1. Uppsala University June 9, 2009
Solar Physics with the
IceTop Air Shower Array
Paul Evenson
University of Delaware (aka New Sweden)
Department of Physics and Astronomy
Paul Evenson June 2009

2. The First Extraterrestrial Event Detected by IceCube
IceTop and Spaceship Earth Observations of the Solar Flare
Dec 13, X3-Class Solar Flare (SOHO)
Paul Evenson June 2009
Dec 14, photograph of auroras near Madison, WI

3. Solar Flares
Somewhere in this picture, particles are being accelerated to GeV energy.
Can you tell where?
I certainly can’t!
Possibly different mechanisms are even operating at the same time.
Paul Evenson June 2009

4. Energy Transport – Conduction and Convection Zones
It takes approximately one million years for the energy to be conducted (by radiation to the outer part of the sun.
Near the surface, convective motion sets in
Approximately 100,000 years of sunlight is stored in the convection zone
Magnetic phenomena partially control release of this energy
“Little Ice Ages”
Paul Evenson June 2009

5. Magnetic Dynamo
The churning of the convection cells, and differential rotation of the sun generate the solar magnetic field
Paul Evenson June 2009

6. Magnetic Energy Also Powers Flares
Exact mechanism is unknown
Production of GeV particles is common, but the mechanism is also unknown
Shocks produced by the release are a popular candidate
We want to study the spectrum and relative timing of the energetic particles
Understanding may also shed light on astrophysical particle accelerators
IceTop was designed to measure the spectrum of high energy cosmic rays
Paul Evenson June 2009

7. The animation illustrates an air shower
This is an aerial view of our small air shower array (SPASE) at the South Pole.
The animation illustrates an air shower
Paul Evenson June 2009

8. IceTop Detectors Blocks of clear ice produced in tanks at the Pole
Cherenkov radiation measured by standard IceCube photon detectors
Sweden is one production site for these “Digital Optical Modules” or DOM
Two tanks separated by 10 meters form a station
2 m
0.9 m
Diffusely reflecting liner
Paul Evenson June 2009

9. Large showers with E ~ 100-1000 PeV
will clarify transition from galactic to
extra-galactic cosmic rays.
Showers triggering 4 stations give
~300 TeV threshold for EAS array
Small showers (2-10 TeV)
associated with the dominant
muon background in the deep detector are detected as 2-tank coincidences at a station.
Paul Evenson June 2009

10. Low Energy (1 GeV) “Showers”
Particles with energy as low as 1 GeV produce secondaries that survive to the surface
Rarely does a single detector see more than one secondary from a primary
Large detectors can have high enough counting rates to make statistically significant measurements of the primary flux
Conventional detectors count muons or neutrons
Energy spectra are determined from detectors located at different geomagnetic cutoffs
Flux anisotropy is a source of error in the spectral measurement
Paul Evenson June 2009

11. Why IceTop Works as a GeV Particle Spectrometer
The IceTop detectors are thick (90 g/cm2) so the Cherenkov light output is a function of both the species and energy of incoming particles
Individual waveform recording, and extensive onboard processing, allow the return of pulse height spectra with 10 second time resolution even at the kilohertz counting rate inherent to the detector
Paul Evenson June 2009

12. Secondary Particle Spectra
At the South Pole, spectra of secondary particles “remember” a lot of information about the primary spectrum.
Paul Evenson June 2009

13. Particle Response Functions (Arbitrary Normalization)
IceTop tank particle response functions change with selection of the threshold
We are now working with FLUKA calculated response functions
Paul Evenson June 2009

14. IceTop Event Overview
A lot of this structure is due to pressure variations
Much is due to cosmic ray variability
The flare event and the “Forbush Decrease” at the end of day 347 are clear
The blast of plasma that produces the decrease is what triggers the anomalous auroral activity
Paul Evenson June 2009

15. Solar Particle Spectrum Determination (I)
Excess count rate (averaged over approximately one hour near the peak of the event) as a function of pre-event counting rate.
Each point represents one discriminator in one DOM.
By using the response function for each DOM we fit a power law (in momentum) to the data
The lines show this fit and the one sigma (systematic) errors
Paul Evenson June 2009

16. Solar Particle Spectrum Determination (II)
IceTop proton spectrum (heavy blue line with one sigma error band).
Black line is the assumed background cosmic-ray proton spectrum
Points are maximum proton fluxes from GOES spacecraft data.
Paul Evenson June 2009

