PROCEEDINGS OF THE 31st ICRC, ŁODZ 2009 1

Precise measurement of the cosmic-ray proton spectrumand the time variation with BESS-Polar I

R. Orito∗,1, K. Abe∗,2, H. Fuke†, S. Haino‡,3, T. Hams§,4, A. Itazaki∗, K. C. Kim¶,T. Kumazawa‡, M. H. Lee¶, Y. Makida‡, S. Matsuda‡, K. Matsumoto‡, J. W. Mitchell§,A. A. Moiseev§, Z. Myers¶,5, J. Nishimura‖, M. Nozaki‡, J. F. Ormes∗∗, M. Sasaki§,6,E. S. Seo¶, Y. Shikaze∗,7, R. E. Streitmatter§, J. Suzuki‡, Y. Takasugi∗, K. Takeuchi∗,

K. Tanaka‡, T. Yamagami†, A. Yamamoto‡, T. Yoshida† and K. Yoshimura‡∗ Kobe University, Kobe, Hyogo 657-8501, Japan

† Institute of Space and Astronautical Science, Japan Aerospace Exploration Agency (ISAS/JAXA),Sagamihara, Kanagawa 229-8510, Japan

‡ High Energy Accelerator Research Organization (KEK),Tsukuba, Ibaraki 305-0801, Japan

§ National Aeronautics and Space Administration,Goddard Space Flight Center (NASA/GSFC), Greenbelt, MD 20771, USA

¶ IPST, University of Maryland, College Park, MD 20742, USA‖ The University of Tokyo, Bunkyo, Tokyo 113-0033 Japan

∗∗ Denver University, Denver, CO 80208, USA

Abstract. The Balloon-borne Experiment with a Su-perconducting Spectrometer over Antarctica (BESS-Polar) had two successful flights in December 2004and 2007. During the first flight in 2004 (BESS-PolarI), 0.9 billion cosmic-ray events have been recordedin an 8.5-days long-duration flight. Here we report onthe primary cosmic-ray proton and helium spectrameasured with BESS-Polar I. Using the spectrometerwith a high rigidity resolution, we measured preciseproton and helium spectra in kinetic energy rangesof 0.22-100 GeV/n and 0.22-54 GeV/n, respectively.With the detailed analysis of detector response andthe correction of atmospheric secondary particlesin short-term period, we also obtained the timevariation of the primary proton and helium fluxes.

Keywords: cosmic-ray proton, superconductingspectrometer, long-duration balloon, solar modula-tion

I. INTRODUCTION

Among cosmic-ray particles, protons and helium nu-clei are the dominant components. The absolute fluxesand energy spectra of primary cosmic-ray protons andhelium nuclei are fundamental information in cosmic-ray physics. The interstellar spectra carry importantinformation on the origin and propagation history ofcosmic-rays in the Galaxy. The spectra are also essentialas an input to calculate spectra of cosmic-ray antiprotonsand positrons which are mainly secondary products ofcosmic-ray interactions with the interstellar gas. At lowenergies, observable spectra in the heliosphere are muchdeformed in a manner that depends on the solar activity.Therefore the low-energy proton and helium spectraprovide significant information on the solar activityand the propagation of cosmic-rays. Here we report onthe measurement of the cosmic-ray proton and helium

spectra in energy ranges of 0.22-100 GeV/n and 0.22-54 GeV/n, respectively, based on the data from thefirst Balloon-borne Experiment with a SuperconductingSpectrometer over Antarctica (BESS-Polar I) which wascarried out in 2004. The BESS-Polar experiment wasproposed as an advanced BESS[1][2][3][4] programwith long-duration balloon flights over Antarctica andhad been prepared since 2001[5][6][7]. The result ofBESS-Polar I antiproton spectrum with discussions forprimary antiprotons, which is the main science objectiveof the BESS-Polar project, is described in Ref. [8].The BESS-Polar spectrometer features high-precisionrigidity measurement with wide rigidity range, large ge-ometrical acceptance, and small material thickness. Wewere successful to perform the precise measurement ofcosmic-ray proton and helium fluxes and its short-termvariation with this long-duration flight, in a transientperiod to the solar minimum. Further discussion andphysics interpretation for proton short-term variation aredescribed elsewhere in these proceedings[9].

