Measurement of high energy cosmic rays above 10 PeV with - - PowerPoint PPT Presentation

measurement of high energy cosmic rays above 10 pev with
SMART_READER_LITE
LIVE PREVIEW

Measurement of high energy cosmic rays above 10 PeV with - - PowerPoint PPT Presentation

Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande Fabiana Cossavella 27 October 2009 Fabiana Cossavella Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 1 Outline overview of the


slide-1
SLIDE 1

Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande

Fabiana Cossavella 27 October 2009

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 1

slide-2
SLIDE 2

Outline

  • overview of the KASCADE-Grande experiment
  • air shower measurement and reconstruction
  • cosmic ray spectrum
  • composition study

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 2

slide-3
SLIDE 3

Cosmic Ray Energy Spectrum

Energy (GeV)

4

10

5

10

6

10

7

10

8

10

9

10

10

10

11

10

12

10

)

1.5

eV

  • 1

sr

  • 1

s

  • 2

J(E) (m

2.5

E

13

10

14

10

15

10

16

10

17

10

18

10

19

10

KASCADE-Grande

direct data

Akeno AGASA HiRes I HiRes II AUGER 2008 (PRL)

EAS-TOP KASCADE (QGSJET 01) µ KASCADE-Grande (QGSJET II) Nch-N

  • break in the single elemental spectra at E ∝ Z · Eknee

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 3

slide-4
SLIDE 4

Detection: Extensive Air Showers

  • interaction primary-air → mainly

π-mesons

  • secondary products interact

(hadrons) or decay (muons, electrons)

  • particles reach ground in a shower

disc

  • longitudinal profile → N particles as function of the

atmospheric depth

  • lateral profile → distribution at ground of the shower particles

as a function of the distance from the shower axis

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 4

slide-5
SLIDE 5

Different primaries

  • dominated by electromagnetic component

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 5

slide-6
SLIDE 6

Different primaries

  • Nµ/Ne larger as compared to proton

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 5

slide-7
SLIDE 7

The KASCADE Experiment

Physics Observables

  • electron
  • gammas
  • muons
  • hadrons

The detectors

  • Array stations: liquid scintillator for e/γ detection and plastic

scintillator for µ detection above 230 MeV

  • Muon Tracking Detector for Eµ ≥ 800 MeV
  • MWPC/LSTs for muon detection above 2.4 GeV
  • Calorimeter, for hadron detection

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 6

slide-8
SLIDE 8

KASCADE-Grande

KASCADE extension to 700 × 700 m2

  • energy spectrum and composition from

1016 eV to 1018 eV

  • galactic to extragalactic transition?
  • Testing hadronic interaction models
  • iron knee ?
  • origin of the knee ?
  • Anisotropies

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 7

slide-9
SLIDE 9

KASCADE-Grande

Physics observables

  • arrival time and energy deposit by

charged particles in each detector ⇒ particle densities

  • muons from the KASCADE

detectors

  • ≈10 m2 of plastic scintillator per

station

  • light detected and coverted into

electronic signal by PMTs

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 8

slide-10
SLIDE 10

Trigger efficiency

KASCADE Piccolo-Stations Grande-Stations

x-coordinate฀[m] y-coordinate฀[m]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 Trigger-Cell฀17(18)

  • 700
  • 600
  • 500
  • 400
  • 300
  • 200
  • 100

100

  • 600 -500 -400 -300 -200 -100

100

  • fiducial area of ∼ 192,000 m2

(orange area),

  • trigger by 7/7 coincidence

/GeV) (E

10

log 6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 trigger efficiency 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 proton - 7/7 iron - 7/7 all - 7/7 trigger all - 4/7 trigger

  • based on air shower simulation
  • full efficiency at E = 1016 eV, for

both proton and iron primaries

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 9

slide-11
SLIDE 11

Shower reconstruction

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 10

slide-12
SLIDE 12

Shower reconstruction

shower core: impact position of the shower axis at ground level = ⇒ Grande measurements arrival direction = ⇒ Grande measurements electron shower size Ne: total number of electrons arriving at

  • bservation level =

⇒ Grande measurements muon shower size Nµ: total number of muons arriving at

  • bservation level =

⇒ KASCADE measurements

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 10

slide-13
SLIDE 13

Reconstruction procedure

X (Meters)

