Measurement of the Cosmic-ray Proton Spectrum with the Fermi Large - - PowerPoint PPT Presentation

SLIDE 1

Measurement of the Cosmic-ray Proton Spectrum with the Fermi Large Area Telescope

David Green, Liz Hays On Behalf of the Fermi-LAT Collaboration UMD/GSFC ICRC 2017 July 14, 2017

SLIDE 2

Energy [GeV]

2

10

3

10

4

10

5

10

]

• 1

sr

• 2

m

• 1

s

• 1

J(E) [GeV ×

2.7

E

8000 10000 12000 14000 16000

ATIC (2003)[1] PAMELA (2006-2008)[4] BESS-TeV (2002)[2] AMS-02 (2011-2013)[5] CREAM-I (2005)[3]

Introduction

• PAMELA and AMS-02
• bserve a spectral break

at ~400 GeV

• Have to reconcile with

“standard” theories for CR origins, acceleration, and propagation

• Additional

measurements extending from GeV to TeV can help understand spectral break

2

LAT Energy Range

SLIDE 3

The Fermi LAT

3

• The Large Area Telescope (LAT) is one 

• f two instruments on the 


Fermi Gamma-ray Space Telescope

• The LAT is a pair conversion telescope

Anti-coincidence Detector (ACD)

• 89 segmented plastic scintillating tiles
• Used for particle identification

Calorimeter (CAL)

• 1536 CsI(Tl) crystals arranged in 8 layers
• Hodoscopic, image shower shape and profile
• Used for energy measurement

Tracker (TKR)

• 18 x-y layers of silicon strip detectors
• Used for direction reconstruction and

particle identification

SLIDE 4

Event Selection

• Proton flux is high enough, we don’t need

a large acceptance to measure the spectrum to TeV energies

• The proton event selection is defined as:
• Event has to trigger and pass onboard

filters

• Require event to have reconstructed

track

• Deposited energy >20 GeV in CAL
• Require a well reconstructed track using

Pass 8 direction classifier

4 Almost 106 events above 1 TeV

Based on AMS-02 proton flux Over 8 years of flight

True Energy [GeV]

2

10

3

10

4

10

sr]

2

Proton Acceptance [m

0.5 1 1.5 2 2.5 3 3.5

Trigger & Filter Track Found Energy > 20 GeV Track Reconstruction PRELIMINARY

SLIDE 5

Charge Measurement

• Cosmic-ray helium and nuclei pose large

contamination source for this study

• We use the TKR and the ACD to independently

measure the charge of incoming cosmic ray in the LAT

• Define a polygon in ACD-charge vs TKR-

charge to select on protons

• Developed using flight data and Geant4

proton/electron/nuclei simulations

• Find a residual contamination from CR helium

and nuclei less than 1%

• CR electrons are under 4%, decreasing with

energy

• We background subtract any residual electron

contamination

5

Recon Energy [GeV]

2

10

3

10

Residual Contamination

4 −

10

3 −

10

2 −

10

1 −

10

Sum Electrons Helium Nuclei

PRELIMINARY

SLIDE 6

Recon Energy [GeV]

2

10

3

10

4

10

True Energy [GeV]

2

10

3

10

PRELIMINARY True Energy = Recon Energy

Energy Measurement

• We use the CAL to measure the energy
• f the proton induced shower
• CAL is up to 2 λI at off axis angles
• Develop event selection to select ideal

event topologies

• Does not fall within gaps between CAL

modules

• Select events with low ‘backsplash’ into

TKR

• Require > 0.5 λI in the CAL
• We fit the profile of energy deposition to

estimate the energy of the incident proton

• Deposited energy primarily from

electromagnetic component of total shower

6

• We unfold the spectrum in true energy

using ROOT’s TUnfold with a Tikhonov regularization term

SLIDE 7

Systematic Uncertainties

• This study is dominated by

systematic uncertainties

• We use two methods to estimate
• ur systematic uncertainties:
• Signal Efficiency
• Selecting events with different

path-lengths

• Alternative GEANT4 models
• Response uncertainties via

alternate hadronic models

• Uncertainty in the energy

measurement is still being finalized

7

True Energy [GeV]

2

10

3

10

Systematic Uncertainty

0.02 0.04 0.06 0.08 0.1 0.12 0.14

Statistical Signal Efficiency Alternative GEANT4 Models Stat + Sys Uncertainties

PRELIMINARY

Does not include energy uncertainties

SLIDE 8

Cosmic-ray Proton Spectrum

• Using 7 years of LAT flight

data, August 4, 2008 to 
 July 30, 2015

• Extends energy of space-

based measurement to 9.5 TeV

• Red markers represent

statistical uncertainty

• Red shaded region includes

systematic uncertainties

• Good agreement with other

cosmic-ray measurements

8 Energy [GeV]

2

10

3

10

4

10

]

• 1

sr

• 2

m

• 1

s

• 1

J(E) [GeV ×

2.7

E

4

10

CREAM-I (2005) ATIC (2003) AMS-02 (2011-2013) PAMELA (2006-2008) Fermi-LAT (2008-2015)

