of quasar
• Cosmological Distance
• Absolute physical quantities (L,M,size)
• Only the optical spectrum (sometimes UV, near-IR)
  gives z
• Easy way to measure z

10
AGN emission line spectra

• Type 1 AGN are those with very broad optical/UV
  permitted emission lines, with FWHM1500-15000
  km/s, while the forbidden lines, like OII3727,
  OIII4959/5007, NII6548/6583, typically have
  FWHMs of order of 500-1000 km/s.
• Type 2 AGN have permitted and forbidden lines
  with approximately the same FWHM, similar to the
  FWHMs of the forbidden lines in Type 1 objects.
• The forbidden lines, while narrower than the
  permitted ones, are usually broader than the
  emission lines in most starburst galaxies.

Optical Type 1
Optical Type 2
11
AGN Unification the standard model
Components Accretion Disk r 10-3 pc
n 1015 cm-3 v 0.3c Broad
Line Region r 0.01-0.1 pc n 1010 cm-3
v few x 103 km s-1 Dusty Torus
r 1-100 pc n 103 -106 cm-3 Narrow
Line Region r 100-1000 pc n 102 -104 cm-3 v
few x 102 km s-1
12
AGN Unification the paradigm
... Much of the variety in AGN types is just the
result of varying orientation relative to the
line of sight. ... We can define an extreme
hypothesis in which there are only two basic AGN
types the radio quiet and radio loud.
(Antonucci 1993)
XBONG
13
AGN Unification a milestone
A classical Seyfert2 galaxy (NGC1068) observed in
polarized light showed a Sy1-like polarized
spectrum, a featureless continuum with high
polarization and position angle perpendicular to
the radio jet. The observed properties could be
explained by reflection of an hidden BLR into the
line of sight, the scattering source being
composed of hot electrons rather than dust grains.
14
BLR vs. NLR

• Photo-ionized by nuclear continuum radiation
• ne 1010 cm-3
• Te few x 103 K
• rBLR lt 0.1 pc (but RM)
• mBLR few x M?
• e 10-2
• Complex/dissimilar line prof.
• Doppler line widths ? ? central object
  gravity

• Photo-ionized by nuclear continuum radiation
• ne 103 cm-3
• Te 1-3 x 104 K
• rNLR gt 102 pc (measured)
• mNLR few x 104M?
• e 10-3-10-4
• Simple/similar line profiles
• Doppler line widths ? ? stellar bulge gravity

15
Spectral classification (em.lines)

• Type 1
• Type 2
• Starforming galaxies

16
Spectral classification (em.lines)

• Type 1
• Type 2
• Starforming galaxies
• How we can discriminate the narrow line objects?

17
Spectral classification (em.lines)

• DIAGNOSTIC DIAGRAM
• Sy 1 galaxies
• Sy 2 galaxies
• Starforming galaxies
• Composite galaxies

SDSS
18
Diagnostic Diagrams
SDSS

• Sy2/StB theoretical separation Kewley et al.
  (1991)

19
Diagnostic Diagrams
SDSS

• Sy2/StB empirical separation Kauffmann et al.
  (1994).

20
Diagnostic Diagrams
The BPT diagrams (Balwin, Phillips Terlevich
1981, Veilleux Ostrebrock 1987), are used in
narrow-line emission systems, to distinguish
between the origins of the photo-ionization, hard
and soft radiation, which is usually ascribed to
non-stellar and stellar activity, respectively.
The general criterium O III / Hß gt 3 could be
wrong !
Shock-heated
Power-law
Sey2
Planetary nebulae
LINERs
H II gal
H II galaxies
(BPT 1981)



(Peterson 1997)
21
Diagnostic Diagrams

• The ten commandments of emission-line diagnosis
  (Veilleux01)
• Thou shalt use lines which emphasize the
  differences between H II regions and AGN i.e.,
  use high-ionization lines or low-ionization lines
  produced in the partially ionized zone.
• Thou shalt use strong lines which are easy to
  measure in typical spectra.
• Thou shalt avoid lines which are badly blended
  with other emission or absorption line features.
• Thou shalt use lines with small wavelength
  separation to minimize sensitivity to reddening.
• Thou shalt use lines from the same elements or
  involving hydrogen recombination lines to
  eliminate or reduce abundance dependence.
• Thou shalt avoid lines from Mg, Si, Ca, Fe
  depleted onto dust grains.
• Thou shalt use lines easily accessible to current
  UV/optical/IR detectors.
• Thou shalt avoid lines affected by strong stellar
  absorption features.
• Thou shalt avoid lines affected by strong
  atmospheric features.
• Thou shalt use lines at long wavelengths to
  reduce the effects of dust extinction.

