ARGO-YBJ: Highlights and Prospects

Transcription

1 ARGO-YBJ: Highlights and Prospects G. Di Sciascio on behalf of the ARGO-YBJ Collaboration INFN Sezione Roma Tor Vergata Frascati Workshop 2013 Multifrequency Behaviour of High Energy Cosmic Sources Palermo May 27-June 1,

2 Questions to the knee energy range Cosmic Ray Sources? Still open Overlap direct indirect measurements? Composition at the knee? Hadronic interaction models? End of galactic spectrum? Transition galactic xgalactic? Anisotropy? Rigidity dependent knee? Still missing Still open Still uncertain Open Totally open Probably established A. Haungs, 2011! 2

4 The ARGO-YBJ experiment Longitude East Latitude North 90 Km North from Lhasa (Tibet) 4300 m above the sea level ~ 600 g/cm 2 The Yangbajing Cosmic Ray Laboratory ARGO Tibet ASγ 4

5 The basic concepts for an unconventional air shower detector HIGH ALTITUDE SITE (YBJ - Tibet 4300 m asl g/cm2) FULL COVERAGE (RPC technology, 92% covering factor) HIGH SEGMENTATION OF THE READOUT (small space-time pixels) Number'of'Fired'Strips1 Real%event%) Space pixels: 146,880 strips (7 62 cm 2 ) Time pixels: 18,360 pads (56 62 cm 2 ) in order to image the shower front with unprecedented details get an energy threshold of a few hundreds of GeV G. Di Sciascio. Frascati Workshop Palermo May 29, 201 5

6 The basic concepts extending the dynamical range up to PeV ANALOG READOUT PeV ( m 2 big pads ) (digital) Strips (digital) (analog) The RPC near the core. analog Well fitted BigPads by (analog) modified readout NKG function s' 2 s' 4.5 r (analog) r Strips Extend the covered energy range (digital) Access the LDF down to the shower core Sensitivity to primary mass Info/checks on Hadronic Interactions Strips RICAP Strips (digital) Lateral Distribution Function Big Pads BigPads With the analog data we can study the LDF without saturation NKG A r M 1 r M RICAP The LDF slope s is related to the shower age. Real event I. De Mitri: Cosmic Ray P 500 I. De Mitri: Cosmic Ray RICAP G. Di Sciascio. Frascati Workshop Palermo May 29, 201 I. De Mitri: Cosmic Ray Physics with ARGO-YBJ 19 /

7 The ARGO-YBJ Collaboration Collaboration Institutions: Chinese Academy of Sciences (CAS) Istituto Nazionale di Fisica Nucleare (INFN) INAF/IASF, Palermo and INFN, Catania INFN and Dpt. di Fisica Università, Lecce INFN and Dpt. di Fisica Universita, Napoli INFN and Dpt. di Fisica Universita, Pavia INFN and Dpt di Fisica Università Roma Tre, Roma INFN and Dpt. di Fisica Univesità Tor Vergata, Roma INAF/IFSI and INFN, Torino IHEP, Beijing Shandong University, Jinan South West Jiaotong University, Chengdu Tibet University, Lhasa Yunnan University, Kunming Hebei Normal University, Shijiazhuang 7

8 Duty-cycle Status and Energy calibration! performance In observation since July 2006 (commissioning phase) Stable data taking since November 2007 End/Stop data taking: January 2013 Average duty cycle ~87% Trigger rate ~ pad threshold N. recorded events: 4 11 from 0 GeV to PeV ARGO-YBJ: perfo 0 TB/year data Intrinsic Trigger Rate stability 0.5% Event Rate ~ 3.5 khz for N hit >20 - Duty cycle ~ 86% - 1 (after corrections for T/p effects) G. Di Sciascio. Frascati Workshop Palermo May 29, 201 Days 8 Days High space/time granularity

10 Why gamma-ray extended sources? TeV gamma-ray extended sources an important tool to investigate the sources of cosmic rays. The observed degree-scale extended emission could be produced by high-energy cosmic rays escaping from the source and diffusing in the interstellar medium. The gamma-ray emission should result from the interaction of these cosmic rays with the ISM particles. 80% of TeV galactic gamma ray sources are extended. Many of them are still unidentified. To study degree-scale sources we need instruments with a large field of view and able to correctly evaluate the cosmic ray background over a large solid angle Sensitivity to an extended source is relatively better for an EAS than an IACT because angular resolution is not as important S S extended point σ σ source detector

