SIGRAV Graduate School in Contemporary

1
Laura Ferrarese Rutgers University Lecture 6 The
Future

• SIGRAV Graduate School in Contemporary
• Relativity and Gravitational Physics

2
Lecture Outline

• A Recap of the Observational Status of SBH
  Research
• Open Questions What Remains to be Done
• Building the Local Sample
• The Redshift Evolution of SBH Scaling Relations
• The Structure of the Broad Line Region
• Black Holes and Globular Clusters
• Binary Supermassive Black Holes.

3
Estimating SBH Masses
Primary Methods
Fundamental Empirical Relationships
Secondary Mass Indicators
Low-z AGNs
4
What We Have Learned, and Open Questions
1. SBHs are fundamental constituents of galaxies
the local SBH mass density is equal to what is
needed to explain the energetics of high redshift
QSOs (Merritt Ferrarese 2001 Ferrarese 2002
Yu Tremaine 2002). 2. The existence of tight
relations between SBHs masses and the large scale
properties of their host galaxies suggests that
the formation and evolution of SBHs and their
hosts must go hand in hand. Understanding how
SBHs form might help us to understand how
galaxies form/evolve (or viceversa).
5
Biases and Systematics
Systematics in the M?? relation (or any other SBH
scaling relation!) have not been fully
investigated ? Slope, zero point
scatter ? 109 M ? regimes
? Dependence on Hubble type ? Dependence on
galaxy environment ? Cosmic evolution
? Reliability of SBH mass measurements
SPIRALS
LENTICULARS
ELLIPTICALS
6
Biases in the Current Mass Estimates

• In the current sample there is only one galaxy
  for which a SBH mass estimate has been obtained
  using two independent methods (IC1459 Verdoes
  Kleijn et al. 2000, Cappellari et al. 2002)
• MBH(stars) (2.6 ?1.1) ? 109 M? (using 3I
  modeling of HST/STIS data with N0/Nc2.0)
• MBH(gas) (0.4 ?1.0) ? 109 M? (depending on
  the assumptions made for the gas velocity field)
• There is only one galaxy for which the same data
  has been analyzed by two different teams using
  the same method (3-I modeling using different
  codes - M32, van der Marel et al. 1998, ApJ, 493,
  613 Valluri et al. 2003, astro-ph/0210379)
• MBH(vdM) (3.4 ?0.7) ? 106 M?
• MBH(Valluri) (1 ? 6) ? 106 M?
• Finally, some of the data being analyzed might
  not be adequate in terms of spatial resolution
  and signal-to-noise ration

7
Resolving the Sphere of Influence

• The quality of the data might very well influence
  the characterization of the scaling relations
  (remember Magorrian et al. 1998)

Kormendy Gebhardt 2001, Gebhardt et al. 2002
8
Resolving the Sphere of Influence
9
1. Addressing the Faint End of the M-s Relation

• How far does the M-s relation extend?

10
1. Addressing the Faint End of the M-s Relation
M33

• M33 is an ideal target
• very nearby (850 kpc) ? tightest limits on a
  small BH.
• very compact nucleus, reaching a stellar mass
  density of several million solar masses per cubic
  parsec ? ideal conditions for BH formation (?)
• Very low central velocity dispersion ( 24 km/s,
  Kormendy McClure 1993) ? very small black hole
• Inconspicuous bulge ? very small black hole (?)
• The M33 nucleus probably contains the most
  luminous ULX in the Local Group (Long et al.
  1981), strengthening the connection with the M82
  ULXs (Ebisuzaki et al. 2001)
• Tightest limit from ground based data MBH 50,000 M?
• The MBH-s relation predicts 2,600 M?
• ? radius of influence
• ? Only an upper limit can be set on the mass.

HST/STIS/0.1 slit
Merritt, Ferrarese Joseph (2001)
11
1. Addressing the Faint End of the M-s Relation
M33

• The upper limit is still consistent with the M??
  relation.

M33
Gebhardt et al. 2001 Merritt, Ferrarese
Joseph 2001
12
1. Addressing the Faint End of the M-s Relation
NGC205

• On-going HST-ACS/NICMOS/STIS project (Ferrarese,
  Merritt, Valluri, Joseph). Of the galaxies for
  expected to harbor a SBH with MBH 205 is the only one for which the sphere of
  influence can be resolved by HST in a finite
  amount of time.

