Contributions to the 32nd International Cosmic Ray Conference (ICRC 2011) by the ANTARES collaboration

Abstract

The ANTARES detector, completed in 2008, is the largest neutrino telescope in
the Northern hemisphere. It is located at a depth of 2.5 km in the
Mediterranean Sea, 40 km off the Toulon shore. The scientific scope of the
experiment is very broad, being the search for astrophysical neutrinos the main
goal. In this paper we collect the 22 contributions of the ANTARES
collaboration to the 32nd International Cosmic Ray Conference (ICRC 2011). At
this stage of the experiment the scientific output is very rich and the
contributions included in these proceedings cover the main physics results
(steady point sources, correlations with GRBs, diffuse fluxes, target of
opportunity programs, dark matter, exotic physics, oscillations, etc.) and some
relevant detector studies (water optical properties, energy reconstruction,
moon shadow, accoustic detection, etc.)

Full text

Page 1

32ND INTERNATIONAL COSMIC RAY CONFERENCE, BEIJING 2011
The ANTARES Collaboration:
contributions to the
32stInternational Cosmic Ray Conference (ICRC 2011),
Beijing, China,
August 2011
Abstract
The ANTARES detector, completed in 2008, is the largest neutrino telescope in the Northern hemisphere. It is
located at a depth of 2.5 km in the Mediterranean Sea, 40 km off the Toulon shore. The scientific scope of the
experiment is very broad, being the search for astrophysical neutrinos the main goal. In this paper we collect the 22
contributions of the ANTARES collaboration to the 32nd International Cosmic Ray Conference (ICRC 2011). At this
stage of the experiment the scientific output is very rich and the contributions included in these proceedings cover the
main physics results (steady point sources, correlations with GRBs, diffuse fluxes, target of opportunity programs,
dark matter, exotic physics, oscillatinos etc.) and some relevant detector studies (water optical properties, energy
reconstruction, moon shadow, accoustic detection, etc.)
arXiv:1112.0478v1 [astro-ph.HE] 2 Dec 2011

Page 2

THE ANTARES COLLABORATION
ANTARES Collaboration
S. Adri´ an-Mart´ ınez1, J.A. Aguilar2, I. Al Samarai3, A. Albert4, M. Andr´ e5, M. Anghinolfi6, G. Anton7, S.
Anvar8, M. Ardid1, A.C. Assis Jesus9, T. Astraatmadja9,a, J-J. Aubert3, B. Baret10, S. Basa11, V. Bertin3,
S. Biagi12,13, A. Bigi14, C. Bigongiari2, C. Bogazzi9, M. Bou-Cabo1, B. Bouhou10, M.C. Bouwhuis9,
J. Brunner3,b, J. Busto3, F. Camarena1, A. Capone15,16, C. Cˆ arloganu17, G. Carminati12,13,c, J. Carr3,
S. Cecchini12, Z. Charif3, Ph. Charvis18, T. Chiarusi12, M. Circella19, H. Costantini6,3, P. Coyle3, A.
Creusto10, C. Curtil3, G. De Bonis15, M.P. Decowski9, I. Dekeyser20, A. Deschamps18, C. Distefano21,
C. Donzaud10,22, D. Dornic2, Q. Dorosti23, D. Drouhin4, T. Eberl7, U. Emanuele2, A. Enzenh¨ ofer7, J-P.
Ernenwein3, S. Escoffier3, P. Fermani15,16, M. Ferri1, V. Flaminio14,24, F. Folger7, U. Fritsch7, J-L. Fuda20,
S. Galat` a3, P. Gay17, G. Giacomelli12,13, V. Giordano21, J.P. G´ omez-Gonz´ alez2, K. Graf7, G. Guillard17,
G. Halladjian3, G. Hallewell3, H. van Haren25, J. Hartman9, A.J. Heijboer9, Y. Hello18, J.J. Hern´ andez-
Rey2, B. Herold7, J. H¨ oßl7, C.C. Hsu9, M. de Jong9,a, M. Kadler26, O. Kalekin7, A. Kappes7, U. Katz7,
O. Kavatsyuk23, P. Kooijman9,27,28, C. Kopper9,7, A. Kouchner10, I. Kreykenbohm26, V. Kulikovskiy29,6,
R. Lahmann7, P. Lamare8, G. Larosa1, D. Lattuada21, D.
Loehner23, S. Loucatos32, S. Mangano2, M. Marcelin11, A. Margiotta12,13, J.A. Mart´ ınez-Mora1, A. Meli7,
T. Montaruli19,33, M. Morganti14,dL. Moscoso10,32,e, H. Motz7, M. Neff7, E. Nezri11, D. Palioselitis9,
G.E. P˘ av˘ alas ¸34, K. Payet32, P. Payre3,e, J. Petrovic9, P. Piattelli21, N. Picot-Clemente3, V. Popa34, T. Pradier35,
E. Presani9, C. Racca4, C. Reed9, G. Riccobene21, C. Richardt7, R. Richter7, C. Rivi` ere3, A. Robert20,
K. Roensch7, A. Rostovtsev36, J. Ruiz-Rivas2, M. Rujoiu34, G.V. Russo30,31, F. Salesa2, P. Sapienza21,
F. Sch¨ ock7, J-P. Schuller32, F. Sch¨ ussler32, T. Seitz
Spurio12,13, J.J.M. Steijger9, Th. Stolarczyk32, A. S´ anchez-Losa2, M. Taiuti6,37, C. Tamburini20, S. Toscano2,
B. Vallage32, C. Vall´ ee3, V. Van Elewyck10, G. Vannoni32, M. Vecchi3,16, P. Vernin32, S. Wagner7, G.
Wijnker9, J. Wilms26, E. de Wolf9,28, H. Yepes2, D. Zaborov36, J.D. Zornoza2, J. Z´ u˜ niga2
1Institut d’Investigaci´ o per a la Gesti´ o Integrada de les Zones Costaneres (IGIC) - Universitat Polit` ecnica de Val` encia. C/ Paranimf 1 , 46730 Gandia, Spain
2IFIC - Instituto de F´ ısica Corpuscular, Edificios Investigaci´ on de Paterna, CSIC - Universitat de Val` encia, Apdo. de Correos 22085, 46071 Valencia, Spain
3CPPM, Aix-Marseille Universit´ e, CNRS/IN2P3, Marseille, France
4GRPHE - Institut universitaire de technologie de Colmar, 34 rue du Grillenbreit BP 50568 - 68008 Colmar, France
5Technical University of Catalonia, Laboratory of Applied Bioacoustics, Rambla Exposici´ o,08800 Vilanova i la Geltr´ u,Barcelona, Spain
6INFN - Sezione di Genova, Via Dodecaneso 33, 16146 Genova, Italy
7Friedrich-Alexander-Universit¨ at Erlangen-N¨ urnberg, Erlangen Centre for Astroparticle Physics, Erwin-Rommel-Str. 1, 91058 Erlangen, Germany
8Dir. des Sciences de la Mati` ere - Inst. de Rech. sur les lois Fondamentales de l’Univers - SEDI, CEA Saclay, 91191 Gif-sur-Yvette Cedex, France
9Nikhef, Science Park, Amsterdam, The Netherlands
10APC - Lab. AstroPart. et Cosmologie, UMR 7164 (CNRS, Univ. Paris 7 Diderot, CEA, O. de Paris) 10, rue Alice Domon et L´ eonie Duquet 75205 Paris Cedex 13,
France
11LAM - Laboratoire d’Astrophysique de Marseille, Pˆ ole de l’´Etoile Site de Chˆ ateau-Gombert, rue Fr´ ed´ eric Joliot-Curie 38, 13388 Marseille Cedex 13, France
12INFN - Sezione di Bologna, Viale Berti-Pichat 6/2, 40127 Bologna, Italy
13Dipartimento di Fisica dell’Universit` a, Viale Berti Pichat 6/2, 40127 Bologna, Italy
14INFN - Sezione di Pisa, Largo B. Pontecorvo 3, 56127 Pisa, Italy
15INFN - Sezione di Roma, P.le Aldo Moro 2, 00185 Roma, Italy
16Dipartimento di Fisica dell’Universit` a La Sapienza, P.le Aldo Moro 2, 00185 Roma, Italy
17Clermont Universit´ e, Universit´ e Blaise Pascal,CNRS/IN2P3, Laboratoire de Physique Corpusculaire, BP 10448, F-63000 Clermont-Ferrand, France
18G´ eoazur - Univ. de Nice Sophia-Antipolis, CNRS/INSU, IRD, Observ. de la Cˆ ote d’Azur and Universit´ e Pierre et Marie Curie, BP 48, 06235 Villefranche-sur-mer,
France
19INFN - Sezione di Bari, Via E. Orabona 4, 70126 Bari, Italy
20COM - Centre d’Oc´ eanologie de Marseille, CNRS/INSU et Universit´ e de la M´ editerran´ ee, 163 Avenue de Luminy, Case 901, 13288 Marseille Cedex 9, France
21INFN - Laboratori Nazionali del Sud (LNS), Via S. Sofia 62, 95123 Catania, Italy
22Univ Paris-Sud , 91405 Orsay Cedex, France
23Kernfysisch Versneller Instituut (KVI), University of Groningen, Zernikelaan 25, 9747 AA Groningen, The Netherlands
24Dipartimento di Fisica dell’Universit` a, Largo B. Pontecorvo 3, 56127 Pisa, Italy
25Royal Netherlands Institute for Sea Research (NIOZ), Landsdiep 4,1797 SZ ’t Horntje (Texel), The Netherlands
26Dr. Remeis-Sternwarte and ECAP, Universit¨ at Erlangen-N¨ urnberg, Sternwartstr. 7, 96049 Bamberg, Germany
27Universiteit Utrecht, Faculteit Betawetenschappen, Princetonplein 5, 3584 CC Utrecht, The Netherlands
28Universiteit van Amsterdam, Instituut voor Hoge-Energie Fysika, Science Park 105, 1098 XG Amsterdam, The Netherlands
29Moscow State University,Skobeltsyn Institute of Nuclear Physics,Leninskie gory, 119991 Moscow, Russia
30INFN - Sezione di Catania, Viale Andrea Doria 6, 95125 Catania, Italy
31Dipartimento di Fisica ed Astronomia dell’Universit` a, Viale Andrea Doria 6, 95125 Catania, Italy
32Dir. des Sciences de la Mati` ere - Inst. de Rech. sur les lois Fondamentales de l’Univers - SPP, CEA Saclay, 91191 Gif-sur-Yvette Cedex, France
33University of Wisconsin - Madison, 53715, WI, USA
34Institute for Space Sciences, R-77125 Bucharest, M˘ agurele, Romania
35IPHC-Institut Pluridisciplinaire Hubert Curien - Universit´ e de Strasbourg et CNRS/IN2P3 23 rue du Loess, BP 28, 67037 Strasbourg Cedex 2, France
36ITEP - Institute for Theoretical and Experimental Physics, B. Cheremushkinskaya 25, 117218 Moscow, Russia
37Dipartimento di Fisica dell’Universit` a, Via Dodecaneso 33, 16146 Genova, Italy
aAlso at University of Leiden, the Netherlands
bOn leave at DESY, Platanenallee 6, D-15738 Zeuthen, Germany
cNow at University of California - Irvine, 92697, CA, USA
dAlso at the Accademia Navale di Livorno, Livorno, Italy
eDeceased
Lef` evre20, G. Lim9,28, D. Lo Presti30,31, H.
7, R. Shanidze7, F. Simeone15,16, A. Spies7, M.

