A review of the past and present MAGIC dark matter search program and a glimpse at the future

0.36 MB·

by  Michele Doro

XXV European Cosmic Ray Symposium, Turin, Sept. 4-9 2016 1
A review of the past and present MAGIC dark matter search program
and a glimpse at the future
Michele Doro
on behalf of the MAGIC Collaboration
University & INFN Padova, via Belzoni 7, 35131 Padova, Italy, [email protected]
The MAGIC TeV gamma-ray telescopes have devoted several hundreds hour of observation time
inaboutadecade,tohuntforparticledarkmatterindirectsignaturesingammarays,fromvarious
candidate targets of interest in the sky: the galactic center, satellite galaxies, galaxy clusters and
unidentifiedobjectsinotherbands. Despitetheeffort,nohintsarepresentinMAGICdata. These
observation are nevertheless not unusable. MAGIC indeed derived the most robust upper limits in
the TeV range than any other instrument. These results, for the time being, only mildly constrain
7
someclassicdarkmattermodels,butareofuseintheconstructionofdarkmattermodelsforthenext
1
searches, that consider also the negative results from accelerator and direct-detection experiments.
0
In the contribution, we discuss and review MAGIC results, putting them into context, and in
2
perspective with the next generation of ground-based Cherenkov telescopes. We will briefly inform
n about future MAGIC projects regarding dark matter searches.
a
J

0 I. INTRODUCTION hundreds of GeV for the mass of the particle [4].
2
In this energy range, ground based Imaging Atmo-

spheric Cherenkov Telescope Arrays (IACTAs), ob-
] The understanding of the nature of Dark Matter
E serving the Cherenkov light produced in Extended
(DM),eitherasanewparticle[1]orasamodification
H Atmospheric Showers (EAS) generated in the strato-
ofthegravitationallaw[2],iskeepinghundredsofsci-
sphere by cosmic gamma rays, are an optimal in-
. entistsandinstrumentsoccupiedinhuntingforsignif-
h strument to search for WIMPs, for the following rea-
p icant signatures, especially in the past two decades.
sons: a) in many scenarios, WIMPs produce a strong
- The need of DM appears from several observations,
o gamma-ray yield in the annihilation or decay prod-
all connected to gravitational effects, at all cosmo-
r ucts, b) the neutrality of photons guarantee that tele-
t logical scales: from the galactic motion of stars, to
s scopes can be pointed to astronomical places where
that of galaxies in clusters, and farther away into the
a DM is expected, c) annihilation or decay spectrum
[ signatures in acoustic oscillations in the cosmic mi-
would be universal, and therefore multiple observa-
crowave background [3]. It is very hard to disprove
1 tionsofsuchspectrafrommultiplesourceswouldpro-
such strong hints of gravitational imbalances, as well
v vide a strong claim, and finally d) the annihilation or
asitiseasiertoexplainthosewiththeintroductionof
2
decayspectraoftenshowlikeacutoffconnectedtothe
0 one (or more) new particle(s), to be the candidate for
DM mass, and sometimes peculiar bumps that could
7 DM.Thisnewparticleshouldbeeitherstable,orvery
disentangle its origin from a standard astrophysical
5 long lived, to guarantee the relic density we see today
0 Ω = 0.259 ± 0.006 [3]. This one parameter, al- one.
. DM The Major Atmospheric Gamma-ray Imaging
1 lowsanyhowforaverylargeparameterspaceinterms
Cherenkov (MAGIC) instrument of the IACTA class,
0 of DM particle mass and annihilation or decay rates.
