TeV Diffuse Galactic Gamma-Ray Emission
Sabrina Casanova
Max Planck Institut fuer Kernphysik, Heidelberg
Surveys of the sky at TeV energies
Milagro: Northern sky
H.E.S.S. : Southern sky
Outline
1. CR origin and propagation are not well known phenomena
2. Diffuse gamma-ray emission: a unique probe of CR origin and
propagation
3. The origin of Galactic Cosmic Rays: searching for the Galactic
Pevatron Sources
4. Cosmic Ray Propagation in the Galaxy
•
Diffuse gamma rays from molecular clouds close to CR sources
•
Large scale TeV diffuse emission
5. On the level of the CR background
6. Summary & Future experimental goals
1. CR origin and propagation are not well
known phenomena
The Cosmic Ray Spectrum at Earth
• Almost featureless
spectrum
δ = -2.7
• Galactic accelerators
have to inject particles up
to at least PeV energies
• Isotropy up to very high
energy: 1 TeV proton in
B=1uG, Gyro-Radius =
200AU= 0.001pc
Galactic Origin
Knee ~ 3 x10 15 eV
δ = -3.0
Extragalactic Origin
Standard Model of Cosmic Rays
Qsource: DSA
 Ein SNRs
Acceleration
-2 to - 2.4
dNobserved
 escape Q source 
dEobserved
 escape  E
Energy dependent
Escape
-0.3
to -0.7
Key Questions in High Energy Astrophysics
The origin of cosmic rays
•
What are the cosmic ray accelerators ?
•
How high in energy can galactic sources produce particles?
•
Are CR accelerators different from electron sources?
•
How are cosmic ray sources distributed in the Galaxy ? Is there
a nearby (<10's of parsecs) source of cosmic rays ?
The character of propagation of cosmic rays
•
How do cosmic rays propagate in the Galaxy ?
•
What is the production rate of protons and electrons ?
2. Diffuse Gamma Ray Emission from the
Galactic CR population :
a unique probe of CR origin and propagation
g-rays from electron and CR Galactic populations
 Electromagnetic Processes:
• Synchrotron Emission
-important for e- energy losses – E g  (Ee/mec2)2 B
• Inverse Compton Scattering
– E f ~ (Ee/mec2)2 E i
• Bremsstrahlung
-more important at GeV energies– E g ~ 0.5 E e
 Hadronic Cascades. Gamma Rays are emitted through decay of
neutral pions produced by CRs interacting with the ambient gas. Also
secondary electrons and neutrinos are produced.
• p + p -> p± +po +… -> e ±+ n + g +…
Diffuse TeV g-rays from the Galaxy:
unique probe of CR origin and propagation
•
g-rays travel rectilinearly through the Galactic magnetic
fields and point back to the production locations
•
The highest energy CR particles produce the highest
energy g-rays.
•
Hadronic and leptonic mechanisms produce different
energy spectra and g-ray morphology. However it is
experimentally not easy to detect such differences.
•
Gamma ray spectrum and spatial distribution of the
diffuse gamma ray emission provide spectra and density
of hadrons and leptons in the different regions of the
Galaxy
3. The origin of Galactic Cosmic Rays :
searching for the Galactic Pevatron Sources
Are SNRs the Galactic PeVatron sources ?
• TeV g-rays and shell type
morphology : acceleration of eand protons
• There is a significant γ-ray flux
(4.8 σ) above 30 TeV. In the
hadronic scenario, this implies
efficient proton acceleration up
to 200 TeV with slope = -2 and
energy in CRs of about 1050
erg/s. In the leptonic scenario
this implies acceleration up to
100 TeV because of KN losses.
Not yet the knee.
• Correlation TeV/X-ray.
• IC and hadronic scenarios can
both explain the g-ray emission.
We do not have a conclusive
proof of CR acceleration.
SNR RX J1713-3946 seen in TeVs
(H.E.S.S. Collaboration)
Γ=
 E 
d

