Galactic Cosmic Rays &
Diffuse Gamma-Ray emission
Igor V. Moskalenko (Stanford U.)
OSU, Columbus, May 1, 2007
GLAST LAT
Cosmic Rays vs Gamma Rays
CR abundances
Yamamoto+ ‘07
Solar system
abundances
pbar
Even an unrealistic model (e.g. Leaky-Box) can be fitted to the CR data
Igor V. Moskalenko 2
May 1, 2007
OSU, Columbus
GLAST LAT
Diffuse VHE γ-ray from the Galactic Center
Igor V. Moskalenko 3
May 1, 2007
OSU, Columbus
2007A Closer
Milagro
Sky
Survey
AtPlane
12 TeV
Look
at the
Galactic
Cygnus region shows three new TeV gamma-ray sources
 Diffuse emission from Cygnus region
 A new TeV source closer to the Galactic Center

AOUS ABDO – PhD. Thesis Defense
Michigan State University, March 28, 2007
Diffuse Emission from Cygnus Region
l(65,85), b (-3,3)
• Exclude a region of 3°×3° around MGRO
J2019+37 and MGROJ2033+42
• Diffuse flux (×10-10 TeV cm-2 s-1 sr-1)
= 4.18 ± 0.52stat ± 1.26sys
~ 2× Crab flux
• Strong & Moskalenko “Galprop” model
– Milagro flux ~ 7x conventional model
of Galprop
– Milagro flux ~3x optimized model
• Hard spectrum cosmic ray sources?
• Unresolved point sources?
Abdo A. A. et al., ApJL 658, L33
AOUS ABDO – PhD. Thesis Defense
Michigan State University, March 28, 2007
EGRET
Milagro
GLAST LAT
Diffuse Galactic Gamma-ray Emission
~80% of total Milky Way luminosity at HE !!!
Tracer of CR (p, e−) interactions in the ISM (π0,IC,bremss):
o
o
o
o
Study of CR species in distant locations (spectra & intensities)
 CR acceleration (SNRs, pulsars etc.) and propagation
Emission from local clouds → local CR spectra
 CR variations, Solar modulation
May contain signatures of exotic physics (dark matter etc.)
 Cosmology, SUSY, hints for accelerator experiments
Background for point sources (positions, low latitude sources…)
Besides:
o “Diffuse” emission from other normal galaxies (M31, LMC, SMC)
 Cosmic rays in other galaxies !
o Foreground in studies of the extragalactic diffuse emission
o Extragalactic diffuse emission (blazars ?) may contain signatures
of exotic physics (dark matter, BH evaporation etc.)
Calculation requires knowledge of CR (p,e) spectra in the entire Galaxy
Igor V. Moskalenko 6
May 1, 2007
OSU, Columbus
GLAST LAT
CR Propagation: Milky Way Galaxy
Optical image: Cheng et al. 1992, Brinkman et al. 1993
Radio contours: Condon et al. 1998 AJ 115, 1693
NGC891
1 kpc ~ 3x1018 cm
Halo
0.1-0.01/ccm
Sun
R Band image of NGC891
1.4 GHz continuum (NVSS), 1,2,…64 mJy/ beam
Igor V. Moskalenko 7
May 1, 2007
Intergalactic space
“Flat halo” model (Ginzburg & Ptuskin 1976)
OSU, Columbus
GLAST LAT
CR Interactions in the Interstellar Medium
SNR RX J1713-3946
42 sigma (2003+2004 data)
ISM
X,γ
+
e-
HESS
B
P diffusion
He energy losses
CNO
reacceleration
+
convection e etc. π +-
IC
ISRF
gas
π0
GLAST
gas
_
P
+
π- p
+
e-
LiBeB
He
CNO
Flux
PSF
Chandra
20 GeV/n
BESS
PAMELA
Igor V. Moskalenko 8
May 1, 2007
helio-modulation
ACE
CR species:
 Only 1 location
 modulation
OSU, Columbus
GLAST LAT
Elemental Abundances: CR vs. Solar System
CR abundances: ACE
O
Si
Na
Fe
S
CNO
Al
Cl
LiBeB
CrMn
F
ScTiV
Solar system abundances
Long propagation history…
Igor V. Moskalenko 9
May 1, 2007
OSU, Columbus
GLAST LAT
Transport Equations ~90 (no. of CR species)