17. Putting the IceTop Observation in Context
Spaceship Earth
Neutron Monitor Array
Spaceship Earth is a network of neutron monitors strategically deployed to provide precise, real-time, 3-dimensional measurements of the angular distribution of solar cosmic rays:
12 Neutron Monitors on 4 continents
Multi-national participation:
Bartol Research Institute, University of Delaware (U.S.A.)
IZMIRAN (Russia)
Polar Geophysical Inst. (Russia)
Inst. Solar-Terrestrial Physics (Russia)
Inst. Cosmophysical Research and Aeronomy (Russia)
Inst. Cosmophysical Research and Radio Wave Propagation (Russia)
Australian Antarctic Division
Aurora College (Canada)
Paul Evenson June 2009

18. Cosmic Ray Detectors at High Latitude
Trajectories are shown for vertically incident primaries corresponding to the 10-, 20-, … 90-percentile rigidities of a typical solar spectrum
Paul Evenson June 2009

19. Station Location is Carefully Chosen
Circles denote station geographical locations.
Average asymptotic direction (squares) and range (lines) are separated from station geographical locations.
STATION CODES
IN: Inuvik, Canada
FS: Fort Smith, Canada
PE: Peawanuck, Canada
NA: Nain, Canada
BA: Barentsburg, Norway
MA: Mawson, Antarctica
AP: Apatity, Russia
NO: Norilsk, Russia
TB: Tixie Bay, Russia
CS: Cape Schmidt, Russia
TH: Thule, Greenland
MC: McMurdo, Antarctica
Paul Evenson June 2009

20. Determination of the Pitch Angle Distribution
Individual station data fitted to an angular distribution of the form
f(μ) = c0 + c1 exp(b μ),
with μ cosine of pitch angle, and c0, c1, and b free parameters. The symmetry axis from which pitch angles are measured was also a free parameter.
Paul Evenson June 2009

21. 13 December 2006 Event Animation
Paul Evenson June 2009

22. Neutron Monitor Response Calculated from IceTop Spectrum
Good agreement (with understanding of viewing direction)
Continuous determination of precise spectrum
All information on anisotropy comes from the monitor network
Paul Evenson June 2009

23. Towards Precision Spectral Information
We are reconfiguring IceTop to provide uniform coverage from 500 to 10,000 Hz
Up to 2000 Hz each DOM will generate a rate histogram
Above 2000 Hz the SPE discriminators will be used, set to a range of thresholds
For larger events this will enable us to go far beyond the simple power spectrum analysis
Exact spectral shape is diagnostic of particle acceleration mechanisms.
Paul Evenson June 2009

24. Standard CalibrationTechnique: Latitude Survey
By observing the change in counting rate as a function of geomagnetic cutoff, response functions can be directly measured
Paul Evenson June 2009

25. Our Plan
Assemble an IceTop tank in a portable freezer in Stockholm / Uppsala / Landskrona (TBD)
Fill with water and freeze
Load on Oden
Take data on voyage to McMurdo and back
If this works, do it again another year but bring it back on a C-17 to do the survey at high altitude
Paul Evenson June 2009

26. Container Used For Latitude Surveys On Polar Star And Polar Sea
Paul Evenson June 2009

27. Example of a Latitude Survey
Left: Course plot with geomagnetic cutoff contours in units of GV
Right: Counting rate of two detectors and geomagnetic cutoff as a function of time
Paul Evenson June 2009

28. Determining a Response Function
Counting rate of a 3NM64 is plotted against cutoff.
To deal with the scatter of the points we fit a “Dorman Function” as indicated.
This function has an analytic derivative as shown. This derivative is exactly the response function for this particular detector.
In the collaboration involving Uppsala we will derive a whole sequence of response functions for different thresholds, but each will be determined by a method similar to this
Paul Evenson June 2009

29. Summary
IceTop, the surface component of the IceCube Neutrino Observatory at the South Pole, is an air shower detector aimed primarily at studying PeV and above cosmic rays.
In this mode the 160 (planned) detectors are operated in coincidence.
Individual detectors are sensitive to the secondary products of particles with energy as low as one GeV, typical of energetic solar particle events.
Information on the spectrum of particles at a few GeV can be extracted from IceTop,
We are working to get a better calibration of these detectors as part of a collaboration including Uppsala University.
Paul Evenson June 2009