Page 3

32ND INTERNATIONAL COSMIC RAY CONFERENCE, BEIJING 2011
Table of Contents
1) Aart Heijboer, ”Recent results from the ANTARES deep sea neutrino telescope”
2) Claudio Bogazzi, ”Searching for point sources of high energy cosmic neutrinos with the ANTARES telescope”
3) Juan Pablo G´ omez-Gonz´ alez, ”Search for point sources with the ANTARES neutrino telescope using the EM
algorithm”
4) Fabian Sch¨ ussler, ”Autocorrelation analysis of ANTARES data”
5) Fabian Sch¨ ussler, ”Search for a diffuse flux of high-energy muon neutrinos with the ANTARES neutrino telescope”
6) Dimitirs Palioselitis, ”Muon energy reconstruction and atmospheric neutrino spectum unfolding with the
ANTARES detector”
7) Corey Reed, Mieke Bouwhuis, Eleonora Presani, ”Searches for neutrinos from GRBs using the ANTARES
telescope”
8) Damien Dornic, ”Search for neutrino emission of gamma-ray flaring blazars with the ANTARES telescope”
9) Damien Dornic et al., ”Search for neutrinos from transient sources with the ANTARES telescope and optical
follow-up observations”
10) Vladimir Kulikovskiy, ”SN neutrino detection in the ANTARES neutrino telescope”
11) Jelena Petrovic, ”Study on possible correlations between events observed by the ANTARES neutrino telescope
and the Pierre Auger cosmic ray observatory”
12) V´ eronique Van Elewyck, ”Searches for high-energy neutrinos in coincidence with gravitational waves with the
ANTARES and VIRGO/LIGO detectors”
13) Guillaume Lambard, ”Indirect dark matter search in the Sun direction using the ANTARES dta 2007-2008 for
the two common theoretical frameworks (CMSSM, mUED)”
14) Goulven Guillard, J¨ urgen Brunner, ”On neutrino oscillations searches with ANTARES”
15) Nicolas Picot-Clemente, ”Search for magnetic monopoles with the ANTARES underwater neutrino telescope”
16) Vlad Popa, ”Nuclearite search with the ANTARES neutrino telescope”
17) Ching-Cheng Hsu, ”Studying Cosmic Ray Composition around the knee region with the ANTARES Telescope”
18) Goulven Guillard, ”ANTARES sensitivity to steady cosmic ray sources”
19) Salvatore Mangano, ”Muon induced electromagnetic shower reconstruction in the ANTARES neutrino telescope”
20) Salvatore Mangano, ”Optical properties in deep sea water at the site of the ANTARES detector”
21) Colas Rivi` ere, Carla Distefano, ”Moon shadow observation with the ANTARES neutrino telescope”
22) Robert Lahmann, ”Status and recent results of the acoustic neutrino detection test system AMADEUS”