 ( r , p, t )
 q( r , p ) sources (SNR, nuclear reactions…)
t
diffusion

   [D


  V ]
xx
  2
  
p D

diffusive reacceleration 
pp
2

p

p

p 
(diffusion in the momentum space)


E-loss 
fragmentation 
(Galactic wind)
  dp
1   


p  V  

p  dt
3






f
radioactive decay
d
+ boundary conditions
Igor V. Moskalenko 10
convection
May 1, 2007
ψ(r,p,t) – density
per total momentum
OSU, Columbus
GLAST LAT
How It Works: Fixing Propagation Parameters
E2 Flux
B/C
Carbon
Radioactive isotopes:
Galactic halo size Zh
Ek, GeV/nucleon
Be10/Be9
Ek, MeV/nucleon
Using secondary/primary nuclei ratio & flux:
•Diffusion coefficient and its index
•Propagation mode and its parameters (e.g.,
reacceleration VA, convection Vz)
Zh increase
Ek, MeV/nucleon
Igor V. Moskalenko 11
May 1, 2007
OSU, Columbus
GLAST LAT
A Model of CR Propagation in the Galaxy
 Gas distribution (energy losses, π0, brems)
 Interstellar radiation field (IC, e± energy losses)
 Nuclear & particle production cross sections
 Gamma-ray production: brems, IC, π0
 Energy losses: ionization, Coulomb, brems, IC, synch
 Solve transport equations for all CR species
 Fix propagation parameters
 “Precise” Astrophysics
Igor V. Moskalenko 12
May 1, 2007
OSU, Columbus
GLAST LAT
Do we understand cosmic ray propagation?
Basic features – Yes
Igor V. Moskalenko 13
May 1, 2007
OSU, Columbus
GLAST LAT
Wherever you look, the GeV -ray excess is there !
EGRET data
4a-f
Igor V. Moskalenko 14
May 1, 2007
OSU, Columbus
GLAST LAT
Reacceleration Model vs. Plain Diffusion
Antiproton flux
B/C ratio
Plain Diffusion
(Dxx~β-3 R0.6)
B/C ratio
Antiproton flux
Diffusive
Reacceleration
Igor V. Moskalenko 15
May 1, 2007
OSU, Columbus
GLAST LAT
Positron Excess ?
HEAT (Beatty et al. 2004)
e+/e
e+/e
E > 6 GeV
GALPROP
HEAT 2000
HEAT 1994-95
10
1
HEAT combined
1
E, GeV
GALPROP
10
E, GeV
Q: Are all the excesses connected?
A: “Yes” and “No”
Systematic errors of different detectors
Same progenitor (CR p or DM) for pbars, e+’s, γ’s
Igor V. Moskalenko 16
May 1, 2007
OSU, Columbus
GLAST LAT
Do we really understand cosmic ray
propagation?!
Not sure…
If these are not instrumental artifacts…
What we can do 
Igor V. Moskalenko 17
May 1, 2007
OSU, Columbus
GLAST LAT
Electron Fluctuations/SNR stochastic events
GeV electrons
100 TeV electrons
E(dE/dt)-1,yr
GALPROP/Credit S.Swordy
107 yr
6
10 yr
Electron energy loss timescale:
1 TeV: ~300 kyr
100 TeV: ~3 kyr
Energy losses
Bremsstrahlung
Ionization
IC, synchrotron
Coulomb
1
GeV
1 TeV
Ekin, GeV
Igor V. Moskalenko 18
May 1, 2007
OSU, Columbus
GLAST LAT
CR fluctuations
Case & Bhattacharya 1998
SNR number density
R, kpc
p
local average
pbar
local average
Igor V. Moskalenko 19
May 1, 2007
OSU, Columbus
GLAST LAT
GeV excess: Optimized/Reaccleration model
Uses all sky and antiprotons & gammas
to fix the nucleon and electron spectra



