Black Hole Masses from Reverberation Mapping

1
Black Hole Masses fromReverberation Mapping
Bradley M. Peterson The Ohio State University

• Collaborators M. Bentz, S. Collin, K. Dasyra,
  K. Denney,
• L. Ferrarese, K. Horne, S. Kaspi, T. Kawaguchi,
  C. Kuehn,
• D. Maoz, K. Metzroth, T. Minezaki, H. Netzer,
  C.A. Onken,
• R.W. Pogge, S.G. Sergeev, L. Tacconi, M.
  Vestergaard,
• A. Wandel, Y. Yoshii

2
Key Points

• The line-width measure used for
  reverberation-based masses should be the line
  dispersion ?line rather than FWHM.
• New observations are leading to improved results,
  better identification of systematics.

3
Reverberation Mapping Results

• Reverberation lags have been measured for 36
  AGNs, mostly for one or more Balmer lines, but in
  some cases for multiple lines.
• AGNs with lags for multiple lines show that
  highest ionization emission lines respond most
  rapidly ? ionization stratification.

4
Evidence for a Virialized BLR

• Gravity is important
• Broad-lines show virial relationship between size
  of line-emitting region and line width, r ? ? ?2
• Yields measurement of black-hole mass
• M f (c?cent? 2 /G)

based on Peterson Wandel (1999)
5
Calibration of the Reverberation Mass Scale

• Determine scale factor f that matches AGNs to
  the quiescent-galaxy MBH-?. relationship
• Current best estimate f 5.5
  1.8
• Scaling factor is empirically determined
• This removes bias from the ensemble
• Equal numbers of masses are overestimated and
  underestimated

M f (c?cent? 2 /G)
based on Onken et al. (2004)
6
Physical Interpretation of f

• The Onken value is an average over the projection
  factors.
• Example thin ring

Aside since unification requires 0 ? i ? imax,
simple disks without a polar component are
formally ruled out.
7
Characterizing Line Widths

• FWHM
• Trivial to measure
• Less sensitive to blending and extended wings

• Line dispersion ?line
• Well defined
• Less sensitive to narrow-line components
• More accurate for low-contrast lines

Some trivial profiles
8

• Reverberation-mapped AGNs show broad range of
  FWHM/?line.
• Mass calibration is sensitive to which line-width
  measure is used!
• Even worse, there is a bias with respect to AGN
  type (as reflected in the profiles)

Extreme examples
9
(No Transcript)
10
(No Transcript)
11
I Zw 1 type
NLS1
12
I Zw 1 type
NLS1
NGC 5548
13
(No Transcript)
14
Mean spectra
RMS spectra
From Collin et al. (2006)
15
Mean spectra
Pop 2
?line-based calibration
RMS spectra
Collin et al.
Pop 1
Pop A
Pop B
similar to Sulentic et al.
From Collin et al. (2006)
16
Mean spectra
Pop 2
FWHM-based
RMS spectra
Collin et al.
Pop 1
Pop A
Pop B
similar to Sulentic et al.
From Collin et al. (2006)
17
Line Width Measures

• Conclusion ?line is probably a less biased
  indicator of the mass than FWHM.
• Use of FWHM will lead us to underestimate the
  masses of NLS1s, I Zw 1-type objects, and
  narrower-line objects in general.
• Can be corrected for empirically, however
  (see Collin, Kawaguchi, Peterson, Vestergaard
  2006).

18
HST ACS images are used to decompose light into
nuclear and starlight components. Effect is to
flatten radius-luminosity relationship. Starlight
components are stronger than previously supposed.
Bentz et al. (2006)
19
Other New Developments

• New reverberation program on bright well-known
  Seyfert galaxies
• Improve time sampling interval over original
  programs by as much as an order of magnitude in
  some cases.
• Ultimate goal a velocity-delay map for at least
  one line in one AGN.
• Secondary goal improve black hole mass
  measurements.

Denney et al., in preparation
Bentz et al., in preparation
20
NGC 4151

• Reanalyzed two UV monitoring data sets from IUE
  archive.
• UV and optical give consistent mass, 5 ? 107 M?

Metzroth, Onken, Peterson (2006) Bentz et al.,
in preparation
21
NGC 4151

• Moreover, the reverberation-based mass is
  consistent with the (highly uncertain) stellar
  dynamical mass based on long-slit spectra of the
  Ca II triplet.

Onken, Valluri, et al., in preparation
22
The AGN MBH ? Relationship
AGNs Ca II triplet
AGNs CO bandhead
(Dasyra Tacconi)
Quiescent (Tremaine et al. 2002)
23
?
24
Could Inclination Play a Role?

• Assume line width ?V ? (a2 sin2i )1/2 Vkep
• Then f ? M? / VP ? 1 / (a2 sin2i )1/2
  
• M? / VP cannot be used to deduce inclination for
  individual sources because NGC 5548 shows that VP
  values can span a factor 3.

Collin et al. (2006)
25
Could Inclination Play a Role?

• However, we can compare the OBSERVED cumulative
  distribution of M? / VP with that predicted by
  this simple model for various values of a.
• Reasonable agreement with simple model if only
  Population 1/A is used.
• Implication is that at least some AGNs have
  narrow lines because of low inclination.

Collin et al. (2006)
26
Summary

• As the database on reverberation mapped AGNs
  improves, identification of systematic biases
  becomes easier.
• Evidence that inclination plays a role.
• Reverberation-masses are less biased with respect
  to profile by using ?line as the line-width
  measure.
• FWHM / ?line is sensitive to Eddington rate and
  inclination.

27
What Do Line Widths Say About Masses?
28
Brads gripe du jour

• For fixed Eddington rate, more massive sources
  have larger line widths
• NLS1 criterion of FWHM lt 2000 km s-1 omits
  higher-luminosity objects from class (I Zw
  1type objects, including, for example, 3C 273)