PoS(ICRC2019)010 Also the gamma-ray http://pos.sissa.it/ s This allowed us to study the spectral and temporal s. 1 [email protected] now on detection of gamma-ray signal from GRB afterglows at very high energies will become one standard the observations. of allows us to constrain some of the key physical factor, parametersminimal electronof energy, thethe ratioGRB ofas thethe bulkradiation Lorentz to magnetic H.E.S.S.field density. imaging atmospheric Cherenkov telescope recentlysignal from the GRB reported180720B, measured in the afterglow phase, 10 hourson after the onset of the a 5 explosion. These observations prove that the recently. BecauseGRBs the areobserved GRBs moredid notpowerful show peculiarthan properties, assumedwe believeuntil that from development of the GRB, revealing a new emission component in the comparable afterglow. Its powerto is that of the synchrotron component.energy distribution of the WeGRB at an found energy of few hundred GeVs. aOur modeling, based on the second peak in data thefrom the two dozen space-born and ground-based instruments spectral that followed GRB 190114C at multiple wavelengths, supports the explanation that the secondCompton peak radiationis mechanism.due Theto thetwo-peaked Inverse structure of the spectral energy distribution So far the highest energy gamma ray measured from a GRB was a single photon energies had ofinstrument.been Emissiontheoreticallyobserved TeV at by higher, the~95 Fermi-LAT GeV, predicted, but never confirmed by observations. Here we report on a detection of a huge signal from GRB 190114C in the energyTeV range by the MAGIC imaging atmospheric Cherenkov telescopes. Starting one minute after the onset of the burst gamma rays in the energy range 0.2 -1 TeV were observed at more than 50 Gamma-ray bursts (GRB) are extremelyradiation in the Universe, occurring at a rate of about violent,once per Dependingday. on the emission serendipitous sources oftime two populations electromagnetic of GRBs can be identified; those flaring longer than“long” 2 s are and labelled as the rest electromagnetic radiation as known in the ”short”.Universe. Their initial prompt Long-duration flashes of MeV GRBsgamma rays are followed areby longer-lasting afterglow emission from theradio waves to GeV gamma most rays. luminous sources of Copyright owned by the author(s) under the terms of the Creative Commons Creative of by author(s)Copyright the terms the the under owned

Speaker Madison, WI, U.S.A. WI, Madison, 1  4.0). 4.0 International (CC BY-NC-ND License Attribution-NonCommercial-NoDerivatives 36th International Cosmic Ray Conference -ICRC2019- Cosmic Ray36th Conference International 1st, 2019 August 24th - July On behalf of behalfOn the Collaboration of MAGIC Abstract Razmik Razmik Mirzoyan PhysicsMax-Planck-Institute (Werner-Heisenberg-Institute) for Ring 6, 808805 Munich,Foehringer Germany E-mail: Major Change in Understanding of GRBs atChange UnderstandingTeV ofin GRBs Major PoS(ICRC2019)010 Razmik Mirzoyan Razmik Long GRBs are The observed observed sub- The emission from GRBs have been about GRBs as, for example, the GRBs appear at random locations and

TeV for questions 2 , which theproduce prompt and emissions,the or afterglow numerous searches particles Despite the big progress, still many basic The initial phase of a long GRB can last up to hundreds of seconds and is characterized GRB emission was anticipated also in the much so-calledhigher, very high energy (VHE) Over the past 2-3 decades, Till the recent detection of TeV gamma rays from GRBs, shown below in this report, so far TeV the recent detection of Till It is interesting to note that such an energetic photon is close to the maximum energy that, Afterglow Afterglow emission was known to extend from radio frequencies up to GeV energies. It is The synchrotron radiation from energetic electrons within the relativistic expanding jet The “afterglow” follows the prompt phase. It is characterized by emission over a very Depending on the duration of the main emission GRBs can be classified as being of GRBs were discovered in late 1960’s as serendipitous sources of extremely intense MeV Introduction carried out by using diverse instruments and observational techniques, but none of them according to theory, the synchrotron process can produce in the details. more afterglow; see §4 below for example, dueforCompton (IC) process. the inverse to, domain the highest energy record-holder was a single photon of ~The 95latter GeV. was measured by the Large Area Telescope (LAT) onboard the Fermi Gamma-ray z [9]. 0.34 ~ at redshift GRB Fermi) 130427A from Space Telescope (hereafter believed to be mainly due to synchrotron radiation from energetic electrons that are accelerated magnetized[2,3]. external shocks plasmaatthewithin further. beprompt to stillbut needs emission, explored this responsible for the could be that the so-called external theafterglow. for responsible shocks, produced when the jets pierce the ambient acceleration mechanisms of gas, are [5]. thejets,openremain of formation massive star collapse launches collimated jets of plasma, which expand with within originates prompt the jets. from emission The space [2,3]. velocities in ultra-relativistic assumption. support photonsthis high-energy timescalesand millisecond associated variability broad wavelength range and smooth decay over much longer timescales. It is widely accepted evidence is supported by supernovae occurring in coincidence. Short GRBs are likely triggered by the mergers of binary [8]. GRB also by as170817A appearing hypothesisthis long-standing supported neutron star systems. The gravitational wave event GW170817 by a “prompt” emission, mostly at energies. OneMeV believes that under certain conditions the cosmological distances and that theseUniverse[6]. in emission the known are the most luminous sources of electromagnetic "short" or "long" type. The border line between initiatedthe in twothe corespopulations of issome dying2s. massive stars undergoing gravitational collapse [7]; such gamma rays.The first publication goes back to 1973 [1]. times in the atsky, a rate of about once per Forday. a short moment they become the brightest sources in the sky but then rapidly fade away. Over the past ~50 years we learned atheir lot about nature [2-5]. Interestingly, it took quite some time to find out that GRBs occur at Major Change in Understanding of GRBs at TeV GRBs in Change at of Major Understanding 1. PoS(ICRC2019)010 seconds. Razmik Mirzoyan Razmik +15912 0 T seconds until /BAT [22]. Very soon other soonother reports Very [22]. /BAT +57 About 4 hours after the burst alert also Swift 0 T s by 3 362 ) [20] and the Gamma-ray Burst Monitor (GBM) Swift /BAT alert, the Major Atmospheric Gamma Imaging Atmospheric Gamma Cherenkov alert, the Major /BAT satellite [21] on 14th January 2019, 20:57:03 Universal Time (time interval containing 90% of the total photon counts) was /GBM [21] and[21] /GBM 90 Swift T Fermi Fermi s by by s ). The 0 T 116 116 Triggered by by the Triggered significance for the first 20 minutes of data. s Figure 1. Energy flux development in time for GRB 190114C as measured space-born as two dozen by in time development 190114C flux forGRB 1. Energy Figure ground-based instruments and abouttwo space born anddozen sparked interest big These GRB 190114C was first identified as a long-duration GRB by the BAT instrument onboard was first identifiedGRB 190114C as a long-duration GRB by the BAT For observing a GRB from the very first moment one needs a telescope with a wide field In the IC mechanism the population of energetic electrons, which produces the results on Fig. 1. The details can be found bedetails canin [25]. found The Fig. 1. results on and ground-based instruments, covering the range from ~ 1.3 GHz to ~1 TeV; see details in [25]. TeV; covering range from ground-based the and 1.3 GHz ~ instruments, to ~1 observed GRB 190114C at different wavelengths, from 1.3 GHz and up to ~1 seeTeV, the than 20 #12390) [23] and a GCN Circular (#(ATel MAGIC23701) AstronomyissuedTelegram [24] an gamma GRB.from a rays TeV thefirstdetection of on (UT) (hereafter to be measured followed. (MAGIC) telescopes observed GRB 190114C from Based on online analysis MAGIC reported detection of gamma rays above 0.3 withTeV more 2.Detection of GRB 190114C 2.Detection the Neil Gehrels Swift Observatory ( instrument onboard the mechanism of a of source [18,19]. mechanism a of view (FoV).Also a narrow FoV telescope has a chance to observe the GRB albeit not from the very first moment. For this it instrument. needs to receive a fast alert from an external wide FoV these will be discussed in the text below. discussed below. willin the betext these synchrotron radiation, scatter on the low-energy photons and boost their energy into the VHE Along rangewith [13-17,46]. synchrotron emission, IC is regularly observed from, for example, the blazar sub-class of active galactic nuclei. Sometimes IC can become the dominant emission Major Change in Understanding of GRBs at TeV GRBs in Change at of Major Understanding succeeded [11,12]. One can find only a couple of weak or ambiguous hints in the literature; PoS(ICRC2019)010

