On the measurement of the proton-air cross section using ...Ralf Ulrich ([email protected]) On...
Transcript of On the measurement of the proton-air cross section using ...Ralf Ulrich ([email protected]) On...
On the measurement of the proton-air crosssection using air shower data
Ralf UlrichJ. Blumer, R. Engel, F. Schussler, M. Unger
KIT, Forschungszentrum Karlsruhe
Aspen 2007
Introduction
Analysis methods
Unaccompanied hadrons
Frequency attenuation
Distribution of Xmax
RMStailshape
Sources of systematic uncertainties
Cosmic ray composition
Methodical
Air shower fluctuations
Hadronic interaction models
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 1 / 25
Overview
(Energy/eV)10
log10 11 12 13 14 15 16 17 18 19
Cro
ss s
ecti
on
(p
roto
n-a
ir)
[m
b]
250
300
350
400
450
500
550
600
650
700
Mielke et al. 1994Aglietta et al. 1999
Baltrusaitis et al. 1984Honda et al. 1999
Knurenko et al. 1999
Nam et al. 1975Siohan et al. 1978Belov et al.
QGSJET01
EPOS
NEXUS3
SIBYLL2.1
QGSJETII.3
Energy [GeV]210 310 410 510 610 710 810 910 1010
[GeV]ppsEquivalent c.m. energy 10 210 310 410 510
Tevatron (p-p)
LHC (p-p)
LHC (C-C)
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 2 / 25
Contributions to total cross section
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 3 / 25
First interaction and fluctuations in air showers
primary particleatmospheric boundary
ground level
fluctu
atio
ns
(flu
ctu
ations)
flu
ctu
ati
on
s
(fluctuations)
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 4 / 25
Air shower profile fluctuations
0 200 400 600 800 1000 1200 1400 1600 1800 2000
6N
/ 1
0
2000
4000
6000
610×
electrons eV19protons, 10
0 200 400 600 800 1000 1200 1400 1600 1800 2000
6N
/ 1
0
10
20
30
10×
muons
]-2slant depth [gcm0 200 400 600 800 1000 1200 1400 1600 1800 2000
(N)
10
RM
S o
f lo
g
0.0
0.1
0.2
0.3 minimum of shower fluctuations
all simulations performed with CONEXRalf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 5 / 25
Air shower profile development (extended Heitler model)
n=1
n=2
n=3
charged neutral
primary
inel
π π
π+ -
o
n(charged)n(neutral) = 2
1 Xmax ∝ λinel + λe.m. · ln( E0
nmult Ec)
Oppenheimer, Heitler, Matthews (2005)
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 6 / 25
High energy hadronic interaction models
total / EmaxE0 0.2 0.4 0.6 0.8 1
pro
bab
ility
0
0.01
0.02
0.03
0.04
0.05
0.06
eV19EPOS, protons at 10eV19QGSJET, protons at 10
eV19QGSJETII, protons at 10eV19SIBYLL, protons at 10eV19NEXUS, protons at 10
total / Ee.m.E0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
pro
bab
ility
-410
-310
-210
-110
eV19EPOS, protons at 10
eV19QGSJET, protons at 10eV19QGSJETII, protons at 10
eV19SIBYLL, protons at 10
eV19NEXUS, protons at 10
(multiplicity)10
log1 1.5 2 2.5 3 3.5
pro
bab
ility
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08eV19EPOS, protons at 10
eV19QGSJET, protons at 10
eV19QGSJETII, protons at 10
eV19SIBYLL, protons at 10
eV19NEXUS, protons at 10dP
dXmax
∼ P(nmult)⊗P(ninel)⊗P(re.m.)⊗ . . .
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 7 / 25
HE models
macroscopic parameters
microscopic parameters
model
cross section multiplicity elasticity . . .
