The Tunka Radio Extension, an antenna array for high-energy cosmic-ray detection

0.5 MB·

by  Y. Kazarina

XXV European Cosmic Ray Symposium, Turin, Sept. 4-9 2016 1
The Tunka Radio Extension, an antenna array for high-energy cosmic-ray
detection
Y. Kazarinaa, P.A. Bezyazeekova, N.M. Budneva, O. Fedorova, O.A. Gressa, A. Haungsb,
R. Hillerb, T. Huegeb, M. Kleifgesc, E.E. Korostelevad, D. Kostuninb, O. Kr¨omerc, V. Kungelb,
L.A. Kuzmichevd, N. Lubsandorzhievd, T. N. Marshalkinaa, R.R. Mirgazova, R. Monkhoeva,
E.A. Osipovad, A. Pakhorukova, L. Pankova, V.V. Prosind, F.G. Schr¨oderb, A. Zagorodnikova
aInstitute of Applied Physics ISU, Irkutsk, Russia;
bInstitut fu¨r Kernphysik, Karlsruhe Institute of Technology (KIT), Germany;
cInstitut fu¨r Prozessdatenverarbeitung und Elektronik, KIT, Germany; and
dSkobeltsyn Institute of Nuclear Physics MSU, Moscow, Russia;
7 The Tunka-Rex (Tunka Radio Extension) has been deployed in autumn 2012 at the territory
1 of the Tunka-133 experiment (Tunka Valley, Republic of Buryatia, Russia), covering an area of
0 approximately 3 km2. Tunka-133 detects the Cherenkov radiation from air showers of cosmic rays
2 atenergies1016.5–1018eV,and63antennasofTunka-Rexmeasuretheradioemissionofthesameair
showers. Three years of joint operation of Tunka-Rex and Tunka-133 have shown that a calibrated
n
radioarraycanbeusedforanindependenttestofthescaleofthecosmic-rayenergy. Furthermore,
a
J by direct comparison of the depth of the shower maximum measured by Tunka-133 and Tunka-
Rex, it was shown that the precision of the radio technique for the shower maximum is at least 40
7
g/cm2. Twothirdsofantennasareconnectedtotherecentlydeployedarraysofscintillationstations
1
Tunka-Grande. Asnextstepthecross-calibrationofTunka-RexandTunka-Grandeisplanned,which

] providesthepossibilityofthecombinedmeasurementsofthemuonandelectromagneticcomponents
E ofair-showers,wheretheradioarraywillprovidesensitivitytotheshowermaximumwithfullduty-
H cycle. Exploiting the complementary muon/radio information, it should be possible to improve
the mass separation in cosmic-ray spectra. This article presents the first results of the combined
.
h measurements of Tunka-Rex and Tunka-Grande as well as studies of the antenna alignment effect
p and an overview of the recent Tunka-Rex results.
-
o
tr I. INTRODUCTION 1000 TunTkau-nGkara-1n3d3e
s
Tunka-Rex
a
[

One of the main puzzles of modern astrophysics
1
are the sources of cosmic rays and their acceleration 500
v
mechanisms. The study of cosmic rays in the en-
7
2 ergy range from 1016 to 1019 eV is of special inter-
7 est. In this range a transition from galactic to ex- m)
4 tragalactic sources is supposed [1–3]. Good sensitiv- h ( 0
0 ity to the mass composition of the primary cosmic Nort
.