Page 4

THE ANTARES COLLABORATION

Page 5

32ND INTERNATIONAL COSMIC RAY CONFERENCE, BEIJING 2011
Recent Results from the ANTARES Deep-sea Neutrino Telescope
AART HEIJBOER1, ANTARES COLLABORATION
1Nikhef, Amsterdam
aart.heijboer@nikhef.nl
Abstract. The ANTARES observatory is currently the
largest neutrino telescope in the Northern Hemisphere.
Located at a depth of 2.5 km in the Mediterranean
Sea, it aims to detect high energy neutrinos that are
expected from cosmic ray acceleration sites. The status
of the experiment will be discussed, including a broad
target-of-opportunity program. The latest results will
be presented, including searches for a diffuse high-
energy cosmic neutrino flux, neutrinos from Gamma
Ray Bursts, and for (galactic) point-like sources.
I. THE ANTARES NEUTRINO DETECTOR
Cosmic Rays are thought to originate in Galactic and
extra- Galactic sources that accelerate protons and other
nuclei up to high energies. Identification of the responsible
objects could be achieved by detecting the distinct signa-
tures of these cosmic accelerators, which are high energy
neutrinos and gamma rays produced through hadronic
interactions with ambient gas or photoproduction on in-
tense photon fields near the source. While gamma rays
can be produced also by directly accelerated electrons,
the detection of high-energy neutrinos from these objects
would provide unambiguous and unique information on
the sites of the cosmic accelerators and hadronic nature
of the accelerated particles.
The ANTARES Collaboration has constructed a neutrino
telescope [1] at a depth of about 2475 meters, offshore
Toulon, France. Neutrinos are detected by photomultiplier
tubes (PMTs), housed in pressure resistant glass spheres,
which are regularly arranged on 12 detection lines. Each
line accommodates up to 25 triplets of PMTs, located
between 100 and 450 m above the sea bed. The lines are
connected to the shore via a junction box and a single,
40 km electro-optical cable, which provides both power
and an optical data link. On shore, a computer farm runs
a set of trigger algorithms to identify events containing
Cherenkov light from high energy muons within the
data stream, which otherwise consists mostly of signals
from radioactive decay and bioluminescence. The selected
events are stored for offline reconstruction. In 2007, the
first 5 detector lines became operational, followed, in May
2008, by the completion of the full 12-line detector.
The reconstruction of muon tracks is based on the arrival
time of the Cherenkov photons on the PMTs. For high
energy neutrinos, the angular resolution is determined
by the timing accuracy, which is limited by the transit
time spread of the PMTs (1.3ns). Time calibration is
performed by a number of independent systems, including
LED and laser beacons [2] located throughout the detector.
The relative inter-line timing has been calibrated using
the time residuals measured in a large number of down-
going reconstructed muon events, in addition to the optical
beacon systems. The positions of the PMTs vary with
time because of the sea currents. Using an acoustic posi-
tioning system, combined with information from internal
compasses and tiltmeters, the positions of the PMTs are
determined every 2 minutes with an accuracy of ∼ 10 cm.
Most of the analyses described here use a muon track
reconstruction algorithm (based on [3]) that consists of
multiple fitting steps. The final step is based on a full
likelihood description of the arrival times of the detected
Cherenkov photons, which also accounts for background
light. The achieved angular resolution is, by necessity,
determined from simulations. However, several aspects
of the simulations were confronted with data in order
to constrain the possible systematic effects in the timing
resolution that would result in a deteriorated angular
resolution. The angular resolution (i.e. the median angle
between the neutrino and the reconstructed muon) was
found to be 0.4 ± 0.1 (sys) degrees for the detector with
all 12 lines operational. Studies of the detector and the
optical water properties [4] are ongoing and may help to
further improve and constrain the angular resolution in the
near future. Moreover, a study to observe the shadow of
the moon using down-going muons might in the future
provide additional information on the (absolute) pointing
accuracy [5].
In the following, a selection of results recently obtained
by the ANTARES experiment will be summarized; many
of them are discussed in more detail in dedicated contri-
butions to this conference.

Page 6

ANTARES COLLABORATION RESULTS
Fig. 1: Upper limit on the diffuse neutrino flux of HE
neutrinos obtained from the 2007-2009 ANTARES data,
compared to theoretical predictions 6 and to limits set by
other neutrino telescopes. (see [6] for references).
II. SEARCHES FOR HIGH ENERGY COSMIC SOURCES
Searches for cosmic neutrinos and their sources com-
prise a main goal of the ANTARES experiment. Various
searches for high energy cosmic neutrinos have been
performed using the first years of data.
A. Search for a diffuse neutrino flux
A search for a diffuse cosmic neutrino flux has been
conducted using 334 live-days of data collected in 2008
and 2009 [6]. Such a flux would result in an excess
of high energy events over the irreducible background
of atmospheric neutrinos. A measure of the energy is
provided by an observable R, which measures the number
of PMTs that detect multiple photons separated in time.
The distribution of the R variable agrees well with the
background-only simulations and shows no evidence for a
contribution from a cosmic diffuse E−2flux, which would
result in an excess of high-R events. Consequently, a 90%
C.L. limit on such a flux is obtained in the energy range
20 TeV - 2.5 PeV. The limit is shown in figure 1 together
with previously published limits from other experiments.
B. Point source search
Cosmic point-like source of neutrinos have been searched
for using 813 live-days of data from 2007 up to and
including 2010 [7]. An earlier version of the analysis is
described in [8]. Event selection criteria have been applied
which optimize both the sensitivity and the discovery po-
tential. Events are required to be reconstructed as upward-
going and to have a good reconstruction quality, quantified
by a variable based on the reduced log-likelihood of
the track fit, and an angular error estimate better than
1◦. The resulting event sample consists of 3058 neutrino
candidates, of which ∼ 84(16)% is expected to be at-
mospheric neutrinos (muons misreconstructed as upward-
going). To search for point sources, the analysis uses an
unbinned maximum likelihood method, which exploits the
knowledge on the angular resolution1of 0.5◦and the rate
of background events as a function of the declination.
Two different versions of the search were conducted:
in the ’full-sky’ search, the full visible sky is searched
for point sources. In the ’candidate search’, neutrinos
are searched for only in the direction of 24 a-priori se-
lected candidate source-locations, corresponding to known
gamma ray objects of interest. Neither search yields a
significant excess of events over the background: the
post-trial p-values are 2.5% (for a cluster of events at
α,δ = (−46.5◦,−65.0◦) for the full sky search and 41%
for the most signal-like source in the candidate source
list (HESS J1023-575). Limits have been extracted on
the intensity2of an assumed E−2neutrino flux from the
candidate sources. They are shown in Figure 2. The limit
computation is based a large number of generated pseudo
experiments in which systematic uncertainties on the
angular resolution and acceptance are taken into account.
These limits are more stringent than those from previous
experiments in the Northern hemisphere (also indicated in
the figure) and competitive with those set by the IceCube
observatory [9] for declinations < −30◦. The various
experiments are sensitive in different energy ranges, even
though they all set limits on E−2
trum, ANTARES detects most events at energies in a
broad range around 10 TeV, which is a relevant energy
range for several galactic source candidates.
An independent point source analysis was performed
using a different search method based on the ’EM-
algorithm’ [10]. This cross-check yielded similar results
as the likelihood based analysis described above.
The sample of neutrino candidates from the previous
search [8] has been used for additional studies, which are
also reported on at this conference:
ν
spectra. For this spec-
• An auto-correlation analysis [11] was performed in
order to test for unexpected (larger scale) structures
in the neutrino candidate sample. No such structures
were found.
• An analysis has been performed to search for di-
rectional correlations between the neutrino candidate
events and 69 published ultra-high energy cosmic
rays events detector by the Pierre Auger Observa-
tory [12]. The data were found to contain no such
correlations.
III. MULTI-MESSENGER ASTRONOMY
Several analyses are performed in ANTARES, which
focus on coincident measurement of neutrinos with a
1. Since part of the data in this analysis was taken by a 5-line
detector, the resolution is slightly worse than the 0.4◦mentioned
earlier for the full detector.
2. The limits are on φ, which is defined by the the fol-
lowing expression for the neutrino flux: dN/dE
(E/GeV)−2GeV−1cm−2s−1.
= φ ×