II. BESS-POLAR I SPECTROMETER

Figure 1 represents a cross-sectional view of BESS-Polar I spectrometer, which consists of a top, bottom,and middle time-of-flight(TOF) hodoscopes[10], a JET-type and two inner drift chambers (JET/IDC)[11][12], anAerogel Cherenkov Counter (ACC)[13] and a supercon-ducting solenoid[14][15]. The rigidity (R) of incomingparticles is measured by the JET/IDC central tracker in a0.8 Tesla uniform magnetic field which is produced bythe thin superconducting solenoid. The TOF countersmeasure the velocity (β), and provide three independentionization energy loss (dE/dx) measurements in thescintillators. Particle identification by mass is performedwith these measurements (R, β, and dE/dx). ACC isalso installed for the background (e±,µ±) rejection.

2 R. ORITO et al. BESS POLAR-I PROTON

Magnet

JET

IDC

IDC

TOF

TOF

MTOF

ACC

Incoming particle

0 1m

Fig. 1. Cross-sectional view of BESS-Polar I spectrometer.

The event trigger is generated by the simple coinci-dence of the signal of top-bottom TOFs or top-middleTOFs. The total cross-sectional mass of the BESS-Polarspectrometer is only 4.5 g/cm2 for top-middle TOFscoincidence events to measure low-energy cosmic-rayparticles precisely.

III. BESS-POLAR I FLIGHT

The BESS-Polar I was launched from Williams Field(77◦51.8′S, 167◦5.4′E) near McMurdo Station on De-cember 13, 2004[16]. During the flight, some PMTsof the TOF hodoscopes had to be turned off becauseof their excessive drawing current. The resulting usablegeometrical acceptance was consequently reduced to 73percent of the design value. The flight was terminatedon December 21 and the payload landed at the southend of the Ross Ice Shelf (83◦6.0′S, 155◦35.4′W) after acontinuous observation period of 8.5 days. The flight tra-jectory was so close to the South magnetic pole that thegeomagnetic cutoff rigidity was below 0.2 GV, as shownin Fig. 2. During a live data-taking time of 507,075seconds at an average floating altitude of 38.5 km(residual atmosphere of 4.3 g/cm2), 894,482,590 cosmic-ray events were accumulated.

IV. DATA ANALYSIS

A. Data Reduction

For the analysis of protons and helium nuclei, only thetop and bottom TOF hodoscopes were used. As the firstselection, events with (1) a single track, a downward-going and a positive rigidity particle fully containedin the fiducial region of the tracking volume, (2) onlyone or two hits each in the top and bottom TOFs, (3)the hit position of r-φ direction at the TOFs consistentwith the extrapolated track inside the JET and IDCs,(4) cosine of zenith angle of the incident particle largerthan 0.8, and (5) either top or bottom TOFs read byPMTs at both ends, are selected. Then as the fiducialselection, events were required to pass through thefiducial volume defined by the central region of the JETchamber and TOFs. Events entering unstable IDC region

Fig. 2. Flight trajectory of BESS-Polar I 2004.

or noisy JET region were also not used. After the fiducialselection, the quality cut was applied, which requiredthe residual between hit position at TOF obtained fromthe time difference in two PMTs and the extrapolatedtrack of the JET to be less than 80 mm. Each analog-to-digital board for TOF has an individual dead timedue to the period of switching reference capacitors usedfor baseline subtraction[10]. Events digitized during thedead time have incorrect charge data. We also rejectedthese events, which are 52% of total data, to ensure thequality of the data.