11 ta Gamma-Ray Astronomy with ARGO-YBJ Crab Nebula 5 years data Energy threshold: few hundreds of GeV Overlaps with Cherenkov detectors Large duty cycle: 86% Large field of view: ~2 sr Declination band from - to 70 Integrated sensitivity in 5 y at ~1 TeV: 0.25 Crab for dec Crab Nebula 5 years da Crab Nebula 5 years data 1 dn de ( stat ) 11 E 1TeV ( stat ) cm 2 s HST 1 TeV 1 (0.5 ) TeV (0.5 ) TeV 11 ( stat )

12 Sky Survey at ~ 1 TeV ARGO-YBJ Sky Survey at 1 TeV Cygnus region Mrk501 Mrk421 CRAB Nebula MGRO J HESS J Final results and detail about possible new TeV sources next ICRC! 12

13 Observation of extended sources with ARGO-YBJ ApJL 745 (2012) L22 HESS J ! MGRO J ! ApJ 767 (2013) 99 ApJ 760 (2012) 1 MGRO J ! FERMI Cocoon spectrum Fermi ARGO-YBJ Milagro G. Di Sciascio. Frascati Workshop Palermo May 29,

14 Comments on HESS J extended sources CRAB MGRO J MGRO J HESS J point source extended extended extended flux agrees with IACTs flux ~ X IACTs flux ~ 4 X IACTs flux ~ 3 X IACTs Systematic disagreement for extended sources! ARGO-YBJ (and MILAGRO) measure higher fluxes ARGO-YBJ Coll., ApJ 767 (2013) 99 Possible systematics in ARGO-YBJ CR background evaluation: checked with the distribution of the excesses (Gauss with s=1) Pointing accuracy (at 0.1 level checked with the Moon Shadow) Error in energy scale < 13% Contribution from the diffuse emission of the Galactic plane < 15% Overall systematics on the flux < 30% The discrepancy could origin from the different techniques used in the background estimation for extended sources. Maybe the extended excess is due to the contribution of different sources G. Di Sciascio. Frascati Workshop Palermo May 29,

16 ] s -1 sr -1 GeV -2 [m 2.75 E Flux The light-component spectrum ( TeV) Light spectrum Extension ARGO-YBJ of Light the Component previous 2008 spectrum - Full data sample ARGO-YBJ light component ARGO-YBJ 5Light spectrum Full data sample - Full 2008 data sample ] s -1 sr -1 GeV -2 [m 2.75 E Flux ] GeV ] sr -1 s -1 sr -1 GeV s -1-2 [m Flux E Measurement in the low and high of energy the light-component (p+he) CR spectrum in the regions energy ARGO-YBJ 5 ARGO-YBJ region Light (2.5 Light Component Component 300) TeV spectrum via spectrum a Bayesian - - Full data sample unfolding Full data procedure sample [m 2.75 The light component spectrum E Flux PAMELA (P) AMS (P) CREAM (P) PAMELA (He) AMS (He) CREAM (He) Energy [GeV] PAMELA (P) PAMELA (He) AMS (P) AMS (He) CREAM (Light) ARGO-YBJ (2012) ARGO-YBJ (2013) Energy [GeV] Direct and ground-based measurements overlap for a wide energy range thus making possible the cross-calibration of the experiments CREAM (P) CREAM (He) Direct and ground-based measurements overlap for a wide energy range thus making possible the crosscalibration of the 3 experiments. PRD 85, (2012) spectrum measurement both = 2.61 ± 0.02 ballons/satellites The spectrum spans a huge energy region: TeV ARGO-YBJ (Light) ARGO-YBJ Light 2013 ARGO-YBJ PRD Energy [GeV] ground-based exp

17 The knee region It is easy to re-implement the idea thinking of the 1961 paper of Bernard Peters stating that both cosmic ray acceleration and propagation in the Galaxy have to be discussed in terms of rigidity (R = p/z). If a proton can be accelerated up to energy E_max then a nucleus of charge Z could achieve Z times higher energy. We did use the Peters cycle trying to fit the shifted air shower spectra. There was no restriction on the number of populations of cosmic rays (presumably due to different types of sources) in the fit. The fitting procedure came up with four population where the fourth one describes the extragalactic cosmic rays. It is highly uncertain because the differences in the UHECR composition derived by HiRes (and TA) and Auger. T. Stanev Ricap