Page 7

32ND INTERNATIONAL COSMIC RAY CONFERENCE, BEIJING 2011
declination (degrees)
-80-60 -40-200 20406080
)
-1
s
-2
flux limit ( GeV cm
-10
10
-9
10
-8
10
-7
10
-6
10
-5
10
-4
10
Antares 2007-2010 preliminary
ANTARES (813 days) ANTARES (813 days) ANTARES (813 days)ANTARES (813 days) ANTARES sensitivity (813 days) ANTARES sensitivity (813 days) ANTARES sensitivity (813 days)ANTARES sensitivity (813 days)
IceCube 40 (375.5 days)IceCube 40 (375.5 days) IceCube 40 (375.5 days) IceCube 40 (375.5 days)IceCube 40 sensitivity IceCube 40 sensitivityIceCube 40 sensitivity IceCube 40 sensitivity
MACRO (2300 days) MACRO (2300 days)MACRO (2300 days) MACRO (2300 days)Super-K (2623 days) Super-K (2623 days) Super-K (2623 days) Super-K (2623 days)
Amanda-II (1387 days) Amanda-II (1387 days) Amanda-II (1387 days)Amanda-II (1387 days)
Fig. 2: Limits set on the normalisation of an E−2
trum of high energy neutrinos from selected candidates.
Also shown is the sensitivity, which is defined as the
median expected limit. In addition to the present result,
several previously published limits on sources in both the
Southern and Northern sky are also shown (see [7] for
references).
ν
spec-
variety of external measurements. A selection is described
below.
A. Neutrinos from flaring blazars
In addition to the time-integrated searches described
above, a time-dependent point source search has been con-
ducted to look for neutrinos in correlation to the variable
gamma-ray emission from blazars measured by the LAT
instrument on-board the Fermi satellite. By restricting the
search to the ’high state’ (typically 1-20 days) of the
gamma emission, the background is reduced compared to
the time-integrated point source search. An analysis using
60 days of live time collected during 2008 is presented in
[13]; no significant excess above the expected background
was observed.
B. Neutrinos from GRBs
Various models predict high energy neutrinos to be emit-
ted by Gamma Ray Burst events. Restricting the neutrino
search to the duration (i.e. T90) of the GRB virtually
eliminates all background events. Hence, the detection of
only a few events could already constitute a discovery.
Two of such searches [14] have been performed. The first
one uses the muon-neutrino channel, exploiting the good
angular resolution of the detector to demand directional
correlation in addition to the time. This search has so far
been performed using 37 GRBs and ANTARES data from
2007 (5 detector lines). No neutrino events were found in
the a-priori-defined search cone and limits on the neutrino
flux were obtained; see figure 3.
The second search is ongoing and searches for ’shower’
events, which are the result of a localized energy depo-
sition in the detector. These events can be produced by
103 108106105104
107
102
10-1
10-3
10-2
1
10
10-2
10-4
10-3
10-5
10-6
E2 Fν [GeV cm-2]
E2 Φν [GeV cm-2 s-1]
Preliminary
E-2 energy spectrum
Waxman&Bahcall energy spectrum
Guetta et al. energy spectrum
Eν [GeV]
Fig. 3: The upper limits (solid lines) for 37 GRBs obtained
by the muon track search for the specified neutrino flux
models (dashed lines) of gamma-ray bursts. The limits
were placed using data taken during 2007, when the
telescope consisted of 5 detector lines.
e.g. electron neutrinos which produce an electromagnetic
shower, or by neutral current interactions of all neutrino
flavours. A reconstruction algorithm for these events has
been developed. The sensitivity of this analysis to GRB
neutrinos of all flavours is presented in [14].
C. Optical follow-up of ANTARES events
To search for transient sources of neutrinos with an optical
counterpart, a system has been setup to enable fast optical
observations in the direction of detected neutrino events.
A reconstruction algorithm that does not require full align-
ment information [5] is run online and alerts are produced
for network of small automatic optical telescopes. Such
alerts are produced for very high energy neutrinos or for
multiple neutrinos that coincide in time and direction.
Since February 2009, ANTARES has sent 37 alert triggers
to the TAROT and ROTSE telescope networks, 27 of
them have been followed. First results on the analysis of
the resulting optical images to search for GRB and core-
collapse SNe will be shown at the conference [13].
Another combination of measurements consists of corre-
lating neutrino events with the signals from the gravita-
tional wave detectors LIGO and VIRGO. A joint analysis
is being performed that searches for a gravitational wave
signal in coincidence with a sample of neutrino candidate
events detected by ANTARES in 2007 [16].
IV. SEARCHES FOR EXOTIC PHYSICS
ANTARES is also searching for signatures of physics
beyond the Standard Model. An analysis is performed
that looks for neutrinos produced by Dark Matter particles
annihilating in the Sun and the Galactic center [17].

Page 8

ANTARES COLLABORATION RESULTS
Fig. 4: Flux upper limit (90 % C.L.) on up-going magnetic
monopole for relativistic velocities (0.55 ≤ β ≤ 0.995)
is compared to the published upper limits set by other
experiments are also shown (see [18] for references) as
well as the theoretical Parker bound.
Magnetic Monopoles with masses between 1010and 1014
traversing the detector volume would be detected as a
very bright track. A search for this signature has been
conducted [18] and a limit on the flux of monopoles with
β > 0.55 has been obtained; see Figure 4. This limit is
more stringent than those from previous experiments.
Another hypothetical form of matter is formed by Nu-
clearites: particles composed of strange quark matter.
The signature is a slow-moving (e.g. 10−3c) bright point
traversing the detector. The searches are described in [19].
V. CONCLUSION
The first deep-sea neutrino telescope, ANTARES, has
been taking data for four and a half years now. A large
number of analysis are being performed, looking for
astrophysical signals of neutrinos, either stand-alone or by
looking for coincident observations with a variety of other
experiments. The geographical position, combined with
the good angular resolution allow ANTARES to explore,
in particular, Galactic neutrino sources in the relevant
energy range. In addition, several analyses are aimed at
detecting signals from non-standard model particles.
The successful operation of ANTARES, and analysis of
its data, is an important step towards KM3NET [20] a
future km3-scale high-energy neutrino observatory and
marine sciences infrastructure planned for construction in
the Mediterranean Sea.
REFERENCES
[1] M. Ageron et al. Nucl. Instr. Meth.,arXiv:1104.1607 [astro-
ph.IM].
[2] J.A. Aguilar et al, Astropart. Phys. 34 (2011) 539549
[3] A. Heijboer, PhD Thesis, university of Amsterdam, 2004.
[4] S. Mangano for the ANTARES collaboration, these proceed-
ings
[5] C. Riviere for the ANTARES collaboration, these proceed-
ings
[6] J.A. Aguilar et al, Physics Letters B 696 (2010) 16; also F.
Sch¨ ussler for the ANTARES collaboration, these proceedings.
[7] C. Bogazzi for the ANTARES collaboration, these proceed-
ings
[8] S. Adri´ an-Mart´ ınez et al, arXiv:1108.0292v1, submitted to
ApjL
[9] R. Abbasi et al, Astrophysical Journal 732 (2011) 18
[10] J.P. Gomez for the ANTARES collaboration, these proceed-
ings
[11] F. Sch¨ ussler for the ANTARES collaboration, these pro-
ceedings.
[12] J. Petrovic for the ANTARES collaboration, these proceed-
ings
[13] D. Dornic for the ANTARES collaboration, these proceed-
ings
[14] C. Reed for the ANTARES collaboration, these proceed-
ings
[15] J.A. Aguilar et al, Astropart. Phys. 34 (2011) 652-662
[16] V. Van Elewyck for the ANTARES collaboration, these
proceedings
[17] G. Lambard for the ANTARES collaboration, these pro-
ceedings
[18] N. Picot-Clemente for the ANTARES collaboration, these
proceedings
[19] V. Popa for the ANTARES collaboration, these proceedings
[20] P. Kooijman for the KM3NeT collaboration, these proceed-
ings; also www.km3net.org