7 However, the current paradigm focuses on a “cold” is a pair of 17-m diameter parabolic dishes, oper-
1 ating since 2009 at the Observatorio Roque de Los
DM scenario, in which the particle has been non rel-
: Muchachos (ORM, La Palma, Canary Islands, Spain)
v ativistic since its decoupling. The velocity of DM
at 2,200 m. a.s.l, in the Northern Hemisphere1.
i particles determines their free-streaming length, and
X MAGIC operates between few tens of GeV and sev-
thus strongly affect the cosmic structures formation,
r eral tens of TeV, with a peak sensitivity of 0.66% of
a at least in the current preferred bottom-up scenario
the Crab Nebula flux above 280 GeV, an energy res-
of merging of smaller structures into larger ones. The
olution of 10−20% and an angular resolution below
best CDM candidate is a particle that does not have
0.1◦ [5]. MAGIC observes during night time, with
anystandardinteraction,andisaWIMP(WeaklyIn-
about1,500hofavailabletimeperyear(1,000hwith
teracting Massive Particles). The WIMP may live in
moonless night). MAGIC, as well as other IACTA in
adarksector(nointeractionwiththestandardmodel
the field, is a wide-scope detector, focused mainly on
particles), which would make the detection prospects
galactic and extragalactic astrophysics with gamma
difficult, or have some channels to the standard mod-
rays, but with good potential also as particle detec-
els, like in the case of Super-symmetrical extensions
of the standard models (SUSY), or Unified Extra Di-
mensions theories. The non-observation of such par-
ticle at LHC, albeit with maximum luminosity, has
pushedthesearchestoamassrangeofseveraltensor 1 MAGICstarteditsoperationsin2004asasingletelescope.
eConf C16-09-04.3
2 XXV European Cosmic Ray Symposium, Turin, Sept. 4-9 2016
tors (electron/positron, proton/antiproton, and per- evolved significantly with time. MAGIC started its
haps neutrino) and exotic physics detectors (see [6] operations in 2004 as single telescope, and was cou-
for a recent review). MAGIC has devoted a signif- pled with a second telescope only in 2009. Both the
icant fraction of its observation time, reaching now first and second MAGIC telescopes had undergone
several hundreds of hours, to DM-related searches at major upgrades along the years, that have substan-
several candidate targets. These include the Milky tiallychangedtheirperformance: fromthefirstsingle-
Way (MW) barycenter region, dwarf satellite galax- telescopesetup,tothecurrent,MAGIChasimproved
ies(DSGs)orbitingintheMWDMhalo,unidentified its sensitivity by a factor of 4 (meaning a factor of 16
Fermi-LAT objects, galaxy clusters, and others (see less time required) at 300 GeV, and a factor of 10 at
also [7] for a recent collection of DM searches from 50GeV.Inthefollowing,wediscusstheearlysearches
IACTA in the past decade). In these 10 years, the by target class.
preferredtargets,campaignsdurationanddatarecon-
struction have changed. In this contribution, we put Dwarf Satellite Galaxies (DSGs) are small galaxies
theMAGICsearchintocontext,showingtheevolution with a common mass scale [9] commonly believed to
ofthissciencefield,andconcludingwithsomeremarks be originated in DM overdensities present in the MW
on the possible future extension of the MAGIC DM DM halo. DSGs have a small stellar content, and
program and the validity of its contribution in the especially are almost depleted of gas, showing no or
field. little stellar activity in the past Gy. Their star veloc-
Thegamma-rayfluxfromDMannihilationordecay ity distribution normally hints to large DM content,
at Earth can be factorized as a product of a particle- withmass-to-lightratioof1000M /L orevenmore,
(cid:12) (cid:12)
physics factor depending on the nature of the DM, dependingonthetargetobject. TheyareoptimalDM
and an astrophysical factor, depending on the target targetsbecauseofnoexpectedastrophysicalradiation
distance and DM distribution, and reads as: and short distance. As such, DSGs have been domi-
¨
natingtheMAGICDMobservationprogramsincethe
P(cid:98) Nγ
Φγ = · ρk dsdω (1) beginning. One can easily see that the first MAGIC
4πk mk
Ω,los DM observations, with the single-telescope, were de-
voted mostly to short (less than 30h) observation of
Annihilating DM:P(cid:98) =(cid:104)σv(cid:105); k =2; DSGs: Draco, Willman 1 and Segue 1. Draco was
consideredoneofthebestcandidateatthattime, be-
Decaying DM:P(cid:98) =τ−1; k =1;
cause of its large mass-to-light ratio, and precise es-
where (cid:104)σv(cid:105) is the velocity average DM annihilation timation of the stellar motion thanks to a sample of
rate and τ is DM particle lifetime, Nγ is the number thousandsofstarmembers. Itwasobservedforabout
of photons promptly produced during an annihilation 8h. In 2007, the Sloan Digital Sky Survey started to
or decay event, and the integrals in the astrophysical produce high-precision photometric data on a newly
factor run over the angular extension of the searched discovered class of DSG, the so-called “ultra-faint”,
region Ω, and along the line of sight. The term ρ duetothefactthatthenumberofmemberstarswere
