Gareth Hughes

1
A Measurement of the Average Longitudinal Shower
Development Profile

• Gareth Hughes
• High Resolution Flys Eye Collaboration
• 2007 30th ICRC Merida Mexico

2
Introduction
3
Motivation

• Highest energy interactions on Earth!
• We dont see 1st 200g/cm2
• Future experiments will be able to see up to 1st
  100g/cm2
• (TALE)
• Must take all we can from these events
• Learn what we can now
• Prepare the way
• Current best method is Fluorescence

4
Previous Work

• Work first done in
• HiRes/MIA Prototype
• T. Abu-Zayyad et al., A Measurement of the
  average Longitudinal Development Profile of CR
  Air showers,
• Reference
• Now
• More statistics
• Improved Monte Carlo
• 2 orders of magnitude in energy range
• Previously lt 1018.0eV
• Now lt 1020.0eV

5
Extensive Air Showers (EAS)

• Cascade Initiated by CRs incident on the
  atmosphere
• Profile often described by Gaisser-Hillas
  function
• Method of Constant Intensity Cuts
• Reference

6
Method
7
Method

• Make quality cuts
• well defined showers
• Standard spectrum cuts
• Track length gt 200g/cm2
• ? lt 110o
• Extra Bracketing -50g/cm2
• Cerenkov Fraction lt 0.35
• Locally Fit Shower Profiles Near Nmax
• Nmax and Xmax
• Normalize

8
Method

• Re-write the Gaisser-Hillas as
• With 2 free parameters

9
Gaussian in Age

• One free parameter
• ? Shower Width
• Symmetric about s1
• Normalized to n(1)1

10
Average Shower

• Black points mean of the blue
• Gaussian fits in bins of age
• Fit to Normalized
• Gaisser-Hillas
• Gaussian in Age

11
Monte Carlo

• We can do the same with Monte Carlo
• Thrown using Discrete Corsika shower library
• QGSJET Proton and Iron
• Parameterized as shown
• Put through the Detector Simulation

12
Results
13
1. Data Monte Carlo
14
Data Monte Carlo Comparison

• Normalized height against shower age
• Top Good agreement between Data and Monte Carlo
• Black Data
• Red Monte Carlo
• Bottom Ratio of Data/Monte Carlo
• Flat from 0.5 to 1.3 in Age

E gt 1018.5eV
15
2. Average Shower
16
Average Showers

• Black points mean of the blue
• Gaussian fits in bins of age
• Make average showers for half decade bins in
  energy
• Good fits above 1018.5eV
• ?2/dof few
• Flat residuals within errors

17
Shower Parameters
18
? Resolution
19
? Resolution

• Energy dependant ? resolution that effects
  reconstruction

20
? Resolution

• Compare Monte Carlo reconstructed with True
  value of ?
• Shows us where we can fit
• Do the Gaussian fits to find the widths

21
3. Shower Widths ?
22
Monte Carlo Result

• 80 mix of Protons and 20 Iron
• Get back what we put in
• Consistent across all energies

23
? Resolution again

• When we look at the width of True ?
• see a constant shift

24
Bootstrap Errors

• Create 100 sets of the same showers with
  replacement
• Measure the widths for the 100 sets
• Find the width of this distribution
• This is the error

25
Data and Monte Carlo Result

• Good agreement
• Except the highest energy bin
• 3.5? significance
• What is this?
• Low statistics

26
Atmospheric Database
27
Atmospheric Database

• Able to throw and reconstruct data and Monte
  Carlo with hourly database
• Looking at the widths of the data we see no
  difference within errors

28
Conclusion

• We have a developed a method to measure the
  Average Longitudinal Shower
• Measured shower parameters as a function of
  energy
• Compared data to Gaisser-Hillas Monte Carlo
• Shows good agreement
• Systematic studies of ? and Atmospheric Database
• ? Gives a constant shift in width
• Atmospheric database no change

29
(No Transcript)
30
(No Transcript)
31
Composition Studies

• CORSIKA Studies by Song
• C. Song Astroparticle Physics 22 (2004) 151158
• Separation of Xmax and Sigma(?)

32
Adding Resolution

• In Monocular Mode
• With better ? resolution ?

33
Compare to Raw QGSJET Showers

• Apply method to raw QGSJET showers
• No detector simulation
• See the same trend
• Different slope
• Top of the mirror?
• Separation is approx. the same

34
Composition

• Could it be that composition changes?
• More Iron like?

35
Previous Composition Measurements

• Not seen in ltXmaxgt
• Including golden events
• Or Dougs HiRes mono composition

36
LPM Effect?

• Landau, Pomeranchuk and Migdal
• Bremsstrahlung and pair production cross section
  ? E-1/2
• Above a sufficiently high energy
• Dense enough medium
• In contrast to Bethe-Heitler energy independence
• ? Showers will be broader
• Now can be included in CORSIKA