Nathan Kelley-Hoskins, for the VERITAS

Nathan Kelley-Hoskins, for the VERITAS

A Likelihood Procedure for Dark
Matter Searches of the Galactic
Center with VERITAS
Nathan Kelley-Hoskins, for the VERITAS Collaboration
TAUP 2015, Torino, Italy
Alliance for Astroparticle Physics
Indirect Detection
Galactic Center
Dark Matter
Improved
DM Limits
Likelihood
Analysis
Decays/
Annihilates
To
Gamma Rays
Are Detected By
Imaging
Atmospheric
Cherenkov
Telescopes
VERITAS
Observes Gamma Rays
Energy Resolution
PSF
Sensitivity
:
:
:
:
85 GeV - >30 TeV
15-25%
Containment68% < 0.1 deg at 1 TeV
1% Crab in ~25 Hours
Monday: “Highlights of recent results from the VERITAS gamma-ray observatory”, Lucy Fortson
Detecting Gamma Rays
http://community.dur.ac.uk/~dph0www4/images/detection.jpg
Galactic Center
Huge J-Factor (2 Orders of Magnitude Higher Than Dwarf Galaxies)
• Many non-DM sources
•
~0.5°: Moon Ø
Λ"90cm
T. N. LaRosa et al. 2000 The Astronomical Journal 119 207
Galactic
Center
VERITASGalactic
Galactic Center
Center O
O
VERITAS
ICRC 2015 : >85 Hrs VERITAS Data
Galactic Longitude [deg]
0.5
VERITAS>>22TeV
TeV
VERITAS
>
>8
G0.9+0.1
G0.9+0.1
SagittariusA*
A*
Sagittarius
-0
-0.5
1
0
0.5
Galactic Longitude[deg]
359.5
A. W. Smith - ICRC 2015
A*
ERITAS > 2 TeV
Galactic Center
2 TeV - 50 TeV
-Position consistent with H.E.S.S., radio
osition
consistent
with+H.E.S.S.,
radio
-Combined
(HESS
VERITAS)
spectral fit favor PL + exp cut
ESS + VERITAS)
spectral
fit favorEcut
PL=+12.8
exp (+/-1.9)
cutoff: TeV
Γ=
2.1 (+/-0.04),
= 2.1 (+/-0.04), Ecut =-No
12.8signs
(+/-1.9)
TeV
of variability
> 2 TeV
-No signs of variability > 2 TeV
A. W. Smith, I
A. W. Smith, ICRC 2015
A. W. Smith - ICRC 2015
Residual Ma
Galactic Center
Subtract Sgr A* Excess within PSF…
VERITAS > 2 TeV
New source:
7.8σ significance!
A. W. Smith - ICRC 2015
Detecting Gamma Rays
Combined Telescope
Shower Images
http://community.dur.ac.uk/~dph0www4/images/detection.jpg
Zenith Angle
Small Zenith
Large Zenith
DISP + Large Zenith Angle
• DISP
VERITAS
Galactic Center Observations
•
•
Better Angular
Resolution
Low Zenith Angle
VERITAS
Center
Observations
• Shifts energyGalactic
threshold
to 2 TeV
•
Both Great for Galactic
Center at Zenith > 60°
Li and Ma
Pick On Sky Region
Pick Off Sky Region
Count Gamma Rays in Each
Region
Off
Off
On
Li and Ma
Great For Point Sources With Clean Off Regions
Off
Not So Great - If Too Many Gamma-Ray Sources
Crowding The Off Regions
- If Background Has Diffuse Sources
Off
On
he observation. Fig. 2 visualizes
which is an evolution of the stannd technique [28] adjusted for this
construction, background regions
ay from the GC than the source
ortant
aspect,
since,2011:
unavoidably,
HESS
paper
M annihilation events would be
ound regions, too, reducing a po• Excluded
Large
Areas
tained
in the source
region.