B. Particle Identification

After the above event selections, the particle iden-tification was performed. The charge of the particleis identified by the ionization loss measurement. BothdE/dx signals from top and bottom hodoscopes wererequired to be proton-like. Then, particles with protonmass were selected by a 1/β-band cut. The selectionof protons with the dE/dx-band cut and 1/β-band cutare shown on top and bottom in Fig. 3, respectively.Particle identification for helium nuclei was performedin the same way. However, 3He nuclei were included in1/β-band cut and were counted as helium-like events. Inconformity with previous experiments, all the helium-like events were treated as 4He nuclei in the analysissuch as the reconstruction from rigidity to kinetic energy,and the efficiency estimation with the Monte Carlo (MC)simulation.

C. Flux determination

1) TOI: The numbers of protons and helium nucleipassing through the BESS spectrometer during the ob-servation were obtained after correcting the detectionefficiency. Then absolute flux at the top of the instrument(TOI) was obtained by taking account of the energy lossinside the detector, live time, and geometrical accep-tance. The energy of each particle at TOI was calculatedby summing up the ionization energy losses inside theinstrument by tracing back the event trajectory. The

PROCEEDINGS OF THE 31st ICRC, ŁODZ 2009 3

Rigidity (GV)1 10

dE/d

x (T

op T

OF

)

0

2

4

6

8

10

Rigidity (GV)1 10

dE/d

x (B

otto

m T

OF

)

0

2

4

6

8

10

Rigidity (GV)1 10

β1/

0.5

1

1.5

2

2.5

3

3.5

4

p

d

Fig. 3. Proton selection by dE/dx-band cut for top (top-left) andbottom (top-right) TOF hodoscopes and 1/β-band cut(bottom).

geometrical acceptance and some of selection efficien-cies were estimated using a MC simulation by applyingthe same selection criteria to the simulated events asto the observed data. The MC simulation was basedon a GEANT3/GHEISHA code[17][18] and tuned toreproduce the results of an accelerator beam test of theprevious BESS spectrometer[19] in which the detectorconfiguration and materials are similar to the BESS-Polar spectrometer. The effective geometrical acceptancewas 0.053 m2sr at 2 GeV/n for protons. The efficienciesfor particle identifications, trigger, track reconstruction,accidental TOF hits, accidental tracks, quality cut, wereestimated from real data.

2) TOA: The flight-averaged flux at TOI was mea-sured under the residual atmosphere of about 4.3 g/cm2.It consists of a primary component surviving withoutany nuclear interactions with air, and of a secondarycomponent produced inside the overlying atmosphere.In order to obtain the flux at the top of the atmosphere(TOA), secondary proton production and interaction lossof primary particles inside the atmosphere were esti-mated by solving simultaneous transport equations[20]with the modified production spectrum to reproducethe observed proton spectra which were measured byprevious BESS flights at several atmosphere depths[21].The primary spectrum at TOA was determined in aniterative procedure so that the estimated spectrum atTOI agrees with the observed one. Figure 4 shows acalculated TOI/TOA correction factor for proton andhelium spectra at the atmospheric depth 4.3 g/cm2.

V. RESULTS AND DISCUSSIONS

We obtained the absolute primary fluxes of protonsand helium nuclei at TOA in the kinetic energy rangesof 0.22-100 GeV/n and 0.22-54 GeV/n, respectively. The

Kinetic Energy (GeV/n)1 10

TO

I/T

OA

Cor

rect

ion

Fac

tor

0.8

1

1.2

1.4

1.6

1.8

Proton

Helium

Fig. 4. Calculated atmospheric correction factor for proton and heliumspectra at the atmospheric depth 4.3 g/cm2.