18 The Cosmic Ray Spectrum in the PeV range ight Component Light Component spectrum - Full Digital data + sample Analog 2008 readout Towards the highest energies and the knee region with the analog readout RGO-YBJ 5 Light Component spectrum - Full data sample ARGO-YBJ: Cosmic Ray Energy Spectrum s -1 sr -1 GeV ] Preliminary! 3 0 [m 2.75 Flux E 3 PAMELA (P) PAMELA (He) AMS (P) AMS (He) CREAM (P) CREAM (He) ARGO-YBJ (Light) ARGO-YBJ Light 2013 ARGO-YBJ PRD 2012 ARGO-YBJ ANALOG 2V ARGO-YBJ ANALOG 20V Energy [GeV] Energy [GeV] 18

19 Cherenkov event The light-component spectrum (0-800 TeV) Hybrid measurement of the light-component (p+he) CR spectrum in the energy region (0 800) TeV Contamination of heavier component < 5 % Energy resolution: ~25% Uncertainty : ~25% on flux Spectra index: γ= ± 0.06 (ARGO: ± 0.02) PRELIMINARY* WFCTA-ARGO Digital, Analog, Hybrid cross-calibrations crucial G. Di Sciascio. Frascati Workshop Palermo May 29,

21 The total p-p cross section The total p-p cross section ARGO-YBJ data in the Review of Particle Properties 2012 Measurement of p-air cross section Use the shower frequency vs (sec -1) I( ) I(0) e h o sec 1 h 0 for fixed energy and shower age. The lenght is connected to the p interaction lenght by the ralation = k int RICAP where k is determined by simulations and G. Di Sciascio. Frascati Workshop 2013 depends - Palermo on: May 29, 2013 ARGO-YBJ Coll., Phys. Rev D 80, (2009) I. De Mitri: Cosmic Ray Physics with ARGO-YBJ 15 / hadronic interactions Constrain X DM = X det X max

22 Shower age and primary mass Np8 (number of particles within 8 m from the core): well correlated with primary energy not biased by finite detector size effects weakly affected by shower fluctuations I. Shower age vs truncated size Shower age vs truncated size RICAP De Mitri: Cosmic Ray Physics with ARGO-YBJ Preliminary PRELIMINARY Preliminary! The s parameter is correlated to the The s parameter is correlated to the X max position, whatever the primary is. The s parameter is correlated to the X Xmax position, whatever the primary is. max position, whatever the primary is. Possibility to get hints on (a) shower age and (b) primary mass Possibility to get hints on (a) shower age and (b) primary mass Possibility to get hints on (a) shower age and (b) primary mass G. Di Sciascio. Frascati Workshop Palermo May 29, RICAP RICAP I. De Mitri: Cosmic Ray Physics with ARGO-YBJ I. De Mitri: Cosmic Ray Physics with ARGO-YBJ 20 / / 21

24 Cosmic Ray Isotropy CRs below 17 ev are predominantly galactic. The bulk of CR is produced by shock acceleration in SN explosions. Diffusion of accelerated CRs through non-uniform, non-homogeneous ISM. At 1 TeV, B ~ 1 μg, Gyro-Radius ~ 200AU, 0.001pc Galactic CRs are expected to be highly isotropic scrambled by galactic magnetic field over very long time. 24

25 Cosmic Ray an-isotropy in arrival direction? The question of a possible variation of the CR flux with time arose soon after its discovery. The earliest evidence of a possible periodicity in the CR intensity was reported in 1932 by Hess and Steinmaurer This observation, interpreted by Compton and Getting as an effect of the Earth motion relative to the sources of CRs, stimulated many measurements to verify this model. Although different ground-based experiments reported evidence of a diurnal variation of galactic CRs long before the 50s, the existence of the anisotropy remained uncertain due to solar modulation geomagnetic field low event rate non uniform time coverage atmospheric effects: the amplitude of the temporal variation of the EAS trigger rate (a few percent), mostly due to the atmospheric pressure and temperature effects, is much larger than the expected amplitude of the CR anisotropy G. Di Sciascio. Frascati Workshop Palermo May 29,

