Multi-Messenger Astrophysics1

Peter Meszaros1,2,3,4, Derek B. Fox1,3,4, Chad Hanna2,1,3,4, KohtaMurase2,1,3,4,5

1Dept. of Astronomy & Astrophysics, Pennsylvania State University, University Park, PA

16802, USA2Dept. of Physics, Pennsylvania State University, University Park, PA 16802, USA3Center for Particle and Gravitational Astrophysics, Pennsylvania State University, Uni-

versity Park, PA 16802, USA4Institute for Gravitation and the Cosmos, Pennsylvania State University, University Park,

PA 16802, USA5Center for Gravitational Physics, Yukawa Institute for Theoretical Physics, Kyoto Uni-

versity, Kyoto 606-8502 Japan

AbstractMulti-messenger astrophysics, a long-anticipated extension to traditional andmultiwavelength astronomy, has recently emerged as a distinct discipline pro-viding unique and valuable insights into the properties and processes of thephysical universe. These insights arise from the inherently complementaryinformation carried by photons, gravitational waves, neutrinos, and cosmicrays about individual cosmic sources and source populations. This comple-mentarity is the basic reason why multi-messenger astrophysics is much morethan just the sum of the parts. Realizing the observation of astrophysicalsources via non-photonic messengers has presented enormous challenges, asevidenced by the fiscal and physical scales of the multi-messenger observato-ries. However, the scientific payoff has already been substantial, with evengreater rewards promised in the years ahead. In this review we survey thecurrent status of multi-messenger astrophysics, highlighting some exciting re-cent results, and addressing the major follow-on questions they have raised.Key recent achievements include the measurement of the spectrum of ultra-high energy cosmic rays out to the highest observable energies; discovery ofthe diffuse high energy neutrino background; the first direct detections ofgravitational waves and the use of gravitational waves to characterize merg-ing black holes and neutron stars in strong-field gravity; and the identificationof the first joint electromagnetic + gravitational wave and electromagnetic +

1Update of arXiv:1906.10212, v1, which appeared in Nature ReviewsPhysics, 1:585 (2019), https://www.nature.com/articles/s42254-019-0101-z

1

arX

iv:1

906.

1021

2v2

[as

tro-

ph.H

E]

8 O

ct 2

019

high-energy neutrino multi-messenger sources. We then review the rationalesfor the next generation of multi-messenger observatories, and outline a visionof the most likely future directions for this exciting and rapidly advancingfield.

Key Points

1. Multi-messenger astrophysics aspires to make use of the informationprovided about the astrophysical universe by all four fundamental forcesof Nature, namely the gravitational, the weak and the strong forces,besides the electromagnetic force which previously had provided almostall our information about the Cosmos. These new channels provide pre-viously untapped, qualitatively different and complementary types ofinformation, capable of probing down to the densest and energy-richestregions of cosmic objects, which were hitherto hidden from astronomers’sights.

2. Diffuse backgrounds of high-energy neutrinos (HENs) with energiesfrom >∼10 TeV to PeV, ultra-high energy cosmic rays (UHECRs) atenergies of >∼ 1018 eV, and γ-rays with energies between MeV and∼TeV have been measured, or upper limits have been provided, byCherenkov detectors, satellites and ground-based air-shower arrays.

3. Gravitational waves (GWs) from merging stellar mass black hole andneutron star binaries have been detected at frequencies in the >∼ 10 Hzto ∼ 1 kHz range with laser interferometric gravitational-wave detec-tors.

4. The sources of the diffuse UHECR and HEN backgrounds remain un-known, although a gamma-flaring blazar (a type of active galaxy witha massive black hole at the center ejecting a relativistic plasma jettowards the observer) has been tentatively identified with observedHENs. While up to ∼85% of the γ-ray background can be attributedto blazars, it appears that at most 30% of the HEN background can bedue to blazars.

5. Formation channels for the observed stellar mass black hole binaries,and their possible role as a cosmologically relevant component of the

2

dark matter, is currently under debate.

6. There is a natural physical connection between high energy cosmic rayinteractions and the resulting very high energy neutrinos and gamma-rays,which needs to be fully exploited to better understand the nature oftheir unknown astrophysical sources. The connection with gravita-tional wave emission, while less direct, can be expected to provide im-portant information about supermassive black hole populations anddynamics.

