Cosmic Ray Electrons and GC Observations with H.E.S.S. · 2009. 7. 20. · Cosmic Ray Electrons Aharonian et al. (2008) Suffer severely from synchrotron and inverse Compton losses

Cosmic Ray Electronsand GC Observations with H.E.S.S.

Christopher van Eldik (for the H.E.S.S. Collaboration)MPI für Kernphysik, Heidelberg, Germany

TeVPA '09, SLAC, July 2009

The Centre of the Milky Way

C. van Eldik • TeV PA 2009

Aharonian et al. (2006)

H.E.S.S. 2004 (55 hours)

G 0.9+0.1

H.E.S.S. J1745-29038 sigma (55h)H.E.S.S. J1745-290point-like < 1.2' (95% CL)



Diffuse emission

H.E.S.S. 2004 (55 hours)

G 0.9+0.1H.E.S.S. J1745-290point-like < 1.2' (95% CL)




● Lack of γ-raysfor l > 1°

● Injection ofprotons at GC

● Assumek = ~3 kpc2 Myr-1for TeV protons→ injection 104 years ago

● Fits age ofSgr A East

Molecular Cloud Association


● Not just passive illumination- enhanced flux for > 1 TeV- photon index ~2.3

● Similar index as HESS 1745-290everywhere in the region


Diffuse Emission Spectrum

VLA Chandra

SNR Sgr A East? SMBH Sgr A*? DM? PWN G359.95-0.04?

10''300''

Possible Counterparts?


VLA Chandra

SNR Sgr A East? SMBH Sgr A*? DM? PWN G359.95-0.04?

10''300''

Possible Counterparts?


● Position?● Variability?● Energy spectrum?

Sgr A*Sgr A East

Best Fit HESS J1745-290 (Aharonian et al. 2004)Best Fit HESS J1745-290 (van Eldik et al. 2007) - preliminary-

VLA 90cm image

0.04 deg

Position: Sgr A East ruled out

CvE et al., Proc. ICRC (2007)

● Dedicated data set using optical guiding telescopes

● 6'' systematic pointing error

● Lack of association with Sgr A East

● Chance probability 10-4 ... 10-11


Variability studies● Sgr A* highly variable at other wavelengths● Quasi-periodic oscillation● Expect correlated VHE variability if emission produced close to BH surface● No obvious variability in VHE lighcurve observed based on 93 hours of data

Aharonian et al. (2009), arXiv:0906.1247

HESS J1745-290, 28 min flux points


Flare Sensitivity


● Maximum needed lightcurve “amplification” for 3σ flare detection● (as usual) statistics limited

Ahar

onia

n et

al.

(200

9)


Variability studies● Simultaneous HESS and

Chandra observations

● X-ray flare detected- 1700s duration- 9x quiescent level

● No increase of gamma flux→ 100% flux increase discarded at 99% CL



Search for QPOs (small time scales)● Quasi-periodic oscillations observed in X-rays and IR● X-ray periodicity 100 s, 219 s, 700 s, 1150 s, 2250 s● Related to accretion disk?● Rayleigh test for continuous 28 min observations (2004-2006 averaged)

→ no hint for QPOs < 1150 s in VHE data

Ahar

onia

n et

al.

(200

9)


Search for QPOs (large time scales)● Lomb-Scargle periodogram

averaged over 2004-2006● Power spectrum compatible

with noise● No indication for QPOs

on 600 s – 1.5 h time scales

Ahar

onia

n et

al.