= 42 . iso E = 0 0.3 z ≥ vertical -ray signal it 1.1 TeV g ≥ Razmik Mirzoyan Razmik , for the 2 Image taken [26]. from . and for energies s flux for Nebula flux Crab for E /GBM at T90 T90 was estimated/GBM atto be Fermi is0.7 5.8TeV . The significance is calculated using the Li & ≥ 4 [26]. Also the measurement of its [26].redshift and by Swift-BAT (15-50 keV) for GRB 190114C for the (15-50 190114C keV) for GRB Swift-BAT by and regions

observed observed by

see Fig. 3 0.3 TeV) TeV) 0.3 s, ≥ ] and references therein. The isotropic-equivalent energy of the The significance for energies Image taken [26]. from [27]. the signal and the background erg. This tells that GRB 190114C was a rather energetic, but not exceptionalnotbutburst. energetic, rather was a tellsThisthat190114C GRB erg.

53 On Fig. 3 we show the distribution of the squared angular distance, θ The subsequent offline analysis of the collected data showed an extremely strong cut for 2 θ Ma method Figure 3. Significance of the γ-ray signal between T0 + 62 s and T0 + 1,227 s measured by MAGIC by 1,227 s measured T0 + from 62 s andT0 + signal between γ-ray 3. Significance the Figure of 190114C. GRB (points) and background events (grey shaded area). The dashed vertical line shows the applied emission emission in the interval 10–1000 keV x 10 3 dashed line shows stable DAQdashed time the start system line the of under conditions detection with significance > 51 has been reported, see [26 Figure 2. Light curves ( 2. Light Figure MAGIC by horizontal MAGIC the The line shows blue-dashed 210 seconds.first The left) solid (1-3, time the The from vertical show sequence for actions MAGIC. lines of TeV. Major Change in Understanding of GRBs at TeV GRBs in Change at of Major Understanding PoS(ICRC2019)010 ≥ EBL E .  European

Razmik Mirzoyan Razmik Several innovative Roque de los Muchachos of the extra-galactic background light (EBL). 5 EBL pair is born. The gamma-ray pair isspectrum in born. close vicinity of sources, + e - measuring in the sub-200 GeV energy domain, down Itto is~10 interestingGeV. to ecause the MAGIC observation started about one minute after the onset of the GRB [26]. This is the most intense signal ever measured from a gamma-ray source since the s One of the main design goals was the repositioning of the telescope within ~20 s to an Since 2009 the MAGIC telescope project operates two 17-m diameter IACTs (MAGIC-I B The observed gamma-ray spectra from cosmologically distant sources gradually change c² can be fulfilled, e an e 1.The MAGIC Telescopes Pioneered Measurements down to few tens oftens GeV few to down Measurements Pioneered Telescopes 1.The MAGIC reflector frame was made from light-weight reinforced carbon-fiber and the novel light-weight individual mirror segments were set on stepping motor driven actuators; these are operated by threshold (use of optimized light guides, nanosecondhigh-efficiency, fast novel hemi-spherical PMTs and analog signal transmission via opticaletc). reflector, parabolic-shape fibers, GSample/s fast signal digitization, arbitrary position in the sky for measuring a possible signal from a GRB. For that purpose the detectors, is defined by the fluctuations of the Light of the Night Sky (LoNS) (see, for example, the review [32]). In mid-1990's aforementioned assumption thewas not correct and in fact a initiatorslarge-size IACT of fast electro-optical of the MAGICdesign can successfully projectoperate above a threshold understoodof few tens of thatGeV [33]. the design features were put forward for MAGIC for achieving the goal of very low energy in 1995 as a stand-alone instrument telescope [31]. It had anote that back in mid pioneering1990's the young community of believed IACTs that design it was impossible to to become lower thethe threshold of the air Cherenkov technique first much The below reason~300 wasGeV. the widely accepted belief that the threshold of an IACT, similar with non-imaging Cherenkov Northern Observatory in La Palma, Canary Islands, Spain [29, 30]. By imaging the Cherenkov light emission from extended air showers within a fieldMAGIC of telescopesview of can~10 square very degrees, efficientlythe discriminatebackground and detect thegamma rays above an vast energy threshold of 30 majority GeV (Sum-Trigger-2, see of isotropic§2.4) or alternatively 50 hadron GeV (Standard TheTrigger). MAGIC-I telescope project was initiated although a partial overlap with the prompt emission phase cannot be excluded [26]. be overlap emission excluded a cannot with the phase although prompt partial 2. and MAGIC-II) in a coincidence (stereoscopic) mode at the 2m where the EBL attenuation plays no role, can be referred to as the sourceas spectrum.Taking a basis a plausible model of the EBL [28], one can infer the source spectrum by de-convoluting attenuation. the anticipated from spectrum the measured 190114C (see Fig. 1), the emission reported TeV should be associated with the afterglow phase which is the stronger the higher is the energy of the photon and the further is the source, is due to its interaction with the low energy photons E The latter is a diffuse, abundant background of infrared, optical and ultraviolet radiation emitted by all stars and galaxies during the existence of the Universe, which fill inAs athe space.result of such interaction, under the condition that the center of mass energy condition √E invent of the ground-based VHE gamma-ray astronomy; in the first 30measured rate ofs gamma ofrays from observationsGRB 190114C wasthe ~130 times higher than that from Crab ourin galaxy. VHE candle for standard the source thestrongestand Nebula, their shape, becoming steeper at high energies above few tens of ThisGeV. flux attenuation, Major Change in Understanding of GRBs at TeV GRBs in Change at of Major Understanding is 2.5 PoS(ICRC2019)010 , 1 h 1 h < <