represents the DM density. afactorof10lessthanthepreviouslydiscovered“clas-
sical”DSG(includingDraco). Twoultra-fainttargets
were observed: Willman 1 and Segue 1. The latter is
II. EARLY SEARCHES an ultra-faint DSG, largely debated in the literature.
Itcurrentlycomprisesabout70memberstars. There-
Before the launch of the Fermi-LAT gamma- sultoftheJ-factorestimationbasedonJeansanalysis
ray satellite-borne instrument (at the time called goes from ranking Segue 1 as the best candidate [24]
GLAST), back in 2004, there were great expectations to an unreliable candidate [25]. MAGIC is recently
about both ground-based and satellite-borne gamma- computing the J-factors using the public CLUMPY
ray instruments to detect gamma rays from DM an- code [26] and the conclusions is that Segue 1 is still
nihilation in a reasonable observation time, at least oneofthecandidatewiththelargestJ-factor. MAGIC
in some optimistic scenarios (see, e.g., [8]). Table I devotedagain alimitedobservation timetothesetwo
reports the full list of MAGIC DM targets from the targets: about 16h for Willman 1 and about 30h for
time of this early estimation until present times. Ob- theSegue1. Asaresult,weputupperlimitsatalevel
servations are ordered by telescopes setup, and year of (cid:104)σv(cid:105)=10−22cm3 s−1. The collection of some of the
and the table provides information about the source discussed MAGIC upper limits is shown in Fig. 1.
class, the observation time, whether results were dis- Luminous structures like star or gas clouds are be-
cussed in terms of annihilating or decaying DM sce- lieved to form by gravitational contraction onto pri-
narios,aswellaslinkstoreferences. However,itisnot mordial DM overdensities. In this sense, “light traces
straightforwardtodrawconclusionsbasedonlyonthe matter”, however, several cases were discussed where
observation time devoted to specific targets in the ta- DM overdensities could be almost completely dark
ble. This is due to the fact that MAGIC performance at all wavelengths but in gamma-rays, due to negli-
eConf C16-09-04.3
XXV European Cosmic Ray Symposium, Turin, Sept. 4-9 2016 3
MAGIC Class Target Year Obs. Time Ann. Decay Ref. Comments
Mono MW Galactic Center 2006/07 25 - - [33]
DSG Draco 2007 7.8 X - [19]
Willman 1 2008 15.5 X - [20]
Segue 1 2008/09 29.4 X - [21]
Unid 3EG1835 2007 25 X - [11]
GC Perseus 2008 24.4 - - [15]
CR All-electrons 2009/10 14 - - [17]
Stereo Unid Many 2009/12 71.3 F - [31] Paper in prep.
DSG Segue 1 2010/13 158 X X [34]
GC Perseus 2009/14 253 F F [30] Paper in prep.
CR All-electrons 2012/14 40 - - [18]
Positrons F - [22] Paper in prep.
MW Galactic Center 2012/16 67 F F [32] Paper in prep.