1 rays is required for the study of this part of the spec-
0 trum. There are two principle methods for cosmic
7 ray detection: direct (satellite detectors measuring -500
1
primary cosmic rays while orbiting Earth) and indi-
:
v rect (ground arrays measuring extensive air-showers
i (EAS) produced by high-energy cosmic rays) for en-
X
ergies above 1014 eV, since the flux of the primary -1000
r
a cosmicraysbecomestoolowfordirectmeasurements. -1000 -500 0 500 1000
Indirect methods for the study of cosmic rays are East (m)
used at the Tunka Valley at the observatory TAIGA
(Tunka Advanced Instrument for cosmic ray physics FIG. 1: Layout of cosmic ray experiments of TAIGA ob-
servatory.
and Gamma Astronomy) [4]. It is a complex, hybrid
detector,basedonarraysofdifferenttypes. Thereare
threedetectorsforcosmicrays: theair-Cherenkovde-
tector Tunka-133, the radio detector Tunka-Rex and
the particle detector Tunka-Grande. TAIGA includes Grande[6]andTunka-Rex[7],conductjointmeasure-
low threshold gamma ray detectors as well: the non- ments of showers from primary cosmic rays with en-
imagingair-CherenkovdetectorTunka-HiSCOREand ergies from 1016 to 1018 eV. FIG. 1 shows the layout
Imaging Atmospheric Cherenkov Telescopes. The ofTAIGAcosmicraydetectors,whicharedistributed
three experiments, namely Tunka-133 [5], Tunka- over 3 km2.
eConf C16-09-04.3
2 XXV European Cosmic Ray Symposium, Turin, Sept. 4-9 2016
II. TUNKA-REX purelyonmeasurements,theotheroneusingCoREAS
simulations for comparison. With both approaches it
was consistently shown that the energy scale of the
The Tunka Radio Extension (Tunka-Rex) is an ar-
ray of 63 antennas distributed on an area of 3 km2. cosmicraysmeasurementsinKASCADE-Grandeand
Tunka-133 experiments agree with each other within
Of them 57 antennas are occupying a denser part of
a relative uncertainty of about 10%.
the detector, an area with radius of 500 m, and 6
satellite antenna stations are placed at a distance of
1 km from the center of the setup. The central an-
IV. FIRST RESULTS OF JOINT
tenna stations are grouped in 19 clusters of 3 an-
MEASUREMENTS OF TUNKA-REX AND
tennas each with distances between the cluster cen-
TUNKA-GRANDE
ters of 200 m. The layout of the setup is given in
FIG. 1. Each Tunka-Rex antenna station consists
During 2015-2016 the detection of air showers has
of two perpendicular short aperiodic loaded loop an-
been conducted by all TAIGA experiments except of
tennas (SALLA) [8, 9]. Before digitalization, signals
thetelescopeswhicharestillunderconstruction. The
are analogically pre-amplified by a low noise ampli-
air-Cherenkov experiments Tunka-133 and Tunka-
fier and processed with a filter-amplifier with an ef-
HiSCORE, whose operation is possible only in moon-
fective band of 30-76 MHz. Each Tunka-Rex antenna
less nights, were operated during about 400 hours.
station is connected either to the Tunka-133 or the
The full duty-cycle experiments, detecting charged
Tunka-Grande local data acquisition and shares the
particlesandradioemissionfromair-showers(Tunka-
sameADCboards. ThefrequencybandofTunka-Rex
Grande and Tunka-Rex, respectively), operated dur-
provides a high signal-to-noise ratio (SNR), and the
ingabout2000hours. Duringsummer(June-August),
atmosphere is transparent for these radio frequencies.
TAIGAwasswitchedoffbecauseoftheriskofdamage
bythunderstorms. Theanalysisofthedatameasured
jointly by the Tunka-Rex and Tunka-Grande experi-
III. THE MAIN TUNKA-REX RESULTS mentsisstillpreliminary. Wefoundabout2000event
candidates for energies above 100 PeV. An example
TheTunka-Rexreconstructionmethodsweredevel- of a reconstructed event is shown in FIG. 2. The lat-
oped and applied for the first two seasons of Tunka- eraldistributionfunction(LDF)ofthisexampleevent
Rex and Tunka-133 joint measurements [10]. Only is shown in FIG. 3. The methods of the combined
events with energies above 1017 eV are taken for the Tunka-Rex and Tunka-Grande reconstruction are in
analysis. The reconstruction of shower parameters is progress.
