Superluminal Velocity Surveys,

1
Superluminal Velocity Surveys, Lorentz Factors,
Doppler Factors, and Brightness Temperatures.
René Vermeulen ASTRON, Dwingeloo, NL
2
The 15 GHz VLBA and MOHAVE Surveys
Work with K.I. Kellermann, M.L.Lister, D.C.
Homan, M.H. Cohen, E. Ros, M. Kadler, J.A.
Zensus, Y.Y. Kovalev.
Kellermann et al., 2004, ApJ,
in press for July 10 Cohen et
al., 2004, Proc. 2003 JENAM, in press
Cohen et al., 2003, ASP Conf. Ser. 300,
27 Zensus et al., 2003, AJ,
124, 662 Kellermann et al.,
1998, AJ, 115, 1295
15 GHz VLBA Survey From complete 5 GHz survey
(Kühr et al., Stickel et al.), plus
additions. Total flux density S(15GHz) gt 1.5 Jy
(2 Jy for Southern sources). Flat spectrum (? gt
?0.5) anywhere above 500 MHz.
MOHAVE Survey (Monitoring Of Jets in Active
galaxies with VLBA Experiments) Sources from 15
GHz survey with total VLBA (compact) flux
density S(15GHz) gt 1.5 Jy (2 Jy for Southern
sources) at any epoch 1995-2003.
3
The 15 GHz VLBA Proper Motion Survey
29 observing sesions 1994 2001 208
components in 110 sources 79 Q, 18 BL, 13 G
Typically 3 7 sessions per source
4
Velocity Histograms
Mostly 0 lt ?app lt 15, tail to ?app 34.
5
Velocity Histograms
Mostly 0 lt ?app lt 15, tail to ?app 34.
0727-115 z1.591 ?app31.2 ? 0.6 2223-052
z1.404 ?app32.5 ? 6.0
6
Velocity Histograms
Mostly 0 lt ?app lt 15, tail to ?app 34.
(0716714 if at z0.3, ?app13)
0727-115 z1.591 ?app31.2 ? 0.6 2223-052
z1.404 ?app32.5 ? 6.0
7
Velocity Histograms
Mostly 0 lt ?app lt 15, tail to ?app 34.
BL-Lacs and Galaxies ?app lt 6 many are not
superluminal differ from Q at 98 confidence.
8
Velocity Histograms
Mostly 0 lt ?app lt 15, tail to ?app 34.
BL-Lacs and Galaxies ?app lt 6 many are not
superluminal differ from Q at 98
confidence. EGRET sources faster than
non-EGRET at 90 confidence.
9
Velocity Histograms
Mostly 0 lt ?app lt 15, tail to ?app 34.
BL-Lacs and Galaxies ?app lt 6 many are not
superluminal differ from Q at 98
confidence. EGRET sources faster than
non-EGRET at 90 confidence. In 5 GHz
CJ-Survey velocity distributions similar in
shape, but roughly a factor 2 slower.
10
Brightness Temperature Histograms
From flux density originating in a given area
derive Tb,struc.
VLBA at 15 GHz N19
VSOP at 5 GHz N59 (Hirabayashi et al. 2000)
11
Brightness Temperature Histograms
From flux density originating in a given area
derive Tb,struc.
VLBA at 15 GHz N19
VSOP at 5 GHz N59 (Hirabayashi et al. 2000)
Can then derive Doppler factor ?struc if
intrinsic temperature is assumed Tb,struc
?struc Tb,int
12
Brightness Temperature Histograms
From flux density originating in a given area
derive Tb,struc.
Can then derive Doppler factor ?struc if
intrinsic temperature is assumed Tb,struc
?struc Tb,int
13
Variability Doppler Factor Histogram
From flux density outburst in a given time derive
Tb,var
Can then derive Doppler factor ?var if intrinsic
temperature is assumed Tb,var ?3var Tb,int
14
Variability Doppler Factor Histogram
From flux density outburst in a given time derive
Tb,var
Lähteenmäki Valtaoja 1999 N81 assuming
Tb,int 5 x 1010 K
Can then derive Doppler factor ?var if intrinsic
temperature is assumed Tb,var ?3var Tb,int
15
Brightness Temperatures and Doppler Factors
Tb,var ?3var Tb,int Tb,struc ?struc
Tb,int Comparison yields Tb,int 1011 K (e.g.
Lähteenmäki et al. 1999)
16
Brightness Temperatures and Doppler Factors
Tb,var ?3var Tb,int Tb,struc ?struc
Tb,int Comparison yields Tb,int 1011 K (e.g.
Lähteenmäki et al. 1999)
Now we can use ?app as a new estimator, involving
?veloc , and depending on Lorentz factor
? angle to the line-of-sight ?
17
Brightness Temperatures and Doppler Factors
Tb,var ?3var Tb,int Tb,struc ?struc
Tb,int Comparison yields Tb,int 1011 K (e.g.