antiprotons
Uses antiprotons to fix
the intensity of CR nucleons @ HE
Uses gammas to adjust
 the nucleon spectrum at LE
 the intensity of the CR electrons
(uses also synchrotron index)
Uses EGRET data up to 100 GeV
electrons
Ek, GeV
protons
x4
x1.8
Ek, GeV
Igor V. Moskalenko 20
Ek, GeV
May 1, 2007
OSU, Columbus
GLAST LAT
Secondary e± are seen in γ-rays !
electrons
Heliosphere:
e+/e~0.2
sec.
IC
positrons
brems
Improves an agreement at LE
Igor V. Moskalenko 21
May 1, 2007
OSU, Columbus
GLAST LAT
Anisotropic IC Scattering in the MW
 Electrons in the halo see anisotropic radiation
 Observer sees mostly head-on collisions
Energy density
e-
R=4 kpc
small boost &
less collisions
head-on:
large boost &
more collisions
γ
γ
Z, kpc
γ
Important @
high latitudes !
Igor V. Moskalenko 22
e-
sun
May 1, 2007
OSU, Columbus
GLAST LAT
Effect of anisotropic ICS
Ratio anisoIC/isoIC
pole
anti-GC
GC
• The anisotropic IC scattering
plays important role in
modeling the Galactic diffuse
emission
• Affects estimates of isotropic
extragalactic background
Igor V. Moskalenko 23
May 1, 2007
Intermediate latitudes
Galactic latitude, degrees
OSU, Columbus
GLAST LAT
Latitude profile of the outer Galaxy
anisotropic IC
0
Total
EG
isoIC
bremsstrahlung
• Agreement with
data impossible
without aniso IC
Latitude
Igor V. Moskalenko 24
• The aniso IC is
maximal (x2) in the
outer Galaxy around
b=20 -30
May 1, 2007
OSU, Columbus
GLAST LAT
Diffuse -ray spectrum of the inner Galaxy
conventional model
optimized model
EGRET
EGRET
COMPTEL
COMPTEL
IC
brems
total
total
brems
EG

IC
EG

The optimized model based on modified cosmic ray spectra
reproduces the EGRET skymaps. GLAST data on diffuse emission are
critical to distinguish between the models and to provide valuable
information on cosmic ray spectra in distant regions of the Galaxy.
Igor V. Moskalenko 25
May 1, 2007 2007, Ann.Rev. 57, in press (astro-ph/0701517)
OSU, Columbus
Strong+
GLAST LAT
Longitude Profiles |b|<5°
50-70 MeV
2-4 GeV
Igor V. Moskalenko 26
May 1, 2007
0.5-1 GeV
4-10 GeV
OSU, Columbus
GLAST LAT
Optimized model
EGRET
COMPTEL
Igor V. Moskalenko 27
May 1, 2007
OSU, Columbus
Diffuse -ray emission model
GLAST LAT
Combined skymaps
Separate components
160 MeV
brems
1.28 GeV
IC
82 GeV
Igor V. Moskalenko 28