26 Measuring the anisotropy Only during 50s, when large detectors were operated monitoring the atmospheric variations, a clear evidence of the existence of the galactic anisotropy was obtained. In 1998 Nagashima, Fujimoto, and Jacklyn reported the first comprehensive observation of a large angular scale anisotropy in the sub-tev CRs arrival direction by combining data from different experiments in the northern and southern hemispheres. The earliest map of the large scale anisotropy Figure 6: The distribution of the tail-in and loss-cone (galactic) anisotropies the equatorial coordinate grid. The excess cone of the tail-in anisotropy, wit which all the directional excess flux is confined, is shown bounded by the thick l including the south pole. A thinner line bounds the deficit cone of the galac anisotropy, in which all the directional flux is confined. PM shows the direct of the proper motion of the solar system. RM shows the direction of the relat motion of the 65 system N to the neutral gas. The galactic equator is represented by thin line labelled by b =0. Tail-in feature directed towards the heliospheric tail peak located at RA ~ 6h (~90 ). 36 N Amplitude and phase change with latitude North-South asymmetry 5 S Tail-in modulated in time: max in Dec. and min in June 43 S G. Di Sciascio. Frascati Workshop Palermo May 29, 201 by Nagashima 1998! 26

27 Most likely m diffusive shoc of protons an fronts associ remnants Source spect In principle any anisotropy reflects a motion: Most likely mechanism: Q(E) α E -2.1:-2 diffusive shock-acceleration Observed po (1) the motion of the Earth/Solar System with respect to the isotropic CRs rest of protons frame and ions spectrum at shockrep (the Compton-Getting effect); fronts associated with propagation SNRs e (2) the motion of CRs from galactic sources to the extragalactic space and/or remnants from extragalactic sources to the Earth. Courtesy by W.Hoffmann Maximum rea Source spectrum: 15 Z ev (Z= Q(E) α E The biggest problem in the quest to find Piera the L. Ghia origin of CRs is the presence CR anisotropies of -2.1:-2.3 magnetic field in the Interstellar Medium (ISM). Observed power-law CR spectrum reproduced when The detection of CRs provides directional information only up to distances propagation as large as effects included their gyro-radius Courtesy by W.Hoffmann Maximum reachable energy E < 15 ev, local galactic magnetic field (GMF) 3μG gyro-radii very 15 Z short: ev (Z=particle 1 pc charge) Cosmic Ray Anisotropy in arrival direction CR production in SNR Piera L. Ghia CR anisotropies 11 isotropy is expected, as no structures of the GMF are known to focus CRs within such horizon. At most, a weak dipolar distribution may exist, reflecting the contribution of the closest CR sources. The possibility that the anisotropy is related to the distribution of CR sources makes its study especially interesting. G. Di Sciascio. Frascati Workshop Palermo May 29,

28 2-dim Maps of CR Intensity The improvement in the measurement of the CR arrival direction, recently allowed the construction of 2-dim maps which confirmed the complicated structure of the CR sky and its analysis on different angular scales. Icecube - 59! 34 9 events" Angular resolution ~ 3 " Median energy ~20 TeV" 28

29 Large scale anisotropy by ARGO-YBJ 2 years data: , E 1 TeV, 3.6 events Tail-in excess region! Loss-cone deficit region! Cygnus region! 29

30 Anisotropy vs energy 0.9 TeV! 3-4 pc 70 AU 1.5 TeV! 2.4 TeV! 3.6 TeV! First measurement with an EAS array in an energy region so far investigated only by underground muon detectors 7.2 TeV! 12.5 TeV! The tail-in broad structure appears to dissolve to smaller angular scale spots 23.6 TeV! 8-3 pc 1900 AU G. Di Sciascio. Frascati Workshop Palermo May 29,

31 Amplitude and phase of the first harmonic ARGO-YBJ results in good agrrement with other experiments. Analysis with the full statistics under way ARGO-YBJ 31 Fig. 3. The Amplitude(A) and the phase(b) of first harmonics of the sidereal daily vari-