(200

9)



Spectrum – a bit of history

● Hard spectrum: Γ = 2.25 ± 0.04 ± 0.10● 10% Crab above 1 TeV● No cut-off: EC > 9 TeV (95% CL)

● 2004-2006 data93 h live time

● 4185 γ-rays (61 σ)160 GeV < E < 70 TeV

● Exponential cut-offΓ = 2.10±0.04±0.10Ec = 15.7±3.5±2.5 TeVχ²/d.o.f. = 23/26

Aharonian et al. (2009)A&A acceptedarXiv:0906.1247

HESS J1745-290Spectrum


Aharonian et al. (2009)A&A acceptedarXiv:0906.1247

HESS J1745-290Spectrum


● 2004-2006 data93 h live time

● 4185 γ-rays (61 σ)160 GeV < E < 70 TeV

● Exponential cut-offΓ = 2.10±0.04±0.10Ec = 15.7±3.5±2.5 TeVχ²/d.o.f. = 23/26

● Broken powerlawΓ1 = 2.02±0.08±0.10Γ2 = 2.63±0.14±0.10EB = 2.57±0.19±0.44χ²/d.o.f. = 20/19

10''

Sgr A*

G359.95

Wang et al. (2005)

Hinton + Aharonian (2007)

● Dense radiation fields● Comparably low

magnetic field→ IC dominant→ plausible candidate


HESS J1745: a pulsar wind nebula?

● All models viablewith current statistics

● CTA/AGIS will help● LAT?

Sgr A* Emission Modelspp interactionsin accretion disk

Aharonian & Neronov (2005)

electron scenariocurvature + IC

Cosmic Ray Electrons


● Suffer severely from synchrotron and inverse Compton losses → steep GeV spectrum ~E-3.3 → steepening at TeV energies ~E-3.9 → TeV electrons must come from local sources

● Compatible with lower-energy measurements: Г = 3.1 with cut-off at 2.1 TeV

● H.E.S.S. can measure electrons at TeV energies → electrons are gamma-like → large detection area

● Large backgrounds - Cosmic ray showers - Galactic diffuse emission - extragalactic diffuse emission

Standard Background Modelling

Ber

ge e

t al.

(200

7)


● Random Forest: train machine learning algorithm on shower image parameters → needs electron simulations and cosmic background for training

● For each shower, RF determines “electron likeness” parameter ζ ε [0;1]

● For ζ>0.9, total background suppression is 10-6

● Signal extraction → Fit ζ-distribution with combination of electron/proton simulations → depends on hadronic interaction model (Sybill/QGSJet)

data

simulatedbackground

Electrons: Background Modelling


● Extrapolation of gamma-ray flux to VHE energies suggests small contribution only

● FERMI preliminary extragalactic diffuse gamma flux softer than EGRET

● Test with first interaction height (only poorly reconstructed)

● At most 50% gamma contamination

Gamma-ray contamination



● Separate fits in energy bands

● Two complementary analyses: - high energies: 600 GeV – 5 TeV (hard cuts for best reconstruction) - low energies: 340 GeV – 700 GeV (looser cuts on image intensity, 2004/2005 data only)

Energy Spectrum


Aharonian et al. (2009) arXiv:0905.0105


● No indication of feature similar to ATIC

● Break in spectrum: Г1 = 3.0±0.1±0.3 Г2 = 4.1±0.3±0.3 EB = 0.9±0.1 TeV

● Compatible to FERMI within energy shift uncertainty

Aharonian et al. 2009


Low Energy Analysis

Aharonian et al. (2009) arXiv:0905.0105

Meade et al. (2009)arXiv:0905.0480

Putting Electrons and GC together

Summary● Solid detection of the GC point source● Sgr A East excluded as a source● After 100 hours of observation, spectrum shows significant deviation simple power-law● No indication for variability● No indication for QPOs

● Measurement of CR electrons (+ extragalactic diffuse gammas)● Implies existence of nearby sources● Energy range 340 GeV – 5 TeV● Consistent with FERMI● No indication for ATIC spectral feature● Significant spectral steepening beyond 1 TeV

Thanks!

Slide 1Slide 2Slide 3Slide 4Slide 5Slide 6Slide 7Slide 8Slide 9Slide 10Slide 11Slide 12Slide 13Slide 14Slide 15Slide 16Slide 17Slide 18Slide 19Slide 20Slide 21Slide 22Slide 23Slide 24Slide 25Slide 26Slide 27

Documents

Cosmic Ray Electrons and GC Observations with H.E.S.S. · 2009. 7. 20. · Cosmic Ray Electrons Aharonian et al. (2008) Suffer severely from synchrotron and inverse Compton losses