Andromeda, NGC 205 and M32 1.5 X 2 degrees
13
2. Building the Local Sample What HST Cannot Do

• Stellar kinematics in giant ellipticals, all of
  which have low central surface brightness.

14
2. Building the Local Sample What HST Cannot Do

• Nearby dwarf systems

NGC 147 (Han et al. 1997, AJ)
WFPC2, F555W
15
2. Building the Local Sample

• What if we wanted to, say, increase the current
  sample size to include significantly statistical
  samples of galaxies belonging to all Hubble types
  and disparate environments?
• Below SBH masses for all galaxies belonging to
  the CfA redshift sample (Huchra et al. 1990,
  ApJS, 72, 433), estimated from the bulge
  luminosity, as a function of distance

16
2. Building the Local Sample

• The sample thins out even more if S/N
  requirements are added. The figure shows the
  requirements for a complete sample of elliptical
  galaxies (from Faber 1989) observed with S/N50
  in the absorption lines at 8500 Å , in the
  equivalent of 3 HST orbits.

17
2. Building the Local Sample SBHs in Low Surface
Brightness Galaxies (?)

• Studies of giant LSB galaxies have found a
  50 incidence of low luminosity AGNs
  (Sprayberry et al. 1995, Schombert 1998).
• This is important for two reasons
• LSBs are not accounted for in studies of
  SBH demographic, but they could contribute
  significantly to the local SBH mass
  function.
• LSBs differ from HSBs in morphological
  appearance, past and current SFR, disk
  kinematics, mass to light ratios, gas ratios and
  molecular content. Their SBH mass function is
  certain to shed light on the mechanisms of
  formation and evolution of SBHs. For instance,
  could LSBs be be remnants of the most massive
  QSOs (Silk Rees 1999)?

Schombert 98
18
2. Building the Local Sample SBHs in Low Surface
Brightness Galaxies (?)

• This is a project that could be possibly done
  with a combination of ground based and HST
  observations. For instance
• There are 73 LSB galaxies with z the Impey et al. (1996) catalogue. For all
• 4m-class telescope low resolution spectra to
  determine incidence of nuclear activity. For the
  AGN subsample
• HST/ACS H? images to determine the extent and
  morphology of the ionized gas in the nuclear
  region. For the most promising candidates
• HST/STIS spectra at 0.1 arcsec spatial resolution
  to determine the nuclear gas kinematics and
  constrain the masses of the central SBHs.

19
3. The Redshift Evolution of the MBH-s Relation

• The most distant objects with a direct SBH mass
  measurement are at 100 Mpc. However, studying the
  MBH-s relation as a function of redshift would
  tell us about the incidence of merging and
  accretion on the evolution of SBHs and their host
  galaxies.
• Even using an 8m diffraction limited telescope,
  resolved stellar/gas dynamics can probe 109M?
  SBHs only up to a few hundred Mpc away. However,
  if we restrict ourselves to AGNs, we do not have
  to rely on resolved kinematics.
• Is targeting AGNs feasible? Is it a good idea?
• AGN samples are quite large. For instance, in the
  Veron-Cetty Veron (2001) catalogue there are
• 13 Seyfert 1s with 1
• 3228 northern QSOs with 1.0
• 1653 northern QSOs with 2.0
• In the local Universe, quiescent galaxies and
  AGNs obey the same scaling relation as far as the
  supermassive black holes at the the center are
  concerned. But is this true at high redshift?
  (there is no answer!)

20
3. The Redshift Evolution of the MBH-s Relation

• Quick and dirty approach
• obtain AGN spectra to 1) estimate the BLR radius
  from the monochromatic luminosity at 5100 Å and
  2) the virial velocity from the width of Balmer
  lines.
• The SBH mass follows by combining the BLR size
  (from the continuum luminosity) with the line
  width.
• In applying this method to high redshift AGNs, we
  make two tacit assumptions
• the RBLR-L5100 relation is valid at high
  redshifts
• The RBLR-L5100 extends at large luminosities.

From Kaspi et al. (2000)
21
3. The Redshift Evolution of the MBH-s Relation

• Quick and dirty approach
• The velocity dispersion can be difficult to
  estimate at high redshifts. It has been proposed
  that the OIII5007 FWHM can act as a surrogate
  for ? (Nelson Whittle 1986, ApJ, 465, 96).