delay delay T T and Razmik Mirzoyan Razmik 1 z < Observations started less than

6 were actually followed up by MAGIC with 500 GeV [35]. Diverse test results 500 support GeV the evidence 1 ≥ z < Only four GRBs were observed under the conditions signal for energies signal for energies s , the zenith angle (higher threshold energy for zenithlarger angles) and ambient z The small number in the latter case is mainly due to adverse weather conditions at

In addition, for some bright GRBs detected by late-timeFermi/LAT, observations . 8 GRBs had a redshift below 1 and 3 GRBs below 0.5. Except forthe GRBrest 190114C, are short This GRBs. is not surprising because it is Below we discuss briefly a few ambiguous hints, which one can findtheliterature. inambiguous cana weone discuss few hints, which Below briefly Interestingly, Interestingly, the measured by MAGIC short GRB 160821B was nearby, it's redshift was Observation of GRB 190114C can be compared with those of other GRBs followed by Despite 15 years of dedicated efforts, no unambiguous evidence for gamma-ray signals MAGIC is responding to GRB alerts since 15th July 2004. For the first 5 years, MAGIC The search for TeV gamma rays from GRBs has been pursued over many years by of a genuine signal but the low signal strength does not allow making a serious claim. allowmakingnot serious thelow doesbutstrength a genuinesignalsignal a of 0.16. MAGIC reached the alerted source position in 24 s and continued observations for ~4 h. The observing weather conditions were not optimal and there was a Threebright Moon independentin the sky. analyses were performed. on 3.1 converged These, after including the number of trials, (Table 1). (Table known that compared to long GRBs the redshift distribution for short ones is shifted to values. Few lower more long GRBs with but the observations were not in found details ]. Morebe [26 can conditions. successful due to technical problems or adverse observing afterglow phase. afterglow MAGIC under similar conditions. The chance to detect a source redshift GRB by an IACT depends on the lighting conditions. of their low-energy spectra. More detailed studies were presented for GRB 080430 and090102 GRB that were simultaneously observed with MAGIC bands. energy Since 2013, GRB observations are performed with andthe below described automatic other instruments in different procedure have been conducted, up to one day after the burst for searching for potential emission in the 30 minutes after the burst for 66 events (of which 33 lack the redshift), and less than 60 seconds for 14 events. the alerts. the time of from GRBs could be detected by MAGIC before GRBs observed in 2005-2006 were found to be consistent withextrapolations simpleGRB power-law 190114C. The flux upper limits for was operated as a single telescope (MAGIC-I). After the added in 2009, GRB observations secondhave been carried out in the coincidence (stereoscopic) mode. telescope (MAGIC-II) was MAGIC observed 105 GRBs from July 2004 toredshifts. February 2019. Of these, 40 have defined 2.2.TeV-band observations of GRBs with MAGIC and other facilities in the in past facilities the MAGIC GRBsand other ofwith observations 2.2.TeV-band employing diverse experimental techniques, but no clear detection was reported, see [26] and therein. the references light-weight telescopes (~ 70 T) can be re-positioned to an arbitrary position in the sky within s. 25 MAGIC, as well as the other IACTs, are tuned to follow external alertsinstruments withprovided large byFoV; satellite some of those provide precise coordinates of the triggered GRBs several within seconds. Major Change in Understanding of GRBs at TeV GRBs in Change at of Major Understanding After many improvements Active overMirror years,Control (AMC) todaysystem. the using the PoS(ICRC2019)010 The after trial

Razmik Mirzoyan Razmik . The red shift of 20 TeV, with a ≥ 650 GeV. it is highly improbable that ≥ GRB 970417A for energies -3 distance scale, higher than that measuredby bymore BATSE 7 1.5 x 10 < 1h) observed by MAGIC under good conditions. Only under 1h) conditions. good < MAGIC by observed delay 650 GeV was ≥ To “rescue” this unusual situation one may assume that the inpeak To It measured a spurious signal from The estimated spectral energy distribution of this event distribution of event looked unusual this The spectral estimated energy energies after trials [36]. Note that the location of the highest excess was at 9° s When an alert is evaluated as observable by AAS, the the telescopes will be automatically An Automatic Alert System (AAS)Automatic has been An developed with the aim of the fastest possible Milagrito, the smaller version of the Milagro, the waspredecessor aof veryHAWC, small The AIROBICC cosmic and gamma-ray detector was part of the HEGRA array. It slewed to Anthe automatictarget position procedure, inimplemented the in sky. 2013, prepares commands to the Central Control (CC) software of check of the visibilitythe of the prioritytarget according list to A waspredefined setcriteria. up for MAGIC telescopes. This includes a the case when several different types of alerts are received Moreover, for quasi-simultaneously. the case of multiple alerts for localization. the same GRB, the AAS will select the one with the best 2.3.The Automatic Alert Alert system Automatic 2.3.The response to GRB alerts. It Coordinates Netwrok (GCN) servers, receives Notices with sky coordinates of isGRBs, and sends a multi-threaded program that connects to the Gamma-ray Table 1. List selected GRBs of t (z 1 and < Table long is type. of GRBthe 190114C any convincing signal.convincing any because the fluence for than one order of magnitude. the X-ray range is at much lower energies than what has been measured, but this is improbable. skeptically possibility. mentioned thatauthors the [37] paper of The Short summaries of the GRB observations H.E.S.S., Milagrito by HAWC, Veritas, and Milagro in the past can be found in [26]. Unfortunately none of those few hundred observations showed the reported excess was due to ato wasdue GRB. excess the reported size (1700 m²) detector. probability of this event was estimated to be was unknown. the source measured a small, spurious excess from the significanceGRB of 920925C2.7 for energies distance from the probable target position, measured mission. byMoreover, keeping inthe mind thehardEURECA/WATCH cited veryX-ray high energy threshold and the strong EBL absorption for the GRB at a possibly cosmological Major Change in Understanding of GRBs at TeV GRBs in Change at of Major Understanding PoS(ICRC2019)010 standard Razmik Mirzoyan Razmik we measured pulsed providing a threshold of signal in less than 6 hours s tracked in wobble mode, the after the receiving alert. after Sum-Trigger, GRB follow-up was achieved for the Sum-Trigger, loaded, the individual electronic trigger implemented in both telescopes, allows allows implemented us telescopes,both in