. . .
diffraction
. . .
parton distribution fragmentation
→ all macroscopic parameters are correlated
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 8 / 25
Composition - impact on Xmax
]-2 [gcmmaxX
600 700 800 900 1000 1100 1200
frac
tio
n
-410
-310
-210
-110proton (58%)helium (30%)iron (10%)gamma (2%)sum
Uncertain composition above ∼ 1015.5eV impacts all air shower observables
SIBYLL 1019eV
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 9 / 25
Unaccompanied hadrons
atmospheric boundary
ground level
hadron calorimeter
satellite
ballone
anti-coincidence array
protons
experimental results from: Nam et al. (1975), Siohan et al. (1978), Mielke et al. (1994), . . .
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 10 / 25
Unaccompanied hadrons
atmospheric boundary
ground level
hadron calorimeter
satellite
ballone
anti-coincidence array
protons
experimental results from: Nam et al. (1975), Siohan et al. (1978), Mielke et al. (1994), . . .
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 10 / 25
Unaccompanied hadrons
atmospheric boundary
ground level
hadron calorimeter
satellite
ballone
anti-coincidence array
protons heavier primaries
experimental results from: Nam et al. (1975), Siohan et al. (1978), Mielke et al. (1994), . . .
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 10 / 25
Unaccompanied hadrons - cross section
φ(X ) = 1λ · e−X/λ
Top of atmosphere
φtop = φ(Xtop = 0gcm−2) by satellites, orφtop = φ(Xtop ∼ 5gcm−2) ballones
Bottom of atmosphere
φbottom = φ(Xground) measured by calorimeter
Flux attenuation
λprod = ∆X/ log(φtop/φbottom) with ∆X = Xbottom − Xtop
σprod = σtot − σel − σq−el − σdiffr
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 11 / 25
Frequency attenuation
→ Attenuation of shower cascades in the atmosphere
ground level
air shower ground array
atmospheric boundary
same energy proton showers
Ne
constant Nµ ⇒ equal E0 constant Ne ⇒ same distance to Xmax
experimental results from: Honda et al. (1993), Hara et al. (1999), Aglietta et al. (1999), ...
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 12 / 25
Frequency attenuation
→ Attenuation of shower cascades in the atmosphere
ground level
air shower ground array
atmospheric boundary
same energy proton showers
Ne Ne
deeply penetra
ting
constant Nµ ⇒ equal E0 constant Ne ⇒ same distance to Xmax
experimental results from: Honda et al. (1993), Hara et al. (1999), Aglietta et al. (1999), ...
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 12 / 25
Frequency attenuation
→ Attenuation of shower cascades in the atmosphere
ground level
air shower ground array
atmospheric boundary
same energy proton showers
Ne
deeply penetra
ting
Ne Ne
constant Nµ ⇒ equal E0 constant Ne ⇒ same distance to Xmax
experimental results from: Honda et al. (1993), Hara et al. (1999), Aglietta et al. (1999), ...