From Nelson Whittle (1986)
Left exactly this procedure applied to a sample
of 107 radio quite QSO and Seyfert 1s (Boroson
2003). Notice the large scatter.
22
3. The Redshift Evolution of the MBH-s Relation

• The only attempt at applying this method at high
  redshift was made by the Shields et al. (2002).
  Although the authors conclude that the MBH-s
  relationship is not a strong function of
  redshift The black hole bulge mass relationship
  is roughly obeyed at a time when much of the
  growth of present day black holes lay in the
  future, notice how
• only very large SBH masses are probed at high
  redshift. Therefore the study contains no
  information about the slope of the relation,
  although it does suggest that the relation is
  pinned at the very high mass end.
• This is to be expected the largest SBHs must not
  have changed much since a redshift 3.

MBH107 ? 3?108 M?
MBH109 ? 1010 M?
From Shields et al. 2002
23
3. The Redshift Evolution of the MBH-s Relation

• There is a more direct but much more involved
  approach to this issue
• Reverberation mapping for selected objects
• needs monitoring at 4d to 3m intervals
  (depending on luminosity/expected SBH mass) over
  periods of 1y to 15y.
• Can be carried out with 4m and 8m class
  telescopes.
• Stellar velocity dispersion in the AGN host
• direct measurements have been obtained for QSOs
  up to z0.3 using conventional 4m class telescope
  (Hughes et al. 2000)
• measurements at z1 can be obtained with an 8m
  class telescope equipped with AO in the near IR.
• measurements at z2 are NGST material.
• A project on this scale would likely require
  international collaboration over at least a
  decade, and is not currently being undertaken.
• Furthermore, the reliability of reverberation
  mapping as a mass estimator ultimately lies in
  probing the the structure of BLR. Are we doing
  anything on this front? (no, but we should be!)

24
4. The Structure of the Broad Line Region

• Based on the experience accumulated so far,
  accurate mapping of the BLR requires a number of
  characteristics
• High time resolution (? 0.2 day)
• Long duration (several months)
• Moderate spectral resolution (? 600 km s-1)
• High homogeneity and signal-to-noise (100)

25
4. The Structure of the BLR KRONOS
Ground-based optical image
26
5. IBH in Globular Cluster Proper Motion Studies

• Confusion in minimized in the U-band (or B-band)
  ? NGST is not ideal
• Observing Constraints

Alcaino et al. 1998, AJ
27
5. IBH in Globular Cluster Proper Motion Studies
28
5. IBH in Globular Cluster But...

• In this case, going to a larger aperture and
  having better resolution only helps up to a
  point! The limit is imposed by the actual number
  of stars in the critical inner region.
• For clusters within 40 kpc, this limit is reached
  for an 8m class telescope.

Paresce de Marchi, 2000, ApJ
29
5. IBH in Globular Cluster But...
M15
NGC 1904
30m
30m
8m
8m
HST
HST
30
5. IBH in Globular Cluster Proper Motion Studies
31
6. Detecting Binary Black Holes

• Milosavljevic Merritt 2001 (ApJ) have conducted
  state of the art N-body simulations aimed at
  following the dynamical evolution of a
  supermassive black hole binary and its
  surrounding stellar system following galaxy
  merging.

32
6. Detecting Binary Black Holes
Separation between the SBHs 0.18 pc (109 M? SBHs)
Difference in Rotational Velocity 95 km/s (109
M? SBHs)
Difference in Velocity Dispersion 136 km/s (109
M? SBHs)
From Milosavljevic Merritt 2001, ApJ
Thick Line SBH Binary Thin Line Single SBH
33
6. Detecting Binary Black Holes
HST
8m
30m
34
Summary
Project Spatial Resol. Aperture FOV
Bandpass l/Dl Comments
35
Suggested Readings

• Thoughts about future work Ferrarese, L. 2002,
  in Hubble's Science Legacy
• Future Optical-Ultraviolet Astronomy from
  Space,
• astroph/0207050
• Peterson 2002, in Hubble's Science Legacy
  Future
• Optical-Ultraviolet Astronomy from Space,
  astroph/0208066