8 high rates and for avoiding the saturation of the system,

The spectrum of the Geminga pulsar was measured Sum-Trigger-2 The steady emission of the Crab Nebula can be measured above t). √ . taking during the telescope slewing; data taken immediately before the about the target from the Central Control software.The rate limiter will be preventing from extremely , providing a threshold of ~50 GeV and with the 1 kHz for Already in mid 1990's the collaboration researchers developed from the the IACT HEGRA Results of those measurements strengthened our fidelity that the follow-up of appropriate For “polishing” the complex, fully automatic procedure used to observe GRBs, once every The goal to measure distant extragalactic sources and GRBs along with pulsars guided the celestial sources albeit, because of the higher noise, at the expense of somewhat higher energy threshold. 2.5.Operating MAGICs in the presence of partial moonlight, at dusk moonlight,and atdawn partial of presence MAGICs the in 2.5.Operating observation technique in the presence of partial moonlight [42]. MAGIC adopted andimproved thisfurther technique, performing regular observations at dusk, dawn and partial moonlight [43]. The use of this technique allows one to prolong by ~30 % the IACT observations of month, during the data taking shift in La Palma we issue a fake GRB alert; only a few persons know about the origin theof fake Typically the location trigger. is chosen to be close to the one of the known intense gamma-ray emitters. In this way we are able to check and further improve the telescopes. as as ofthe system entire diverse of of well subsystems the performance above the energy threshold of 16 GeV [41]. GeV 16 of threshold the energy above GRB candidates will lead MAGIC to a successful measurement above the energy threshold of matter timefindthe first andshouldbeto asignal. of it just GeV tens of few ~30 GeV. Already in 2008, by using the firstgamma raysversion from the ofCrab pulsar above the energy threshold of 25 GeV with the stand-alone improved The [39]. MAGIC-I to measure the Crab pulsar above the threshold of ~20 GeV with 5 (the quality factor Q=2.1 the energy threshold of ~30 GeV [40]. respond to alerts for measuring GRBs measuring alerts to for respond the design of the MAGIC telescope. Today we can operate thetrigger telescopes with the set to DAQ system. When the repositioning completes, the targetstandard observing mode foris MAGIC.date, To the fastest onlyseconds thedatataking 160821B, 24 started GRB when promptly to and its readiness MAGIC of threshold low energy very2.4.The operational alert will be saved, relevant trigger thresholds willtables be set andAMC systemthe will adjust the individual mirrorswill to provide the best be parabolic shape at the target position [38]. While moving, the calibrationimaging cameras procedure.continue the The Data Acquisition (DAQ)receives information system continues taking data while it Major Change in Understanding of GRBs at TeV GRBs in Change at of Major Understanding the subsystems for data PoS(ICRC2019)010 ) and the DAQ Razmik Mirzoyan Razmik + 57 s 0 T /BAT /BAT distributed an alert Swift ), +22 s 0 ), as shown in Fig. 2. Observations were T + 62 s 9 0 T This 60°.detection marked a new Accordingera to in the understanding of GRBs. ). After starting the slewing, the telescopes reached the target position in ≥ The telescopes can observe sources starting from few tens ofat low GeV zenith angles and While we were anticipating to measure a gamma-ray signal from the next best remote Due to upgrade of the telescopes in 2013 and a number of following improvements + 50 s 0 MAGIC observations of GRB 190114C. MAGIC T minutes) minutes) when data taking started. Data acquisition started at 20:58:00 ( system was operating stably from 20:58:05 ( approximately 27 seconds.At the end of the slewing, the cameras on the telescopes oscillated for a short time. Later on we reproduced the exact motion of the telescopes, and measured the oscillations of the imaging cameras by using four high-rate CCD and CMOS cameras, watching the imaging cameraverified behaviorthat of the both duration We telescopes.of the oscillations was less than 2 seconds after the start of tracking, and its amplitude was below 1mm (0.6 arc- reporting the first estimated coordinates of GRB 190114C (RA: +03h 38m 02s; Dec: -26d 56m AAS validated Theit18s). as observable and triggered the automatic re-pointing procedure, and the telescopes began slewing in fast mode from the position before the alert. The MAGIC-I and MAGIC-II telescopes arrived on the target and began tracking GRB190114C at 20:57:53( UT only ~10% from the yearly available time. Operating an IACT at partial theprobability GRB. a %, thus to detect increasing ~30 thisby another resource Moondawn increase and dusk and 3. On the night of 14 January 2019, at 20:57:25 UT ( of an IACT at a large zenith angle. zenith ata large IACT an of Because the energy threshold of an IACT increases with observation zenith threshold. to energy than close rather its athigher a measuretoenergies GRB probable angle, it is more The probability of a GRB to happen in the field which makes are operating at ofdark nights with a clear sky, the IACTs Typically viewoperational time. of a telescope is proportional to its large observational angles. observational large zenithanglehappenedduringa and at large moonlight GRB 190114C why 2.7.As to The number of GRB candidates that an canIACT follow is proportional to the solid angle it can Therefore itcover isin themuch moresky. probable that a GRB will appear in the field of view GRB candidate above a very low energy threshold, we measured instead a gigantic signal from the GRB 190114C at TeV energies, in the presence of partial moonlight angle range and from the zenith Monte Carlo simulations we measured a signal from GRB 190114C above the energy threshold of ~200 GeV; the relatively high threshold was due to the presence of partial Moon and the observation of sources in the zenith angle range 70° - 80° can provide a shower collection area [44,45]. km² 1-2 theorder of on zenith 80°. angle at of the TeV ~100to up understanding of GRBs of understanding MAGIC has arrived at its best sensitivity of ~0.6 % Crab [30]. The sensitivity has been further boosted at the lowest energies due to the use of the Sum-Trigger-2 and at the highest energies (~100 TeV) due to the use of the so-called very large zenith angle observation technique; Major Change in Understanding of GRBs at TeV GRBs in Change at of Major Understanding marked a the in beginning ofera new the detecting GRB of190114C the 2.6.The surprise PoS(ICRC2019)010 ]. No 43 [26]. s ]. The final 26 of the data was of Razmik Mirzoyan Razmik event image, set to 80 were derived by assuming a times the level for dark observations. Figure 4 6 cleaning was applied to both the real and