22
Diagnostic Diagrams

• DIAGNOSTIC DIAGRAMS
• Em.lines in the red range
• NIII6584/Ha vs. OIII5007/Hß
• SII6717-31/Ha vs. OIII5007/Hß
• OI6300/Ha vs. OIII5007/Hß
• Em.lines in the blue range
• OII3727/Hß vs. OIII5007/Hß
• OII3727/Hß vs. continuum index
• Other line ratios in UV, NIR and FIR spectral
  ranges
• NV1240/Lya, NV1240/HeII1640, CIV /Lya
• Si VI 1.962µm/Paa
• NeV14µm/NeII12.8µm,
• OIV26µm/NeII12.8µm,
• EW(PAH 7.7µm)

23
AGN taxonomy LINERs
LINER Low-Ionization Narrow-Line Region They
are characterized by O II ?3727Å / O III
?5007Å 1
O I ?6300Å / O III ?5007Å
1/3 Most of the nuclei of nearby galaxies are
LINERs. A census of the brightest 250 galaxies in
the nearby Universe shows that 5075 of giant
galaxies have some weak LINER activity (Phillips
et al. 1986, Ho, Filippenko Sargent 1993, ).
They are the weakest form of activity in the AGN
zoo.
(Heckman 1980)
24
AGN taxonomy BAL QSOs
BAL QSOs Broad Absorption Line QSOs Otherwise
normal QSOs that show deep broad absorption
lines, blueward of the corresponding emission
resonance lines of CIV, SiIV, NV. The
interpre-tation is that they are intrinsic and
arise from clouds outflowing the nucleus. They
mainly are at z 1.5 because the phenomenon is
observed in the rest-frame UV. At these
redshifts, they are 10 of the observed
population.
Mean QSO spectrum PG 0946301.
Arav et al. (1999)
25
AGN taxonomy BL Lacs Blazars

• BL Lacertae is the prototype of this class an
  object, stellar in appearance, with very weak
  emission lines and variable, intense and highly
  polarized continuum. The weak lines often just
  appear in the most quiescent stages. BL Lacs,
  along with optically violent-variable (OVV) QSOs,
  constitute the class of Blazars these are
  believed to be objects with a strong
  relativistically beamed jet in the line of sight.

26
AGN taxonomy XBONGs
XBONGs X-ray Bright Optically Normal
Galaxies This AGN class consists of luminous hard
X-ray sources hosted by "normal" galaxies with
optical spectra typical of early-type systems
(Comastri et al. 2002). Why the relatively
bright X-ray emission, typical of moderately
luminous (1042-43 erg s-1) Active Galactic Nuclei
, does not leave any optical signature of the
presence of a nuclear source is still matter of
debate.
27
AGN taxonomy XBONGs

• The upper limits on the optical emission lines
  (OIII, Ha), expected from the nuclear activity,
  in some XBONGs are tight enough to place these
  sources outside the typical AGN properties.
• Possible interpretations
• Dilution from the host galaxy light
• Radiatively inefficient accretion flow
• Heavy obscuration by Compton-thick nuclear gas
• NLR obscured on galaxy scale (i.e., Kpc dust
  lanes, see Malkan et al. 1998, Rigby et al. 2006)
• Extreme BL Lacs objects

Hellas2XMM
28
AGN taxonomy the type-2 QSOs

• Definition
• Quasar (high L) analogue of Sy2 galaxies
• OPTICAL high bolometric luminosity (? high z)
  objects with high ionization, narrow (FWHMlt1500
  km/s) emission lines, no broad lines. (expected
  according to the Unification models of AGN)
• X-RAYS high-luminosity (Lx gt 1044 erg/cm/s2)
  and obscured (NHgt1022 cm-2) AGN
  (required by XRB synthesis models).
• Do obscured (type II) quasars exist? To date,
  only a small number of candidates have been found
  (but NLRG). First examples did not confirm their
  classification after follow-up observations (high
  S/N and/or redward).