32 x-check: Compton-Getting effect Expected CR anisotropy due to Earth s orbital motion around the Sun: when an observer (CR detector) moves through a gas which is isotropic in the rest frame (CR gas ), he sees a current of particles from the direction opposite to that of its own motion A benchmark for the reliability of the detector and the analysis method. In fact, all the features (period, amplitude and phase) of the signal are predictable without uncertainty, due to the exquisitely kinetic nature of the effect. I = CR intensity γ = power-law index of CR spectrum (2.7) v = detector velocity 30 km/s!"!!" θ = angle between detector motion and CR arrival direction #$!"! A detector on the Earth moving around the Sun scans various directions in space while the Earth spins. Maximum at 6 hr solar time (when the detector is sensitive to a direction parallel to the Earth s orbit) galactic cosmic ray anisotropy - Paolo De The first clear observation of the SCG effect with an EAS array was reported by EAS-TOP (LNGS) in 1996 at about 14 ev. 32

33 Compton-Getting effect by ARGO-YBJ PRELIMINARY Solare Time (UT) data Nhit TeV to avoid solar effects on low energy CRs Evidence for an additional new anisotropy component at lower energy (solar effects?) under study 33

34 medium / small scale angular scale in cosmic ray medium/small scale anisotropy anisotropy arrival direction loss-cone region! Abdo A.A. et al., 2008, Phys. Rev. Lett., 1, 2211 tail-in excess region! 2 hr = 30 Milagro events median CR energy ~ 1 TeV = 12 ev average angular resolution < 1o 5 Particle acceleration in reconnection regions - Paolo Desiati 2hr time window smoothing 2 hr = 30 Abdo et al., Phys.Rev.Lett.1 (2008) 2211 filter all angular features > 30 technique used in γ-ray searches filter all angular features > angular 30 scale anisotropy region contained inside a larger one relies on The observation of a possible small the capability for suppressing the anisotropic structures at larger scales without, simultaneously, introducing eﬀects of the analysis scales. technique usedoninsmaller gamma ray searches R. Iuppa and G. Di Sciascio, ApJ 766 (2013) 96 6 Particle acceleration in reconnection regions - Paolo Desiati 34

35 Medium Scale Anisotropy How to focus on medium scale structures? Traditional background estimation methods: Time swapping/scrambling (3 hrs,) Direct integration (3 hrs) (consistent each other within 0.3 s.d.) An effective high-pass filter for structures narrower than 3 hrs 15 /hrs = 45 in R.A. (35 safety-limit) every feature larger than ΔT is brought to zero (apart from the peak) First systematic study of the time average-based methods G. Di Sciascio. Frascati Workshop Palermo May 29, 201 R. Iuppa and G. Di Sciascio, ApJ 766 (2013) 96 35

36 Medium Scale Cygnus Region Anisotropy CRAB Data: November 8, May 20, events dec. region δ Map smoothed with the detected PSF for CRs Galactic center Galactic plane Proton median energy 1 TeV CRs excess 0.1 % with signifincance up to 15 s.d. 4 Strip-multiplicity number of E 50 p [TeV] interval events (38%) (48%) (%) (3%) 7.3 more than (1%) 20 TABLE I: Multiplicity intervals used in the analysis. The central columns report the number of events collected. The right column shows the corresponding isotropic CR proton median energy. the overall absolute pointing accuracy is 0.1.Theabsolute Di Sciascio pointing. Frascati of the detector Workshop is stable 2013 at - Palermo a level ofmay , 201 G. and the angular resolution is stable at a level of % on mercoledì a monthly 29 maggio basis. 13The absolute rigidity scale uncertainty 36

37 MSA vs energy N = relative intensity The size spectra look quite harder than the CR isotropic flux multiplicity G. Di Sciascio. Frascati Workshop Palermo May 29,

38 Origin/explanation of these hot spots There is currently no explanation for these local enhancements in the CR flux Composition: Not photons or electrons (Milagro) Neutrons from a star? Unlikely TeV neutrons decay in 0.1 pc much closer than the nearest star. Gyro-radius of a TeV proton in a 2μG magnetic field is only ~0.005 pc (00 AU). - Magnetic field must connect us to the source and be coherent out to it ( 0 pc). Tips: Connection to heliosphere? Region 1 coincides with the direction of the heliotail. The direction of both regions is nearly perpendicular to the expected Galactic magnetic field direction Multiple explanations were proposed: Salvati & Sacco, A&A 485 (2008) 527 Drury & Aharonian, Astrop. Phys. 29 (2008) 420. K. Munakata AIP Conf Proc Vol 932, page 283 Salvati, A&A 513 (20) A28 Lazarian & Desiati, ApJ 722 (20) young nearby SN? first look inside the heliotail? 38