10 , Power-law photon index is taken from time-resolved b The spectra in ]. 26 . 12 h 12 . 4 . A A higher. threshold cut on the integrated charge of the ] 30 and the analysis chain tuned for data taken under moonlight conditions [ [30] Light curves in KeV, GeV and TeV, andLight spectral curves GeV evolutioninand TeV, in KeV, bandthe TeV for GRB 190114C. MAGIC data analysis for GRB190114C for data analysis MAGIC . , Light curves in energy flux (left axis) and apparent luminosity (right axis), for MAGIC at 0.3–1 TeV Figure 4. a (red atsymbols), 0.1–10Fermi-LAT GeV (purple) and Swift atXRT 1–10 keV (green). For the MAGIC data the source flux is corrected for EBL [26]. horizontal showsThe −2. dashed value are statistical, line the Shown errors 1 spectra.intrinsic compared compared to standard analysis, a higher level of image the MC data [ photo-electrons, was used for evaluating the photon flux. Details of the analyses, including the assessment of the absolute energy scale as well as the small impact severalof EBL models uncertaintiescan be offound intested [ spectrum. thesourceintrinsicfunction law for simple power DAQ DAQ rates. the simulationgamma-rayproduced for analysis, Carlo(MC) of data was Monte set dedicated A distribution, thezenith-azimuth and settingsthresholds), (discriminator matching the trigger the LoNS level during the GRB 190114C observations; details can be found in [ data-set comprises events starting from 20:58:05 UT. Due to the higher level of LoNS, software detailed information on the atmospheric transmission was availablethat of the Becausequality during thenight the observation.operating notof was since the micro-LIDAR assessed by checking the other auxiliary weather monitoring devices the the clouds well stability as thepresence as ofof theheat-sensing for infrared camera camera, such as the star-guider Data taking for GRB190114C stopped on 15 January 2019, 01:22:15 UT. The was time GRB190114C for total exposure 3.1 Data collected from GRB190114C was analyzed by using the standard MAGIC analysis Major Change in Understanding of GRBs at TeV GRBs in Change at of Major Understanding performed in the presence of moonlight, approximately PoS(ICRC2019)010 ≥ Razmik Mirzoyan Razmik ]. The origin of 5 ]. Interestingly, we 26 -rays of energy higher g and the derived source spectra The errors on the flux correspond the temporal decays of the energy 0.07. ± 11 (gray open circles) = -1.60 -1.60 =

b and energy loss, the latter primarily due to synchrotron 0.25) (statistical errors only). Use of other EBL models .

- ] The observed spectrum can be fitted with a power withpower fittedlaw a with be observed spectrum can The with b 26 t ≈ ], we obtained the GRB spectrum, which can be well described by F(t) F(t) 28 0.22 (only the statistical error). Obviously the “culprit” for such a very =-2.22 (+0.23 ± source after unfolding and correcting for the EBL absorption for “true” energies E energies “true” for absorption theEBL unfolding correcting afterfor and 5.43 a - see, for example, example, for [ see, = B . It can be shown that the maximum energy of synchrotron emission is obs a . [10] On Fig. 5 one can see the observed If the acceleration of electrons and protons in the external shock occurs in a correlated The striking feature seen on Fig. 1 and Fig. 4 is that . The upper limits at 95% confidence level are shown for the first non-significant bin at s in the TeV and X-ray bands show similar behavior. One may assume that the VHE emission VHE that Onemayassume the similarbands X-ray behavior. show and TeV in the Synchrotron burnoff limit for the afterglow emission the afterglow burnoff limit for Synchrotron independent ofindependent ambient medium, producing shocks, where electrons afterglowcan is bethe acceleratedradiation of[ those electrons. The attainedmaximum energyin ofthe electrons referencethat framecan co-movingbe withequating the timescalesthe of acceleration post-shock region can be estimated emissionby TeV energy range. range. energy TeV 4. The expanding relativistic jets from GRBs decelerate and dissipate their kinetic energy in the a power law with showed that the spectral index changes within the could citednot observe any abovecut-off or break in errorsthe spectrum at [ the highest energies; with confidence The fact thatlevel the ofspectrum 95 can % be described it by extendsa beyondpower 1TeV. law of ~2 tells that there is equal power radiated in the measured energy range 0.2-1 andTeV Thisthat ascertains a possiblysignificant beyond. fraction of the GRB radiation is emitted in the steep spectrum is the strong absorption by the EBL. By estimated,using thefor plausibleexample, EBL modelsthat we the gamma-ray flux of 1Nevertheless, because TeV of thephotons extremely strong issignal we could attenuatedobserve 300 times. By de-convoluting the measured spectrum from thanthe 1TeV. absorption effect of the byEBL using the plausible model [ high energies. Also shown is the best-fit model for The solid grey curve function. for the observed spectruma assuming is power-law a convolutionthe source spectrum (black curve) when of this curve with EBL. The grey function. power-law with aspectrum dashed curve is the forward-folding fit an exponentto the observed assumption filled circles) (blue averaged over the initial ~40 minutes of observation. TeV, 0.2 to 1 external shock scatter on ambient low-energy photons, boosting these to VHE energies, could viable a scenario. offer manner, also the ultra-high energy protons could offer an explanation for the observed spectra. But the low radiative efficiency of hadron-induced processes does not support such an a simple power-low function simplepower-low a flux is correlated with the electron mechanism. synchrotronThe well-known inverse emission. Compton radiation, This where the hintsenergetic electrons in to the a “leptonic” emission Major Change in Understanding of GRBs at TeV GRBs in Change at of Major Understanding The light curve for the correctedEBL MAGIC flux in thecan berange wellTeV 0.3-1 fitted by PoS(ICRC2019)010 . and its ] 26 max [ ; syn e

Razmik Mirzoyan Razmik mechanism may be see for details [26] and . A more plausible [26] see [25,26] and references references therein. and [25,26] see ], which allows one to estimate 12 , which depends only on the time-dependent bulk 12 [ )]

z + (1 / ) t ( b G [

energy of photons detected by MAGIC; see for detailsforsee by photonsdetectedMAGIC; of energy x

the MeV of the external shock. The latter can be derived from the solution of the 100