based on the lateral distribution, i.e., the dependence
of the radio amplitude on the distance to the shower
axis. The amplitude parameter of the lateral distri- V. EFFECT OF ANTENNA ALIGNMENT
bution is correlated with the primary energy and the
slopeofthelateraldistributionissensitivetotheposi- Besides the main goals we studied the influence of
tion of the shower maximum, X , [11]. The energy the antenna alignment for cosmic-ray measurements.
max
and X reconstructed by Tunka-Rex have a strong The radio signal from cosmic ray air showers is pre-
max
correlation with the same parameters reconstructed dominantly polarized along the geomagnetic Lorentz
by Tunka-133. Theachievedresolution forthe energy force, whose direction depends on the direction of
reconstruction for Tunka-Rex is 15%. When exploit- the shower axis and the geomagnetic field (it corre-
ingtheshowergeometryreconstructedbythehostde- spondstoeast-westdirectioniftheshoweraxisisver-
tector Tunka-133, the energy can be estimated even tical)[19],thus,wehadsupposedthattheefficiencyof
with a single antenna station to about 20% preci- a radio detector should depend on the antenna align-
sion [12]. The X resolution of Tunka-Rex is ap- ment. This is the reason why the Tunka-Rex anten-
max
proximately 40 g/cm2 - for high-quality events. This nasarerotatedby45◦ tothegeomagneticnorth-south
can be improved by increasing the number of anten- axis,sincewewantedtohavemoreantennaswithsig-
nas and by a stricter event selection using high qual- nal in both channels, tolerating to have less events
ity cuts. It means that the shower reconstruction by with signal in at least one channel. To check the
Tunka-Rex is reliable and provides a similar precision correctness of this assumption the vector product of
as other modern radio experiments (AERA [13], LO- the shower axis vector and the magnetic field vector
FAR [14, 15]) and as the established air-shower tech- v×B is used, where v is the shower axis vector and
niques. One of the main Tunka-Rex results is the en- B is the magnetic field vector. Two configurations
ergy scale comparison of the KASCADE-Grande [16] of antennas were considered: first, when antennas are
and Tunka-133 [2] experiments via their radio exten- aligned strictly along the north-south and east-west
sions, LOPES [17] and Tunka-Rex, respectively [18]. axes, which used in most experiments for detection of
Weusedtwodifferentanalysisapproaches: onerelying radio emission A(Ch , Ch ), and second, the align-
1 2
eConf C16-09-04.3
XXV European Cosmic Ray Symposium, Turin, Sept. 4-9 2016 3
arms for both alignments. Then, the difference be-
tween configurations was quantified by the difference
between the highest amplitude in channels for each
configuration (see Eq. 1):
∆max(θ,φ)=max[(E,Ch )2,(E,Ch )2]−
1 2
−max[(E,Ch(cid:48))2,(E,Ch(cid:48))2], (1)
1 2
where θ is zenith angle of the shower axis, φ its
azimuth, and ∆max(θ,φ) is the difference between
configurations A and A(cid:48), E = v×B is E-field vector
of the radio signal emitted by the shower. Large val-
ues of ∆max, mean that the alignment has a large
effect on the number of events detected in at least
one channel or in both channels, respectively. The
average dependence on the antenna alignment can be
seen as function of the zenith angle θ when ∆max is
integrated over the azimuth angle φ (see Eq. 2):
FIG. 2: Example of a reconstructed event: footprint of
theevent,wherethesizeofthecrossesindicatesthesignal
strength in each polarization, the color code is the arrival (cid:90) 2π
time, and the line and star are the direction and shower I(θ)= ∆max(θ,φ)dφ (2)
core, respectively. Grey stations are below threshold, and 0
small crosses indicate stations not operating during this FIG.4showstheresultsofthecalculationmadefor
event.
the geomagnetic field of various modern radio experi-
ments: Tunka-Rex, AERA and LOFAR. From FIG. 4
Event 2016-04-13 17:40:08 it can be concluded that the dependence on align-
800 ment vanishes on average when the zenith angle of
Tunka-133 ant. anairshower θ isgreater orequaltothe geomagnetic
Tunka-Grande ant.