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 12 / 25
Frequency attenuation - examples
AGASA like experiment (Xobs=920gcm−2, ∆(Ne )Ne
= 0.05,∆(Nµ)
Nµ
= 0.1)
(toy detector simulation, Φ ∼E−2.7, 0◦
< θ < 60◦, protons only)
)θsec(1.05 1.1 1.15 1.2 1.25 1.3
freq
uen
cy
[ar
bit
rary
un
its]
310
) binµ/Ne
frequency of events in (N-2=87.7138gcmΛexponential attenuation,
hist_0Entries 3390Mean 889.2RMS 58
]-2 [gcm1-XobsX600 700 800 900 1000 1100 1200 1300 1400
entr
ies
0
50
100
150
200
250
hist_0Entries 3390Mean 889.2RMS 58
/deg < 21θ 0 < /deg < 29θ 21 < /deg < 36θ 29 < /deg < 42θ 36 <
>1GeV)) < 6.25µ
(Eµ
(N10
6.05 < log
>1MeV)) < 8.4e
(Ee
(N10
8.2 < log
→ observed attenuation is only partly due to cross section
compare for example Alvarez-Muniz et al. (2002)
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 13 / 25
Ne sensitivity
]-2(attenuation) [gcmΛ
0 20 40 60 80 100 120
(p
roto
n-a
ir)
[m
b]
tot
σ
0
200
400
600
800
1000
1200
1400 qgsjetqgsjetIIsibyll
Energy reconstruction/selection accurate (E0 = 10EeV), ∆(Ne )Ne
= 0.05, Xobs=920gcm−2
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 14 / 25
Ne sensitivity
]-2(attenuation) [gcmΛ
0 20 40 60 80 100 120
(p
roto
n-a
ir)
[m
b]
tot
σ
0
200
400
600
800
1000
1200
1400 qgsjetqgsjetIIsibyll
Energy reconstruction/selection accurate (E0 = 10EeV), ∆(Ne )Ne
= 0.05, Xobs=920gcm−2
∆Λ ∼ 5gcm−2 ⇒ ∆σ ∼ 300mb
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 14 / 25
Xmax distribution: tail and fluctuations
]-2 [gcmmaxX
600 700 800 900 1000 1100 1200
frac
tio
n
-410
-310
-210
-110
))Λtail (exp(-X/
RMSλint = k · Λobs = <M>
σint
all detector effectsand fluctuations arecontained in k
RMS
Walker & Watson (Havera Park, 1982)Linsley (1985)
tail
Baltrusaitis et al. (Fly’s eye, 1984)Knurenko et al. (Yakutsk, 1999)
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 15 / 25
Xmax sensitivity
]-2) [gcmmax
(XΛ
40 50 60 70 80 90 100 110 120
(p
roto
n-a
ir)
[m
b]
tot
σ
0
200
400
600
800
1000
1200
1400
]-2) [gcmmax
RMS(X
40 50 60 70 80 90 100 110 120
(p
roto
n-a
ir)
[m
b]
tot
σ
0
200
400
600
800
1000
1200
1400qgsjet
qgsjetII
sibyll
Xmax-resolution is 30gcm−2. Energy reconstruction is accurate (E0 = 10EeV).
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 16 / 25
Xmax sensitivity
]-2) [gcmmax
(XΛ
40 50 60 70 80 90 100 110 120
(p
roto
n-a
ir)
[m
b]
tot
σ
0
200
400
600
800
1000
1200
1400
]-2) [gcmmax
RMS(X
40 50 60 70 80 90 100 110 120
(p
roto
n-a
ir)
[m
b]
tot
σ
0
200
400
600
800
1000
1200
1400qgsjet
qgsjetII
sibyll
Xmax-resolution is 30gcm−2. Energy reconstruction is accurate (E0 = 10EeV).
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 16 / 25
Deconvolution of Xmax-distribution
Xmax = X1 + ∆X
-2 / gcmmaxX600 700 800 900 1000 1100 1200 1300 1400
pro
bab
ility
-510
-410
-310
-210
-110
-2 / gcm1X0 100 200 300 400 500 600 700
pro
bab
ility
-510
-410
-310
-210
-110
-2 X /gcm∆600 700 800 900 1000 1100 1200 1300
pro
bab
ility
-510
-410
-310
-210
-110
⊗=
PXmax(Xmax) =
∫∞
0
dX1 PX1(X1) · P∆X (∆X |X1; HEM)
method proposed by Belov et al. (HiRes)
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 17 / 25
Independence of ∆X -distribution from X1
QGSJET01, protons 1018eV (∼200.000 profiles):
-2 X / gcm∆600 700 800 900 1000 1100 1200 1300
Pro
bab
ility
-610
-510
-410
-310
-210 < 13 (n=53876)2/gcmfirst
0< X
< 39 (n=70785)2/gcmfirst
13< X
< 78 (n=52277)2/gcmfirst
39< X
< 533 (n=38958)2/gcmfirst
78< X
X1 and ∆X are independent parameters P∆X (∆X |X1) = P∆X (∆X )
(confirmed for NEXUS3, SIBYLL2.1, QGSJETII.3, QGSJET01 for protons at 1018eV and 1019eV)
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 18 / 25
Dependence on HE model
sibyll_0.5Entries 9800Mean 766.6RMS 54.6
]-2 X [gcm∆600 650 700 750 800 850 900 950 1000 1050 1100
X∆d
n /
d
-410
-310
-210
sibyll_0.5Entries 9800Mean 766.6RMS 54.6
QGSJETII, factor=0.5 QGSJETII, factor=0.6 QGSJETII, factor=0.7 QGSJETII, factor=0.8 QGSJETII, factor=0.9 QGSJETII, factor=1.0 QGSJETII, factor=1.2 QGSJETII, factor=1.4 QGSJETII, factor=1.6 QGSJETII, factor=1.8 QGSJETII, factor=2.0
→ similar dependence on multiplicity expected
P∆X is a function of σ and other correlated HE model parameters (multiplicity,...)
