Very high energy g-ray observations
of the Galactic Center with H.E.S.S.
Matthieu Vivier
IRFU/SPP CEA-Saclay
On behalf the H.E.S.S. collaboration
Outline
Context
The H.E.S.S. observations of the Galactic Center
Position & mophology of HESS J1745-290
Spectrum & variability
Models for TeV g-rays at the GC (if time)
Conclusions
Moriond 2009
Matthieu Vivier
2
Context
MWL source in the central parsecs of our Galaxy: emitting from radio to TeV grays
From radio to X-rays: originates from the SMBH Sgr A*
Origin of the hard X-rays/Tev g-rays?
X-rays
IR
Radio
TeV
Sgr A*
Moriond 2009
?
Matthieu Vivier
3
Context
MWL source in the central parsecs of our Galaxy: emitting from radio to TeV grays
From radio to X-rays: originates from the SMBH Sgr A*
Origin of the hard X-rays/Tev g-rays? Possible counterparts: SNR Sgr A East,
PWN G359.95-0.04, Sgr A*…
G359.95-0.04
≈ 10pc×10pc
H.E.S.S. angular resolution
+
Moriond 2009
Matthieu Vivier
4
Context
Highly variable in radio, IR & X-rays
Moriond 2009
Matthieu Vivier
5
Context
Highly variable in radio, IR & X-rays
→106 days periodic modulation in the radio flux: accretion disk?
Fourier power spectrum: VLA data
Zhao, Bower & Goss (2001)
Moriond 2009
Matthieu Vivier
6
Context
Highly variable in radio, IR & X-rays
→106 days periodic modulation in the radio flux: accretion disk?
→IR/X-ray flares of ≈ 1h time duration + QPOs: strongly supports the idea
of an accretion disk around Sgr A*
Fourier power spectrum: VLA data
Zhao, Bower & Goss (2001)
Moriond 2009
Matthieu Vivier
IR flare of June 16th, 2003: VLT data
Genzel et al. (2003)
7
The observations of the GC in 2004
Detection of a point-like source: power-law spectrum + non variable emission
on timescales < 1 year
Moriond 2009
Matthieu Vivier
8
The observations of the GC in 2004
Detection of a point-like source: power-law spectrum + non variable emission
on timescales < 1 year
Spectrum incompatible with DM particle annihilations: the fits are bad in the
low & high energy parts.
14 TeV
5 TeV
10 TeV
« HESS Observations of the Galactic Center Region and Their
Possible Dark Matter Interpretation » , PRL, 97 (2006) 221102.
Moriond 2009
Matthieu Vivier
9
The observations of the GC in 2004
Detection of a point-like source: power-law spectrum + non variable emission
on timescales < 1 year
Spectrum incompatible with DM particle annihilations: the fits are bad in the
low & high energy parts.
14 TeV
5 TeV
10 TeV
« HESS Observations of the Galactic Center Region and Their
Possible Dark Matter Interpretation » , PRL, 97 (2006) 221102.
A DM contribution is not excluded: estimated to be < 10%
Moriond 2009
Matthieu Vivier
10
l (deg)
The 2004-2006 observations
2004-2006 dataset: 100h
(≈ 0.1 year of data collection)
3 sources in the 5°×5° FoV:
GC source HESS J1745-290: pointlike, 60s detection.
SNR G0.9+0.1: point-like.
extended source 3EG J1744-3011
(EGRET).
Moriond 2009
Matthieu Vivier
~ 500 pc X 500 pc
b (deg)
11
l (deg)
The 2004-2006 observations
2004-2006 dataset: 100h
(≈ 0.1 year of data collection)
3 sources in the 5°×5° FoV
Diffuse emission along the galactic plane
« Discovery of very high energy g-rays from
the Galactic Centre ridge », Nature 439
(2006) 695-698
~ 500 pc X 500 pc
b (deg)
Moriond 2009
Matthieu Vivier
expected level of
background events
12
l (deg)
The 2004-2006 observations
2004-2006 dataset: 100h
(≈ 0.1 year of data collection)
3 sources in the 5°×5° FoV
Diffuse emission along the galactic plane
« Discovery of very high energy g-rays from
the Galactic Centre ridge », Nature 439
(2006) 695-698
~ 500 pc X 500 pc
statistic of ≈ 4000 events in a circular
0.1° region centered on the GC.
ON source region
Moriond 2009
Matthieu Vivier
b (deg)
expected level of
background events
13
l (deg)
The 2004-2006 observations
2004-2006 dataset: 100h
(≈ 0.1 year of data collection)
3 sources in the 5°×5° FoV
Diffuse emission along the galactic plane
« Discovery of very high energy g-rays from
the Galactic Centre ridge », Nature 439
(2006) 695-698
~ 500 pc X 500 pc
statistic of ≈ 4000 events in a circular
0.1° region centered on the GC.
b (deg)
Position/morphology
Spectrum/variability
ON source region
Moriond 2009
Matthieu Vivier
expected level of
background events
14
Position & morphology
point-like source: intrisic size < 1.2’ (≈ 2.9 pc) at the 99% C.L.
position:
l=359°56’41.1’’± 6.4’’± 6’’
b=-0°2’39.2’’ ± 5.9’’ ±6’’
centroid emission located at 7’’ ± 12’’
from Sgr A*
Sgr A East excluded at the 7s C.L.