39 Beyond ARGO-YBJ: LAWCA project An L-shaped water pool: North-East of ARGO-YBJ hall; 23,000 m2; 4.5 m depth; 900 cells, with an 8 PMT in each cell; Cells are partitioned with black curtains. Sensitivity: TeV; TeV ~x better than ARGO-YBJ 39

40 The LHAASO experiment LHAASO: a 1 km2 multi-component EAS array at 4300 m a.s.l. in Shangri-La, Yunnan province, China, mostly driven by the Chinese community. G. Di Sciascio. Frascati Workshop Palermo May 29, k m2 Water Cerenkov Array 1 km2 Scintillator Array Wide FoV Cerenkov detectors 400 burst detectors 40

41 Conclusions and outlook First Northern sky survey (- < δ < 70 ) at 0.25 Crab Units. Study of the CR anisotropy at different angular scales Detailed study of flaring and extedend TeV gamma-ray sources Measurement of CR light component energy spectrum up to PeV Measurement of the CR antip/p flux ratio in TeV energy range Measurement of the p-air and p-p cross sections up to 0TeV A ground based large and complexγ/cr observatory at high altitude (4300 m a.s.l.) within 5 years Complementary to CTA in γ-astronomy Unique for CR measurements at the knee Useful for exploring new physics All technology that will be used in LHAASO are tested with engineering arrays at scales of 1% -% of the full project in Tibet with ARGO-YBJ The construction of LHAASO is expected to start in a couple of years, LAWCA is on its way this year. 41

42 BACKUP SLIDES 42

43 originally measured as solar diurnal variation of Solar vs Sidereal day Solar vs Sidereal day muon count rate with a seasonal modulation Sidereal day time it takes a star at the meridian to return to the meridian. 23 hours 56 min 4 sec atmospheric daily/seasonal temperature variation Compton-Getting effect due to Earth s motion Solar day time it takes the Sun at meridian (noon) to return to the meridian. noon to noon or 24 hours around the Sun Why the 4-minute difference? 4 as it rotates, the Earth also orbits the Sun. Earth must rotate an extra degree (4 Sidereal day time it takes a star at the meridian to min) each day...for any observer on Earth to be at noon again return to the meridian. 23 hours 56 min 4 sec Solar day time it takes the Sun at meridian (noon) to return to the meridian. noon to noon or 24 hours Why the 4-minute difference? as it rotates, the Earth also orbits the Sun. Earth must rotate an extra degree (4 min) each day for any observer on Earth to be at noon again Piera L. Ghia CR anisotropies 88 G. Di Sciascio. Frascati Workshop Palermo May 29,

44 Heliosphere scale : 0-,000 AU pc solar system moves wrt IS medium at 26 km/s solar wind diverts interstellar plasma at km/s termination solar pressure ~ interstellar pressure : ~ 0 AU ~ AU ~ 5 - TeV solar and interstellar medium (& magnetic field) separated by heliopause : ~ AU heliotail size up to ~,000 AU? 3 µg 0 AU 0 AU > 1,000 -,000 AU ~1.4 TeV ~1.4 TeV ~ TeV G. Di Sciascio. Frascati Workshop Palermo May 29,

45 Time-swapping method Random background events are generated for each observed event by associating the event local coordinates (δ, α) with times selected randomly from all event times recorded over a 3 hr period. New values of celestial coordinates are then calculated for each background event. The new triplet constitues a new, fake, background event. Because the selection of a new event time amounts to a rotation of the celestial sphere with respect to the Earth, this method changes only the α of an event, but not its δ ( δ, α, t) ( δ, α, t! ) event 1 BKG ( δ, α, t! ) BKG To reduce statistical fluctuations fake events are generated for each real events. 45

46 Light Component spectrum - Selection efficiency Selection efficiency with digital readout -1-1 Proton Helium CNO -2-2 Iron Log(E [GeV]) Log(E [GeV]) 46