May 1, 2007
OSU, Columbus
GLAST LAT
Synchrotron, Northern Galaxy
WMAP
B-field
Igor V. Moskalenko 29
May 1, 2007
OSU, Columbus
GLAST LAT
Porter & Strong
Igor V. Moskalenko 30
May 1, 2007
OSU, Columbus
GLAST LAT
•
Distribution of interstellar gas
Neutral interstellar medium – most of the interstellar gas mass
– 21-cm H I & 2.6-mm CO (surrogate for H2)
– Differential rotation of the Milky Way – plus random motions, streaming,
and internal velocity dispersions – is largely responsible for the spectrum
– Rotation curveV(R)  unique line-of-sight velocity-Galactocentric distance
relationship
CO
Rotation Curve
Dame et al.
(2001)
HI
Kalberla et al.
(2005)
Clemens (1985)
W. Keel
•
•
This is the best – but far from perfect – distance measure available
Column densities: N(H2)/WCO ratio assumed; a simple approximate
correction for optical depth is made for N(H I); self-absorption of H I
remains
Igor V. Moskalenko 31
May 1, 2007
OSU, Columbus
Column densities of gas
GLAST LAT
•
•
WCO
•
N(H I)
Igor V. Moskalenko 32
May 1, 2007
Here are examples of the
resulting ‘rings’
For the local (7.5-9.5
kpc) annulus we are
incorporating new
intermediate latitude CO
survey data (Dame 2007)
and additional coverage
from the NANTEN survey
in the south (Onishi,
Mizuno, & Fukui 2004)
We are also investigating
incorporating a ‘dark’
component of molecular
gas not traced by CO
(Grenier, Casandjian, &
Terrier 2005)
OSU, Columbus
GLAST LAT
Gammas from neutral pion decay pp0
 New parameterization
(Kamae+ 2005, 2006) is
based on Pythia Monte
Carlo event generator and
includes diffraction
dissociation
Pion decay -ray spectra for different
regions on the sky
Kamae+ 2006
GALPROP old
 New parameterization
shows some improvement
over the old formalism
employed in GALPROP
 Galprop now has a
parameter to choose a
formalism
Igor V. Moskalenko 33
May 1, 2007
OSU, Columbus
GLAST LAT
Dark Matter package & DarkSUSY
• DM package is now a part of the GALPROP distribution
– Allows a user to define the DM density profile and spectra
for annihilation products (pbar, e±, ) and propagate them
throughout the Galaxy, calculate the skymaps of -rays
produced in DM annihilation
DarkSUSY – GALPROP interface
(Baltz―Moskalenko, available soon)
– GALPROP can now be called from the
DarkSUSY to calculate the Green’s functions
of particle propagation in the Galaxy
• Can work down to 1 keV in electron kinetic energy (not in
the public version yet)
– Allows particle propagation down to very low energies
Igor V. Moskalenko 34
May 1, 2007
OSU, Columbus
GLAST LAT
Extragalactic Gamma-Ray Background
E2xF
EGRB in different
directions
Sreekumar+ ‘98
Dermer’07
Predicted vs. observed
Elsaesser & Mannheim,
astro-ph/0405235
Strong+ ‘04
E, MeV
• Blazars
• Cosmological
neutralinos
Igor V. Moskalenko 35
May 1, 2007
OSU, Columbus
GLAST LAT
GALPROP Web-site
galprop.stanford.edu
 This Web site is dedicated to
research in astrophysics of
cosmic rays and diffuse gamma
rays. It is designed to be a
communication forum between
researchers in different
disciplines.
 Systematic work on evaluation of
the codes and data posted on the
Web-site, cross tests of
different propagation models and
approaches, should make the
calculations of propagation in the
interstellar space and in the
heliosphere more reliable.
Igor V. Moskalenko 36
May 1, 2007
OSU, Columbus
GLAST LAT
What Pamela & GLAST can clarify
Igor V. Moskalenko 37
May 1, 2007
OSU, Columbus
GLAST LAT
Igor V. Moskalenko 38
May 1, 2007
E.Bloom’05
OSU, Columbus
GLAST LAT
Where is the DM ?!