  • 700
  • 600
  • 500
  • 400
  • 300
  • 200
  • 100

100 Y (Meters)

  • 700
  • 600
  • 500
  • 400
  • 300
  • 200
  • 100

100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37

Time (ns) 4845 5554

(x,y)=(-115 m, -223 m), (θ, φ)=(11◦, 338◦)

Distance From Shower Axis [m] 100 200 300 400 500 )

2

*m ρ (

10

log

  • 1

10 1 10

2

10

Charged (Grande) (KASCADE) µ + e) µ ( ) µ Lagutin (

log10(Ne) = 7, log10 Nµ = 5.7

Core position and total number of charged particle (Nch) by fitting the lateral distribution with a modified NKG function: ρch = Nch · C(s) · r r0 s−1.6 ·

  • 1 + r

r0 s−3.4 r0 = 30m

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 11

slide-14
SLIDE 14

Reconstruction procedure

X (Meters)

  • 700
  • 600
  • 500
  • 400
  • 300
  • 200
  • 100

100 Y (Meters)

  • 700
  • 600
  • 500
  • 400
  • 300
  • 200
  • 100

100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37

Time (ns) 4845 5554

(x,y)=(-115 m, -223 m), (θ, φ)=(11◦, 338◦)

Distance From Shower Axis [m] 100 200 300 400 500 )

2

*m ρ (

10

log

  • 1

10 1 10

2

10

Charged (Grande) (KASCADE) µ + e) µ ( ) µ Lagutin (

log10(Ne) = 7, log10 Nµ = 5.7

Total number of muons (Nµ) by fitting the muon lateral distribution with a Lagutin function: ρµ = Nµ· 0.28 3202 · r 320 −0.69 ·

  • 1 +

r 320 −2.39 ·

  • 1 +
  • r

10 · 320 2−1

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 11

slide-15
SLIDE 15

Reconstruction procedure

X (Meters)

  • 700
  • 600
  • 500
  • 400
  • 300
  • 200
  • 100

100 Y (Meters)

  • 700
  • 600
  • 500
  • 400
  • 300
  • 200
  • 100

100

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37

Time (ns) 4845 5554

(x,y)=(-115 m, -223 m), (θ, φ)=(11◦, 338◦)

Distance From Shower Axis [m] 100 200 300 400 500 )

2

*m ρ (

10

log

  • 1

10 1 10

2

10

Charged (Grande) (KASCADE) µ + e) µ ( ) µ Lagutin (

log10(Ne) = 7, log10 Nµ = 5.7

Total number of electrons (Ne) by subtracting the estimated muon content from the estimated charged particles content.

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 11

slide-16
SLIDE 16

Reconstruction accuracies sim/data

)

true e

(N

10

log 5 5.5 6 6.5 7 7.5 8 8.5

true e

)/N

true e

  • N

rec e

(N

  • 0.2

0.2 0.4 0.6

proton iron

)

true e

(N

10

log 5 5.5 6 6.5 7 7.5 8 8.5 Core accuracy [m] 5 10 15 20 25 30

proton iron

  • statistical uncertainty for Ne at full efficiency around 25 %
  • systematics change from less than 5% at threshold to 15%
  • small dependence on primary particle type
  • core accuracy ≈ 6 m

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 12

slide-17
SLIDE 17

Reconstruction accuracies sim/data

(Size_KASCADE)

10

log 5.8 6 6.2 6.4 6.6 6.8 7 7.2 Size ∆

  • 0.1
  • 0.05

0.05 0.1 0.15 (Size_KASCADE)

10

log 5.8 6 6.2 6.4 6.6 6.8 7 7.2 Core accuracy 2 4 6 8 10 12

  • Trigger from Grande
  • Maximum density in the central station of the

cluster overlapping with KASCADE

  • KASCADE reconstruction used as reference

KASCADE Piccolo-Stations Grande-Stations x-coordinate฀[m] y-coordinate฀[m]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 Trigger-Cell฀17(18)