flight-averaged fluxes which were obtained from all 8.5-days data are shown in Fig. 5, 6 together with resultsfrom previous BESS flights[12][21][22]. The uncertain-ties including statistical and systematic errors are ±7.5%at 0.23 GeV/n and ±3% at 93 GeV/n for protons, ±4%at 0.23 GeV/n and±5.5% at 50 GeV/n for helium nuclei.In low energies, the resultant proton and helium fluxesare between BESS-1999 and BESS-2002 as shown inFig. 5. The effect of the solar modulation was clearlyobserved. In high energies, the resultant fluxes of BESS-Polar I were consistent with BESS-1998 and BESS-2002. Figure 7 shows a time variation of proton andhelium fluxes normalized to the flight-averaged fluxesin kinetic energy ranges of 0.22-0.50, 0.50-1.0, 1.0-3.5,3.5-10 GeV/n, together with the residual air during theflight and Bartol South Pole neutron monitor[23]. Herethe time-divided data analysis has been done with thesame way of the flight-averaged flux. The residual airabove the instrument and its TOI/TOA correction factorfor protons varied between 3.8-5.3 g/cm2 and 1.45-1.63at 0.23 GeV/n respectively with these time bins. Furtherdiscussion for the short-term variation of protons and itsrelation to the solar activity is described elsewhere[9].

VI. CONCLUSION

The BESS-Polar experiment had a first successfullong-duration flight in December 2004. We reportedthe absolute fluxes and energy spectra of the primarycosmic-ray protons and helium nuclei in kinetic energyranges of 0.22-100 GeV/n and 0.22-54 GeV/n respec-tively, during the transient period to the solar minimum.Using the advantage of long-duration flight and largegeometrical acceptance, the fluxes with short-term pe-riod have been also calculated. These precisely measureddata are fundamental information on the origin andpropagation history of cosmic-rays and also importantto study the effect of solar activity.

ACKNOWLEDGEMENTS

The authors thank NASA Headquarters for the con-tinuous encouragement in this U.S.-Japan cooperativeproject. Sincere thanks are expressed to the NASA Bal-loon Programs Office at GSFC/WFF and CSBF for theirexperienced support. They also thank ISAS/JAXA andKEK for their continuous support and encouragement.

4 R. ORITO et al. BESS POLAR-I PROTON

Kinetic energy per nucleon (GeV/n)-110 1 10

n)

-1 G

eV-1

sec

-1 s

r-2

Pro

ton

and

Hel

ium

flu

x (m

-210

-110

1

10

210

310

BESS-1997

BESS-1998

BESS-1999

BESS-2000

BESS-2002

BESS-Polar 2004

Proton flux

Helium flux x 1/10

Fig. 5. Proton and helium fluxes at the top of atmosphere ob-tained with BESS-Polar I together with results from previous BESSflights[21]. Note that helium fluxes are plotted after being divided bya factor of 10.

Kinetic energy per nucleon (GeV/n)1 10 210

)1.

5 (

GeV

/n)

-1 s

ec-1

sr

-2 (

mk2.

5F

lux

x E

210

310

410

BESS-1998

BESS-2002

BESS-Polar 2004

BESS-1998

BESS-2002

BESS-Polar 2004

Proton

Helium

Fig. 6. Proton and helium fluxes at the top of atmosphere ob-tained with BESS-Polar I together with results from previous BESSflights[12][22] which measured cosmic-ray fluxes up to high energies.

Special thanks go to the National Science Foundation(NSF), U.S.A., and Raytheon Polar Service Company fortheir professional support in U.S.A. and in Antarctica.The BESS-Polar experiment is being carried out as aJapan-U.S. collaboration, and is supported by a KAK-ENHI(3001004 and 18104006) in Japan, and by NASAgrants in U.S.A.

1Present address: Max-Planck-Institut fur Physik, Munchen 80805,Germany

2Present address: Kamioka Observatory, ICRR, The University ofTokyo, Kamioka, Gifu 506-1205, Japan

3Present address: Istituto Nazionale di Fisica Nucleare (INFN),Perugia 06123, Italy

4Also at University of Maryland Baltimore County, Baltimore, MD,21250

5Present address: Physics Department, Technion Israel Institute ofTechnology, Technion City, Haifa 32000, Israel