700 LDF fit zenith (i.e. inclination in the coordinate system of
EAS)θ (forTunka-Rexθ =18.2◦,forAERAθ =
B B B
600 53.4◦, and for LOFAR θB = 22◦). For an individual
m) shower,thealignmentcanstillbeimportant,buttak-
V/µ 500 ingintoaccountthattheshowerdirectionsareequally
h ( distributed over azimuth, the choice of antenna align-
ngt 400 ment becomes unimportant for zenith angles θ > θB.
e
str This implies that for Tunka-Rex the antenna align-
al 300 ment is important only for vertical events (when the
gn zenithangleoftheshowerissmallerthanthegeomag-
Si 200 netic zenith of 18.2◦), and for more inclined events it
does not matter. Only for the AERA experiment the
choice of antenna alignment is important for a signif-
100
icant range of zenith angles, because the geomagnetic
field at AERA is much more inclined.
0
0 100 200 300 400 500 Theeffectofantennaalignmentwascheckedonthe
Distance to shower axis (m) reconstructed Tunka-Rex data of 2012-2013. The to-
tal time of measurements was about 400 hours. The
FIG. 3: Lateral distribution of example event shown in number of reconstructed events is 146 and there is
FIG. 2 with preliminary calibration of the antennas con- only one event with zenith angle below the geomag-
nected to Tunka-Grande. The curve indicates best LDF netic one. Thus, no significant effect is expected of
fit. the antenna alignment on the detection efficiency for
Tunka-Rex. Furthermore, the effect of the antenna
alignment on the efficiency of Tunka-Rex was stud-
ment as in Tunka-Rex and LOFAR where antennas ied by using CoREAS simulations [20] in the pres-
are rotated by 45◦ with respect to the geomagnetic ence of noise. About 300 events for proton and 300
north-south axis A(cid:48)(Ch(cid:48), Ch(cid:48)). The next step is to events for iron as primary particle (where the energy
1 2
evaluatetheefficiencyofeachconfigurationbytaking and direction were taken from Tunka-133) were simu-
theprojectionofthevectorproductv×Bonantenna lated. Fortheseeventsexperimentallymeasurednoise
eConf C16-09-04.3
4 XXV European Cosmic Ray Symposium, Turin, Sept. 4-9 2016
1200 theshowermaximum. Duetotheabsolutecalibration
Tunka(cid:45)Rex
oftheradioantennas,theenergyscalesofKASCADE-
AERA
1000 LOFAR Grande and Tunka-133 could be compared and were
found consistent within 10%, which enables a better
a.u. 800(cid:76) comparison of features observed in the energy spec-
(cid:72)
max(cid:68) 600 trum. After preliminary analysis of the experimental
Integrated 400 dRaetxa,aonbdtaTinuendkab-yGcraonindcei,dewnetmfoeuansduraebmoeuntts20o0f0Tuevneknat-
candidates. Thedevelopmentofreconstructionmeth-
200 odsisinprogressandfurtherTunka-Rexanalyseswill
focusonamass-compositionstudyjointlywithTunka-
0 Grande. Studies of the antenna alignment show that
0 15° 30° 45° 60° 75° 90°
Zenith the efficiency of a radio detector does not dependent
ontheantennaorientationwhentheairshowerzenith
FIG. 4: The effect of antenna alignment I(θ), see Eq. 2, angle is larger than the geomagnetic zenith at the de-
dependingontheshowerzenithangle,forthegeomagnetic tector location. In particular for future arrays aim-
fields at Tunka-Rex, AERA and LOFAR.
ing at inclined air showers, this gives the freedom to
choose the antenna orientation based on any other
technical criteria.