TABLEI. CompoundofobservationaltargetsforindirectdarkmattersearcheswithMAGIC.Observationsaregrouped
in two blocks for MAGIC results when in single-telescope (mono) configuration and as telescope-pair (stereo). Classes
are: MW(MilkyWay),DSG(DwarfSatelliteGalaxy),Unid(UnidentifiedHEsource),GC(GalaxyCluster),CR(Cosmic
Ray). Theobservationtimeisgiveninhours. FortheDMmodels,“X”meansthereferenceprovidesconstraintsonthat
model, “-” that the reference does not discuss that model, “F” means that constraints about that model are foreseen.
gible stellar activity. This was the case that moti- in all, robust predictions in this case are very hard
vated our search for Intermediate Mass Black Holes to achieve. Less problematic is the case of the de-
(IMBHs), following [10]. It was at the time believed caying DM in galaxy clusters, because of the linear
that IMBHs in the range 102 −106 M(cid:12) could have dependence on the DM density (see Eq. 1). MAGIC
formed at the place of DM overdensities and have in- preliminary results in single-telescope were presented
creased the DM concentration (and hence the annihi- in [15]. In terms of annihilating DM, the constraints
lation rate) due to the gravitational pull not strongly were weak. In Sec. III, we will update this considera-
affectedbythepossibledilutingeffectofstellaractiv- tionwiththevery-largePerseuscampaignsperformed
ity. Such objects, yet undiscovered, could have been with the stereoscopic MAGIC.
bright in gamma-rays and dark in other wavelengths.
Possible candidates were the (back then) unidentified
MAGIC is also an instrument capable of measur-
EGRET sources (EGRET is a precursor of Fermi-
ing cosmic ray particles [6]. Cosmic electrons, con-
LAT). The source 3EG1835 was, at that time, the
stituting a few percents of the total cosmic ray flux
brightest of the unidentified EGRET sources. Unfor-
arriving at the Earth, initiate EAS totally similar to
tunately, the MAGIC data taken on that target were
gamma-ray induced ones. The trick to separate them
affectedbytelescopetechnicalproblemthatprevented
is to consider only sky regions were no gamma-rays
a journal publication. The results were discussed in
aredetected,andconsiderascontrolbackgroundthat
conference [11].
obtained with MC simulations. Clearly the analysis
Moving to non-galactic targets, galaxy clusters are is complex because of: a) of the very precise con-
expected to host enormous amount of DM. Consider- trol of the MC-instrument matching required, and b)
ingthat80%ofthetotalmasscontentoftheUniverse of the fact that e-induced showers have an isotropic
isintheformofDM,andconsideringthetotalmassof origin, and images are more complex to reconstruct.
agalaxyclusters,forexamplethePerseusclustermay The interest stems from the fact that an excess in
hostsomethinglike1014−15M(cid:12)inDM.Answeringthe the positron spectrum was detected and confirmed in
question whether galaxy clusters are optimal target the past years by several instruments [16, and refer-
forDMsearchesisnottrivial,becausetherearemany ences therein]. This discrepancy can be explained by
processesatwork. Ononeside,thathugeDMcontent standard astrophysical mechanisms like the emission
may hint to large concentrations at the barycenter, from local pulsar(s), or by secondary electrons pro-
however, the same central region maybe affected by ducedinDMdecayorannihilationproducts. MAGIC
strongoutwardwindsofbaryonicmatterduetostan- can contribute significantly in an energy range hardly
dard astrophysical activity (GRBs, supernova, etc) at reach of satellite instruments such as AMS-II or
which may counteract the gravitational pressure and Fermi-LAT. MAGIC reported some preliminary re-
reduce the central DM content. On the other side, sults at conferences [17]. The follow-up study was
in the case of annihilating DM, the contribution of performed in stereoscopic mode, however, our data
the DM substructures could be extremely high, with are strongly dominated by systematic uncertainties,
authors claiming a factor of 10 to 1000 higher flux whichmaketheall-electronspectrumonlymoderately
with respect to the “smooth-halo” case [12–14]. All informative [18].
eConf C16-09-04.3
4 XXV European Cosmic Ray Symposium, Turin, Sept. 4-9 2016
III. RECENT RESULTS than with DSG, however, for the decaying DM case,
Perseusisexpectedtodeliverthestrongestlowerlim-
Figure 1 show the collection of some DM related its in decay lifetime for DM particles above few hun-
MAGIC results and their evolution. With the advent dreds GeV [30]. The full paper is in preparation.