29
AGN taxonomy the type-2 QSOs

• Without the low order Balmer lines, a secure
  type2 classification is a moot point.

Boyle et al. (1999)
Type 2
Halpern et al. (1999)
Type 1
30
AGN taxonomy the type-2 QSOs
Norman et al. (2002)
Stern et al. (2002)
Martinez-Sansigre et al. (2005)

• The high-z type-2 QSOs do exist! Selected both in
  optical, X-rays and FIR. But they are demanding
  targets for optical spectroscopy. (but see SDSS)

31
The redshift desert for type-2 QSOs
High-z range
Low-z range
32
Optical spectra of different AGN types
33
Points to Take Away (1)

• The optical spectrum offers a wealth of
  information about BLR/NLR and the AGN structure,
  but it is mainly composed by secondary
  radiations.
• The unification scheme is a clear and simple way
  to classify AGN, but dont tell us all the truth.
• ty1/ty2 ? f(L),f(z),f(?),f(env),f(host-type),
  f(t)
• Multi-wavelength approach to AGN study

34
(No Transcript)
35
The Luminosity Function

• To study galaxy/AGN evolution we need to compare
  extragalactic objects today with the same kind of
  objects in the past
• To perform a fair comparison we need to compare
  the emitted power at the same wavelengths, hence
  K-corrections k(z)
• We cannot observe the same galaxy at different
  times, so we must look at the statistical
  properties as populations, hence we study
  Luminosity Functions (LF)

36
The Luminosity Function

• The luminosity function characterises the number
  density of (active) galaxies as a function of the
  luminosity L
• Luminosity function, usually written ?(L),
  is defined as the co-moving number density
  (number of objects per co-moving volume) in some
  luminosity range (usually logarithmic)

37
The Luminosity Function

• In order to allow easy comparison between
  different determinations and different types of
  LF it is usual to use simple parameterisations
• Schecter ? common for galaxies
• Two-Power-Law ? common for AGN
• Power-Law with exponential cut-off ? infrared
  galaxies

38
LF Schecter Function (for galaxies)
Exponential Cut-off
39
Parametric Evolution of Luminosity Functions
Pure density evolution
Density Evolution
Pure Luminosity evolution
Luminosity Evolution
40
Parametric Evolution of Luminosity Functions
41
Luminosity Function methods of calculation

• LF is just the number density of galaxies
• so just have to count the number of galaxies...
• and divide by the observable volume of each
  one.
• simple but ....
• Only complication is the magnitude limits of the
  survey
• 1/Vmax (move the galaxy)
• Move the galaxy forwards and backwards in
  redshift, to find the maximum volume it could
  have inhabited and still been observed.
• Maximum likelihood (change its brightness)
• At a fixed redshift, what is the range of
  luminosities that a galaxy could have - and still
  be observed need to assume a form for the LF.

42
The QSO LF goals

• The optical luminosity functions of quasars
  (OLF), as well as different types of AGN, hold
  important clues about the demographics of the AGN
  population, providing strong constraints on
  physical models and evolutionary theories of AGN.
• The QSO OLF at high redshifts provides important
  constraints on the ionizing UV radiation field of
  the early universe.
• The faint end of the QSO LF has not been measured
  at high redshift until now. Instead, low-z
  measurements of the faint end were combined with
  high-z measurements of the bright end to estimate
  the entire LF at high z.