Page 9

32ND INTERNATIONAL COSMIC RAY CONFERENCE, BEIJING 2011
Searching for Point Sources of High Energy Cosmic Neutrinos with the ANTARES telescope
CLAUDIO BOGAZZI1ON BEHALF OF THE ANTARES COLLABORATION
1FOM Instituut voor Subatomaire Fysica Nikhef, Science Park 105, 1098 XG Amsterdam, The Netherlands
claudiob@nikhef.nl
Abstract. ANTARES is currently the largest neutrino
detector on the Northern Hemisphere. It consists
of a tri-dimensional array of 885 photomultipliers
arranged on 12 vertical lines, placed at a depth of
2475 meters in the Mediterranean Sea near Toulon,
France. The telescope, completed in 2008, detects
the Cherenkov radiation of muons produced by high
energy neutrinos interacting in or around the detector.
Muon tracks are then reconstructed using a likelihood-
based algorithm. One of the main goals of the exper-
iment is the search for high-energy neutrinos from
astrophysical point-like sources. Due to its location,
ANTARES is sensitive to up-going neutrinos from
many potential galactic sources in the TeV to PeV
energy regime. New results from an unbinned method
as well as the sensitivity of the detector are presented.
I. INTRODUCTION
One of the main questions in astroparticle physics is
the origin of high energy Cosmic Rays (CRs). In the
last decade progress has been made related to energy
spectrum and composition [1]. However, the origin of
CRs remains unknown. Many acceleration sites have been
suggested, such as supernova remnants, microquasars and
active galactic nuclei [2]. The final signature of these
cosmic accelerators are gamma rays and high energy
neutrinos produced through hadronic interactions. The
observation of a point-like source of neutrinos would then
offer a unique occasion to study the mechanism of CRs
acceleration.
A. The ANTARES detector
The ANTARES detector is located at a depth of 2475
m in the Mediterranean Sea, 42 km from Toulon in the
south of France (42◦48N,6◦10E). It consists of a tri-
dimensional array of 885 optical sensors arranged on 12
vertical lines. Each line comprises up to 25 detection
storeys each equipped with 3 downward-looking 10-inch
photo-multipliers (PMTs), oriented 45◦to the line axis.
The spacing between storeys is 14.5 m while the lines are
spaced by 60-70 m. A buoy at the top of the line keep
them to stay vertical.
The telescope operates by detecting Cherenkov light
emitted by charged particles that result from neutrino
interactions in or around the detector. The arrival time
and amplitude of the Cherenkov light on the PMTs are
digitized into ’hits’ [3] and transmitted to shore.
II. DATA SELECTION
The data analysed in this work were collected between
January 31st 2007 and December 30th 2010. The total
livetime of the analysis is 813 days of which 183 days
were with 5 lines, while for the remaining 630 days the
detector consisted of 9, 10 and 12 lines
The reconstruction of the muon track is achieved using the
time and position information of the hits. The algorithm is
based on a maximum likelihood method [4] where a multi-
stage fitting procedure is applied in order to maximise
the likelihood of the observed hit times as a function
of the muon direction and position. The quality of the
reconstruction is defined by the variable Λ, which is based
on the maximisation of the log-likelihood [4]. Figure 1
shows the cumulative distribution of Λ for upward-going
events with the simulated contributions of atmospheric
muons and neutrinos. Atmospheric muons are simulated
with the MUPAGE package [4]; neutrinos are instead
generated with the GENNEU [5] package and the Bartol
model [7].
Neutrino candidates events are selected requiring an up-
ward going track, i.e. zenith angle < 90◦, and a value
for the lambda variable Λ > −5.2. The latter is obtained
by optimizing the background reduction and the signal
efficiency, in terms of the discovery potential. Another
cut is then applied in order to reject mis-reconstructed
atmospheric downward going muons using the informa-
tion of the uncertainty on the reconstructed muon track
direction obtained from the fit. This value is required to
be ≤ 1◦. The final sample consists of 3058 events; from
the simulations 84% of them are estimated to be neutrinos,
while the rest are mis-reconstructed atmospheric muons.

Page 10

CLAUDIO BOGAZZI ON BEHALF OF THE ANTARES COLLABORATION
cumulative number of events
1
10
2
10
3
10
4
10
5
10
6
10
7
10
Antares 2007-2010 preliminary
data
MC
ν
MC
µ
Λ
quality variable
-7
0
-6.5-6 -5.5-5-4.5-4 -3.5 -3
ratio
0.5
1
1.5
Fig. 1: Cumulative distribution of the reconstructed quality
variable Λ. The dashed line is for simulated atmospheric
muons while the solid line corresponds to simulated
atmospheric neutrinos. The bottom plot shows the ratio
between data and Monte Carlo.
III. DETECTOR PERFORMANCE
The angular resolution and the acceptance of the detec-
tor have been obtained from simulation. The systematic
uncertainty on the angular resolution has been computed
following the procedure described in [9] by smearing the
hit times according to a Gaussian with a width of σtin
order to artificially deteriorate the simulated timing accu-
racy. The study leads to exclude an additional smearing
of 3 ns which was found to be incompatible with data at
the 2σ level where σ is the uncertainty on the flux model
[8]. The best agreement between data and Monte Carlo is
obtained for σt= 2 ± 0.5 ns. This value is indeed used
for the simulations presented in this paper.
A. Angular resolution
Figure 2 shows the cumulative distribution of the angle
between the reconstructed muon direction and the gen-
erated neutrino direction for neutrino events where we
assumed an energy spectrum proportional to E−2
Eνthe neutrino energy. The median of this angular error
is 0.46 degrees.
ν
with
B. Acceptance
The acceptance for signal neutrinos is also estimated using
simulations. In the search, we deal with fluxes of the form
of
dN
dE= φ(Eν
where φ is the flux normalisation. The acceptance, A, is
defined as the constant of proportionality between φ and
the number of selected events. Figure 3 shows exactly
this proportionality: for a source at a declination of -90
(0) degrees, A = 8.8(4.8) × 107GeV cm2s. Systematic
uncertainties on the acceptance are constrained by the
agreement between the simulated atmospheric neutrino
GeV)−2GeV−1s−1cm−2,
(1)
log10( angle / deg )
-2 -1.5-1 -0.50 0.51 1.52
cumulative distribution
0
0.2
0.4
0.6
0.8
1
neutrino MC, preliminary
-2
ANTARES 2007-2010 E
Fig. 2: Cumulative distribution of the angle between
the reconstructed muon direction and the true neutrino
direction for simulated upward going neutrinos that pass
the cuts described in Section II assuming a E−2
spectrum.
ν
neutrino
sin(delta)
-1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.81
s)
2
GeV cm
7
(10
sig
N
-1
φ
0
1
2
3
4
5
6
7
8
9
neutrino MC, preliminary
-2
ANTARES 2007-2010 E
Fig. 3: Acceptance, i.e. the constant of proportionality
between the normalisation factor for an E−2flux and the
selected number of events.
sample and data. For the computation of the flux limits
an uncertainty of 15% is assumed.
IV. SEARCH METHOD
The algorithm used on the analysis is based on the
likelihood of the observed events which is defined :
?
+ Bi× N(Ni,bkg
where the sum is over the events, F is the point spread
function, i.e. the probability density of reconstructing an
event i at a distance βifrom the true source declination
and right ascension δs,αs; B is a parametrization of the
background rate, obtained from the observed declination
distribution of the events; µsigis the mean number of de-
tected events that the source produces and µtotrepresents
the total number of expected events and N(Ni
logLs+b=
i
log[µsig× F(βi(δs,αs)) × N(Ni,sig
hits)
hits)] + µtot,
hits) is the

