Документ взят из кэша поисковой машины. Адрес оригинального документа : http://nuclphys.sinp.msu.ru/conf/lpp14/210809/Panov.pdf
Дата изменения: Wed Sep 16 16:30:30 2009
Дата индексирования: Tue Oct 2 00:39:35 2012
Кодировка:
Systematics and fine structure in the cosmic ray electron spectrum measured by ATIC

2009.08.21 14th Lomonosov conference on elementary particle physics

1


ATIC (Advanced Thin Ionization Calorimeter) flights around South Pole
Trajectory of ATIC4
ATIC1 28.12.2000­13.01.2001 ATIC2 29.12.2002­18.01.2003

ATIC3 2005 failed to rich altitude ATIC4 26.12.2007­15.01.2008
2


ATIC spectrometer

1 -- Silicon matrix 80 56 pixels, 1.5 2cm 2 -- Scintillator hodoscopes 3 -- Carbon target 4 -- BGOcalorimeter View from above: 50 50 cm BGO crystal: 2.5 2.5 25 cm 8 layers in ATIC2 10 layers in ATIC4
3


Proton event Hodoscopes energy deposits

Calorimeter energy deposites Electron or positron event Hodoscopes energy deposites Calorimeter energy deposites
4


Spectrum of cosmic rays electrons measured by ATIC:
J. Chang, J. H. Adams Jr, H. S. Ahn et. al. An excess of cosmic ray electrons at energies of 300­800 GeV. Nature V.456 (2008), P.362365

Dark matter (KaluzaKlein particles of mass 620GeV) fit

5


Kfilter
Layers of the calorimeter from the top:

l = 0, 1, 2, 3, 4, 5, 6, 7 Root Mean Squares (RMS):
Electrons Proton background

Cascade curve:

K
(relative deposit of energy in a layer l )

Fparameter:

Kfilter:
6


The purpose of this work: 1. To chek the results with alternate techniques of separation of electrons from protons 2. To study systematic errors.

The work was carried out comletely independently of previous analysis star ting from lowlevel calibration procedures up to the final results in the MSU subgroup of the ATIC collaboration

7


We define five new (main) filters: 1. Four simple filters

8


Selection of electrons with simple filters
R
Electrons

Proton
background

L1

L2

R

9


Likelihood filter LogP.
Proton background Electrons Electrons

Protons

Projection of the simulated probability distribution on plate

Selection of electrons with LogPfilter

Simulated probability distribution for the shower parameters for incident electrons (FLUKA,. GEANT4):

10


Systematics in proton background calculation
Proton background calculation using simulation is similar to illdefined problem and may produce large systematic errors.

?

One must have very exact simulation of behaviour of filters for proton events but it is not the case. L1filter
11


Inadequate background calculations with FLUKA for different filters

Different backgrounds

Different results after background subtraction

We do not try to subtract proton backgrounds from the spectrum of selected 'electronlike' events. We could not provide exact values of absolute intensity and slope, but we still could study structrures in the electron spectrum.
12


"Product" filter J

13


Systematics in the accounting for the scattering of the electrons in the atmosphere
bremsstrahlung

Small loss of energy (should be studied) Inefficency may exist (should be studied)
Supposition: No loss of energy, no inefficiency. The supposition is expected to be valid at least at energses > n 100 GeV

14


Comparison of the spectrum of this work and the published one in Nature

This work ­ backgrounds are not subtracted! Chang et. al ­ backgrounds are subtracted! But the spectra at mean energies are the same. The background is low in present work or it was underestimated in Nature paper.
15


ATIC2 (this work) and Fermi/LAT

There are no signs of contradiction ATIC2 and Fermi/LAT
16


Energy resolution of ATIC for electrons is about 2%

17


Short scale systematically independent features
ATIC's excess

Rough energy step

ATIC's excess resolved into fine structure (?)

Fine energy step

200

600

18


Fine structure in the electron spectrum?

ATIC2 and ATIC4, step(in log energy)=0.03

The probability of accidental correlation of ATIC2 and ATIC4 fine structures was estimated as P = 0.00064
19


Selection of electrons with the measured level of current of albedo particles in the hodoscopes (not calorimeter).

The fine structure does not depend on the level of background! It is systematic independent effect.

20


ATIC2 + ATIC4, step(in log energy)=0.02

If the fine structure is real, what it may be?

Preliminary

D. Malyshev et al. arXiv:0903.1310

21


ATIC2 electron spectrum for (number of events)/bin

22


Electron + positron spectrum from pulsars of ATNF catalog
Dmitry Malyshev, Ilias Cholis, Joseph Gelfand arXiv:0903.1310

23