• The luminosity function of quasars is one of the
  principal constraints on the accretion history of
  the most massive black holes

43
The QSO LF goals

• Derive the density of AGNs as function of
  bolometric luminosity, redshift
• ?(Lbol z type)
• Relates to
• Characterizing accretion history
• Distribution functions of black hole activity as
  function of MBH, accrection rate and radiative
  efficiency and redshift
• Probing galaxy/BH coevolution
• Test unification model

44
The QSO LF basic issues

• Instead of ?(Lbol z type), we observe
• N(f?, z, AGN type, selection criteria)
• Selection effect
• Incompleteness due to selection criteria
  (correctable)
• Selection bias (e.g., optical survey missing
  obscured sources)
• Bolometric correction
• Redshift effect
• Flux-limited vs. volume limited, truncated data
  set
• Limited luminosity range at any given redshift
• K-correction

45
Accretion history of AGNs
Quasar space density as a function of redshift.
46
Accretion history and luminosity function of AGNs

47
Black hole mass density and the OLF of AGNs

Local black hole mass density mainly comes from
the accretion during bright AGN phases.
48
Formation history of the SMBHs mass density

• Total accreted mass (hence BH mass density) can
  be estimated by LF assuming the mass-to-energy
  conversion factor e(Soltan 1982).
• Compare with a local BH mass density estimated
  from the relation between BH mass and velocity
  dispersion to constrain e.
• Previous studies
• Fabian and Iwasawa (1999) , Elvis et al (2002)
• use the XRB intensity assuming z2 for all
  the XRB source
• Yu and Tremaine (2002)
• use the QSO optical LF from 2dF survey
• Marconi et al. (2004)
• use both the XRB and SDDS data (galaxies
  quasars)

49
Lifetime of bright AGNs
50
The scenario for AGN formation and evolution
51
The QSO LF Parameterization

• Quasar LF double power-law
• Luminosity-dependent density evolution (Schmidt
  and Green 1983)
• ?(L,z) ?(L,z) ?(L,z0)
• Overall density evolves
• Shape (bright and faint end slopes) evolves as
  well

52
QSO Luminosity Function from 2dF Quasar Survey
www.2dfquasar.org
53
3 Lya 2
CIV
CIII

MgII 1


OIII 0 4000 Å
observed wavelength
8000 Å
redshift
54
Properties of the 2dF Quasar Survey

• Quasars selected from pointlike sources using
  U-BB-R colours (UVX)
• 0.3ltzlt2.5
• 25000 Blt21 QSOs in final catalogue
• Volume probed 4 x109h-3Mpc3
• www.2dfquasar.org

Croom et al. 2002, MNRAS, 322, L29 Croom et al.
2004, MNRAS, 349, 1397
55
Optical Counts from 2dF Quasar Survey
56
Optical Luminosity Function from 2dF Quasar Survey
Boyle et al. 2001
57
Optical Luminosity Function from 2dF Quasar Survey

• Best fit model pure luminosity evolution (PLE)
• ? cosmic look-back time ? 6 ? -3.3 ?
  -1.0

M constant apparent mag - Selection
effect?? Faint end slope poorly determined
58
PLE
PDE
59
SDSS Quasar Survey
QSOs selected from imaging in 5 wavebands u g r
i z Multi-colour selection ? Sensitive to QSOs at
high redshift (zlt6.5) Currently 50000 QSOs in
DR3 ilt19 (main sample) ilt20 (high-z sample)
NGP SGP
Schneider et al. 2003, AJ, 126, 2579 www.sdss.org
60

46,420 Quasars from the SDSS Data Release Three
5
Ly? forest
3
Ly?
2
CIV
redshift
CIII
MgII
FeII
1
FeII
H?
OIII
0
wavelength
4000 A
9000 A
61
Hubble diagram for SDSS
62
SDSS quasar OLF
63
SDSS quasar OLF evolution at high redshift ?

• The SDSS counts and LF agree with the results of
  the 2QZ at redshifts and luminosities at which
  the two samples overlap, but the SDSS data probe
  to much higher redshifts than does the 2QZ
  sample.
• Strong evolution in bright end slope at zgt3
  (cant be PLE all the way) the slope flattens
  with redshift.
• But SDSS doesnt go faint enough at low-z to
  differentiate PLE from PDE or else.....

64
COMBO-17
Due to the relatively bright magnitude limits of
the SDSS and 2QZ surveys, the LF analysis is
restricted to relatively bright QSOs especially
at high redshift. What about fainter QSOs? The
COMBO-17 survey, using multi-band photometry in
17 filters within 350 nm lt obs lt 930 nm,
simultaneously determine photometric redshifts
with an accuracy of ?z lt 0,03 and obtain spectral
energy distributions. Photometric selection of
192 1.2ltzlt4.8 QSOs, reaching R24
Wolf et al. (2003)
65
COMBO-17 OLF at zgt1

• The evolving LF can be adequately described by
  either PLE (red dashed line) or PDE (black solid
  line) largely due to the absence of an obvious
  break.