Flavors:
Neutrinos ~ visible matter
Super-heavy relics: “wimpzillas”
Axions
Topological objects “Q-balls”
Neutralino-like, KK-like
Places:
 Galactic halo, Galactic center
 The sun and the Earth
Tools:
 Direct searches
– low-background experiments
(DAMA, EDELWEISS)
– neutrino detectors
(AMANDA/IceCUBE)
– Accelerators (LHC)
 Indirect searches
– CR, γ’s (PAMELA,GLAST,BESS)
from E.Bloom presentation
Igor V. Moskalenko 39
May 1, 2007
OSU, Columbus
GLAST LAT
PAMELA antiprotons
After 3 years
• Dark Matter signatures
• Diffuse Galactic gamma-ray emission
• CR propagation (pbar spectrum is
different from other CR species) – their
source spectrum is well known (if we
know CR protons)
• Heliospheric modulation
Igor V. Moskalenko 40
May 1, 2007
OSU, Columbus
GLAST LAT
PAMELA positrons
After 3 years
• Dark Matter signatures
• Diffuse Galactic gamma-ray emission
• CR propagation (pbar spectrum is
different from other CR species) – their
source spectrum is well known (if we
know CR protons)  A factor of 2 will become
statistically significant
• Heliospheric modulation
 Measuring absolute flux not ratio
 Solar minimum
conditions
• Local sources of primary
positrons
Igor V. Moskalenko 41
May 1, 2007
OSU, Columbus
PAMELA: Secondary to Primary ratios
 LE: sec/prim peak:
one instrument -no
cross calibration
errors
 HE: Dxx(R)
Data plots: M.Simon
GLAST LAT
GLAST Large Area Telescope (LAT)
This is an animation that steps from 1.
EGRET (>100 MeV), to 2. LAT (>100 MeV),
to 3. LAT (>1 GeV)
EGRET
Simulated LAT
(>1 GeV, 1 yr)
(>100 MeV)
Seth Digel
Igor V. Moskalenko 43
May 1, 2007
OSU, Columbus
GLAST LAT
A.Morselli
Igor V. Moskalenko 44
May 1, 2007
OSU, Columbus
GLAST LAT
Electron Spectrum in the Heliosphere
 GLAST LAT is expected to detect ~107
electrons/yr above 20 GeV, 4×105 electrons/yr
above 100 GeV, and ~2,500 electrons/yr above
500 GeV assuming a steep power law electron
spectrum with power index -3.3.
 Energy range ~20 GeV--2 TeV
 Local CR sources (pulsars, SNRs)
 Diffuse emission & CR propagation
 IC scattering in the heliosphere
Igor V. Moskalenko 45
May 1, 2007
OSU, Columbus
GLAST LAT
The Excess: Clues from the Local Medium
Positions of the local clouds
Observations of the local medium in different
directions, e.g. local clouds, will provide a clue to the
origin of the excess (assuming it exists).
Inconclusive based on EGRET data
Will GLAST see the excess?
sun
Yes
No
Pohl et al.2003
EGRET data
Poor knowledge of
π0-production cross
section:
better understanding of
π0-production
Possibility:
cosmic-ray spectral
variations.
Further test: look at
more distant clouds
Dark Matter signal:
look for spectral signatures
in cosmic rays (PAMELA,
BESS, AMS) and in
collider experiments (LHC)
Digel et al.2001
Igor V. Moskalenko 46
May 1, 2007
OSU, Columbus
GLAST LAT
Inverse Compton scattering
e
•
•
•
•
QED
AGN
SNR
Accretion disks
ISM
©UCAR
The heliosphere is filled with Galactic CR
electrons and solar photons
•electrons are isotropic
•photons have a radial angular distribution
IM,Porter,Digel: ApJ 648 (2006)L65
Igor V. Moskalenko 47
May 1, 2007
OSU, Columbus
GLAST LAT
Heliosphere
FluxIC ~1/r


r1 (AU) = sin, <90°
r1 (AU) = 1,
>90°
r
r2=10r1
r2
Looking in different
directions one can
probe the e-spectrum
at different distances
from the sun!
Igor V. Moskalenko 48
r1
May 1, 2007
e
OSU, Columbus
GLAST LAT
Found in EGRET data !
Thompson+ 1997:
Upper limit 2x10-7 cm-2 s-1
Reanalysis by Petry,
Orlando, Strong 2007:
Discovery of both solar
disk pion-decay emission
and extended inverse
Compton-scattered
radiation in combined
analysis of EGRET data
from June 1991!!
Igor V. Moskalenko 49
May 1, 2007
OSU, Columbus
GLAST LAT
Conclusion
Combined efforts of PAMELA, GLAST,
and other missions will bring together
astronomers, astrophysicists, and
particle physicists and provide
valuable contributions to:
•Search for Dark Matter signatures
•Studies of cosmic particle accelerators
(Galactic & extragalactic)
•Studies of cosmic ray propagation in the
Galaxy & the heliosphere
•Understanding the Milky Way galaxy and
other normal galaxies
•Understanding the Universe
Igor V. Moskalenko 50
May 1, 2007
OSU, Columbus