 I 0 E -  exp  dE
E
0


2.04±0.04
E0 = 17.9 ± 3.3 TeV
 E 
d

 I 0 E - exp  dE
E
0 

Variability of X-rays on year timescales –
witnessing particle acceleration in real time
•flux increase - particle acceleration
• flux decrease - synchrotron cooling
• it requires B-field of order 1 mG in
hot spots and, most likely, 100G
outside. This supports the idea of
amplification of B-field by in strong
nonlinear shocks through non-resonant
streaming instability of charged energetic
particles (Bell, 2000
Uchiyama, Aharonian, Tanaka, Maeda, Takahashi, Nature 2007
Zirakashvili, Ptuskin Voelk 2007)
Are SNRs the Galactic PeVatron sources?
DSA in SNRs accelerates electrons -> DSA is likely to
accelerate protons. However :
 Quantify the electron vs proton content in SNRs : is VHE
emission from SNRs leptonic or hadronic in nature ?
 What is the max proton energy achieved ?
 How do CRs escape SNRs and diffuse in the ISM ?
– PeV particles are accelerated at the beginning of Sedov phase
(~200yrs), when the shock speed is high. PeV particles quickly
escape as the shock slows down. Lower energy particles are
released later.
– Even if a SNR accelerates particles up to PeV energies, a SNR
is a PeVatron for a very short time
– There is still no evidence for the existence of escaping CRs
4. Cosmic Ray Propagation in the Galaxy
Diffuse gamma rays from molecular
clouds close to CR sources
Gamma Rays from Molecular Clouds close to CR sources
 MCs are ideal targets to amplify the emission produced by CRs
accelerated by nearby sources (Aharonian & Atoyan, 1996,
Aharonian, 2001)
 Highest energy protons quickly escape the accelerator and
therefore do not significantly contribute to gamma-ray production
inside the proton accelerator-PeVatron. Gamma rays from nearby
MCs are therefore crucial to probe the highest energy protons .
Cosmic ray flux in MCs close to a SNR:
dependence on SNR age and location
Gabici et al, 2009
1 = 500 yr
2 = 2000 yr
3 = 8000 yr
4 = 32000 yr
Power in CRs: ECR = 3 X 1050 erg,
Diffusion coefficient: D = D0 E0.5
GeV versus TeV emission spectra in MC
1 = 500 yr
2 = 2000 yr
3 = 8000 yr
4 = 32000 yr
GeV
steady
CR sea
TeV
SNR CRs
t-dependent
Concave shape
Steep spectrum
at GeV energies
and hard at TeV
M=105 solar masses; R=20pc ;n =120 cm-3 ; B=20G; D=1kpc ;D =1028 cm2 /s
Gabici et al 2009
MWL implications
• GeV-TeV connection
• ...and PeV-hard X connection (pp
interactions produce secondary electrons)
Ee = 100 TeV
 B  Ee 

