Marcus Niechciol a for the Pierre Auger Collaboration b a Department - - PowerPoint PPT Presentation

SLIDE 1

Diffuse and targeted searches for ultra-high-energy photons using the hybrid detector of the Pierre Auger Observatory

Marcus Niechciola for the Pierre Auger Collaborationb

a Department Physik, Universität Siegen, 57068 Siegen, Germany b Observatorio Pierre Auger, Av. San Martín Norte 304, 5613 Malargüe, Argentina

E-mail: auger_spokespersons@fnal.gov Full author list: http://www.auger.org/archive/authors_icrc_2017.html

[Photo by S. Saffi, Univ. of Adelaide]

SLIDE 2

2 / 12 Marcus Niechciol (Pierre Auger Collaboration) | ICRC 2017 (Busan) | CRI183 18.07.2017

Windows to the Universe

X-rays Radio γ-rays Visible light

Eγ [eV]: 10-3 1 104 1012

Highest-energy photons detected: ≈1014 eV

[ESO] [NRAO] [ESA] [H.E.S.S.]

1018 ?

• Cosmic rays: energies up to

3 × 1020 eV observed

• UHE photons and cosmic rays

are intimately connected Observation may help to answer some of the most pressing questions about cosmic rays

→

SLIDE 3

3 / 12 Marcus Niechciol (Pierre Auger Collaboration) | ICRC 2017 (Busan) | CRI183 18.07.2017

Pierre Auger Observatory

• Two independent and complementary detector systems

§ SD: array of 1660 water Cherenkov detectors (covering 3000 km2) § FD: 27 fluorescence telescopes, overlooking the SD § Hybrid data combine information from the FD and the SD

]

2

slant depth [g/cm 400 600 800 1000 1200 )]

2

dE/dX [PeV/(g/cm

• 2

2 4 6 8 10 12

/Ndf= 162.4/214

2

χ

r [m]

500 1000 1500 2000 2500 3000

Signal [VEM]

1 10

2

10

Stage: 4.5 /Ndf: 8.5/ 8

2

χ candidates non-triggering removed

azimuth [deg] elevation [deg]

5 10 15 20 25 30 30 40 50 60 70 80 90 100 110 120

FD SD

[The Pierre Auger Collaboration, NIM A 798 (2015) 172-213]

SLIDE 4

4 / 12 Marcus Niechciol (Pierre Auger Collaboration) | ICRC 2017 (Busan) | CRI183 18.07.2017

Photon-induced air showers

• Main characteristics of photon-induced air showers compared to those

initiated by hadrons:

§ Delayed shower development (multiplicity of electromagnetic processes smaller than that of hadronic processes) larger depth of the shower maximum Xmax § Lack of muons (photo-nuclear interactions suppressed compared to electromagnetic interactions)

→

E [eV]

16

10

17

10

18

10

19

10

20

10

21

10

〉

µ

N 〈

3

10

4

10

5

10

6

10

7

10

8

10

9

10 Proton Iron Photon

CONEX v2r5p40, QGSJET-II-04

Proton Iron Photon E [eV]

16

10

17

10

18

10

19

10

20

10

21

10

]

• 2

[g cm 〉

max

X 〈

500 600 700 800 900 1000 1100 Proton Iron Photon

LPM effect Preshower effect

CONEX v2r5p40, QGSJET-II-04

Photon Proton Iron

LPM effect Preshower effect

SLIDE 5

5 / 12 Marcus Niechciol (Pierre Auger Collaboration) | ICRC 2017 (Busan) | CRI183 18.07.2017

• Experimental observables:

§ Xmax (FD related) § Parameter Sb (SD related) Sb exploits the different lateral distributions of photon- and hadron-induced showers § Number of triggered surface detectors Nstat (SD related)

[The Pierre Auger Collaboration, JCAP 04 (2017) 009]

→

Signal in surface detector i Distance between surface detector i and the shower axis r0 = 1000 m as reference distance Parameter chosen for the best photon/hadron separation

[G. Ros et al, Astropart. Phys. 35 (2011) 140]

