Advances in Reverberation Mapping

1
Advances in Reverberation Mapping

• Shai Kaspi
• Tel-Aviv University Technion Haifa
• Israel

Line
Continuum
q
The Central Engine of AGNs Xian, China ,16
October 2006
2
Outline

• Introduction to reverberation mapping- Current
  BLR Size Luminosity Relation- Broadening the
  luminosity range- Mapping low luminosity AGNs-
  Mapping of high-luminosity quasars preliminary
  results- Dust reverberation mapping- Summary
  and future prospects

3
Reverberation Mapping
4
Reverberation Mapping

• Continuum luminosity vary.
• BLR respond to the variations (via
  photoionization).

5
Reverberation Mapping
Blandford McKee (1982) coined the term
reverberation mapping and put it into
mathematical formulation
6
Two decades of data acquisition
From Peterson (1988) Object Radius
(ld) Akn 120 lt30 NGC 4151 6 NGC
5548 30 3C 390.3 25-45
Early studies used relatively few, poorly spaces
observations.
Peterson Gaskel 1986

• Monitoring campaigns during the 1990s
• Individual monitoring of Seyfert I AGNs
  (NGC5548, NGC4151, NGC7469 Maoz et al. 1990 -
  and many more by the AGN Watch projects
  Peterson 1999).
• The Lovers of Active Galaxies (LAG) campaign
  (Robinson 1994).
• The Ohio State monitoring program (Peterson et
  al. 1998).
• The Wise Observatory and Steward Observatory 17
  PG quasars monitoring program (Kaspi et al. 2000).

Altogether 35 objects with data sufficient for
reverberation
7
AGN Spectrum
Balmer lines
PG0804761
Ha
Flux
Hb
Other lines
Continuum
Observed Wavelength
8
Light curves
Continuum Flux
Hb
Line Flux
Time
Kaspi et al. 2000
9
One Dimension Reverberation Mapping
Due to the sparse data and inability to measure
precisely subtle profile changes the search for
the transfer function
is collapsing to the one dimension reverberation
mapping, which is just a cross correlation
between the continuum and line light curves. The
peak/centroid of the cross correlation is a
measure to the size of the BLR RBLR.
10
Central AGN Mass

Finding the central (black hole) mass is one of
the holy grails of reverberation mapping in the
past decade. (but the sample might be biased.)
..More in the next talk by Brad Peterson.
11
BLR Size Luminosity Relation

• The BLR size luminosity relation
• Both are fundamental measured quantities.
• Peterson et al. (2004) compiled all studies to
  date.
• 35 objects with Balmer (mainly Hb) lines time
  lag.
• Characteristic BLR size Time Lag speed of
  light.
• Luminosities in the Optical, UV, and X-rays.
• BLR size from averaging all Balmer lines
• time lags per object.

12
Linear Regression
Uncertainties in both quantities And Intrinsic
scatter in the relation Two regression methods
1. FITEXY from Press et al. (1992) implemented by
Tremaine et al. (2002).
2. BCES (Bivariate Correlated Errors and
intrinsic Scatter) by Akritas Bershady (1996).
and also outlier points
13
Hb RBLR Optical luminosity (5100 A)
RBLR ? lLl(5100 Å) (0.690.05)
Kaspi et al. 2005
14
Hb RBLR Optical luminosity (5100 A)
RBLR ? lLl(5100 Å) (0.5180.039)
Bentz et al. 2006
15
Hb RBLR UV luminosity (1450 A)
RBLR ? lLl(1450 Å) (0.560.05)
Kaspi et al. 2005
16
RBLR X-ray luminosity (2-10 keV)
RBLR ? lLl(2-10 keV) (0.700.14)
Kaspi et al. 2005
17
RBLR luminosity Relation, conclusions

• Though small differences exist between the
  different regression methods the results are
  generally consistent.
• Average slope is 0.670.05 for the optical
  continuum and broad Hß luminosity, about
  0.560.05 for the UV luminosity, and about
  0.700.14 for the X-ray luminosity.
• We find in these relations an intrinsic scatter
  of about 40 .
• In some energy bands the slope is roughly like
  the naive theoretical prediction of 0.5. This
  prediction is naively based on the assumption
  that all AGNs have the same ionization parameter,
  BLR density, column density, and ionizing SED.