Days of Dec.200414 15 16 17 18 19 20 21 22

Rel

ativ

e F

lux

0.9

0.95

1

1.05

1.1

0.22 - 0.50 GeV/n

Proton

Helium

Days of Dec.200414 15 16 17 18 19 20 21 22

Rel

ativ

e F

lux

0.9

0.95

1

1.05

1.1

0.50 - 1.00 GeV/n

Proton

Helium

Days of Dec.200414 15 16 17 18 19 20 21 22

Rel

ativ

e F

lux

0.9

0.95

1

1.05

1.1

1.00 - 3.50 GeV/n

Proton

Helium

Days of Dec.200414 15 16 17 18 19 20 21 22

Rel

ativ

e F

lux

0.9

0.95

1

1.05

1.1

3.50 - 10.00 GeV/n

Proton

Helium

Days of Dec.200414 15 16 17 18 19 20 21 22

]2R

esid

ual A

ir [

g/cm

3.5

4

4.5

5

5.5

Residual Air

Days of Dec.200414 15 16 17 18 19 20 21 22

cou

nts/

h)5

Rat

e (1

0

10

10.2

10.4

10.6

10.8

Bartol South Pole Neutron Monitor

Fig. 7. Short-term variation of the proton and helium fluxesnormalized to the flight-averaged fluxes in kinetic energy ranges of0.22-0.50, 0.50-1.0, 1.0-3.5, 3.5-10 GeV/n. The data of residual airduring the flight and Bartol South Pole neutron monitor[23] are alsoshown.

6Also at University of Maryland, College Park, MD, 207427Present address: Japan Atomic Energy Agency (JAEA), Tokai-

mura, Naka-gun, Ibaraki 319-1195, Japan

REFERENCES

[1] S. Orito, in: Proceedings of ASTROMAG Workshop, KEKReport KEK87-19, 1987, p. 111.

[2] A. Yamamoto, et al., IEEE Trans. Magn. 24 (1988) 1421.[3] Y. Ajima, et al., Nucl. Instrum. Meth. A 443 (2000) 71.[4] S. Orito, et al., Phys. Rev. Lett. 84 (2000) 1078.[5] A. Yamamoto, et al., Adv. Space Res. 30 (5) (2002) 1253.[6] J.W. Mitchell, et al., Nucl. Phys. B (Proc. Suppl.) 134 (2004)

31.[7] T. Yoshida, et al., Adv. Space Res. 33 (10) (2004) 1755.[8] K. Abe, et al., Phys. Lett. B 670 (2008) 103.[9] T. Hams, et al., “Short-term and diurnal proton flux variation

during the BESS-Polar I balloon flight”, in these proceedings.[10] M. Sasaki, et al., in: Proc. 29th Int. Cosmic Ray Conf. (Pune),

vol. 3, 2005, p. 421[11] S. Haino, et al., Nucl. Instrum. Methods A 518 (2004) 167.[12] S. Haino, et al., Phys. Lett. B 594 (2004) 35.[13] T. Hams, et al., in: Proc. 29th Int. Cosmic Ray Conf. (Pune),

vol. 3, 2005, p. 69.[14] A. Yamamoto, et al., IEEE Trans. Appl. Supercond. 12 (1) (2002)

438.[15] Y. Makida, et al., IEEE Trans. Appl. Supercond. 16 (2) (2006)

477.[16] T. Yoshida, et al., in: Proc. 29th Int. Cosmic Ray Conf. (Pune),

vol. 3, 2005, p. 33.[17] R. Brun et al., GEANT3.21. Detector Description and Simulation

Tool, CERN Program Library, Long Writeup W5013.[18] H. Fesefeldt, PITHA 85/02, Aachen, 1985.[19] Y. Asaoka, et al., Nucl. Instrum. Methods A 489 (2001) 170.[20] P. Papini, et al., Nuovo Cimento 19C (1996) 367.[21] Y. Shikaze, et al., Astropart. Phys. 28 (2007) 154.[22] T. Sanuki, et al., Astrophys. J. 545 (2000) 1135.[23] Bartol Neutron Monitor, http://neutronm.bartol.udel.edu/