ofthesecondtelescope,thestrategytohuntDMwith TheGalacticCenterwasobservedbyMAGIC-mono
MAGIC improved because: for 25 h in 2006 [33], however, the paper focused only
on the astrophysical interpretation. A larger cam-
1. As mentioned in Sec. II, besides the obvious
paign was made with MAGIC-stereo in year 2012-
improvement from mono to stereo, the perfor-
16[32]. TheGCisobservableonlyatlargezenithan-
manceoftheinstrumentreceivedastrongboost
gles from MAGIC. This has a double effect of largely
along the years by instrumental upgrades;
increasing the energy threshold (because of stronger
2. It was clear that substantial effort in observa- extinction of the Cherenkov light in the atmosphere),
tion time should be devoted in order to gain in- but at the same time increasing the effective area at
teresting results. From Table I one can see that high energies (due to a larger footprint of Cherenkov
hundredsofhoursweredevotedtosometargets; photons at the ground). More than 70 h were col-
3. The data reconstruction and analysis was op- lected. The GC presents difficulties connected to the
timized in several ways: a) using a full likeli- presence of one or more bright and extended astro-
hood approach that takes into account the DM physicaltargetsatitscenter,aswellasdiffusegamma-
spectral features, the instrument response func- ray emission. For this reason, we expect MAGIC re-
tion, the uncertainties in the background and sults to be competitive. The DM related paper is
the J-factor model, all resulting in a perfor- under preparation.
mance boost of a factor of 2 as well as preciser Furthermore,about50hweredevotedtothesearch
results [23], b) enlarging the search region to a ofunidentifiedHEtargets,asdescribedinSec.II.This
larger range of DM masses, c) providing model- time,theFermi-LATall-skycatalogsweresearchedfor
independent results for pure annihilation/decay stable,unassociatedsourcesoffthegalacticplane,op-
channels instead of benchmarks models (that timalcandidatestobeDMoverdensities. Resultswere
werebasedonLHCsearchesandlostimportance shown in conferences [31]. No detection was found.
as long as LHC was ruling them out) The paper is in preparation too.
As a result, the MAGIC exclusion curves have
grown substantially better. MAGIC devoted 160 h
to the best DSG candidate (at that time): Segue 1. IV. DISCUSSION AND OUTLOOK
The Segue 1 stereo paper [34] provided the absolute
strongest upper limits from DSG for DM particles TheprevioustwosectionshaveshownthatMAGIC
above few hundreds GeV (at lower DM masses, the devoted a substantial effort toward DM searches,
results from Fermi-LAT are the most constraining). which has increase from first “snapshot” observation
TheseMAGICresultsmadeintothePDG’sReviewof to long-term observation campaigns. The capability
Particle Physics [27]. In addition, MAGIC data were ofextractingrobustandoptimizedresultshasalsoin-
usedforthefirsttimeincombinationwithFermi-LAT creased, specially by using a tailored full likelihood
data to provide stronger constraints [28]. It is impor- method as well as combining results with other in-
tant to stress that any further DSG observed with struments. However,Fig.1showsthatMAGICupper
eitherFermi-LATorMAGICcanbesimplycombined limitsarestillsometwoordersofmagnitudeabovethe
with previous observations thus resulting in a global thermal relic annihilation rate.
evolution toward stronger limits. One could wonder whether we are too far from de-
With the successful multi-year campaign on the tection. The opinion of the author is that the plot
Perseus Galaxy Cluster, originally motivated also by does not bear this information, due to the following
cosmic ray astrophysics, MAGIC collected a total of reasons: a) there are several mechanisms for which
morethan250h. Theseallowedtoprovideverystrong a DM particle satisfying all known constraints (e.g.