Sb =

N

X

i

Si ✓ ri r0 ◆b

Search for a diffuse photon flux: observables

SLIDE 6

6 / 12 Marcus Niechciol (Pierre Auger Collaboration) | ICRC 2017 (Busan) | CRI183 18.07.2017

Search for a diffuse photon flux: MVA

• The three observables are combined in a boosted decision tree (BDT)
• Training of the BDT using simulated samples of photon- and proton-induced air showers

for signal and background

• Background contamination 0.14 % at 50 % signal efficiency
• Apply the analysis to data collected between 01/2005 and 12/2013

§ 8178 events with Eγ > 1018 eV after cuts for good geometry and profile reconstruction § Three events pass the photon candidate cut

BDT response 0.4 − 0.2 − 0.2 0.4 0.6 Entries 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18

Photon (training) Photon (test sample) Proton (training) Proton (test sample) eV)

18

> 10

γ

data (E Median cut

[The Pierre Auger Collaboration, JCAP 04 (2017) 009]

Photon candidate cut at the median of the photon distribution

SLIDE 7

7 / 12 Marcus Niechciol (Pierre Auger Collaboration) | ICRC 2017 (Busan) | CRI183 18.07.2017

Search for a diffuse photon flux: results

• Number of candidates compatible with the background expectation:

determine upper limits on the integral photon flux

• Upper limits to the integral photon fraction assuming the Auger energy

spectrum: 0.1 %, 0.15 %, 0.33 %, 0.85 % and 2.7 % at E0 = 1, 2, 3, 5 and 10 EeV

Φ0.95

UL (Eγ > E0) =

N0.95

γ

(Eγ > E0) Eγ(Eγ > E0|EΓ

γ ) Number of candidate events at 95 % C.L. Integrated exposure assuming a power law spectrum

Several sources of systematic uncertainties have been studied, e.g. detector, interaction model, spectrum Upper limits in the EeV range improved by at least a factor of 4 Severe constraints for top-down models Some GZK scenarios are in reach

[eV] E

18

10

19

10

20

10 ]

• 1

yr

• 1

sr

• 2

[ km > E

γ

Integral photon flux E

3 −

10

2 −

10

1 −

10 1

GZK proton I GZK proton II

Hy 2011

+syst. Hy 2016

Y 2010 TA 2015

SD 2015

HP 2000 A 2002

Z-burst TD SHDM I SHDM II

upper limits 95% CL

→ → →

[The Pierre Auger Collaboration, JCAP 04 (2017) 009]

→

SLIDE 8

8 / 12 Marcus Niechciol (Pierre Auger Collaboration) | ICRC 2017 (Busan) | CRI183 18.07.2017

Targeted photon search

• Photons are not deflected in magnetic fields: the signature of a photon

point-source is an accumulation of events from the direction of the source

• Select photon-like events using a BDT with 5 input observables
• FD-related: Xmax, χ2 of a Greisen fit to the longitudinal profile, ratio of EGreisen and EFD
• SD-related: Sb, ratio of early and late signals in the detectors
• Selection cut in the BDT response β is optimized for each target direction by minimizing

the upper limit, taking into account the expected background

• Apply the analysis to data collected between 01/2005 and 12/2013
• Energy range: 1017.3 – 1018.5 eV (average angular resolution in this energy range: 0.7°)
• Use top-hat counting with radius 1° for each target direction

[The Pierre Auger Collaboration, ApJ 837 (2017) L25]

• 1
• 0.8
• 0.6
• 0.4
• 0.2

0.2 0.4 0.6 0.8

• 1

10 1 10

ß

(1/N) dN / d ß

photon (testing sample) proton (testing sample) photon (training sample) proton (training sample) 1

MC simulations Background map obtained using the scrambling method

[G.L. Cassiday et al., Nucl. Phys. B (Proc. Suppl.) 14 A (1190) 291]

SLIDE 9

9 / 12 Marcus Niechciol (Pierre Auger Collaboration) | ICRC 2017 (Busan) | CRI183 18.07.2017