) t . . Image taken fromImage taken [26]. ( ≈ b G max ; syn e , it isstill well it below , The spectral energy distributions (SEDs) of the radiation detected by MAGIC in five The extremely strong signal measured by MAGIC allowed us to follow both the temporal Although proton synchrotron emission may possibly explain the GeV emission observed It can be shown that even under extreme assumptions, resulting in very high values of Also synchrotron emission by protons accelerated to ultrahigh-energies in the external max ; Spectral and temporal developments of GRB 190114C from observations at observations from developments of GRB 190114C andSpectral temporal syn show that the synchrotron component peaks in the AtX-ray band.higher energies, up to 1 GeV, the SED is a decreasing function of dataenergy (Fermi-LAT at 0.1-0.4 GeV). On the contrary, and spectral developments of the GRBof190114C. spectral the developments and consecutive time intervals are shown in Fig. 6. For the first two intervals also the observations in the GeV and X-ray bands are available. During the first interval of 42 s duration (68–110 s; blue data points and blue confidence regions), Swift-XRT, Swift-BAT and Fermi-GBM data 5. multiple wavelengths references therein references in some GRBs, due to its low radiative itefficiency, is strongly disfavored as the origin of the luminous emissionTeV observed in GRB 190114C electrons, accelerated Comptonemission inverseby e emission in GRBshock potentiallyhasafterglows, been proposed as a mechanism for GeV-TeV at energies above the burn-off limit for electron synchrotron emission, The observed spectrum of afterglow synchrotron emission is expected to display a cutoff below the energy Lorentz factor dynamical equations of the external shock evolution. Figure 5. MAGIC spectrum averaged above over 0.2 TeV the period between T0 + 62 s and T0 + 2,454 s GRB 190114C. for Major Change in Understanding of GRBs at TeV GRBs in Change at of Major Understanding PoS(ICRC2019)010

Razmik Mirzoyan Razmik Image taken [25]. from 13 0.2 TeV shows a harder spectrum. This evidence, ], shows that the TeV radiation cannot be a mere ≥ As already mentioned, the synchrotron self-Compton [25 25,26]. SSC emission has been predicted for GRB theafterglows, forecasted spectrum Our studies show that with the development of the GRB afterglow the SSC peak moves to Though Though the lower energies (see the development of the MAGIC SED andFig. theMAGIC6)development in consecutive time ofbins on (see the energies lower crosses the MAGIC energy band. It is remarkable to observe a possible softening of the spectra minutes. tensof of thetime few spanMAGIC a during by measured and luminosity suffered from large uncertainties of poorly known physical emissionparameters inregion, the as, for example, the magneticdetection offield MAGIC madestrength. it possibleFor to theinfer the firstimportant physical timeparameters for the SSC model. adoptingthe strong Figure 6. SED development of GRB 190114C in time measured by FermiSwift GBM/LAT XRT/BAT, and by MAGIC.The MAGIC measurement is split into 5 consecutive periods of time. For the first 2 time also from data satellitesbins the above-mentioned are shown. the (SSC) radiation in the external forward shock can offer a natural explanation. In that scenario a is produced. component,peakingenergies, at very high spectral second independent of the chosen EBL model higher energy extension of the known afterglow component.synchrotron The extended durationemission; and the smooth,it power-law temporal decayis of the radiation a new spectral detected by MAGIC (see Fig. 1 and Fig. 4) suggest a emissioncorrelation betweenand theTeV the broadband emissionafterglow [ Major Change in Understanding of GRBs at TeV GRBs in Change at of Major Understanding the MAGIC flux at much higher energies PoS(ICRC2019)010 s . The Razmik Mirzoyan Razmik taken from taken [25]. in Fig. 7 (blue solid curve; see [25]) 14 modeling is shown

Modeling of the broadband spectra in the time intervals 68–110 s and 110–180 s. Thick blue vation of GRB 180720B by the H.E.S.S.IACT the GRB 180720B vation of by [46]. remarkable that the signal from GRB 180720B has been so measured many hours theafter onset of the burst, obviously deep in the afterglow phase of the GRB. It came as a real surprise that a gamma-ray signal in the range of few 100's of GeVs could be measured in such a late phase of a GRB published inand thetheATels GCN circulars in mid January 2019, the H.E.S.S. collaboration reported on a detection of a gamma-ray signal from GRB 180720B at thesymposium CTA in Bologna on May 8th 2019. The H.E.S.S. observations were performed 10 hours after the onset of the burst for ~4 hours in the zenith angle range of 40°-25° The degrees. signal strength of 5 The redshift of the GRB is Itz=0.653. is beenhas interval measured100-440 GeV. in the energy components are similar. components 6.Obser Five months after the detection of GRB 190114C by MAGIC and two dozen other instruments, dashed line shows the SSC spectrum when internal absorption is neglected. The thin solid line shows the model spectrum(empty observations circles). including EBL attenuation,Our model indicates that comparable amount of radiated energy has been channeled through the in comparison toAlso prompt theand the radiated following phases.power afterglow in the synchrotron and SSC the MAGIC curve, modeling of the multi-band data in the synchrotron and SSC afterglow scenario. Thin solid lines, synchrotron and SSC (observed spectrum) components. Dashed lines, SSC when internal γ–γ opacity is neglected. The adopted parameters for modeling data are spectrum. and 6. Image Contour points inobserved MAGIC regions Fig. as can be found in [MWL]. Empty circles show the An example of the theoretical Figure 7. Major Change in Understanding of GRBs at TeV GRBs in Change at of Major Understanding PoS(ICRC2019)010 Razmik Mirzoyan Razmik