Lähteenmäki et al. 1999)
Now we can use ?app as a new estimator, involving
?veloc , and depending on Lorentz factor
? angle to the line-of-sight ?
18
Brightness Temperatures and Doppler Factors
Tb,var ?3var Tb,int Tb,struc ?struc
Tb,int Comparison yields Tb,int 1011 K (e.g.
Lähteenmäki et al. 1999)
Now we can use ?app as a new estimator, involving
?veloc , and depending on Lorentz factor
? angle to the line-of-sight ?
? and ? not known in individual cases but for
sources with both ?app and ?var can match
statistical distribution to Monte Carlo.
19
Motion Statistics, Upper Envelopes, and Beaming
Doppler favouritism in simple beaming models
combines with solid angle available to predict
many sources oriented at ? 1/ ? , corresponding
to maximal velocities Velocity
distributions sharply peaked at the high end
Crowding towards well-defined upper envelopes
This is not seen !
20
Motion Statistics, Upper Envelopes, and Beaming
Doppler favouritism in simple beaming models
combines with solid angle available to predict
many sources oriented at ? 1/ ? , corresponding
to maximal velocities Velocity
distributions sharply peaked at the high end
Crowding towards well-defined upper envelopes
This is not seen !
Instead, the actual distribution could indicate
Broad distribution of jet bulk Lorentz
factors in the population Reflected
by shape of velocity-apparent luminosity diagram ?
21
Velocity is Correlated with Apparent Luminosity
Upper envelope of ?app distribution rises
with Lobs(15GHz). ?app and Lobs could really
be correlated pattern and bulk motion in the
same jet flow, with similar Lorentz factor
?. Or it could be the result of Malmquist
bias if high ? jets are fairly rare, e.g. N(?)
? ? -1.5 , none may be seen at low L low z,
where there is not much volume.
22
Motion Statistics, Upper Envelopes, and Beaming
Doppler favouritism in simple beaming models
combines with solid angle available to predict
many sources oriented at ? 1/ ? , corresponding
to maximal velocities Velocity
distributions sharply peaked at the high end
Crowding towards well-defined upper envelopes
This is not seen !
Instead, the actual distribution could indicate
Broad distribution of jet bulk Lorentz
factors in the population Reflected
by shape of velocity-apparent luminosity diagram ?
23
Motion Statistics, Upper Envelopes, and Beaming
Doppler favouritism in simple beaming models
combines with solid angle available to predict
many sources oriented at ? 1/ ? , corresponding
to maximal velocities Velocity
distributions sharply peaked at the high end
Crowding towards well-defined upper envelopes
This is not seen !
Instead, the actual distribution could indicate
Broad distribution of jet bulk Lorentz
factors in the population Reflected
by shape of velocity-apparent luminosity diagram ?
Decoupling between source selection and
component motions Pattern motions
(shocks moving in fluid) Jet
curvature (between core and moving knot)
Different beaming cones between core and jet
components
24
Motion Statistics, Upper Envelopes, and Beaming
Doppler favouritism in simple beaming models
combines with solid angle available to predict
many sources oriented at ? 1/ ? , corresponding
to maximal velocities Velocity
distributions sharply peaked at the high end
Crowding towards well-defined upper envelopes
This is not seen !
Instead, the actual distribution could indicate
Broad distribution of jet bulk Lorentz
factors in the population Reflected
by shape of velocity-apparent luminosity diagram ?
Decoupling between source selection and
component motions Pattern motions
(shocks moving in fluid) Jet
curvature (between core and moving knot)
Different beaming cones between core and jet
components
6cm velocities statistically 2x slower than 2cm !
25
Brightness Temperatures and Doppler Factors
Tb,var ?3var Tb,int Tb,struc ?struc
Tb,int Comparison yields Tb,int 1011 K (e.g.
Lähteenmäki et al. 1999)
Now we can use ?app as a new estimator, involving
?veloc , and depending on Lorentz factor
? angle to the line-of-sight ?
? and ? not known in individual cases but for
sources with both ?app and ?var can match
statistical distribution to Monte Carlo.
26
Brightness Temperatures and Doppler Factors
Top-left panels Monte-Carlo simulation using
N(?) ?-1.5 , ?max30 (N100) Other panels
Measured ?app
(N30)
against ?var , assuming Tb,int
4x109 K, 2x1010 K, 1011 K
27
Lorentz Factors
Using Tb,int 2x1010 K, can then derive Lorentz
factor distribution
28
Summary
At 15 GHz, mostly 0 lt ?app lt 15, tail to ?app
34. Motions are statistically slower at 5
GHz, possibly faster at higher frequencies
deceleration, decollimation, bending,
banana-trunk jets, ... Motion distributions,
correlations with luminosity may show a steep
intrinsic Lorentz factor distribution, N(?)
?-1.5 , ?max30. If variability and 15 GHz
superluminal motions are closely linked, then
perhaps typically Tb,int 2x1010 K.