MAGIC
A first synoptic blazar study comprising eleven
blazars visible in E>100 GeV gamma rays
Major Atmospheric
Gamma Imaging
Max-Planck-Institut
für Physik
Werner-Heisenberg-Institut
Robert Marcus Wagner
Cerenkov Telescope
Max-Planck-Institut für Physik, Föhringer Ring 6, D-80805 München, Germany
ABSTRACT
Since 2002, the number of detected blazars at g-ray energies above 100 GeV has more than doubled. I present a study of all currently known highfrequency peaked BL Lac-type objects, for which photon energy spectra at E>100 GeV have been inferred. The intrinsic energy spectra of the sources are
reconstructed using a contemporary model for the extragalactic background light. Then, the properties of the observed emission are compared and
correlated among each other, with X-ray data, and with the individual estimated black hole masses. In addition, I consider temporal properties of the VHE
g-ray flux. Key findings concern the connection of observed photon flux and estimated black hole mass as well as the correlation between the spectral
slope and the luminosity. As a specific application, the synoptic study allows to constrain the redshift of PG 1553+113, a TeV blazars whose distance is
unknown to date.
• 3rd generation Imaging Air Cherenkov
Telescopes with unprecedented
sensitivity and energy thresholds
online
• Blazar detection rate increased
180
from 6 in 10 years to now
2/yr and increasing
• For almost all VHE blazars are now VHE
γ-ray spectra available
• No comparative study exists yet that
includes the whole enlarged blazar
sample
• Of some of them, different flux states are
known
• Relations to the BH/jet environment
expected: Probe correlation with BH
mass
• In SSC models, X/VHE correlations
expected
• First such study (Krawczynski et al.
2004): comprised Mkn421, Mkn501,
1ES2344, 1ES1959, 1H1426
Identified EGRET AGN
Tentatively id EGRET AGN
Mkn501
PG1553
1ES2344
-180
BL Lacertae
PKS0548
PKS2005
PKS2155
1ES0347
H2356
12
1118+3
03401- 04
7-1232
21
09
-3
56
PK
S2
15
5
14
02226+
9+ 428
200
-90
0.1
1553+113
Methodology
0.15
Redshift
0.2
0.25
BH mass estimations
Here we use:
• Differential luminosity at 500 GeV (no
extrapolations necessary)
• Spectral slope (no extrapolations)
• 1 keV X-ray luminosity
• VHE timescale
• Black hole mass estimations
• Redshift
Parameters on which jet emission
(probably) depends:
• Size of black hole, spin of black hole
• Accretion rate ~ jet luminosity
• Heavy objects in accretion disk:
modulation?
Scaling laws are thought to govern BH
physics. Simulation suggest that:
• Jet luminosity increases with spin
• Poynting flux increases with spin
• Counter-/co-rotating BH produce
different jets
1ES1101
1ES1959
Time scales & Duty Cycles
We used σ measurements by Barth et al., Falomo
et al., Wu et al. (2002) and the M-σ relation
given by Tremaine et al. (2002)
Redshift z
BH/jet connection
0.05
HBL in this study
HBL/LBL not in study
Mkn180
Mk
Mkn 42
23 n 501
Mk44+51
19 n 1814
59 0
+65
0
PKS
P K 054
S 28
00
5
0
1ES1218
H1426
• Now probing higher distances
and fainter sources; presumably
low-state sources, too
HBL objects included in this study
θ
Motivation
90
Mkn421
1ES0229
M
B L87
Lac
• Supermassive central object of 106 to 1010 solar
masses - black holes
• Relativistically rotating accretion disk (thermal
Black
radiation up to keV): 3-20 AU
Hole
• Gas cloud around accretion disk
Can we see the
Jet
• High variability at all frequencies imprint of some
of those in the
• Jets viewed under small angle
VHE γ-ray
• High doppler factors expected
emission?