15 GRB 180720B. GRB Multi-wavelength light curve of GRB 180720B. a, Energy-flux light-curves detected by Fermi Image taken fromImage taken [46]. Taking Taking as the basis our measurement and the modeling work we anticipate that the TeV The model developed by the MAGIC collaboration, which is based on multi-wavelength On the night January 14/15 2019 MAGIC followed an alert from the Swift satellite Summary What concerns the measurement of a signalTeV from GRBs, we anticipate that as long as the energetic bursts will happen at relatively low redshifts (no extreme absorption by the EBL), the data from about two dozen space-born and ground-based instruments, allowed us to reveal the second peak in the SED of GRB assume 190114C atthatWe energies. thefrom (sub)-TeV now on the GRB afterglows will regularly manifest themselves with the appearance of peakthe in second their SED. Such measurements will help us to significantly extend our assessment and GRB. as phenomena dubbed thecomplex understanding ofimprove strongly amount of released energy. energy. releasedof amount emission shall be a rather common process. This thedetection H.E.S.S. of ofon report anticipation found its confirmation by the although some contribution from the prompt phase cannot be excluded. This was identified as an SSC emission whose power is comparable to that of the synchrotron at lower energies. One may assume that the majority of GRBs behave similar to GRB 190114C. This is supported by the fact that most parameters inferred from our model fall within from thepast afterglow studies. Onerange may conclude ofthat researchers in thosethe past missedinferred a significant mission and observed the GRB 190114C, starting from the first minute after theburst, onsetfor ofthe the next four hours. It discovered intensitythe s of the first 30 the In most TeV. to 1 above GeV from 200 intensein the energy GRB 190114C from ever gamma-ray signal at VHE emission was on the level of ~130 Crab. The signal was measured in the afterglow phase can be taken from the top-right insert. The black dashed temporalSwift decay andwith H.E.S.S. α (1σ error = line −1.2. b, Photon index for the spectra of Fermi-LAT, is for eye-guiding purposes; it shows a bars). 7. Figure 8. H.E.S.S.GBM/LAT, (red arrow shows the upper limit from the second(data observation),extrapolated betweenSwift the BAT/XRT two instruments) and the optical r-band; the relevant colour assignment Major Change in Understanding of GRBs at TeV GRBs in Change at of Major Understanding PoS(ICRC2019)010 (2018). , v. 44, , v. (2002). Science Science , 850, L40, , 850, L40, 561, 1–109 Astropart. Astropart. Razmik Mirzoyan Razmik ApJL , 1842003 27 , 137–169 Phys. Rep. Phys. Ann. Rev. A&A Ann. Rev. 40 . , 1143–1210 (2005). , 1143–1210 76 , 932–936 (2012). . Int. J. Mod. J. D Int. Phys. 337 182, L85 (1973). 182, L85 Science Ap.J. 16 Annu. Rev. Astron. Astrophys Astron. Annu. Rev. Rev. Mod. Phys Mod. Rev. , 181–193 (1994). 432

The Connection, Supernova Gamma-Ray Burst ApJ , 63–67 (2010). On the Progenitor of Binary Neutron Star Merger GW170817, GW170817, NeutronOn Binary Merger Star of Progenitor the , 718

ApJL , 252–275 (2013). v. 387, Issue (1997). 783-785 6635, pp. v. 43 . , 42–47 (2014). gamma-ray burst sources. burst gamma-ray Inoue, S. et al., Gamma-ray burst science in the era of the Cherenkov Telescope Array, Array, Telescope Gamma-ray al., in et era science burst the Inoue, S. Cherenkov the of Phys from bursts. emission gamma-ray Nava, High-energy L., Spectral blast-wave of H., properties M. J. & of Papathanassiou, Rees, models P., Mészáros, (2017). observations gamma-ray the 130427A. of GRB burst al., M. et Ackermann, Fermi-LAT 343 GeV external for the shock E., On gamma-ray synchrotron model bursts’ Nakar, & T. Piran, emission. Costa, E., et al., Discovery of the X-Ray Afterglow of the Gamma-Ray February the Burst of of 28 1997, Afterglow the of Costa, Discovery al., E., et X-Ray Nature, Bloom, S.E., J.S., Woosley, 1, pp.507-556 (2006). Issue al. et Abbott, B.P., Piran, T., The physics of physics bursts. The gamma-ray T., Piran, bursts, Gamma-ray P., &Gehrels, N. Mészáros, & bursts relativistic jets. The gamma-ray of physics B., & Zhang, P. Kumar, (2015). Klebesadel R.W., Strong I.B., and Olson R.A., Strong and I.B., R.W., Klebesadel bursts. Theories gamma-ray of P., Mészáros, [9] [6] [7] [8] [3] [4] [5] [1] [2] [11] [12] [13] [10] References this further enhances the chance probability. And thethis furtherlast enhances but notthe chance least probability. answer is that one needed a from GRB signal athefirsttimeisdo that measureto (sub)-TeV possible in fact, it to show for and this is just what MAGIC observational targetsdid. at (sub)-TeVs and Thethe successful measurementsgood will soon providething wealth is that aboutthese explosions. questions clues monstrous usmany topuzzling data allowing of to find the GRBs will become regular happen not so frequently at relatively low red-shifts (EBL absorptionanswer isis strong). thatThe thesecond probability is higher that energiesthese muchwill higherhappen thanat thelarge thresholdzenith ofangles a(i.e. correspondinglygiven at plan the reactionground-based of instruments to instrument);alerts. The third reasonone is that one needsneeds to to observe a evenGRB atat any, barely acceptable outdoor lighting conditions, once it is relevant; phase, even after many hours from the onset of the burst. from hoursthe the burst. onset afterof even many phase, 8.Afterword The frequent question that in the past year we asked ourselves is why ityears, took soto long, measuresay 15 a gamma-ray signal from a GRB at TeV ? The first answer is that these Major Change in Understanding of GRBs at TeV GRBs in Change at of Major Understanding current ground-based instrumentsTeV shall be able to measure these, at least in the afterglow PoS(ICRC2019)010 ,

ApJ ICRC using using th burst. burst. , 317–324 improvements improvements Proc. 24 Proc. Razmik Mirzoyan Razmik GCN Circular GCN 272 afterglows. afterglows. 23724

, 1073–1085 (2015). org/?read=12390 org/?read=12390 ApJ 454

264 (1988) 55-63. -ray burst GRB 190114C. GRB burst -ray 190114C. g MNRAS GCN Circulars Circulars GCN Proc. “Towards A Major Major A “Towards Proc. , 61–75 (2016). 72 . GCN Circulars 23707 GCN Circulars , 76–94 (2016). 72 . , ed. O.C. De Yager, Kruger South Park, National Yager, O.C. De, ed. 17 Nucl. Instr. Meth. Phys. Res. A Meth. Res. Phys. Nucl. Instr. . 12390 http://www.astronomerstelegram. Astropart. Phys Astropart. , 880, L27 (10pp), 2019 August 1 August (10pp),, 880, L27 2019 Astropart. Phys Astropart. ApJL High Energy Radiation from Black Holes: Gamma Rays, Cosmic Rays, Rays, Gamma Holes: Black Cosmic Radiation Rays, from Energy High , 2556–2578 (2011). https://gcn.gsfc.nasa.gov/gcn3/23688.gcn3 (2019). November 2019, p. 459 November 2019, p. 458. 410