PRELIMINARY

Does not include energy uncertainties

SLIDE 9

Conclusions and Future

• Space-based spectral

measurement to from 54 GeV to 9.5 TeV

• Additional cosmic-ray proton

studies with the LAT

• Cosmic-ray Proton Anisotropy

with LAT by Matt Meehan - CRD092

• Testing methods to estimate

energy uncertainty

9

Courtesy of Matt Meehan

SLIDE 10

References

1.A. D. Panov et al. Energy spectra of abundant nuclei of primary cosmic rays from the data of ATIC-2 experiment: Final results. Bulletin of the Russian Academy of Sciences: Physics, 73(5): 564–567, 2009. 2.Y. Shikaze et al. Measurements of 0.2–20 GeV/n cosmic-ray proton and helium spectra from 1997 through 2002 with the BESS spectrometer. Astroparticle Physics, 28(1):154 – 167, 2007. 3.Y. S. Yoon et al. Cosmic-ray proton and helium spectra from the first CREAM flight. The Astrophysical Journal, 728(2):122, 2011. 4.O. Adriani et al. PAMELA measurements of cosmic-ray proton and helium spectra. Science, 332(6025):69–72, 2011. 5.M. Aguilar et al. Precision measurement of the proton flux in primary cosmic rays from rigidity 1 GV to 1.8 TV with the alpha magnetic spectrometer on the international space station. Phys.

• Rev. Lett., 114:171103, Apr 2015.

10

SLIDE 11

Backup Slides

SLIDE 12

Anti-Coincidence Detector (ACD)

• ACD’s main purpose is to detect CRs
• Consists of 89 plastic scintillating tiles and 8

plastic scintillating ribbons that cover the TKR

• Top tiles arranged in a 5 x 5 grid
• Side tiles arranged in 5 x 3 grid with single

large tile on the bottom row

• Signal in each tile read by two PMTs
• Each PMT has a dual range, linear low range

and non-linear high range

• Energy deposition in ACD described by

ionization

• Can use this to identify charge of incident

particle 12

ACD Base Electronics Assembly

arXiv:0902.1089v1

SLIDE 13

The Tracker (TKR)

• 16 layers of high Z tungsten foil
• Convert photon to e+ e− pair
• Last 4 conversion layers about 6 times thicker

than previous 12

• 18 layers of silicon strip detectors
• Measure position of charged particle
• TKR is 1.5 radiation lengths thick
• TKR is used to measure direction of incident

cosmic-ray

• Direction used to path-length correct signal and

in reconstruction of several variables

• Additionally, energy deposited via ionization
• Can use TKR as independent measure of CR

charge 13

Layer 1 Layer 2 Layer 3 Photon Tray Structural Material Tungsten x silicon strips y silicon strips Tungsten x silicon strips y silicon strips Tungsten x silicon strips y silicon strips

• Astropart. Phys., 28, 422

SLIDE 14

Calorimeter (CAL)

• Use CAL to measure CR energy and direction
• Composed of 16 modules; each module has 96

CsI(Tl) crystals

• Arranged in 8 layers in alternating x-y directions
• This allows for not only measuring energy

deposition but also imaging of shower shape and direction

• Shower shape can be used for particle

identification

• 8.6 radiation lengths deep (0.5 nuclear

interactions) at normal incidence

• 2.5 nuclear interactions lengths for maximum off

angle axis

• At higher energies shower leakage crystal

saturation needs to be corrected and accounted 14

CDE: CsI Detectors + PIN diodes (both ends) Carbon Cell Array Al Cell Closeout Al EMI Shield Readout Electronics

Atwood 2009 arXiv:0902.1089v1

SLIDE 15

Hadronic Showers in the LAT

15

• We can estimate how proton induced shower

look like in the CAL 100 GeV 1 TeV 100 GeV 1 TeV

• Same can be seen for radial profile, EM

core with hadronic extension

• EM component dominates early

longitudinal profile and radial core

SLIDE 16

Recon Energy [GeV]

2

10

3

10

4

10

True Energy [GeV]

2

10

3

10

PRELIMINARY

Unfolding The Spectrum

16

Recon Energy [GeV]

2

10

3

10

4

10

]

• 1

Total event rate [s

7 −

10

6 −

10

5 −

10

4 −

10

3 −

10

2 −

10

1 −

10

PRELIMINARY

True Energy [GeV]

2 10 3 10

]

• 1

Unfolded event rate [s

3 − 10 2 − 10 1 − 10

PRELIMINARY

True Energy [GeV]

2 10 3 10

sr]

2

Acceptance [m

0.05 0.1 0.15 0.2 0.25 0.3

PRELIMINARY

True Energy [GeV]

2

10

3

10

]

• 1

sr

• 2

m

• 1

s

• 1

J(E) [GeV

6 −

10

5 −

10

4 −

10

3 −

10

2 −

10

1 −

10

PRELIMINARY

Unfold via
 ROOT’s TUnfold Divide by 
 acceptance 
 and bin width

Event Rate Response Matrix Unfolded Event Rate Acceptance Proton Spectrum

SLIDE 17

Signal Efficiency

• Primary measure of systematic uncertainty in acceptance
• Test stability of spectral measurement over different path-lengths through LAT
• Probes shower development through different geometric cross-sections of LAT
• Find energy dependent quantiles of path-length and produces cuts for 90% - 30%

quantiles

• Produce different IRF for each quantile cut and reconstruct the spectrum
• The maximum variation of all spectra determines the uncertainty

17

SLIDE 18

GEANT4 Hadronic Models

• Main measure of systematic uncertainties 


in energy measurement

• Produce dedicated proton simulations with alternative hadronic models in GEANT4
• Alternative models change shower development and deposited energy
• Tested 3 alternative models
• Checked data/MC agreement from beam-test data
• Produce IRFs for each alternative models and unfold the spectrum
• Uncertainty is set from maximum variation of each alternative hadronic model

18