7. Even before the arrival of the next generation of gravitational wave,neutrino, and cosmic ray detectors, the present advanced LIGO/VIRGOdetectors will be able to detect hundreds of binary mergers up to ∼Gpcdistances; yet electromagnetic (EM) counterpart searches rely primarilyon the aging space-based facilities Swift and Fermi, currently operatingwell beyond their design lifetimes. These EM counterparts have beenfound mainly in gamma- or X-rays, and there is an urgent need for anew generation of EM detectors, also extending into other frequenciesincluding the UV, optical/IR, and radio.

1 Introduction

Of the four fundamental forces in nature – the electromagnetic, gravitational,weak and strong nuclear forces – until the middle of the 20th century it wasonly messengers of the electromagnetic force, in the form of optical photons,which allowed astronomers to study the distant universe. Subsequently, ad-vancing technology added to these radio, infrared, ultra-violet, X-ray andgamma-ray photons. Finally, in the last few decades the messengers of theother three forces, namely gravitational waves (GWs), neutrinos, and cosmicrays (CRs), began to be used in earnest. Thus, we are now finally usingthe complete set (as far as known) of forces of Nature, which are revealingexciting and hitherto unknown details about the Cosmos and its denizens.

These new non-photonic messengers are generally more challenging todetect and to trace back to their cosmic sources than most electromagneticemissions. When detected, they are usually associated with extremely highmass or high energy density configurations, e.g. the dense core of normalstars, stellar explosions occurring at the end of the nuclear burning life of

3

massive stars, the surface neighborhood of extremely compact stellar rem-nants such as white dwarfs, neutron stars or black holes, the strong and fastvarying gravitational field near black holes of either stellar mass or the muchmore massive ones in the core of galaxies, or in energetic shocks in high ve-locity plasmas associated with such compact astrophysical sources. This as-sociation with the most violent astrophysical phenomena known means thatthe interpretation of multi-messenger observations requires, and can haveimplications for, our theories of fundamental physics, including strong-fieldgravity, nuclear physics, and particle interactions.

Figure 1: Examples of current instruments observing cosmic messengers via the elec-tromagnetic, weak, gravitational, and strong forces, showing their location. Clockwisefrom top left: the LIGO Hanford gravitational wave interferometer; one of the MAGICair Cherenkov telescopes; the Fermi gamma-ray space telescope; a schemaric of the PierreAuger cosmic ray observatory in Malargue, Argentina; a schematic of the IceCube cubickilometer neutrino detector in Antarctica;

The study of such high energy compact objects started in earnest in the1950’s, after decades of a slower build-up with increasingly large ground-based optical telescopes. The first major breakthroughs came from the de-ployment of large radio-telescopes, followed by the launching of satellites

4

equipped with X-ray and later gamma-ray detectors, which established theexistence of active galactic nuclei, neutron stars, and black holes, and re-vealed dramatic high-energy transient phenomena including X-ray novae,X-ray bursts, and gamma-ray bursts (GRBs). Starting in the late 1960’s,large underground neutrino detectors were built, measuring first the neutri-nos produced in the Sun and later those arising from a supernova explosion;and it was only in the current decade that extragalactic neutrinos in the TeV-PeV range were discovered. Cosmic rays in the GeV energy range startedbeing measured in the 1910s, but it was only in the 1960s that large de-tectors started measuring higher energies implying an extragalactic origin,and only in the last decade has it become practical to start investigating thespectrum and composition in the 1018 − 1020 eV range. Gravitational wavedetectors started being built in the 1970s, but it was not until the 1990s thatnew technologies and large enough arrays began to be built approaching thesensitivity required for detections, the first successes starting in 2015. Forthe first time, detectors covering all four fundamental forces of Nature (Fig,1) have been thrown into the breach to explore all the previously hiddenaspects of the Cosmos.

2 Mono- and Multi-Messenger Advances

The exciting experimental results listed in this section confirm many of thetheoretical/phenomenological predictions and expectations that had beenformulated over the past several decades, while also bringing up new sur-prises, which we discuss in the subsequent section.