st st Very High Energy Cosmic Gamma Radiation: A Crucial Window on the Extreme Window Crucial A Gamma Radiation: Cosmic Energy High Very Princeton University Press, November Press, University Princeton 2009. The Astronomer’s Telegram Telegram Astronomer’s The World Scientific Publishing Co Pte Ltd (1762) (2004). (1762) Ltd Pte Co Publishing Scientific World MNRAS v. 575, 21 v. 575, 21 v. 460, Issue 2, 2036-2049 (2016). 460, Issue gcn.gsfc.nasa.gov/gcn3/23724.gcn3 (2019). https://gcn.gsfc.nasa.gov/gcn3/23701.gcn3 (2019). , 110–122 (2001). , 110–122 Atmospheric Cherenkov Detector - V” Detector - workshop Cherenkov Atmospheric 298-314 August, (1997) 8-11 Africa, Bradbury, S., et al., A Project for a 17 m Diameter Imaging Cherenkov Telescope, Telescope, for Diameter Project a Imaging 17 m Cherenkov A S., al., et Bradbury, 1, p. 1051 (1995). Iucci, E. Lamanna, N. ed. v. Astronomy, Gamma-Ray TeV T.C., Weekes, IACTs, Constraints of in threshold the Mirzoyan, Reducing R., fractions. upgrade major MAGIC hardware the the The I: of telescopes, part al., J. et Aleksić, of commissioning the system. the and upgrade major a MAGIC the The II: study performance of telescopes, part al. J. et Aleksić, of Nebula.observations Crab the Acciari, V.A., Ansoldi, S., et al., Teraelectronvolt emission from the emission Teraelectronvolt Ansoldi, S., al., et V.A., Acciari, Nature, methods for astronomy. Analysis in results gamma-ray Q., Y. & Ma, P. T. Li, (1983). galaxy-SED-typeAEGIS Extragalactic al., et inferred A. light background from Domínguez, Mirzoyan, R. et al., MAGIC detects the GRB 190114C in the TeV energy domain. energy TeV in the al., et 190114C detects GRB Mirzoyan, R. the MAGIC 23701 a Collaboration Observation al., Compton et of from long MAGIC γ-ray inverse emission Nature, Krimm, H. A. et al., GRB 190114C: Swift-BAT refined analysis. Swift-BAT A. GRB al., et 190114C: Krimm, H. https:// detects GRB the MAGIC energies; a al., et at time of GRB Mirzoyan, detection R., First sub-TeV 190114C. (2019). and Neutrinos,and a of Swift detection very bright optical counterpart. a bright with burst J. D., 190114C, Gropp, GRB 23688 Circulars GCN Fermi GBM detection. GRB 190114C, R., Hamburg, https://gcn.gsfc.nasa.gov/gcn3/23707.gcn3 (2019). Derishev, E.V., Piran, T., The physical conditions of the afterglow implied by MAGIC’s sub-TeV sub-TeV implied MAGIC’s by physical ofThe afterglow conditions the T., Piran, E.V., Derishev, of 190114C, observations GRB Aharonian, F.A., Universe, C.D., Menon, G., Dermer, Beniamini, P., Nava, L., Duran, R. B. & Piran, T., Energies of GRB of prompt blast waves and Energies T., & Piran, Duran, Nava, L., R. B. P., Beniamini, afterglows. X-ray of GeV implied and as modelling by efficiencies acceleration, magnetization in Particle relativistic shocks, radiation and T., Piran, E.V., Derishev, MNRAS Zhang, B. & Mészáros, P., High-energy spectral components in spectral gamma-ray components burst High-energy B.P., &Zhang, Mészáros, 559 [31] [32] [33] [29] [30] [26] [27] [28] [24] [25] [22] [23] [20] [21] [17] [18] [19] [15] [16] [14] Major Change in Understanding of GRBs at TeV GRBs in Change at of Major Understanding PoS(ICRC2019)010 , 94 . , ICRC Razmik Mirzoyan Razmik th 36 Astropart. Phys Astropart. , Borisov, A. S. et al., 809 (POS, 2015). A. al., et S. 809 (POS, Borisov, eds.

18 , 583, p. 824 (2003). 12, 65-74 (1999). 12, 65-74 ApJ 489, 1073–1077 (2008). Extending the observation limits of Imaging Air CherenkovAir observationExtending the Imaging limits of Nucl. Instr. Meth. Phys. Res. A Phys. Meth. Res. Nucl. Instr. 34th ICRC (ICRC2015), , Busan, South 2017. Korea, A&A Astropart. Phys. Astropart. 337, 43–50 (1998). 337, 43–50 Proc. Proc. ICRC November 2019, p.464. th st . A&A 35 , Madison, USA, PoS(ICRC2019)645. USA, , Madison, Observation of Pulsed γ-Rays Above 25 GeV from the Crab Pulsar with from Pulsar MAGIC, Crab the 25 GeV Above Pulsed γ-Rays Observation of ICRC 21 Nov 2008, v. 322, Issue 5905, pp. 1221-1224. 21 Nov 2008, v. th v. 575, 21 v.

36 the external shock model, external shock the Acciari, V.A., et al., MAGIC very large zenith angle observations of the Crab Nebula up to 100 TeV, zenith observations Crab the 100 angle of Nebula up to large MAGIC al., et very V.A., Acciari, A&A submitted to deep afterglow, burst component in γ-ray the very-high-energy A R., al., et Adam, H., Abdalla, H. Nature, the GRB emission of from in context Piro, Prospects detectionA., L., very for of high-energy Galli, Ahnen, M. L. et M. L. MAGICAhnen, moonlight. the Performance al., of telescopes under 29–41 (2017). al., et I., Vovk, Mirzoyan, R., toward horizon, Telescopes https://doi.org/10.1016/j.nima.2018.11.046 Lopez, M., et al., Detection of the Geminga pulsar withLopez, Detection Geminga MAGIC al., the M., et the of pulsar telescopes, PoS(ICRC2019)728. USA, Madison, observations Crab the y-ray Mkn 501 during of and TeV R., Mirzoyan, Kranich, D., al., et twilight, moonshine and MAGIC Telescopes. Telescopes. MAGIC Aliu, al., E., et Science 20 above the Nebula studies G., for of and spectra Pulsar Ceribella, energies MAGIC Crab al., et GeV, Padilla, L., et al., Search for gamma-ray bursts above 20 TeV with the HEGRA AIROBICC with HEGRA the TeV L., above bursts Search forPadilla, al., 20 et gamma-ray Cherenkov array, Spectrum GRB of 970417a Gamma-Ray Fluence Energy and High-Energy The R., al., et Atkins, MILAGRITO, With Observations From the band with A. in very the Recent high al. energy et observations GRBs follow-up of Carosi, Inoue, S., et al., al., et Inoue, S., [44] [45] [46] [42] [43] [40] [41] [38] [39] [35] [36] [37] [34] Major Change in Understanding of GRBs at TeV GRBs in Change at of Major Understanding