Page 11

32ND INTERNATIONAL COSMIC RAY CONFERENCE, BEIJING 2011
mean number of signal events
0510 15202530
discovery probability
0
0.2
0.4
0.6
0.8
1
), preliminary
o
= −70
δ
ANTARES 2007 − 2010 MC, full sky search (
info
hits
not using N
σ
5
info
hits
using N
σ
5
info
hits
using N
σ
3
info
hits
not using N
σ
3
Fig. 4: Probability for a 3σ (dashed and solid lines) and 5σ
(dotted and dashed-dotted lines) discovery as a function
of the mean number of signal events for the case where
we use the number of hits information in the likelihood
(dashed and dotted lines) and for the case where we do
not use it (solid and dashed-dotted lines) for the full sky
search. In this case the signal was added at a declination
of -70◦.
probability for an event i to be reconstructed with Nhits
number of hits (this probability was not included in the
analysis with 2007 and 2008 data [9]).
In order to compute the test statistic the free parameters of
the likelihood are maximized. We have now to distinguish
between the two different analysis presented in this paper:
in the candidate list search only the µsigare fitted while
in the full sky search we have in addition the source
coordinates (δs,αs) to fit. In both cases the results of
the fit are the maximum likelihood value Lmax
estimates of the free parameters. The test statistic is then
defined as:
s+band the
Q = Lmax
s+b− Lb
(2)
where Lbis the likelihood computed for the background
only case. The higher Q the more the data are compatible
with signal.
Just using the number of hits information in the likelihood
let us to gain a 25% (22%) factor for the 3 (5) σ discovery
probability as shown in Figure 4 for the full sky search.
V. RESULTS
As mentioned above two different analyses have been
done. The first one is a full sky survey with no assump-
tions about the source position. In the second analysis, we
made a search for a signal excess in an a priori defined
spot in the sky corresponding to the position of some
interesting astrophysical objects.
-150-100-50050 100150
-80
-60
-40
-20
0
20
40
60
80
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Antares 2007-2010, preliminary
=-90
δ
=+90
δ
Fig. 5: Galactic skymap showing the 3058 data events. The
position of the most signal-like cluster is indicated by the
circle. The stars denote the position of the 51 candidate
sources.
A. Full sky search
In the full sky search, no significant clusters of neutrino
candidates were found. The most signal-like cluster is
located at αs,δs = (−46.5◦,−65.0◦). The fit assigns
5 events above the background. The value of the test
statistic for this cluster is 13.0 which yields to a p-value
of 2.6%. Figure 4 shows a sky map of the selected events
in galactic coordinates with the location of the most signal
like cluster.
B. Candidate list search
The results of the search in the direction of 51 pre-
defined candidate sources are shown in Table I. None
of the sources have a significant excess of events. The
most signal-like candidate source is HESS J1023-575
where the post-trial p-value is of 41%. Figure 6 shows
the 90% confidence level limits on φ using the Feldman-
Cousins prescription [10] and assuming an E−2neutrino
spectrum for each of the source candidates as a function
of the source declination. The sensitivity of this analysis
is also presented, defined as the median expected limit
and resulting in a factor 2.7 better than the one obtained
with data collected during 2007 and 2008 only [9]. Limits
from other experiments are also shown. However, it should
be noted that for this spectrum, ANTARES detects most
events at energies around 10 TeV while the limits in the
Southern Hemisphere published by the IceCube Collabo-
ration [11] apply to the PeV region.
VI. CONCLUSIONS
A search of high energy cosmic neutrinos has been
performed. Data were taken during the first four years of
operation when ANTARES consists of 5 line for most of
the first year considered and 9, 10 and 12 for the rest. A

Page 12

CLAUDIO BOGAZZI ON BEHALF OF THE ANTARES COLLABORATION
source
HESS J1023-575
3C 279
GX 339-4
Cir X-1
MGRO J1908+06
ESO 139-G12
HESS J1356-645
PKS 0548-322
HESS J1837-069
PKS 0454-234
IceCube hotspot
PKS 1454-354
RGB J0152+017
Geminga
PSR B1259-63
PKS 2005-489
HESS J1616-508
HESS J1503-582
HESS J1632-478
H 2356-309
MSH 15-52
Galactic Center
HESS J1303-631
HESS J1834-087
PKS 1502+106
SS 433
HESS J1614-518
RX J1713.7-3946
3C454.3
W28
HESS J0632+057
PKS 2155-304
HESS J1741-302
Centaurus A
RX J0852.0-4622
1ES 1101-232
Vela X
W51C
PKS 0426-380
LS 5039
W44
RCW 86
Crab
HESS J1507-622
1ES 0347-121
VER J0648+152
PKS 0537-441
HESS J1912+101
PKS 0235+164
IC443
PKS 0727-11
αs(◦)
155.83
-165.95
-104.30
-129.83
-73.01
-95.59
-151.00
87.67
-80.59
74.27
75.45
-135.64
28.17
98.31
-164.30
-57.63
-116.03
-133.54
-111.96
-0.22
-131.47
-93.58
-164.23
-81.31
-133.90
-72.04
-116.42
-101.75
-16.50
-89.57
98.24
-30.28
-94.75
-158.64
133.00
165.91
128.75
-69.25
67.17
-83.44
-75.96
-139.32
83.63
-133.28
57.35
102.20
84.71
-71.79
39.66
94.21
112.58 -
δs(◦)
-57.76 0.41
-5.79 0.48
-48.79 0.72
-57.17 0.79
6.27 0.82
-59.94 0.94
-64.50 0.98
-32.27 0.99
-6.95 0.99
-23.43 1.00
-18.15 1.00
-35.67 1.00
1.79 1.00
17.01 1.00
-63.83 1.00
-48.82 1.00
-50.97 1.00
-58.74 1.00
-47.82 1.00
-30.63 1.00
-59.16 1.00
-29.01 1.00
-63.20 1.00
-8.76 1.00
10.52 1.00
4.98 1.00
-51.82 1.00
-39.75 1.00
16.15 1.00
-23.34 1.00
5.81 1.00
-30.22 1.00
-30.20 1.00
-43.02 1.00
-46.37 1.00
-23.49 1.00
-45.60 1.00
14.19 1.00
-37.93 1.00
-14.83 1.00
1.38 1.00
-62.48 1.00
22.01 1.00
-62.34 1.00
-11.99 1.00
15.27 1.00
-44.08 1.00
10.15 1.00
16.61 1.00
22.51 1.00
11.70 1.00
pφ90%CL
6.6
10.1
5.8
5.8
10.1
5.4
5.1
7.1
8.0
7.0
7.0
5.0
6.3
7.3
3.0
2.8
2.7
2.8
2.6
3.9
2.6
3.8
2.4
4.3
5.2
4.6
2.0
2.7
5.5
3.4
4.6
2.7
2.7
2.1
1.5
2.8
1.5
3.6
1.4
2.7
3.1
1.1
4.1
1.1
1.9
2.8
1.3
2.5
2.8
2.8
1.9
TABLE I: Results of the candidate source search. The
source coordinates and the p-values (p) are shown as well
as the limits on the flux intensity φ90%CL; the latter has
units 10−8GeV−1cm−2s−1.
declination (degrees)
-80-60 -40-2002040 6080
)
-1
s
-2
flux limit ( GeV cm
-9
10
-8
10
-7
10
-6
10
-5
10
-4
10
Antares 2007-2010 preliminary
ANTARES (304 days)
ANTARES (813 days)
IceCube 40 (375.5 days)
MACRO (2300 days)
Amanda-II (1387 days)Amanda-II (1387 days)Amanda-II (1387 days)Amanda-II (1387 days)
ANTARES sensitivity (304 days)
ANTARES sensitivity (813 days)
IceCube 40 sensitivity
Super-K (2623 days) Super-K (2623 days) Super-K (2623 days) Super-K (2623 days)
ANTARES (304 days)
ANTARES (813 days)
IceCube 40 (375.5 days)
MACRO (2300 days)MACRO (2300 days) MACRO (2300 days)
ANTARES sensitivity (304 days)
ANTARES sensitivity (813 days)
IceCube 40 sensitivity IceCube 40 sensitivityIceCube 40 sensitivity
ANTARES (304 days)
ANTARES (813 days)
IceCube 40 (375.5 days)IceCube 40 (375.5 days)
ANTARES sensitivity (304 days)
ANTARES sensitivity (813 days)ANTARES sensitivity (813 days)
ANTARES (304 days)
ANTARES (813 days)
ANTARES sensitivity (304 days)
Fig. 6: Limits set on the normalisation φ of an E−2
spectrum of high energy neutrinos from selected candi-
dates (see Table I). Also shown is the sensitivity, which
is defined as the median expected limit. In addition to
the present result, several previously published limits on
sources in both the Southern and Northern sky are also
shown.
ν
measurement of the angular resolution using MonteCarlo
data yields to 0.46 degrees. Neither the full sky search
nor the candidate list search show a significant excess of
events therefore limits have been obtained on the neutrino
flux.
REFERENCES
[1] E. Waxmann, to be published in Astronomy at the Frontiers
of Science, 2011, ed. J. P. Lasota, Springer.
[2] J. K. Becker, Physics Reports, 2008, 458: 173-246.
[3] J. A. Aguilar et al. , Nucl. Instr. Meth., 2010, A622: 59-73.
[4] A. Heijboer,2004,
http://antares.in2p3.fr/Publications.
[5] G. Carminati, M. Bazzotti, A. Margiotta and M. Spurio,
Comp. Phys. Comm., 179: 915.
[6] D. Bailey,2002,
http://antares.in2p3.fr/Publications.
[7] V. Agrawal, T. K. Gaisser, P. Lipari, T. Stanev Phys. Rev.
D, 1996, 53: 1314-1323.
[8] G. D. Barr et al. , Phys. Rev. D, 2006, 74(9): 094009.
[9] S. Adrian-Martinez et al. (ANTARES Collaboration),
arXiv:1108.0292v1 [astro-ph.HE], submitted to Astrophysical
Journal Letters.
[10] G. J. Feldman and R. D. Cousins, Phys. Rev. D, 1998,
57(7):3873-3889.
[11] R. Abbasi et al. , Astrophys. J., 2011, 732: 18.
PhDThesis,
PhDThesis,