CONEX, QGSJETII, 10 EeV
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 19 / 25
∆X -distribution for different HE-models
-2 X / gcm∆600 650 700 750 800 850 900 950 1000 1050 1100
pro
bab
ility
-610
-510
-410
-310
-210 qgsjetsibyllnexusqgsjetII
Position of maximum shifts up to ∼ 60gcm−2
Exponential slope after maximum changes up to a factor of ∼ 1.36
→ Model-dependence
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 20 / 25
Sensitivity of deconvolution
[mb]modelσ - modifiedσ-500 -400 -300 -200 -100 0 100 200 300 400 500
[
mb
]tr
ue
σ -
re
cσ
-500
-400
-300
-200
-100
0
100
200
300
400
500 (QGSJET)max
X(QGSJET), X∆
(QGSJETII)max
X(QGSJETII), X∆
E0 = 1019eVRalf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 21 / 25
Sensitivity of deconvolution
[mb]modelσ - modifiedσ-500 -400 -300 -200 -100 0 100 200 300 400 500
[
mb
]tr
ue
σ -
re
cσ
-500
-400
-300
-200
-100
0
100
200
300
400
500 (QGSJET)max
X(QGSJET), X∆
(QGSJETII)max
X(QGSJETII), X∆
(SIBYLL)max
X(QGSJET), X∆(QGSJETII)
max X(QGSJET), X∆
model uncertainty band
E0 = 1019eVRalf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 21 / 25
Summary
measuring σp−air means measuring EAS fluctuations
assuming CR primary interaction length the only source of EASfluctuations is not enough. At least needed:
Additional HE interaction characteristics (multiplicity,...)
The first few interactions at still extreme energy
diffraction
Meaningful estimate of uncertainty needs: HE model dependence and CRcomposition x
Future
→ better understanding of fluctuations
→ measurement of composition
→ new experiments (fluorescence/cherenkov, muons, ...)
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 22 / 25
Summary
measuring σp−air means measuring EAS fluctuations
assuming CR primary interaction length the only source of EASfluctuations is not enough. At least needed:
Additional HE interaction characteristics (multiplicity,...)
The first few interactions at still extreme energy
diffraction
Meaningful estimate of uncertainty needs: HE model dependence and CRcomposition x
Future
→ better understanding of fluctuations
→ measurement of composition
→ new experiments (fluorescence/cherenkov, muons, ...)
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 22 / 25
Summary
measuring σp−air means measuring EAS fluctuations
assuming CR primary interaction length the only source of EASfluctuations is not enough. At least needed:
Additional HE interaction characteristics (multiplicity,...)
The first few interactions at still extreme energy
diffraction
Meaningful estimate of uncertainty needs: HE model dependence and CRcomposition x
Future
→ better understanding of fluctuations
→ measurement of composition
→ new experiments (fluorescence/cherenkov, muons, ...)