• Strong nonthermal radiation
Accretion
from the center of the galaxy
Disk
Obscuring
Torus
• Jet scales: 10-1011 AU (10-100ly)
• Jets may attain high luminosities: Lobs ~ L × δ4
Illustration: NASA/CXC/M.Weiss
• In leptonic acceleration models: Connection of
synchrotron (X-ray) an IC peak (Gev/TeV γ-rays) expected
Currently known HBL sample
23
High-peaked BL Lac-type AGN
1
0.1
0.01
L(500 GeV) [erg sr-1 s-1]
Follow an approach by Krawczynski et al.:
0.001
6
6.5
7
7.5
8
8.5
9
9.5
10
log(M
/M
)
• Study long-term 2-10 keV RXTE ASM light
AGN class
Seyfert1
BL Lac
Radio-loud quasar
curves (1996-2006)
VHE γ emitter
Seyfert2
Radio galaxy
Radio-quiet quasar
• All RXTE ASM data points >20% over
Here we compare the VHE γ emitters with a
Reconstruction of VHE spectra
VHE vs. X-ray luminosity
average are thought to be on-duty time sample of 452 AGNs by Woo & Urry:
• Our refined strategy: Require significant • Only M>108M objects show VHE γ -ray
• Systematic assession of all known TeV HBLs
sun
activity - 5σ above baseline
emission
• EBL absorption using Kneiske et al.'s lowCorrelations with X-ray duty cycle
EBL model
• So far consistent with BL Lac nature
• Previously seen: indication for an anti• Extract slope in region in which spectrum
A
redshift
limit
for
PG1553
Correlation?
correlation (Krawczynski et al.)
was measured
• Luminosity distribution can be used to
•
Now:
no
clear
trend,
Mkn
421
seems
to
• Modeling partially elusive: Extract
constrain redshift of PG1553: Assume it is
be
quite
outstanding
differential luminosity at 500 GeV
not too different from the most extreme
6´ 10
10
2´ 10
10
• Anticorrelation mainly tainted by
BL Lacs -or- that its large distance implies
After EBL de-absorption: Γ=1.5-3.3
recently discovered 1H2356, PKS0548
unusually
high
luminosity/jet
Doppler
factor
•
From
SSC
one
might
expect
a
correlation
Intrinsically-hard spectra plus newly
Duty cycle vs. M
Duty cycle vs. luminosity
Kr
• 30 times more luminous: z<0.48 (2σ)
soft-spectrum population tapped
aw
• Not very clear: 1 keV might be quite
cz
yn
sk
contaminated
• 1000 times more luminous: z<0.68 (2σ)
ie
ta
l.
50
Correlation with BH mass?
• VHE luminosity spans over almost 3
distance + EBL Kneiske et al. best
49
distance + EBL Kneiske et al. low
decades, corresponding X-ray luminosity
48
distance only
47
only over one decade
BH
sun
Mkn 421
Mkn 501
2344+514
1044
Mkn 180
1959+650
2005-489
43
10
2155-304
1426+428
2356-309
1042
1218+304
1101-232
1553+113
44
45
45
46
-1
L(1 keV) [erg sr s-1]
4
3.5
3
2.5
2
1044
10
43
Luminosity vs. photon index
1042
8.2 8.4 8.6 8.8
9
log(M
2344+514
Mkn 180
9.2
BH
1959+650
2005-489
7.6 7.8
/Msun)
1426+428
8
8.2 8.4 8.6 8.8
9
log(M
9.2
/M sun)
BH
1218+304
2356-309
• No correlation (yet?)
• Is the VHE emission rather independent of
the BH mass?
• Does the acceleration environment play
the dominant role?
Redshift dependency?
Γ = Γ0 + m ⋅ log10(L )
4
Γ 0 = 24.60 ± 4.84
m = -0.52 ± 0.11
3.5
χ2
= 21.57/14
red
3
2.5
2
4.5
3.5
3
2.5
2
1.5
1.5
1
1
0.5
1042
10
43
1044
0.5
-1
L(500 GeV) [erg sr s-1]
1042
10
43
1044
L(500 GeV) [erg sr-1 s-1]
Mkn 421
2344+514
1959+650
2155-304
2356-309
1101-232
Mkn 501
Mkn 180
2005-489
1426+428
1218+304
1553+113
Hint for an anti-correlation:
The harder the spectrum, the higher
the luminosity
L(500 GeV) [erg sr-1 s -1]
Intrinsic photon index Γ
• Not expected for low z and small samples
• EBL overprediction?