Targeted photon search: target classes

• Restrict analysis to pre-defined target classes (reduce the trial factor)
• Galactic set: contains e.g. different classes of pulsars and X-ray binaries
• Similar to the targeted neutron search previously published by Auger
• Extragalactic set: include nearby extragalactic targets
• Centaurus A (d = 3.8 Mpc): core region
• Large Magellanic Cloud (d = 50 kpc): three powerful γ-ray emitters
• 90
• 60
• 30

GC 30 60 90 60 120 180 180 240 300

msec Pulsars

• ray Pulsars

γ Low-mass X-ray Binaries High-mass X-ray Binaries H.E.S.S. Pulsar Wind Nebulae H.E.S.S. other identified sources H.E.S.S. unidentified sources Microquasars Magnetars Galactic Center Large Magellanic Cloud Centaurus A (Core)

[The Pierre Auger Collaboration, ApJ 789 (2014) L34] [The H.E.S.S. Collaboration, Science 347 (2015) 406] [The Pierre Auger Collaboration, ApJ 837 (2017) L25]

SLIDE 10

10 / 12 Marcus Niechciol (Pierre Auger Collaboration) | ICRC 2017 (Busan) | CRI183 18.07.2017

Targeted photon search: results

• No target class reveals compelling evidence for sources emitting photons

with EeV energies

• No evidence for one outstanding target in any target class

[The Pierre Auger Collaboration, ApJ 837 (2017) L25]

p-values (unpenalized and penalized)

Class N P Pw p p⇤ f 0.95

UL [km2 yr1]

msec PSRs 67 0.14 0.57 0.010 0.476 0.043 γ-ray PSRs 75 0.98 0.97 0.007 0.431 0.045 LMXB 87 0.74 0.13 0.014 0.718 0.046 HMXB 48 0.84 0.33 0.040 0.856 0.036 H.E.S.S. PWN 17 0.90 0.92 0.104 0.845 0.038 H.E.S.S. other 16 0.52 0.12 0.042 0.493 0.040 H.E.S.S. UNID 20 0.45 0.79 0.014 0.251 0.045 Microquasars 13 0.48 0.29 0.037 0.391 0.045 Magnetars 16 0.89 0.30 0.115 0.858 0.031

• Gal. Center

1 0.59 0.59 0.471 0.471 0.024 LMC 3 0.62 0.52 0.463 0.845 0.030 Cen A 1 0.31 0.31 0.221 0.221 0.031 Most significant source candidate

combined p-values (unweighted and weighted) Flux upper limit

Combined analysis

SLIDE 11

11 / 12 Marcus Niechciol (Pierre Auger Collaboration) | ICRC 2017 (Busan) | CRI183 18.07.2017

Targeted photon search: galactic center

• Interpretation of H.E.S.S. PeVatron results for the galactic center region
• Constrain the naive extrapolation to EeV energies
• Upper limit on the cutoff energy of 2 EeV

[The Pierre Auger Collaboration, ApJ 837 (2017) L25]

E [TeV]

• 1

10 1 10

2

10

3

10

4

10

5

10

6

10

7

10

]

• 1

s

• 2

x flux [TeV cm

2

E

• 16

10

• 15

10

• 14

10

• 13

10

• 12

10

• 11

10 H.E.S.S. measurement confidence band of the best-fit spectra σ 1 Auger photon GC limit = 2.0 EeV)

cut

=2.32 and E Γ H.E.S.S. extrapolation ( 0.11) ± =2.32 Γ H.E.S.S. extrapolation ( 0.11) ± =2.32 (Γ

[The H.E.S.S. Collaboration, Nature 531 (2016) 476]

SLIDE 12

12 / 12 Marcus Niechciol (Pierre Auger Collaboration) | ICRC 2017 (Busan) | CRI183 18.07.2017

Summary

• Two searches for ultra-high-energy photons using data from the hybrid

detector of the Pierre Auger Observatory have been presented

• No photons with EeV energies detected so far
• Search for a diffuse flux of photons:

Upper limits impose severe constraints on non-acceleration models for the origin of ultra-high-energy cosmic rays Theoretical predictions from some GZK-based models are within reach

• Targeted search:

No evidence for EeV photon emitters in any of the studied source classes Connection to the TeV energy regime enables new multi-messenger studies