66
Faint end of the quasar LF ? fainter qso surveys
2dF-SDSS QSO survey (2SLAQ)
VVDS type-1 AGN sample

• SDSS photometry 2dF spectroscopy
• 1m deeper than 2QZ
• Multicolor selection
• 5645 qso in 100deg2
• 2SLAQ counts and LF are steeper in the faint end
  (34 more faint quasars)
• LF PLE evolution of a double power-law but no
  well-defined break (PDE not rejected)

• Broad Line objects in the pure magnitude selected
  VVDS
• 130 AGN down to I 24
• Bongiorno et al. (2007) merge VVDS SDSS samples
  to find Luminosity Dependent Density Evolution
  (LDDE ) model for the OLF of broad line AGN.
• The peak of AGN space density shifts to lower
  redshifts for the lower luminosity objects (AGN
  cosmic downsizing) see X-rays !!

67
Unaccounted AGNs
Cosmic X-ray Background

• Significant obscured population
• e.g. Seyfert 2s, Narrow line radio galaxies.
  How many type 2 QSOs?
• Low luminosity AGNs
• e.g. Seyfert 1s, LINERs.
• host dilution, lack of all emission lines for
  diagnostic tests (z gt 0.5)

NH 1020 cm-2 NH 1024 cm-2
Gilli, Comastri Hasinger 2006
Do they contribute significantly to the
cumulative emissivity of AGNs? Yu and Tremaine
(2002) match local BH mass density using the OLF
of QSOs.
68
AGN X-ray LF motivation

• X-rays AGN Luminosity Function
• a goal of X-ray surveys
• the most fundamental measure to understand the
  cosmological evolution of AGNs
• How many AGNs in the universe (as a function of
  time)?
• How supermassive black holes form?

see Roberto Gilli Lesson
69
AGN X-ray LF

• Quasar density peaks at z2-3
• Low-L AGN density peaks at z0.5 - 1
• Paradox 1
• Most of BH accretion happens in quasars at high-z
• Most of X-ray background in Seyfert 2s at low-z

70
The AGN Number Density as a Function of Redshift

• Luminous AGNs has a peak (cutoff redshift)
  earlier than less luminous AGNs
• Miyaji et al (2006)
• Fiore et al (2003)
• Related to the star forming activity?
• e.g., Francheschini et al. (1999)
• The evoluiton of the number density of high (low)
  luminosity AGNs is simlar to that of the star
  forming rate of early (late) type galaxies.
• A strong link between formation of the SMBH and
  that of the spheroid component of galaxies.

(1z) 4 (zgtzc)
71
AGN X-ray LF
Miyaji et al. 2006

• PLE doesnt work need luminosity-dependent
  density evolution (LDDE) to characterize
  evolution of the XLF

72
The 0.52 keV luminosity function for type 1 AGN
Hasinger, Miyaji Schmidt (2005).
73
The HXLF of All the Compton-thin AGNs (Type-I
Type-II)

• Best described with a luminosity dependent
  density evolution (LDDE) where the cut-off
  redshift increases with the luminosity

74
AGN X-ray LF

• Type 2 fraction a strong function of luminosity
• Paradox 2
• At high (quasar) luminosity type 2 lt20 optical
  color selection is highly complete since almost
  all are type 1s, and includes most of luminosity
  AGN population emitted in the Universe.
• At low (Seyfert) luminosity type 2 80 optical
  color selection miss most of the AGNs in the
  Universe in terms of number .

75
What do the optical surveys tell us about quasar
LF ?

• The LF is a powerful tool for the study of AGN
  population (demography and structure)
• LF determination bright end/high-z needs large
  survey area (rare objects) faint end /high-z
  needs deep surveys (faint objects). Requisite to
  populate the bins ? large N surveys. NO single
  survey
• Survey e f ( band z AGN type selection
  criteria ) ? Multiwavelength approach to AGN
  study (again)
• For some applications (e.g. BH mass density) the
  OLF of bright quasar is adequate.