was added. It was confirmed that neither the number
of events nor the number of stations depends signif-
icantly on the azimuthal alignment of the antennas. Acknowledgments
Consequently, for Tunka-Rex either choice of antenna
alignment is fine, as long as the exact alignment is
The construction of Tunka-Rex was funded by the
known and can be taken into account during data
GermanHelmholtzAssociationandtheRussianFoun-
analysis.
dationforBasicResearch(grantHRJRG-303). More-
over, this work was supported by the Helmholtz Al-
liance for Astroparticle Physics (HAP), by Deutsche
VI. CONCLUSION
Forschungsgemeinschaft (DFG) grant SCHR 1480/1-
1, and by the Russian Federation Ministry of Edu-
Tunka-RexistheradioextensionoftheTAIGAex-
cationandScience(agreement14.B25.31.0010). Also,
periment, agrowinginfrastructureforallcomponents
thisworkwassupportedbytheRussianFundofBasic
of air showers produced by high-energy cosmic rays
research grants 16-32-00329 and 16-02-00738.
aswellasgamma-rayastronomy. Tunka-Rexprovides
competitiveprecisionoftheenergyandthepositionof
[1] W. D. Apel, et al. (KASCADE-Grande Collabora- [10] P.A.Bezyazeekov,etal.-Tunka-RexColl.,JCAP01,
tion), Astroparticle Physics 36, 183 (2012). 052 (2016).
[2] L. Sveshnikova, L. Kuzmichev, E. Korosteleva, [11] D. Kostunin et al., Astropart. Phys. 74, 79 (2016).
V. Prosin, and V. Ptuskin, Nucl.Phys.Proc.Suppl. [12] R. Hiller, et al. - Tunka-Rex Coll., EPJ Web of Con-
256-257, 218 (2014). ferences proc. of ARENA 2016, in press (2016),
[3] F.G. Schr¨oder, Prog. Part. Nucl. Phys. 93, 1 (2017). arxiv:1611.09614.
[4] N.M.Budnev,etal.-TAIGAColl.,JINST9,C09021 [13] Pierre Auger Coll., Phys. Rev. D 93, 122005 (2016).
(2014). [14] A. Nelles et al., JINST 10, P11005 (2015).
[5] B.A.Antokhonovetal.,Nucl.Instrum.Meth.A628, [15] S. Buitink et al., Phys. Rev. D 90, 082003 (2014).
124 (2011). [16] W. D. A. et al. (KASCADE-Grande Collaboration),
[6] N. M. Budnev, A. L. Ivanova, N. N. Kalmykov, Nucl. Instrum. Meth. A 620, 202 (2010).
L. A. Kuzmichev, V. P. Sulakov, and Yu. A. Fomin, [17] W.D. Apel, et al. - LOPES Coll., Phys. Rev. D 90,
Moscow Univ. Phys. Bull. 70, 160 (2015), [Vestn. 062001 (2014).
Mosk. Univ.2015,no.2,8084(2015)]. [18] W.D. Apel, et al. - LOPES and Tunka-Rex Colls.,
[7] P.A.Bezyazeekov,etal.-Tunka-RexColl.,Nucl.Inst. Physics Lett. B 763, 179 (2016).
Meth. A 802, 89 (2015). [19] F.D.KahnandI.Lerche,inProceedingsoftheRoyal
[8] O. Kro¨mer et al. (LOPES Collaboration), Society of London. Series A, Mathematical and Phys.
Proc. of the 31st ICRC, L(cid:32)o´d´z, Poland (2009), Sciences (1966), vol. 289, p. 206.
http://icrc2009.uni.lodz.pl/proc/html/. [20] T.Huege,M.Ludwig,andC.James,AIPConf.Proc.
[9] P. Abreu et al. (Pierre Auger), JINST 7, P10011 1535, 128 (2013), 1301.2132.
(2012), 1209.3840.
eConf C16-09-04.3