Page 13

32ND INTERNATIONAL COSMIC RAY CONFERENCE, BEIJING 2011
Search for point sources with the ANTARES neutrino telescope using the EM algorithm
JUAN PABLO G´ OMEZ-GONZ´ ALEZ1, ON BEHALF OF THE ANTARES COLLABORATION.
1IFIC - Instituto de F´ ısica Corpuscular, Edificios de Investigaci´ on de Paterna, CSIC - Universitat de Val` encia, Apdo.
de Correos 22085, 46071 Valencia, Spain.
jpablo@ific.uv.es
Abstract. The ANTARES detector, currently the largest
deep-sea neutrino telescope in the Northern Hemi-
sphere, consists of a three-dimensional array of 885
optical modules arranged over 12 detection lines an-
chored at a depth of 2475 m in the Mediterranean
Sea, 40 km offshore from Toulon (France). The pho-
tomultiplier tubes detect the Cherenkov light induced
by the charged particles produced in the interaction
of cosmic neutrinos with the matter surrounding
the detector. The trajectories of the resulting muons
are reconstructed with high precision, revealing the
direction of the incoming neutrinos. The main scientific
goal of ANTARES is the search for high energy
neutrinos coming from astrophysical sources. This
contribution describes a point source search using a
dedicated clustering algorithm, based on the analytical
maximization of the likelihood. The results of shuch
analysis using four years of data will be presented.
I. INTRODUCTION
The ANTARES neutrino telescope [2] started data taking
in 2007 and is fully operational since 2008. Located at 40
km off the coast of Toulon it consists of 12 detection lines
anchored to the seabed at a depth of 2475 m and sustained
vertically by means of buoys. Each line has 25 floors (or
storeys) composed by a triplet of photomultiplier tubes
(PMTs) housed in pressure resistant glass spheres called
optical modules (OMs). The OMs are facing downward at
45◦from the vertical for an increased detection- efficiency
for up-going neutrinos.
This three-dimensional photo-detector array detects the
Cherenkov light emitted by the charged leptons originated
in the interaction of high energy neutrinos with the matter
surrounding the detector. The tracks of the produced
muons can be reconstructed using the position and timing
information of the hits arriving to the PMTs. An accurate
timing and position calibration [2] of the detector OMs is
necessary in order to achieve the best attainable angular
resolution.
The main goal of the experiment is the detection of high
energy neutrinos from extraterrestial origin, and one of
the most promising ways of establishing their existence
is the search for point sources. Here, we present such
a search using data collected between years 2007 and
2010 for a total of 813 days of livetime. In section 2
the track reconstruction method and data selection criteria
are described. The detector performance is reviewed in
section 3. The clustering algorithm applied in this analysis
is explained in section 3. Finally the search results are
presented in section 4.
II. DATA SELECTION AND TRACK RECONSTRUCTION
Data runs used in this analysis were recorded in the first
four years of detector operation. Taking into account the
time spent on sea operations (like the deployment of new
lines) and sporadic data taking problems of the detector,
the total livetime of the analysis is 813 days; about 77%
of this data were collected using 9, 10 and 12 detection
lines, while the remaining 183 days correspond to data
gathered with the initial 5-lines configuration.
The reconstruction method [3] is based on the maximiza-
tion of the likelihood function describing the probability
density function (PDF) for the residuals, defined as the
difference between the measured hit time and the ex-
pected arrival time of the hits. The goodness of the track
reconstruction is described by the Λ parameter, which
is basically the log-likelihood of the fitted track. This
parameter can be used to eliminate badly reconstructed
tracks by selecting an appropriate cut on the Λ value. The
corresponding cumulative Λ distribution for events recon-
structed as upgoing is shown on Figure 1. The contribution
from the different components of the expected background
is also included. The simulation reproduces well the data.
For this analysis atmospheric muons were simulated us-
ing the MUPAGE package [10], while the atmospheric
neutrinos were generated with the GENNEU [5] package
according to the Bartol model [6].
A cut on the quality of the reconstruction at Λ ≥ -5.2
was found to be the optimal for the search discovery

