EGRETs Excess of Diffuse Gamma Rays as Dark Matter Tracer

1
EGRETs Excess of Diffuse Gamma Raysas Dark
Matter Tracer

• X.-J. Bi, J. Zhang, Q. Yuan, J.-L. Zhang, and H.
  Zhao, astro-ph/0611783
• W. de Boer, C. Sander, V. Zhukov, A. V.
  Gladyshev, and D. I. Kazakov,
• Astron. And Astrophysics 444 (2005) 51

• Pearl Sandick
• University of Minnesota

2
Energetic Gamma Ray Experiment Telescope

• Compton Gamma Ray Observatory
• Flight Duration 4/5/91 to 6/4/00
• EGRET Energy Range 20 MeV to 30 GeV

3
Principle obective detailed study of high
energy gamma ray emitting sources
4
EGRET Energy Spectrum

• Background Components
• p0 ? ??
• Inverse Compton Scattering
• Bremsstrahlung from electrons
• Extragalactic backgrounds

Flux
De Boer et al., astro-ph/0408272
E (MeV)
5
Possible Explanations for Excess

• Harder nucleon spectrum M. Mori, ApJ 478 (1997)
  225
• Harder electron spectrum Moskalenko, Strong,
  Reimer, AA 338 (1998) L75
• Modified nucleon and electron spectra
• Strong, Moskalenko, Reimer, ApJ 537 (2000) 763

4. Dark matter annihilation (DMA) in the
galactic halo
6
Dark Matter Annihilation
W. de Boer, C. Sander, V. Zhukov, A. V.
Gladyshev, and D. I. Kazakov, Astron. And
Astrophysics 444 (2005) 51

• DM produced in thermal equilibrium with other
  particles in the early universe
• Number density of DM (n?) decreased to present
  value through annihilations
• All enhancements to annihilation rate (from
  clumping) calculated w.r.t generic cross section
• boost factor

7
DMA Products

• Stable DMA products are ?, ?, p, p, e-, e

8
DMA and BG Spectra

• Uncertainties in cosmic ray fluxes and gas
  densities for a given sky direction (20), so
  normalization of background is uncertain
• For a cosmic ray spectrum, spectral shape of
  gamma rays known
• Electron-induced gamma production can be
  calculated
• Nuclei-induced gamma production understood from
  accelerator experiments
• Spectral shape is well measured by EGRET because
  errors are correlated between neighboring points
• -gt Leave normalization as free parameter and use
  spectral shapes to disentangle background from
  DMA signal!

9
GALPROPbest estimate of galactic backgrounds

• Conventional
• Assumes local p,e spectra are representative of
  galactic spectra
• (not a great assumption for electrons because
  they have larger energy losses, so only use E? gt
  .07 GeV where electron-induced component is small)

Optimized Spectra optimized to explain EGRET
GeV excess without DM (does not work well for all
sky regions)

• Extragalactic backgrounds?
• Can only be obtained iteratively (Sander, 2005)
• Assumed uniform for all sky directions
• Contribution becomes important towards galactic
  poles, where galactic background and DMA are
  small

10
Sky Regions
11
Background and DMA Signal
Boost Factor ? 100 M? 60 GeV
de Boer et al. (2005)
12
Does DMA make sense?
WIMP mass dependence
flux minus background
13
Does BG normalization from fit agree with
absolute prediction from Strong et al. (2004)?
14
Optimized GALPROP (no DMA)
Fit probability lt 10-7
15
Optimized GALPROP with DMA
Boost Factor ? 30
Fit probability 0.8
16
Optimized GALPROP
Strong, Moskalenko, and Reimer, ApJ 613 (2004) 962

• Assume excess can be explained by modifying
  proton and electron injection spectra
• Average electron intensity (gt 20 GeV) is 4 times
  local intensity
• Average proton intensity (gt 10 GeV) is 1.8 times
  local intensity
• Strength of model reproduces locally measured
  positron and proton fluxes