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 22 / 25
Final rating
method HE-model composition fluctuationsunaccompanied hadrons ⊙ ⊙ ⊕frequency attenuation ⊖ ⊖ ⊖⊖Xmax RMS ⊖ ⊖⊖ ⊖Xmax tail ⊖ ⊙ (⊕) ⊖Xmax deconvolution ⊖⊖ ⊖ ⊙
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 23 / 25
Final - ’conservative’ picture
(Energy/eV)10
log10 11 12 13 14 15 16 17 18 19
Cro
ss s
ecti
on
(p
roto
n-a
ir)
[m
b]
250
300
350
400
450
500
550
600
650
700
Mielke et al. 1994
Nam et al. 1975
Siohan et al. 1978
QGSJET01
EPOS
NEXUS3
SIBYLL2.1
QGSJETII.3
Energy [GeV]210 310 410 510 610 710 810 910 1010
[GeV]ppsEquivalent c.m. energy 10 210 310 410 510
Tevatron (p-p)
LHC (p-p)
LHC (C-C)
lower limits
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 24 / 25
Final - ’progressive’ picture
(Energy/eV)10
log10 11 12 13 14 15 16 17 18 19
Cro
ss s
ecti
on
(p
roto
n-a
ir)
[m
b]
250
300
350
400
450
500
550
600
650
700
Mielke et al. 1994
Baltrusaitis et al. 1984
Nam et al. 1975
Siohan et al. 1978
QGSJET01
EPOS
NEXUS3
SIBYLL2.1
QGSJETII.3
Energy [GeV]210 310 410 510 610 710 810 910 1010
[GeV]ppsEquivalent c.m. energy 10 210 310 410 510
Tevatron (p-p)
LHC (p-p)
LHC (C-C)
HE
mo
del
⇒ weak constraints on HE interaction models
98% confidence limit
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 25 / 25
Changing composition - data
Energy [eV]10
1710
1810
1910
20
]2>
[g/c
mm
ax<
X
500
550
600
650
700
750
800
850
900
950
1000
QGSJET01SIBYLL 2.1
DPMJET 2.55
HiRes-MIA
HiRes
Yakutsk 2001
Fly’s Eye
Yakutsk 1993
proton
iron
-ray
γ
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 26 / 25
Composition - model by Hillas
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 27 / 25
Equal intensity cut
)θsec(1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9
inte
nsi
ty
[arb
itra
ry u
nit
s]
-910
-810
-710
-610) < 8.25
ch(N
10 8 < log
) < 8.5ch
(N10
8.25 < log
) < 8.75ch
(N10
8.5 < log
) < 9ch
(N10
8.75 < log
constant intensity → same primary energy (isotropic flux)
relates shower size Nch at different zenith angle with primary energy
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 28 / 25
Equal intensity cut
)θsec(1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9
inte
nsi
ty
[arb
itra
ry u
nit
s]
-910
-810
-710
-610) < 8.25
ch(N
10 8 < log
) < 8.5ch
(N10
8.25 < log
) < 8.75ch
(N10
8.5 < log
) < 9ch
(N10
8.75 < log
constant intensity
constant intensity → same primary energy (isotropic flux)
relates shower size Nch at different zenith angle with primary energy
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 28 / 25
Equal intensity cut
)θsec(1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9
inte
nsi
ty
[arb
itra
ry u
nit
s]
-910
-810
-710
-610) < 8.25
ch(N
10 8 < log
) < 8.5ch
(N10
8.25 < log
) < 8.75ch
(N10
8.5 < log
) < 9ch
(N10
8.75 < log
constant intensity
constant intensity → same primary energy (isotropic flux)
relates shower size Nch at different zenith angle with primary energy
Ralf Ulrich ([email protected]) On the measurement of the proton-air cross section using air shower data 28 / 25