0.520.04
How can we determine a better relation?
By taking more data
18
Broadening the Luminosity range - Hb
Current studies span 4 orders of magnitude.
There are 4 more orders of magnitude to be
explored. Extrapolation does not necessarily
give the real situation.
We need to expand the luminosity range with
reverberation mapping studies.
19
Broadening the Luminosity range C IV
Up to 2004 only four AGNs with C IV BLR size
measurements
20
Broadening the Luminosity range C IV
Peterson et al. (2005) added NGC4395 four
orders of magnitude in luminosity lower
21
Broadening the Luminosity range C IV
Still there are the high luminosity quasars at
three orders of magnitude higher
22
Higher luminosities Quasar monitoring
Photometrically monitoring 11 quasars for the
past decade. 7 of which are spectroscopically
monitored for the past 5 years. 2.1 lt z lt
3.2 1045.6 lt ?L?(5100 Å) lt 1047
erg/s Photometric observation at the 1m Wise
Observatory. Spectroscopic observation at 9m
Hobby-Eberly Telescope (HET) and at the Wise
Observatory. Lines monitored are C IV and Ly?
using the method of a comparison star
simultaneously with the quasar in the slit.
Some preliminary results (Kaspi et al., ApJ,
Submitted)
23
SBS 1116603
?L?(5100 Å) 1.41046 erg/s z 2.628
?R0.34 ?B0.44
24
SBS 1116603 C IV Continuum CCF
No measurable time lag is found
25
S5 083671
?L?(5100 Å) 1.11046 erg/s z 2.172
?R0.34 ?B0.44
26
S5 083671 C IV Continuum CCF
27
Broadening the Luminosity range C IV
A preliminary result suggests a correlation
between the C IV size and the luminosity
28
Mass Luminosity Relation
Peterson et al. (2005)
29
Two Dimension Reverberation Mapping
Transfer function simulation assuming BLR
geometry and dynamics (Welsh Horne 1991).
30
Dust Reverberation Mapping
31
Summary of Reverberation Mapping

• Size of the BLR BLR size scales with
  Luminosity.
• Radial ionization stratification.
• BLR clouds motion are virial and primarily
  orbital.
• Mass of the Black Hole.
• Dust reverberation mapping to map the dusty
  region.
• Geometry and Kinematics of the BLR (?)

- Finding the transfer function and deciphering
the geometry and dynamics of the BLR still wait
for high-spectral resolution, high S/N monitoring
campaigns.
- Reverberation mapping of the broad Iron 6.4 keV
emission line to map the accretion disk
(Reynolds 1999).
32
Future prospects

• In the 1042 to 1046 erg/s luminosity range a
  firm relation exist between the BLR size and
  luminosity. Slope ranging from 0.5 to 0.7.
• Expanding the luminosity range is important and
  first steps are being taken
• - Low luminosity Seyferts and LINERS to cover the
  luminosity range of 1040 to 1042 erg/sec.
• - High luminosity quasars preliminary results
  are encouraging.
• Better measurement of the AGN luminosity.
• Dust reverberation mapping.
• Better time coverage and S/N spectra are needed
  for 2D TF.
• Fe ka line reverberation in the X-ray.

33
Cross Correlation Methods

• ICCF Interpolated Cross Correlation Function
  interpolating one light curve to the observation
  of the other and vice versa, then using the
  average CCF (Gaskell Sparke 1986 White
  Peterson 1994).
• DCF Discrete Correlation Function binning
  the actual time delay between points of the light
  curve (Edelson Krolik 1988).
• ZDCF z-transform Discrete Correlation Function
  doing the DCF in a z-transformed space to get a
  better handle of errors (Alexander 1997).

Uncertainty of the peak/centroid of the CCF (the
time lag) is done using FR/RSS (Flux
Randomization / Random Subset Selection) of
Peterson et al. (1998)
34
Continuum light curves
Z log L ?R ?B 2.628 46.14
0.34 0.44 3.177 46.88 0.16
0.22 2.172 46.05 0.34 0.44 2.824
46.62 0.25 0.27 3.200 46.96 0.14
0.19 2.722 47.04 0.16 0.20
35
S4 063668
?L?(5100 Å) 7.61046 erg/s z 3.117
?R0.16 ?B0.22
36
HS 17006416
?L?(5100 Å) 1047 erg/s z 2.722
?R0.16 ?B0.20
37
Scaling C IV size to Hb size
S5 083671 CIV lag days
NGC4395 CIV lag min
? Hb lag about min
In two weeks, optical monitoring of NGC4395
spectroscopically from KPNO and photometrically
from four observatories around the world cover 22
hours a day for four days.
38
Title
Text