 20
 keV
 30G  100TeV 
2
Esyn
Multiwavelength emission close to MCs
M=105 solar mass; R=20pc ;n=120cm-3 ; B=20G ; d (SNR/MC)=100pc ; D=1kpc
Gabici et al, 2009
t = 500 yr
t = 2000 yr
t = 8000 yr
t = 32000 yr
t = ∞ yr
Synchrotron of
secondaries and
primary e-
RADIO
X-RAYS
GAMMA RAYS
The gamma-ray emission is an order of magnitude higher than the emission
at other wavelengths. So far unidentified TeV sources (dark sources) could
be MCs illuminated by CRs from a nearby SNR.
Modeling particle escape from RXJ1713-3946
Zirakashvili & Aharonian, 2010
CRs
CRs
CLOUD
SNR
gammas
CRs
gammas
gammas
Assumptions :
1050 erg
CRs of about 150
TeV are escaping
from the shell now
•
Energy in CRs: ECR = 3 X
•
Injection spectrum: f(E) = K E-2
•
Diffusion coefficient: D = D0 E0.5
•
Source age and location t= 1600 yr, d = 1kpc
Variety of CR spectra close to RXJ1713
c
GeV
steady
TeV position
dependent
Variety of g-ray spectra close to RXJ1713
Constraints on
diffusion regime
Casanova et al, 2010
The emission from the environment of RX J1713
provides clues concerning the SNR accel history.
D = 1026 E0.5 cm2/s at 10 GeV
6 TeV
SNR RXJ1713
SNR RXJ1713
CR sea
(Runaway CRs + sea CRs) /sea CRs
Morphological studies of the emission close to CR
sources: constraining CR diffusion regime
D = 1027 E0.5 cm2/s at 10 GeV
D0 = 1028 E0.5 cm2/s at 10 GeV
Casanova et al, 2010
Molecular Clouds close to the SNR W 28
H.E.S.S. collaboration, F. Aharonian et al.,
and Y. Moriguchi, Y. Fukui, NANTEN, 2008
W28 : an old SNR and nearby molecular
clouds.The VHE/molecular cloud
association could indicate a hadronic
origin for HESS J1801-233 and HESS
J1800-240
HESS observations of the diffuse emission from the GC
(Aharonian et al, 2006)
Spectral index
2.29 ± 0.07 ± 0.20
implies harder
CR spectrum than in
solar neighborhood  Proximity
of accelerator and target
Molecular Clouds close to the SNR IC 443
•seen by MAGIC & VERITAS
• 12CO emission line intensity by Dame
et al, 2001 (cyan line)
•contours of 20 cm VLA radio data
from Condon et al. (1998) (green)
•1720 MHz OH maser indication for a
shock in a high matter density
environment (black point)
•Xray contours from Rosat (purple)
(Asaoka & Aschenbach 1994)
•Gamma –ray contours from EGR-ET
(Hartman et al. 1999) (black)
•gamma-ray excess coincides with
cloud and masers
•soft spectrum: Γ = 3.1 ± 0.3
[Albert et al. ApJ 664L 87A 2007
4. Cosmic Ray Propagation in the Galaxy
Large scale TeV diffuse emission
Milagro analysis of Galactic TeV diffuse emission
Flux profiles ← Source subtraction
+ offset/base
Galactic Latitude Profile
GALPROP Model
hadronic
leptonic
total
(Abdo et al, 2008)
The narrow distributions of Milagro data points agrees
better with a dominant hadronic component.
Galactic Longitude Flux Profile
Longitude Profile |b|<2°
Cygnus
Region
1.5x GP
2.7x GP
Abdo et al, 2008
Sources Subtracted
GALPROP (optimized)
Below Horizon
There is an excess of diffuse TeV gamma rays from the Galactic plane
• Additional unresolved sources (Casanova & Dingus, 2008)?
• Cosmic-ray acceleration sites? Most significant discrepancy
between data and model predictions in the Cygnus region.
Diffuse Emission from the Cygnus Region
Strong & Moskalenko
GALPROP model of
Cygnus Region
Abdo et al, 2007
Cygnus Region: Matter Density
Contours overlaying Milagro Obs.
Pevatrons in the Cygnus Region ?
MGRO J2031+41
100 pc
The extension of the entire Cygnus region is about 300 pc, and thus a
single accelerator might influence strongly the entire region
In fact only one or at most a few strong young accelerators in the Cygnus
region are needed to explain the excess emission measured by Milagro
with respect to GALPROP
5. On the level of the cosmic ray background
On the level of the cosmic ray background
 Average CR flux (“CR sea”) is produced by the effective
mixture of CRs from individual sources diffusing in the
Galaxy for times of 107 years
 CR at Earth is representative of the average CR flux ?
 Observational test : MCs as cosmic ray barometers
• Assuming CR sea = CR at Earth
• An observed gamma-ray flux from MC less than
predicted would imply a CR density less than that
observed at Earth
The level of the CR background
only one or at most few massive MCs
sr-1 cm-2 s-1 TeV-1
gammas
CLOUDS
CR sea
gammas
many low mass MCs
100 GeV
Casanova et al, 2010
Casanova et al, 2010
Localisation of the peak emission along the line of sight
MASS
DENSITY
Flux >1 Gev
The emission from MCs constrains the CR flux
in specific regions of the Galaxy
Casanova et al, 2010
CR= g MH2 /dH22
Mass and integral flux as a function of
the distance for the region 345.3<l<346
and 1<b<1.7
85% of the emission for the region
345.3<l<346 and 1<b<1.7 is produced
between 0.5 and 3 kpc
Summary
 The CR origin and propagation are still uncertain.
 TeV diffuse gamma-ray emission is a unique probe
of CR origin and propagation
 TeV diffuse emission has been observed from MCs
close to CR sources by HESS,MAGIC and Veritas
and on a larger scale by Milagro
 MCs close to CR sources are ideal targets to amplify
the emission produced by CRs penetrating the cloud
and to produce very specific observable features in
the gamma-ray emission
 Diffuse gamma-ray from MCs far away from known
CR sources can help constraining the level of the
CR sea.
Future experimental goals
 Detect large (>>10G) magnetic field : x-ray telescopes
such as Chandra and Astro-H
 Observe g-ray at hundreds TeV : CTA, AGIS, HAWC
 Observed X-rays produced by secondary electrons
 Observe neutrinos : Icecube and KM3NET
 Image the spectrum and spatial distribution of the Galactic
diffuse emission
 Compare the source g-ray images, spectra, and variability
with those from other wavelengths (for instance with radio)
 Detect many sources of many different classes : population
studies with CTA, AGIS, HAWC
Thanks for the invitation !