SNR SgrA East
(90 cm)
G359.95-0.04 still inside error bars
(8.7’’ from Sgr A*)
van Eldik et al., ICRC (2007)
Moriond 2009
Matthieu Vivier
15
The 2004-2006 spectrum
curvature in the spectrum: deviates from a power-law in the high energy part.
2 dN
E
 f(E
)
dE
power-law with an
exponential cut-off
broken power-law
G = 2.10 ± 0.04
G2 = 2.63 ± 0.14
G1 = 2.02 ± 0.08
Ebreak = 2.57 ± 0.19 TeV
Flux
Flux
Ecut = 15.7 ± 3.40 TeV
power-law
power-law
G1
G2
Ecut
Moriond 2009
log(E)
Matthieu Vivier
Ebreak
log(E)
16
The 2004-2006 spectrum
curvature in the spectrum: deviates from a power-law in the high energy part.
2 dN
E
 f(E
)
dE
power-law with an
exponential cut-off
broken power-law
G = 2.10 ± 0.04
G2 = 2.63 ± 0.14
Ecut = 15.7 ± 3.40 TeV
G1 = 2.02 ± 0.08
Ebreak = 2.57 ± 0.19 TeV
Fit residuals
Moriond 2009
Matthieu Vivier
17
Variability
run-by-run light curve (integrated fluxes per 28 min intervals) compatible with a
constant
c2/dof = 233/216
flare sensitivity study:
Moriond 2009
Matthieu Vivier
18
Variability
Is the TeV signal correlated with the X-ray signal?
→simultaneous observations with the Chandra satellite in 2005.
Limit on the TeV flux increase during the flare: < factor 2 (99% C.L)
In agreement with the flare sensitivity study.
« Simultaneous H.E.S.S. and Chandra observations of Sgr A*
during an X-ray flare », A&A 492L, 25 (2009)
Moriond 2009
Matthieu Vivier
J.Hinton, M.V, et al.,
(HESS) ICRC 2007
19
Variability
QPO structures in the IR & X-rays flares: likely to correspond to oscillation
modes of an accretion disk around Sgr A* (♭: oscillations not confirmed in IR
by the Keck telescopes)
Moriond 2009
Matthieu Vivier
20
Variability
QPO structures in the IR & X-rays flares: likely to correspond to oscillation
modes of an accretion disk around Sgr A* (♭: oscillations not confirmed in IR
by the Keck telescopes)
2 cases:
1) The coherence time of the disk oscillations is short: Rayleigh test on
photon time arrival distribution in a data run (< 28 min) + average of the
Rayleigh power at one frequency over the whole runs
Rayleigh test
c2/dof = 35/29
1150s 700s
Moriond 2009
219 s
100 s
Matthieu Vivier
M.V et al., ICRC (2007)
21
Variability
QPO structures in the IR & X-rays flares: likely to correspond to oscillation
modes of an accretion disk around Sgr A* (♭: oscillations not confirmed in IR
by the Keck telescopes)
2 cases:
2) The coherence time of the disk oscillations is of the order of a few
hours: Lomb-Scargle periodogram on 5 min integrated fluxes in a night
of data collection (≤ few hours) + average of the Fourier power at each
tested frequency over the whole nights.
Lomb-Scargle periodogram
M.V et al., ICRC (2007)
Moriond 2009
Matthieu Vivier
22
VHE emission models
Leptonic models: inverse Compton scattering of leptons on the dense photon
fields (UV/Optical/IR) in the vicinity of Sgr A*. Origin of the leptons?
→From the nearby pulsar G359.95-0.04: Hinton & Aharonian (2007)
→Stochastic acceleration of leptons in an accretion disk around Sgr A*:
Liu et al. (2006), Atoyan & Dermer (2004)
Black Hole Plerion model
Atoyan & Dermer (2004)
Hinton & Aharonian (2007)
cut-off caused by the Klein-Nishina effect. No expected variability.
Moriond 2009
Matthieu Vivier
23
VHE emission models
Hadronic models: acceleration of protons in the vicinity of the supermassive
black hole, Ballantyne et al. (2007)
p0
p0
0
p
→collision on the surrounding
accretion disk
molecular clouds. p0 then gs
molecular
clouds
Aharonian & Neronov (2005): possible origins
for the g-ray cut-off:
p+
p+
→cut-off in the initial proton
p0
p0
injection spectrum
→cut-off originates from the
p+
p+
g
competition between the injection
Sgr A*
of protons in the surrounding medium
and their escape from the Central
Molecular Zone (≈leaky box model with energy-dependent diffusion)
p0
Models favoured by the observation of the g-ray diffuse emission along the
Galactic plane (very well interpreted by cosmic rays interactions with giant
molecular clouds, see Nature paper).
Moriond 2009
Matthieu Vivier
24
Conclusions
Sgr A East is now spatially excluded: Sgr A* and G359.95-0.04 still remain as
possible counterparts.
Curvature in the spectrum
No variability in the H.E.S.S. signal on timescales ranging from a few minutes
to one year:
→no flaring activity
→no periodic modulation of the TeV flux (QPOs)
TeV emission decorrelated from the other wavelengths
Analysis and results soon published in A&A
Moriond 2009
Matthieu Vivier
25
Conclusions
Sgr A East is now spatially excluded: Sgr A* and G359.95-0.04 still remain as
possible counterparts.
Curvature in the spectrum
No variability in the H.E.S.S. signal on timescales ranging from a few minutes
to one year:
→no flaring activity
→no periodic modulation of the TeV flux (QPOs)
TeV emission decorrelated from the other wavelengths
Analysis and results soon published in A&A
Thank you!