Page 14

JUAN PABLO G´ OMEZ GONZ´ ALEZ, SEARCH FOR POINT SOURCES WITH ANTARES
cumulative number of events
1
10
2
10
3
10
4
10
5
10
6
10
7
10
Antares 2007-2010 preliminary
data
MC
ν
MC
µ
Λ
quality variable
-7
0
-6.5 -6-5.5-5 -4.5-4 -3.5-3
ratio
0.5
1
1.5
Fig. 1: Cumulative distribution of the quality of the
reconstruction parameter for data and MC upgoing events.
potential using muon tracks recostructed as upward going
(θ < 90◦). Additionally, the uncertainty on the muon
direction estimated from the fit is required to be ≤ 1◦.
The selected sample contains 3058 events, out of which
it is estimated from MC simulations that about 84% are
neutrino events and only 16% downgoing atmospheric
muons mis-reconstructed as upgoing.
III. DETECTOR PERFORMANCE
The two main parameters describing the performance
of a neutrino telescope are the angular resolution and
the acceptance. Both parameters are estimated from
simulations. Figure 2 shows the cumulative angular
resolution for upgoing neutrino events following an E−2
spectrum and complying the selection criteria described
in Section 2. The plot shows that roughly 80% of the
signal events are reconstructed with an angular error less
than 1◦, being the median value of the reconstruction
error equal to 0.46 ± 0.1◦. The uncertainty on this value
has been computed considering all the effects leading
to a deterioration of the detector timing resolution [7].
In addition, the uncertainty on the absolute orientation
of the detector, which is estimated to be of the order of
0.1◦, is also taken into account in the limits computation
(see section 4).
The acceptance allows us to relate the detected event-
rate with the neutrino flux of the source at the Earth,
and it is shown on Figure 3 as a function of the si-
nus of the declination considering a flux normalization
φ = 10−8GeV−1cm−2s−1. Based on the agreement
between data and simulations a 15% systematic error on
the detection efficiency has been considered for the limits
calculation.
IV. CLUSTERING METHOD
Point source clustering techniques try to identify and sep-
arate events coming from real sources from background
events. The Expectation-Maximization (EM) algorithm [8]
] °
) [
α
(
10
log
-2 -1.5-1 -0.50 0.51 1.52
cumulative distribution (a.u.)
0
0.2
0.4
0.6
0.8
1
, preliminary
ν
-2
ANTARES 2007-10 E
Fig. 2: Cumulative angular resolution for E−2upgoing
neutrinos selected for this analysis.
) δ
sin(
-1-0.8 -0.6 -0.4 -0.20 0.2 0.4 0.6 0.81
s)
2
cm
-1
flux (GeV
-2
acceptance for E
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
neutrino MC
-2
ANTARES 2007-2010 E
Fig. 3: Detector acceptance as a function of the sinus of
the declination.
used in this analysis is a pattern recognition algorithm
that analytically maximizes the likelihood in finite mixture
problems. These mixture models are different groups of
data described by different density components. In the
case of a search for neutrino point sources [9] the mixture
problem can be expressed as:
p(x) = πbgPbg(δ) + πsgPsg(x;µ;Σ) · (Pnhits
sg
/Pnhits
bg
)
(3)
where πbgand πsgare the so-called mixing proportions,
x = (α,δ) is the position of the signal event in equatorial
coordinates, µ = (µα,µδ) and Σ = (σα,σδ) are, respec-
tivly, the mean and the covariance vector of the Gaussian
distribution, and Pnhitsis the probability for an event to
be reconstructed using nhits number of hits.
In this analysis the expected density distributions of
background and signal events are parametrized. The pdf
describing the background is obtained from the declination
distribution of data events, while the source signals are
supposed to follow a two-dimensional Gaussian distribu-
tion.
The EM algorithm works in two steps. In the first step
called “Expectation” the expected value of the complete

Page 15

32ND INTERNATIONAL COSMIC RAY CONFERENCE, BEIJING 2011
050100150200250300350
-100
-80
-60
-40
-20
0
=-180
20
40
60
80
°α
°
=180
α
°
=-90
δ
°
=60
δ
ANTARES 2007-2010, preliminary
Fig. 4: Skymap with the equatorial coordinates of the 3058
events selected. The position of the most significant cluster
is denoted by the green circle and the coordinates of the
candidate list sources are indicated with the stars.
data log-likelihood is computed for a given set of param-
eters. In the “Maximization” step a new set of parameters
that maximizes the likelihood is found. In our case, the
parameters to be maximized are the two components of the
Gaussian width, the expected number of signal events and,
in the full sky search (see next section), the coordinates
of the signal source.
After likelihood maximization the so called test-statistic,
defined as the likelihood ratio of the two mixture models,
is computed. Lower values of this quantity indicate that
data is more likely to be produced by the background,
while larger values are more likely to be produced by the
presence of the searched signal.
V. RESULTS
Two different searches for point sources have been consid-
ered in this analysis. In the first approach a blind survey is
done looking everywhere in the whole ANTARES visible
sky. The second search used a catalog of candidate sources
to look for presence of signal at particular locations
in the sky. The candidate list of sources includes both
galactic and extra-galactic sources known to be gamma-
ray emitters. The detector visibility and PSF was taking
into account when defining the list.
No significant excess of events was found neither in the
full sky search, nor in the candidate list search. The
most signal-like cluster was found at (δ = −64.87◦,
α = −46.49◦) in the all sky survey. For this cluster a
2.6% excess p-value is found.
The locations (in equatorial coordinates) of the most
significant cluster, the 3058 events selected and the 51
candidate sources are shown on figure 4. Upper limits1on
the E−2neutrino flux spectrum are reported in table I and
in figure 5 as a function of the declination for the sources
-80-60-40-200204060
) °
(
80
-9
10
-8
10
-7
10
-6
10
-5
10
-4
10
ANTARES sensitivity (813 days)
ANTARES (813 days)
MACRO Source List (2300 days)
Super-K Source List (2623 days)
AMANDA-II Source List (1387 days)
IceCube 40 (376 days)
]
-1
s
-2
[GeV cm
90%
φ
µ
ν
δ
ANTARES 2007-2010 (preliminary)
Fig. 5: Limits on the E−2
candidate list search. Upper limits previously reported by
other neutrino experiments for both Northern and South-
ern sky are shown. The ANTARES sensitivity computed
in this analysis is also included.
ν
flux for the 51 sources in the
in the candidate list. The ANTARES sensitivity (defined
as the median value of the expected limit) it is shown,
as well as limits reported by other neutrino experiments
included for comparison.
VI. CONCLUSIONS
This contribution presented the analysis of 813 days of
livetime using data collected in the first four years of the
ANTARES neutrino telescope operation. No statistically
significant excess of events has been found neither in the
search using a candidate list of interesting sources, nor in
the full sky search. The most significant cluster, with a
post-trial probability of 2.6 % was found at coordinates
δ = −64.87◦, α = −46.49◦. Some of the most stringent
limits to E−2
ν
flux were obtained for sources located in the
ANTARES field of view. Using a different search method
the results presented here are consistent with the main
analysis [10] reporting upper limits for the 51 candidate
sources using the Feldman-Cousins prescription.
Igreatfullyacknowledge
MICINN(FPA2009-13983-C02-01
CSD2009-00064) andof
(Prometeo/2009/026).
thefinancialsupport
MultiDark
Valenciana
of
and
Generalitat
REFERENCES
[1] M. Ageron et. al., ANTARES: the first undersea neutrino
telescope, 2011, arXiv:1104.1607v1.
[2] J. A. Aguilar et al., Time Calibration of the ANTARES
neutrino Telescope., Astroparticle Physics, 2011, 34, 539-549.
[3] A.Heijboer,Track
sourcesearcheswith
UniversitetvanAmsterdam,
http://antares.in2p3.fr/Publications/index.html#thesis.
reconstruction
ANTARES,
andpoint
thesis,PhD
available2004,at
1. Here we follow the Neyman prescription.