  • 700
  • 600
  • 500
  • 400
  • 300
  • 200
  • 100

100

  • 600 -500 -400 -300 -200 -100

100

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 12

slide-18
SLIDE 18

Reconstruction accuracies sim/data

)

true µ

(N

10

log 4.5 5 5.5 6 6.5 7

true µ

)/N

true µ

  • N

rec µ

(N

  • 0.2

0.2 0.4 0.6

proton iron

)

true e

(N

10

log 5 5.5 6 6.5 7 7.5 8 8.5 Angular accuracy [deg] 0.5 1 1.5 2

proton iron

  • statistical uncertainty for Nµ at full efficiency around 20 %
  • bias of 10% to -5% in energy (muon) dependence
  • Angular accuracy < 0.5◦ up to 40◦

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 13

slide-19
SLIDE 19

Spectra

SIZE (electron) SPECTRA

)

e

(N

10

Log

6 6.5 7 7.5 8

]

  • 1

sr

  • 1

s

  • 2

[m

e

dI/dN

  • 21

10

  • 20

10

  • 19

10

  • 18

10

  • 17

10

  • 16

10

  • 15

10

  • 14

10

  • 13

10

  • 12

10 ° 0-17 ° 17-24 ° 24-30 ° 30-35 ° 35-40

  • 987 days of data taking
  • fiducial area of ≈ 0.2 km2
  • ≈ 880, 000 events

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 14

slide-20
SLIDE 20

Ne − Nµ correlation

µ

lg N

4.5 5 5.5 6 6.5 7

e

lg N

5 5.5 6 6.5 7 7.5 8 8.5

number of showers

1 10

2

10

3

10

4

10

Proton

Carbon

Iron

10 PeV 30 PeV 100 PeV 300 PeV 1 EeV

(Ne, Nµ) = f (E, m)

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 15

slide-21
SLIDE 21

Energy spectrum

  • correlation Nch with E0 from simulation → good

reconstruction accuracy, BUT very much composition dependent

  • correlation Nµ with E0 from simulation → problems with

reconstruction accuracy, less composition dependent

  • estimation of the primary energy as E0(Nch, Nµ) → no

composition dependence! THEY ALL REQUIRE ASSOMPTION ON HIGH ENERGY HADRONIC INTERACTION MODELS

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 16

slide-22
SLIDE 22

Energy estimation

log10 (Nch/Nµ)p,Fe = cp,Fe · log10(Nch) + dp,Fe

0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 5.5 6 6.5 7 7.5 8 8.5

log10(Nch) log10(Nch/Nμ)

r = log10(Nch/Nµ) − log10 (Nch/Nµ)p log10 (Nch/Nµ)Fe − log10 (Nch/Nµ)p

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 17

slide-23
SLIDE 23

Energy estimation

6 6.5 7 7.5 8 8.5 9 5.5 6 6.5 7 7.5 8 8.5

log10(Nch) log10(E/GeV)

log10(Eest) = (ap + ∆a · r) · log10(Nch(θ)) + (bp + ∆b · r)

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 17

slide-24
SLIDE 24

Energy estimation

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 18

slide-25
SLIDE 25

Mass composition analysis method

  • statistical methods to infer cosmic rays mass
  • search of mass-sensitive observables
  • classification of an air shower through kNN method

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 19

slide-26
SLIDE 26

Parameters relevance

)

rec e

(N

10

log 5.5 6 6.5 7 7.5 8 8.5 9 )

rec µ

(N

10

median log 4 4.5 5 5.5 6 6.5 7 7.5 8

proton iron data

error bars represent the spread of the distribution in each bin

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 20

slide-27
SLIDE 27

KNN

classification method: comparison of each data point with the “K nearest neighbours” in reference samples

  • data point, proton, iron
  • characterization of each point

through Nch and Nµ

  • Mahalanobis distance:
  • (x1 − x2)T(S)−1(x1 − x2)

with S covariance matrix

  • S for each reference sample separately
  • K number to be chosen on analysis base
  • assignment through majority vote

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 21

slide-28
SLIDE 28

Training

  • misclassification rates Pωp→ωFe
  • misclassification matrix

Pωp→ωp PωFe→ωp Pωp→ωFe PωFe→ωFe n∗

p

n∗

Fe

  • =

np nFe

  • separability index G = Pωp→ωp · PωFe→ωFe
  • 5 zenith angle bins: 0◦ − 40◦
  • different energy ranges: 1016 eV to 1018 eV