• 3.3σ correlation (excluding PG1553)
• Why do we only see rather far-away hard• Decrease of ∆Γ=0.5 per decade of
spectrum sources? Selection effect?
luminosity
• Also intra-source behavior shows
spectral hardening
ti y
4.5
4
3.5
3
2.5
2
44
10
10
43
1042
IA
1
0
1041
Photon index vs. redshift
0.03
Mkn 421
Mkn 501
0.2
0.1
2344+514
Mkn 180
0.3
Redshift z
1959+650
2005-489
e
s
T
Luminosity vs. redshift
0.03
1426+428
1218+304
2356-309
1101-232
0.1
1553+113
0.2
Duty cycl e
0.2
0.15
0.15
0.1
0.1
0.05
0.05
0.3
Redshift z
9
log(M
9.2
BH
43
1042
/Msun)
10
44
10
-1 -1
L(500 GeV) [erg sr s ]
Mkn 421
2344+514
1959+650
2155-304
2356-309
1553+113
Mkn 501
Mkn 180
2005-489
1426+428
1101-232
0548-322
VHE time scale vs. M BH
VHE time scale vs. luminosity
3
10
2
3
10
2
10
10
10
10
1
1
-1
10
-2
10
-3
10
-3
10
7.6 7.8
Mkn 421
8
8.2 8.4 8.6 8.8
9
log(M
2344+514
2005-489
BH
9.2
/Msun)
1042
43
10
2155-304
1044
-1
L(500 GeV) [erg sr s-1]
1553+113
REFERENCES
Barth, Ho, & Sargent, 2003, ApJ 583, 134 - Cui, 2004, ApJ 605, 662 - Falomo, Kotilainen, &
Treves, 2002, ApJL 569, 35 - Kneiske et al., 2004, A&A 413, 807 - Krawczynski et al., 2004, ApJ
601, 151 - Tremaine et al., 2002, ApJ 574, 740 - Woo & Urry, 2002, ApJ 579, 530 - Wu, Liu, &
Zhang, 2002, A&A 389 ,742
1959+650
45
44
currently known
VHE HBLs
mit
IACT sensitivity li
43
42
1553+113 luminosity evolution
41
40
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Redshift z
Conclusions
First study to encompass 11 known HBLs to date
First study that also assesses intercorrelations of
VHE emission properties
• No obvious correlations of VHE γ properties, flare
duty cycles and flare time scales with BH masses
• BH mass might not be as influential to VHE γ-ray
emission as e.g. environment in acceleration region
• Indications that in variable sources the observed
spectra become harder with increasing luminosity
-2
10
46
• VHE luminosity is marginally correlatied with
spectral hardness, also found for individual sources
-1
10
Mkn 501
C
1.5
0.5
iv
it
ns
0.2
8.2 8.4 8.6 8.8
5σ duty cycle
VHE variability time scales might still be
dominated by (previously inferior)
sensitivity of IACT
... and by chance: IACT were lucky in
observing Mkn 501 (MAGIC Jul 2005) and
PKS 2155 (H.E.S.S. Jul 2006) flares
... there is probably more to it than just one
simple number “duty cycle” or the
minimum variability time (cf. variability
hierarchy studies, Cui 2004)
Sources observed
in different flux states
4
t [d]
Mkn 421
Mkn 501
8
BH
Intrinsic photon index Γ
0 7.6 7.8
Luminosity vs. M
1041
Intrinsic photon index Γ
Photon index vs. M BH
0.5
4.5
0.25
8
3σ duty cycle
0.3
0.25
7.6 7.8
1.5
1
0.3
0.35
log(L(500 GeV) [erg sr -1 s-1])
4.5
0.35
t [d]
L(500 GeV) [erg sr-1 s -1]
Intrinsic photon index Γ
Duty cycl e
BH
• Luminosity distribution constrains redshift of
PG1553 to z<0.68: Large distance implies high
luminosity or jet Doppler factor
• Marginal correlation between intrinisic spectral
hardness and source distance is likely due to an EBL
overprediction by the current EBL models