For
profiles,(source
the expected
DM annihicrowding)
er in the background regions than
Fig. 1), making the measurement
• Complex
Off Regions
on flux
possible. Note,
however,
halo profile, the signal would be
As far as the background from
n is concerned, its predicted flux
w the current analysis sensitivity,
not further considered in the anale its intensity is believed to drop
Galactic Latitude (deg)
GC with HESS
Search Region
Excluded Region
Background Region
1
3
0
2
0
-1
1
-2
-3
2
Abramowski, 2011, arxiv:1103.3266
0
-2
Galactic Longitude (deg)
<σv>
3 -1
[cm s ]
-23
10
<σv>
Einasto (this work)
NFW (this work)
Sgr Dwarf
Willman 1
Ursa Minor
Draco
10-24
10
3 -1
[cm s ]
GC with HESS
10-22
-25
-22
10-26
1
-27
-23
10
10
-28
-24
10
Einasto (this work)
χ NFW (this work)
Sgr Dwarf
Willman 1
Ursa Minor
Draco
m [TeV]
(at 95% CL) on the velocity-weighted
10
1
10
mpar[TeV]
ion
⟨σv⟩
as
a
function
of
the
DM
Abramowski,
2011,
arxiv:1103.3266
10
-29
10-25
-1
-26
χ
…
GC
with
Fermi
PHYSICAL REVIEW D 89, 042001 (2014)
PHYSICAL REVIEW
Lower WIMP masses
further excluded
Ackerman, 2014, PHYSICAL REVIEW D 89, 042001
Higher masses still viable
Likelihood
• Two model groups
(hypotheses)
• Maximize Likelihoods
• Test Statistic
predicted # gamma rays
Background
Source +
Background
space/energy/time
T estStatistic = −2∗ln
!
Lbackground
Lsource+background
"
Modelling the Backgrounds
predicted # gamma rays
Proton Backgrounds, Pulsar Wind Nebulae,
Supermassive Black Hole, Diffuse Emission,
Supernova Remnants, Dark Matter? (Ein/NFW/Isotherm)
along galactic plane
Gammalib and CTOOLS
•
Likelihood Toolbox for Gamma-Ray Data
•
Binned & Unbinned
•
Multiple Spatial/Spectral Models
•
•
Gaussian Point Source, Shell, Power Law, Broken Power
Law, etc
Currently in Testing by VERITAS, MAGIC and HESS
http://cta.irap.omp.eu/ctools/
Gammalib and CTOOLS
Models
Data
VERITAS Data
CTA Data
HESS/MAGIC
Fermi Data
Dark Matter
Other Gamma-Ray Sources
Halo Profile
Diffuse Emission
Mass/Cross-section
Spectrum
Supernova Remnants
Molecular Clouds
Pulsar Wind Nebula
Ctools Analyses
Likelihood Tests
Gammalib
Skymaps
Spectra
Light Curves
http://cta.irap.omp.eu/gammalib/
http://cta.irap.omp.eu/ctools/
Conclusions
•
New source near GC at 7.8σ above 2 TeV
•
Likelihood Modelling of the Galactic Center with
VERITAS can probe higher WIMP masses due to
the Large Zenith Angle
•
Checking Likelihood Results (spectra/significance)
With Standard Analysis (Li&Ma) Results
•
Next steps: DM and Non-DM Source Modelling
3
< σ v > [cm s-1]
10-22
-23
10
10-24
10
-25
Abramowski 2011
10-26
1 log (E[TeV])1.5
10
-27
10
-28
10
-29
Einasto (this work)
NFW (this work)
Sgr Dwarf
Willman 1
Ursa Minor
Draco
10-1
1 log (E[TeV])1.5
ial flux FSrc/Bg ,
ed for the source
t. The units are
endent selection
ined from γ-ray
dified compared
ured on Earth.
re Fres = FSrc −
s . The residual
Comparable null
us of the source
nt time periods
alf of the source
1
mχ [TeV]
10
FIG. 4. Upper limits (at 95% CL) on the velocity-weighted
annihilation cross-section ⟨σv⟩ as a function of the DM particle mass mχ for the Einasto and NFW density profiles.
The best sensitivity is achieved at mχ ∼ 1 TeV. For comparison, the best limits derived from observations of dwarf
galaxies at very high energies, i.e. Sgr Dwarf [10], Willman 1, Ursa Minor [15] and Draco [9], using in all cases
NFW shaped DM profiles, are shown. Similar to source region of the current analysis, dwarf galaxies are objects free
of astrophysical background sources. The green points represent DarkSUSY models [32], which are in agreement with
WMAP and collider constraints and were obtained with a
random scan of the mSUGRA parameter space using the
following parameter ranges: 10 GeV < M0 < 1000 GeV,
10 GeV < M1/2 < 1000 GeV, A0 = 0, 0 < tanβ < 60,
sgn(µ) = ±1.