• More details on the two analyses can be found in:
• The Pierre Auger Collaboration (A. Aab et al.), Search for photons with energies above

1018 eV using the hybrid detector of the Pierre Auger Observatory, JCAP 04 (2017) 009

• The Pierre Auger Collaboration (A. Aab et al.), A Targeted Search for Point Sources of EeV

Photons with the Pierre Auger Observatory, Astrophys. J. 837 (2017) L25

→ → → →

SLIDE 13

13 / 12 Marcus Niechciol (Pierre Auger Collaboration) | ICRC 2017 (Busan) | CRI183 18.07.2017

Backup

SLIDE 14

14 / 12 Marcus Niechciol (Pierre Auger Collaboration) | ICRC 2017 (Busan) | CRI183 18.07.2017

Attenuation length

• 2
• 1

1 2 3 4 5 12 14 16 18 20 22 24 lg(Datt/Mpc) lg(E/eV)

γ p

redshift CMB IR URB pair production photo-pion

Fe

production

[M. Risse and P. Homola, Mod. Phys. Lett. A 22 (2007) 749]

SLIDE 15

15 / 12 Marcus Niechciol (Pierre Auger Collaboration) | ICRC 2017 (Busan) | CRI183 18.07.2017

Photon candidates

• Closer look at Event 6691838: longitudinal profile and dedicated proton

simulations

Event ID Eγ [EeV] Zenith [] Xmax [g/cm2] Sb [VEM] Nstat l [] b [] 3218344 1.40 ± 0.18 34.9 ± 0.9 851 ± 31 2.04 ± 0.77 2 218.21 ± 1.29

• 25.67 ± 0.36

6691838 1.26 ± 0.05 53.9 ± 0.3 886 ± 9 4.94 ± 1.21 2 100.45 ± 0.57

• 46.25 ± 0.25

12459240 1.60 ± 0.14 49.4 ± 0.4 840 ± 21 9.57 ± 2.56 3 324.94 ± 0.37

• 24.70 ± 0.60

]

2

slant depth [g/cm

400 500 600 700 800 900 1000 1100 1200 1300

)]

2

dE/dX [PeV/(g/cm

0.5 1 1.5 2 2.5

/Ndf= 76.4/90

2

χ

[VEM] )

b

( S

10

log 0.5 1 1.5 2 2.5 3 ]

2

Xmax [ g/cm 600 700 800 900 1000 1100 1200 proton data proton - selected

[The Pierre Auger Collaboration, JCAP 04 (2017) 009]

SLIDE 16

16 / 12 Marcus Niechciol (Pierre Auger Collaboration) | ICRC 2017 (Busan) | CRI183 18.07.2017

Hybrid exposure

[The Pierre Auger Collaboration, JCAP 04 (2017) 009]

[eV]) (E

10

log 17.8 18 18.2 18.4 18.6 18.8 19 19.2 19.4 sr yr]

2

Integral exposure [km

2

10

3

10

= 2.0 ) Γ 2005 - 2013 (

• syst. uncertainties

SLIDE 17

17 / 12 Marcus Niechciol (Pierre Auger Collaboration) | ICRC 2017 (Busan) | CRI183 18.07.2017

Systematic effects

[The Pierre Auger Collaboration, JCAP 04 (2017) 009]

E0 [EeV] 1 2 3 5 10 Hadronic model (Epos LHC) Nγ 7 1 Φ95%C.L. 0.043 0.015 0.008 0.008 0.008 Mixed composition Nγ 2 Φ95%C.L. 0.041 0.019 0.008 0.007 0.007 Spectral Index Γ = 2.5 Nγ 6 1 Φ95%C.L. 0.046 0.017 0.010 0.009 0.009 Spectral Index Γ = 1.5 Nγ 3 Φ95%C.L. 0.025 0.008 0.008 0.007 0.006

Detector systematic uncertainties Source

• Syst. uncert.