constrainonthePerseuscoredynamics, andthermal- theSommerfeldeffect, whichisexpectedbecauseDM
to-nonthermal radiation balance [29]. The DM ex- is cold) can have (cid:104)σv(cid:105)larger than the thermal value,
pectations in Perseus were computed by several au- b) the presence of more than one DM particle would
thors [12–14] and disagree of a factor of 100 one an- translate into a higher relic annihilation rate that
other due to different computation of the additional what computed considering only 1 thermal relic, c)
boost due to substructures of the DM halo, as dis- MAGIC and the other IACTA are the most sensi-
cussed in Sec. II. For MAGIC, computing limits from tive instruments at a TeV DM mass range, with no
Perseus is hindered by the presence of a bright source other competitors in the field, and as such, regardless
at the center, the radio galaxy NGC1275, and there- of the true nature of DM (to be discovered!), these
fore an optimized analysis was performed. The anni- results constitute long-lasting unique information, d)
hilating DM case can be constrained less effectively evennullresultsfrompresentandfuturedirectdetec-
eConf C16-09-04.3
XXV European Cosmic Ray Symposium, Turin, Sept. 4-9 2016 5
several years) with an astrophysical factor 10 times
largerthanofSegue1(asinRef.[28])2. Thesecurves
would be then very close to the thermal annihilation
rate curve, even crossing it in the hard τ+τ− annihi-
lation channel. Considering that such a target may
exist, this is definitely a motivating point for MAGIC
to maintain this effort.
Consequently, MAGIC will continue pursuing DM
searches. The most promising targets remain the
DSGs,becauseoftheircleanenvironment,andstrong
expectedsignal,which-ifdetected-wouldconstitute
a clear detection compared to targets with strong as-
trophysical background. This is also justified by the
factthatmanynewDSGshavebeendiscoveredinthe
recent times [36–43], and more are expected in future
campaigns. Some of these are also modeled with ex-
tremelylargeDMcontent(expectedfromsimulations,
see Fig.3 of [44]). All future MAGIC DSGs observa-
tion could be combined together as done in [28].
FIG.1. CollectionofupperlimitsobtainedwithMAGIC.
With Draco and Willman1 we used benchmark models. V. CONCLUSIONS
For Segue 1 we report results in the case of annihilating
In this contribution, we have shown the evolution
DMfortheb¯bandτ+τ−channelsincaseofMAGIC-mono,
of MAGIC results for dark matter searches at differ-
MAGIC-stereo and an extrapolation of MAGIC perfor-
ent target of interests. The dedication of hundreds of
manceforatarget10timesbrighterthanSegue1observed
hours allowed MAGIC to place the most robust up-
for500h. AdditionalFermi-LAT[45]exclusioncurveand
thethermalreliccross-section[46]areshown. Furtherde- per limits above few hundreds GeV on annihilating
tails and discussion are given in the text. dark matter particle models through the observation
of highly dark matter dominated satellite galaxies.
The long campaign on Perseus will allow to put simi-
tion or accelerator experiments for DM, there is al- larstrongconstraintsonthelifetimeofdecayingdark
ways a portion of the parameter space only accessible matter. MAGIC continues its dark matter program
to TeV instruments [35] e) Fig. 1 also plots an esti- in order to provide legacy results before the times of
mationofwhatexclusioncurveMAGICcouldprovide CTA.
if observing a DSG for 500 h (this can be done over
[1] J. L. Feng, “Dark Matter Candidates from Particle using observations of the Crab Nebula”, Astropart.
Physics and Methods of Detection,” Ann. Rev. As- Phys. 72 (2016)
tron. Astrophys. 48 (2010) 495 [6] M. Doro, “MAGIC. A gamma-ray and particle-
[2] M. Milgrom, “A Modification of the Newtonian dy- physics detector”, Procs. of the 11th Scineghe Con-
namics as a possible alternative to the hidden mass ference, Pisa (2016).
hypothesis,” Astrophys. J. 270 (1983) 365. [7] M. Doro, “A decade of dark matter searches with
[3] P. A. R. Ade et al. [Planck Collaboration], “Planck ground-based Cherenkov telescopes,” Nucl. Instrum.