47 The Sidereal Time distribution: harmonic analysis Finally: the sidereal time distribution The CR arrival distribution in sidereal time was never found to be purely dipolar, First σ level Second 3 σ level The use of 2 harmonics means that the spatial distribution of CR intensity has a rather complicated structure No simple kinetic nature of the anisotropy (CG effect) The origin of the Large Scale Anisotropy is still unknown Piera L. Ghia CR anisotropies 37 The CR plasma is supposed to co-move with the solar system and the origin of the observed anisotropy is probably related to harder effects, to be searched for in unknown features of the local ISM, either for the magnetic field and the closest CR sources. 47

48 Amplitude and Phase of the first harmonic -3-3 Consistency between different sets of data. Slow increase of A with increasing energy to a maximum at few TeV. Slow fall of A to a minimum at about 0 TeV. Evidence of increasing A above 0 TeV. Phase nearly constant (slowly decreasing) around 0 hrs. Clear change (decrease) of phase above 0 TeV. 1-dim approach: Morphological study precluded Analysis on different angular scales precluded G. Di Sciascio. Frascati Workshop Palermo May 29, 201 A [hrs] φ 1 φ A 1-5 [hrs] Norikura1973 Musala1975 Baksan1981 Morello1983 Norikura1989 EasTop1995 EasTop1996 Tibet1999 Tibet2005 Milagro2009 EasTop2009 Baksan2009 Norikura1973 ARGO2011 Musala1975 Utah1981 Baksan1981 Ottawa1981 Morello1983 Holborn1983 Norikura1989 Bolivia1985 EasTop1995 Misato1985 EasTop1996 Budapest1985 Tibet1999 Hobart1985 Tibet2005 Yakutsk1985 Milagro2009 London1985 EasTop2009 Socorro1985 Baksan2009 HongKong1987 ARGO2011 Baksan1987 Utah1981 Artyomovsk1990q Ottawa1981 Sakashita1990 Holborn1983 Utah1991 Bolivia1985 Liapootah1995 Misato1985 Matsushiro1995 Budapest1985 Poatina1995 Hobart1985 Kamiokande1997 Yakutsk1985 Macro2003 London1985 SuperKamiokande2007 Socorro1985 IceCube20 HongKong1987 IceCube2012 Baksan1987 Artyomovsk1990q Sakashita1990 Utah1991 Liapootah1995 Matsushiro1995 Poatina1995 Kamiokande1997 Macro2003 SuperKamiokande2007 IceCube20 IceCube energy [ev] 13 energy (a) [ev] (a) energy [ev] 13 (b) energy [ev] (b) Norikura1973 Musala1975 Baksan1981 Morello1983 Norikura1989 EasTop1995 EasTop1996 Tibet1999 Tibet2005 Milagro2009 EasTop2009 Baksan2009 ARGO2011 Utah1981 Ottawa1981 Holborn1983 Norikura1973 Bolivia1985 Musala1975 Misato1985 Baksan1981 Budapest1985 Morello1983 Hobart1985 Norikura1989 Yakutsk1985 EasTop1995 London1985 EasTop1996 Socorro1985 Tibet1999 HongKong1987 Tibet2005 Baksan1987 Milagro2009 Artyomovsk1990q EasTop2009 Sakashita1990 Baksan2009 Utah1991 ARGO2011 Liapootah1995 Utah1981 Matsushiro1995 Ottawa1981 Poatina1995 Holborn1983 Kamiokande1997 Bolivia1985 Macro2003 Misato1985 SuperKamiokande2007 Budapest1985 IceCube20 Hobart1985 IceCube2012 Yakutsk1985 London1985 Socorro1985 HongKong1987 Baksan1987 Artyomovsk1990q Sakashita1990 Utah1991 Liapootah Matsushiro Poatina Kamiokande Macro2003 SuperKamiokande2007 IceCube20 IceCube Figure 1: First harmonics of the sidereal daily variations measured by underground 15

49 IceCube: large scale anisotropy cosmic ray anisotropy large scale NOTE: anisotropy is not a dipole IceCube topology changes at high energy 20 TeV 400 TeV mixed composition relative intensity equatorial coordinates relative intensity - equatorial coordinates IceCube TeV -90 IceCube TeV -90 deficit 6.3 σpost NOTE: anisotropy is not a dipole topology changes at high energy IC59 C59 Abbasi Abbasi et al., et al., ApJ, ApJ, 746, 746, 33, 33, IC22 Abbasi et al., ApJ, 718, L194, 20 49