UL0.95 change (Eγ > 1 EeV)

Energy scale

± 14% (+18, -38)%

Xmax scale

± 10 g/cm2 (+18, -38)%

Sb

± 5% (-19, +18)%

Exposure

± 6.4% (-6.4, +6.4)%

SLIDE 18

18 / 12 Marcus Niechciol (Pierre Auger Collaboration) | ICRC 2017 (Busan) | CRI183 X.07.2017

p-values

P · A p-value pi is assigned to each candidate source of a target set as follows. The p-value for the target i is defined by pi ≡ [Poisson(ni, bi) + Poisson(ni + 1, bi)]/2, where Poisson(ni, bi) is the probability of getting ni or more arrival directions in the target when the observed value is ni, and the expected number from the back- ground is bi. Averaging the values for n and n+1 avoids a bias toward low or high p-values for pure background fluctuations. fluctuations. The combined weighted probability Pw is the frac- tion of isotropic simulations yielding a weighted product Q

i pwi i,iso that is not greater than the measured weighted

product Q

i pwi i :

Pw = Prob Y

i

pwi

i,iso ≤

Y

i

pwi

i

! , (2) where pi,iso denotes the p-value of target i in an isotropic

• simulation. The combined unweighted probability P is

given by the same formula with wi = 1 for all targets

[The Pierre Auger Collaboration, ApJ 837 (2017) L25]

18.07.2017

SLIDE 19

19 / 12 Marcus Niechciol (Pierre Auger Collaboration) | ICRC 2017 (Busan) | CRI183 18.07.2017

Full table of results

Table 1. Combined Unweighted probabilities P and Weighted Probabilities Pw for the 12 Target Sets.

• Note. In addition, information on the most significant target from each target set is given. The number of observed (Obs)

and expected (Exp) events and the corresponding exposure are shown. The numbers in brackets in the observed number of events column indicate the number of events needed for a 3σ observation unpenalized and penalized (∗). Upper limits (UL) are computed at 95% confidence level. The last two columns indicate the p-value unpenalized (p) and penalized (p∗). Due to the discrete distribution of p-values arising in isotropic simulations, P can differ from p in the sets that contain only a single target. Class No. Pw P R.A. Decl. Obs Exp Exposure Flux UL E-flux UL p p∗

[] [] [km2 yr] [km2 yr1] [eV cm2 s1]

msec PSRs 67 0.57 0.14 286.4 4.0 5 (7,9∗) 1.433 236.1 0.043 0.077 0.010 0.476 γ-ray PSRs 75 0.97 0.98 312.8

• 8.5

6 (8,10∗) 1.857 248.1 0.045 0.080 0.007 0.431 LMXB 87 0.13 0.74 258.1

• 40.8

6 (8,11∗) 2.144 233.9 0.046 0.083 0.014 0.718 HMXB 48 0.33 0.84 285.9

• 3.2

4 (7,9∗) 1.460 235.2 0.036 0.066 0.040 0.856 H.E.S.S. PWN 17 0.92 0.90 266.8

• 28.2

4 (8,10∗) 2.045 211.4 0.038 0.068 0.104 0.845 H.E.S.S. other 16 0.12 0.52 258.3

• 39.8

5 (8,10∗) 2.103 233.3 0.040 0.072 0.042 0.493 H.E.S.S. UNID 20 0.79 0.45 257.1

• 41.1

6 (8,10∗) 2.142 239.2 0.045 0.081 0.014 0.251 Microquasars 13 0.29 0.48 267.0

• 28.1

5 (8,10∗) 2.044 211.4 0.045 0.080 0.037 0.391 Magnetars 16 0.30 0.89 257.2

• 40.1

4 (8,10∗) 2.122 253.8 0.031 0.056 0.115 0.858

• Gal. Center

1 0.59 0.59 266.4

• 29.0

2 (8,8∗) 2.048 218.9 0.024 0.044 0.471 0.471 LMC 3 0.52 0.62 84.4

• 69.2

2 (8,9∗) 2.015 180.3 0.030 0.053 0.463 0.845 Cen A 1 0.31 0.31 201.4

• 43.0

3 (8,8∗) 1.948 214.1 0.031 0.056 0.221 0.221

[The Pierre Auger Collaboration, ApJ 837 (2017) L25]