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 22

slide-29
SLIDE 29

Training results

proton relative error 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 ° < 17.6 θ ≤ ° ° < 24.99 θ ≤ ° 17.6 ° < 30.73 θ ≤ ° 24.99 ° < 35.63 θ ≤ ° 30.73 ° < 40 θ ≤ ° 35.63 )

est

(E

10

log 7.2 7.4 7.6 7.8 8 8.2 8.4 8.6 8.8 iron relative error 0.05 0.1 0.15 0.2 0.25

  • misclassification better

than 25%

  • separability better than

80%

  • best k=9

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 23

slide-30
SLIDE 30

kNN on KASCADE-Grande data

)

rec µ

(N

10

log 4 4.5 5 5.5 6 6.5 7 )

rec e

(N

10

log 5 5.5 6 6.5 7 7.5 8 8.5 9

proton iron data

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 24

slide-31
SLIDE 31

kNN on KASCADE-Grande data

/GeV)

est

(E

10

log

7 7.5 8 8.5 9

relative mean number

0.2 0.3 0.4 0.5 0.6 0.7 0.8 proton iron

change to heavier composition up to E = 3 · 1016 eV

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 24

slide-32
SLIDE 32

kNN on KASCADE-Grande data

/GeV)

est

(E

10

log

7 7.5 8 8.5 9

relative mean number

0.2 0.3 0.4 0.5 0.6 0.7 0.8 proton iron

  • between 3 · 1016 eV to 1018 eV: ≈ 70% iron-like
  • E > 1018 eV: 6 events per zenith bin, large fluctuation,

compatible with constant composition at 1.5 σ level.

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 24

slide-33
SLIDE 33

kNN: systematic sources

  • method dependent on hadronic interaction model used
  • contribution of other primaries to the measured frequencies of

proton-like and iron-like air showers:

  • He → 80% proton, 20% iron
  • C → 50% proton, 50% iron
  • Si → 20% proton, 80% iron
  • systematic underestimation of energy ≈ −5%

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 25

slide-34
SLIDE 34

Summary

  • Shower reconstruction is accurate enough for analysis
  • Methods to reconstruct the energy spectrum available
  • Increase of iron fraction up to E = 3 · 1016 eV → consistent

with a decrease of the light primaries

  • at E > 3 · 1016 eV no evident change of composition observed:

≈ 70% iron-like air showers

  • no transition to extragalactic (light) component for

E < 1018 eV

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 26

slide-35
SLIDE 35

kNN on other primaries

/GeV)

est

(E

10

log

7 7.5 8 8.5 9

relative mean number

  • 0.2

0.2 0.4 0.6 0.8 1 1.2 proton iron

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 27

slide-36
SLIDE 36

kNN on other primaries

/GeV)

est

(E

10

log

7 7.5 8 8.5 9

relative mean number

0.2 0.3 0.4 0.5 0.6 0.7 0.8 proton iron

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 27

slide-37
SLIDE 37

kNN on other primaries

/GeV)

est

(E

10

log

7 7.5 8 8.5 9

relative mean number

0.2 0.4 0.6 0.8 1 proton iron

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 27

slide-38
SLIDE 38

Shower profiles

]

2

atmospheric depth [g/cm

100 200 300 400 500 600 700 800 900 1000

number of electrons

4

10

5

10

6

10

7

10

8

10 hydrogen iron 10PeV 100PeV

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 28

slide-39
SLIDE 39

Shower profiles

]

2

atmospheric depth [g/cm

100 200 300 400 500 600 700 800 900 1000

number of muons

3

10

4

10

5

10

6

10 hydrogen iron 10PeV 100PeV

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 28

slide-40
SLIDE 40

Shower profiles

core distance [m]

200 400 600 800 1000

]

  • 2

particle densities [m

  • 3

10

  • 2

10

  • 1

10 1 10

2

10

3

10

4

10 hydrogen iron 10PeV 100PeV

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 28

slide-41
SLIDE 41

Shower profiles

core distance [m]

200 400 600 800 1000

]

  • 2

particle densities [m

  • 3

10

  • 2

10

  • 1

10 1 10 hydrogen iron 10PeV 100PeV

Fabiana Cossavella — Measurement of high energy cosmic rays above 10 PeV with KASCADE-Grande 28