2015results.XIII.Cosmologicalparameters,”Astron. Meth. A 742 (2014) 99
Astrophys. 594 (2016) A13 [8] L. Bergstrom and D. Hooper, “Dark matter and
[4] J.L.Feng,“NaturalnessandtheStatusofSupersym- gamma-raysfromDraco: MAGIC,GLASTandCAC-
metry,” Ann. Rev. Nucl. Part. Sci. 63 (2013) 351 TUS,” Phys. Rev. D 73 (2006) 063510
[5] MAGIC Collaboration, “The major upgrade of the [9] L.E.Strigarietal.,“Acommonmassscaleforsatellite
MAGIC telescopes, Part II: A performance study galaxies of the Milky Way,” Nature 454 (2008) 1096
[10] G. Bertone et al.,“A new signature of dark matter
annihilations: gamma-rays from intermediate-mass
black holes,” Phys. Rev. D 72 (2005) 103517
2 Thereadershouldbewarnedthatthisisasimplerescalingof [11] M. Doro et al., “Indirect Dark Matter Search at In-
theRef.[28]curve,whichreferstoaspecificdataset(withits termediate Mass Black Holes with the MAGIC Tele-
fluctuations)anddoesnotmakeafullcomputationusingthe scope,” Procs. 30th ICRC (2007)
expected sensitivity, for simplicity. The order of magnitude
oftheresultisstillvalid.
eConf C16-09-04.3
6 XXV European Cosmic Ray Symposium, Turin, Sept. 4-9 2016
[12] M. A. Sanchez-Conde et al.,“Dark matter searches [30] J.Palacioetal.,“ConstrainingtheDarkMatterdecay
with Cherenkov telescopes: nearby dwarf galaxies or lifetime with very deep observations of the Perseus
local galaxy clusters?,” JCAP 1112 (2011) 011 clusterwiththeMAGICtelescopes,”PoSICRC2015
[13] A. Pinzke et al.,“Prospects of detecting gamma-ray (2016) 1204
emission from galaxy clusters: cosmic rays and dark [31] K. Satalecka et al.,“Observations of hard spectrum
matterannihilations,”Phys.Rev.D84(2011)123509 Unassociated Fermi Objects with MAGIC,” PoS
[14] L. Gao, C. S. Frenk, A. Jenkins, V. Springel and ICRC 2015 (2016) 724.
S. D. M. White, “Where will supersymmetric dark [32] M. L. Ahnen et al., “Observations of Sagittarius A*
matter first be seen?,” Mon. Not. Roy. Astron. Soc. during the pericenter passage of the G2 object with
419 (2012) 1721 MAGIC,” Accepted to A&A. arXiv:1611.07095
[15] MAGIC Coll., “MAGIC Gamma-Ray Telescope Ob- [33] MAGIC Coll., “Observation of gamma-rays from the
servationofthePerseusClusterofGalaxies: Implica- galacticcenterwiththemagictelescope,”Astrophys.
tionsforCosmicRays,DarkMatterandNGC1275,” J. 638 (2006) L101
Astrophys. J. 710 (2010) 634 [34] MAGIC Coll., “Optimized dark matter searches in
[16] AMS Coll., “First Result from the Alpha Magnetic deep observations of Segue 1 with MAGIC,” JCAP
Spectrometer on the International Space Station: 1402 (2014) 008
Precision Measurement of the Positron Fraction in [35] M. Cahill-Rowley, “Complementarity between col-
Primary Cosmic Rays of 0.5350 GeV,” Phys. Rev. lider, direct detection, and indirect detection experi-
Lett. 110 (2013) 141102. ments,” PoS Scineghe 2014 (2015) 027
[17] D. Borla Tridon et al., “Measurement of the cos- [36] K. Bechtol et al. [DES Collaboration], “Eight New
mic electron spectrum with the MAGIC telescopes,” Milky Way Companions Discovered in First-Year
Procs. ICRC (2011) DarkEnergySurveyData,”Astrophys.J.807(2015)
[18] K.Mallotetal.,”TheHighEnergyElectronPositron no.1, 50
Spectrum measured by the MAGIC Telescopes” [37] S. E. Koposov et al.,“Beasts of the Southern Wild:
TeVPa Conference, Tokyo, (2015) DiscoveryofnineUltraFaintsatellitesinthevicinity
[19] MAGIC Coll., “Upper limit for gamma-ray emis- of the Magellanic Clouds,” Astrophys. J. 805 (2015)
sionabove140-GeVfromthedwarfspheroidalgalaxy no.2, 130
Draco,” Astrophys. J. 679 (2008) 428 [38] B. P. M. Laevens et al., “A New Faint Milky Way
[20] MAGICColl.,“UpperlimitsontheVHEgamma-ray Satellite Discovered in the Pan-STARRS1 3 pi Sur-
emissionfromtheWillman1satellitegalaxywiththe vey,” Astrophys. J. 802 (2015) L18
MAGIC Telescope,” Astrophys. J. 697 (2009) 1299 [39] B. P. M. Laevens et al., “Sagittarius II, Draco II and
[21] MAGICColl.,“SearchesforDarkMatterannihilation Laevens3: ThreenewMilkywaySatellitesDiscovered
signatures in the Segue 1 satellite galaxy with the in the Pan-starrs 1 3π Survey,” Astrophys. J. 813
MAGIC-I telescope,” JCAP 1106 (2011) 035 (2015) no.1, 44
[22] P. Colin et al., “Probing the CR positron/electron [40] D. Kim et al.,“A Heros Dark Horse: Discovery of an
ratioatfewhundredsGeVthroughMoonshadowob- Ultra-faint Milky way Satellite in Pegasus,” Astro-
servation with the MAGIC telescopes,” Procs. ICRC phys. J. 804 (2015) no.2, L44
(2011) [41] N. F. Martin et al., “Hydra II: a faint and compact
[23] J. Aleksic et al.,“Optimized analysis method for Milky Way dwarf galaxy found in the Survey of the
indirect dark matter searches with Imaging Air MagellanicStellarHistory,”Astrophys.J.804(2015)
Cherenkov Telescopes,” JCAP 1210 (2012) 032 no.1, L5
[24] R.Essigetal.,“IndirectDarkMatterDetectionLimits [42] D. Kim and H. Jerjen, “Horologium II: a Second
from the Ultra-Faint Milky Way Satellite Segue 1,” Ultra-faint Milky Way Satellite in the Horologium
Phys. Rev. D 82 (2010) 123503 Constellation,” Astrophys. J. 808 (2015) L39
[25] V. Bonnivard et al., “Dark matter annihilation and [43] A. Drlica-Wagner et al. [DES Collaboration], “Eight
decay in dwarf spheroidal galaxies: The classical and Ultra-faint Galaxy Candidates Discovered in Year
ultrafaint dSphs,” Mon. Not. Roy. Astron. Soc. 453 Two of the Dark Energy Survey,” Astrophys. J. 813
(2015) no.1, 849 (2015) no.2, 109
[26] V. Bonnivard et al., “CLUMPY : Jeans analysis, - [44] M. Hu¨tten et al., “Dark matter substructure mod-
ray and fluxes from dark matter (sub-)structures,” elling and sensitivity of the Cherenkov Telescope Ar-
Comput. Phys. Commun. 200 (2016) 336 raytoGalacticdarkhalos,”JCAP1609(2016)no.09,
[27] C. Patrignani et al. (Particle Data Group), Chin. 047
Phys. C, 40 (2016) [45] Fermi-LAT Coll., “Searching for Dark Matter Anni-
[28] MAGICandFermi-LATColl.,“Limitstodarkmatter hilation from Milky Way Dwarf Spheroidal Galaxies
annihilationcross-sectionfromacombinedanalysisof withSixYearsofFermiLargeAreaTelescopeData,”
MAGIC and Fermi-LAT observations of dwarf satel- Phys. Rev. Lett. 115 (2015) no.23, 231301
lite galaxies,” JCAP 1602 (2016) no.02, 039 [46] G.Steigmanet al.,“PreciseRelicWIMPAbundance
[29] MAGICColl.,“DeepobservationoftheNGC1275re- and its Impact on Searches for Dark Matter Annihi-
gion with MAGIC: search of diffuse γ-ray emission lation,” Phys. Rev. D 86 (2012) 023506
fromcosmicraysinthePerseuscluster,”Astron.As-
trophys. 589 (2016) A33
eConf C16-09-04.3