17
Optimized GALPROP
Strong, Moskalenko, and Reimer, ApJ 613 (2004) 962
Solid lines are conventional GALPROP, dashed
lines are optimized GALPROP. Upper curves are for
interstellar spectrum, lower are with propagation
into the solar system.
18
DM Profile

• Can be obtained from directional dependence of
  DMA signal
• Finer sampling of sky directions
• Longitude 8 degree bins (45)
• Latitude 0-5, 5-10, 10-20, 20-90 degrees (4)
• 180 independent sky regions
• Assume B independent of r and n2 (no clustering)
• Plot flux as a function of longitude for each
  latitude bin

19
DM Halo Profile
cored isothermal a5 kpc (a,ß,?)(2,2,0)
20
Where is the dark matter?

• Allowed for up to 40 substructures
• Free parameters
• Radii
• Gaussian widths for size of ring (in and out of
  galactic plane)
• Found two ring-like structures (4 and 14 kpc)
• Enhanced radiation at these radii was already
  noted in original excess discovery paper
  Hunter et al. (1997)

21
DM Profile
cored isothermal with ring-like substructures at
4 and 14 kpc
22
Matter Distribution
?0
?inner ring
DM
?outer ring
disk
Baryonic Matter
bulge
23
Rotation Curve
24
Summary
W. de Boer, C. Sander, V. Zhukov, A. V.
Gladyshev, and D. I. Kazakov, Astron. And
Astrophysics 444 (2005) 51

• Excess has same spectrum in all sky directions,
  implying a common source
• Shape corresponds to expected spectrum from DMA
  of WIMPS with mass of 50-70 GeV
• Excess traces the DM distribution in our galaxy,
  proven by reconstruction of rotation curve
• Galactic Parameters (upper limit)
• Radius containing an average density of 200 ?c
  is R200 310 kpc
• Total DM mass inside R200 is M200 3.0 1012
  Msun (compared with visible mass of 5.5
  1010 Msun)
• Inner (outer) ring contribute 0.3 (3) of total
  DM mass

25
What about the antiproton flux?
L. Bergstrom, J. Edsjo, M. Gustafsson, and P.
Salati, JCAP 0605 (2006) 006

• de Boers model with a neutralino WIMP
  overproduces the antiproton flux observed by BESS
  by a factor of 2-10
• Did not find any MSSM scenarios that were
  compatible with both the BESS and EGRET
  measurements!

26
A Different Approach Bi et al., arXiv
astro-ph/0611783

• Directly compute background spectrum and DMA flux
• Best fit by adjusting propagation parameters in
  GALPROP (dist. of cosmic ray sources, scaling
  factor for density of CO)
• MSSM neutralino with m? 48.8 GeV
• Oh2 0.09 (de Boer et al. used Oh2 0.11)
• Subhalos enhanced ? flux dependent on sky region
• Spherically symmetric subhalo distribution
• DMA gamma ray flux supressed at small
  galactocentric radius by tidal disruption of
  subhalos
• Whole halo contributes to diffuse ? ray
  intensity, but only diffusion region (Rlt30 kpc)
  contributes to antiproton flux
• Different from universal boost factor of de Boer
  et al.

27
CO scaling factor - XCO

• (1.9 0.2)1020 cm-2/(K km s-1) from EGRET with
  E? 0.1 10 GeV
• .9 1.651020 cm-2/(K km s-1) from particular
  local clouds
• Fit to EGRET background finds smaller value
  XCO
  0.6 1.0 1020 cm-2/(K km s-1)

28
Gamma Ray Spectra
2.6
20
14
1.0
6.0
3.0
29
Antiproton Flux
Taking rings into account enhances smooth
component by about a factor of two.
30
Summary

• Smaller XCO favored
• Rings (de Boer et al.) necessary
• Very cuspy profile for subhalos
• Annihilation of MSSM neutralino DM with
  M 40 - 50 GeV can simultaneously
  explain the GeV excess of gamma rays measured by
  EGRET and the antiproton flux measured by BESS.