2.1 Recent Non-Photonic Mono-Messenger Results

• Ultra-High Energy Cosmic Rays.- The Pierre Auger cosmic ray observa-tory (PAO) [1], located in Argentina, is a 3,000 km2 array of 1660 waterCherenkov stations and 27 air-fluorescence telescopes (one of the tanks anda set of fluorescence telescopes is in Fig. 1, lower right), designed to detectUHECRs at energies between 1017 eV to 1021 eV. Its measurements of thediffuse UHECR flux energy spectrum, starting in 2009, confirmed conclu-sively the existence of a spectral steepening setting in near 6× 1019 eV, e.g.[2], which had been first observed by the HiRes instrument [3]. This is con-sistent with the so-called GZK (Greisen, Zatsepin, Kuz’min, [4, 5]) feature

5

expected from CR proton energy losses or from heavy ion photo-dissociation[6] due to interactions with cosmic microwave background photons, althoughit could also be due to reaching a maximum acceleration at the sources. From2010 onwards, Auger also started showing evidence for an UHECR chemicalcomposition becoming heavier above >∼ 1018.5 eV. The statistical significanceof these results has become stronger over the years [7, 8, 9]. The spectralresults are consistent, within statistical uncertainties with those obtainedwith the smaller Telescope Array (TA) UHECR observatory [10], which isimportant because Auger is in the Southern hemisphere while TA is in theNorthern. The chemical composition issue [11] is still under debate, al-though a joint Auger-TA paper [12] shows results which agree within theerrors. The angular resolution in the arrival direction of UHECRs is below1o above ∼ 1019 eV for both protons and heavy elements, although the mag-netic deflection increases with mass; around 1019 eV it is <∼ 5o for protons,while for heavy nuclei it could be tens of degrees. At these energies, due tothe energy losses caused by the GZK effect mentioned, these UHECRs musthave originated within distances of <∼ 100Mpc. So far, all attempts at findingangular spatial correlations between UHECRs and any type of known cosmicsources have been unsuccessful [2].

• High Energy Neutrinos.- The IceCube neutrino observatory [13] consistsof a cubic kilometer (roughly a Gigaton) of ice at a depth between 1.4 and2.4 km below the South Pole, instrumented with 86 strings connecting 5,160optical phototubes (see schematic in Fig. 1, lower right), which mea-sure the light radiated from charged particles produced by passing highenergy neutrinos interacting with the ice. Its construction was finished in2010, and in 2012-2013 it discovered a diffuse flux of neutrinos in the range100 TeV <∼ Eν <∼ ‘1PeV [14, 15], later extended down to <∼ 100 TeV. Theenergy spectrum dN/dEν can be fitted with a ∼ −2.5 index power law, butthere may be an indication for two components, steeper below ∼ 200 TeVand flatter (index ∼ −2 above that, the highest energy so far being ∼ 10PeV.IceCube detects all neutrino flavors, with muon neutrino charged current in-teractions resulting in elongated Cherenkov tracks and all other neutrinoflavors and interactions largely producing near-spherical optical Cherenkovsignals from secondary particle cascades, the direction of arrival being uncer-tain by ∼ 10o − 15o for cascades and ∼ 0.5o − 1.0o for tracks. Tau neutrinoscan also be identified at sufficiently higher energies, where the statistics arelower, and these have not yet been identified, although a suggested tau-

6

like candidate has been discussed [16]. The observed flavor distribution iscompatible with complete flavor mixing having occurred due to the neutrinooscillation phenomenon over cosmological distances [17, 18]. So far there isno evident correlation of the observed neutrinos with any type of known cos-mic objects [19], except for one interesting case discussed below. The smallerunderwater Cherenkov telescopes ANTARES [20] and Baikal-GVD [21] havealso been in operation and providing upper limits. At much higher energies,the high altitude balloon experiment ANITA [22], flying in a circumpolar or-bits in Antarctica, has used a radio technique to measure neutrinos at >∼ 1017

eV, which is starting to provide constraints on cosmological neutrino sourcesand the GZK-related cosmogenic neutrino fluxes, complementary with thoseprovided by Auger [23].

• Gravitational Waves.- The Laser Interferometric Gravitational Wave Ob-servatory (LIGO) consist of two detectors, in Louisiana and Washingtonstate, each with 4 km long L-shaped arms, (Fig. 1, upper left) which in2015 began operation in the ∼ 10 − 103 Hz frequency range [24]. Anotherarray, VIRGO [25], located near Pisa, Italy, and similarly L-shaped with 3Km long arms, has been operating at epochs coincident with LIGO. Both areactively being commissioned and will achieve design sensitivity in the comingyears. The long-awaited first discovery of gravitational waves from a stellarmass binary black hole merger (labeled GW150914) was announced by LIGOin 2016 [26] (Fig. 2(a). This was soon followed by a number of other binaryblack hole (BBH) mergers detected both by LIGO and, with lower statisticalsignificance, by VIRGO as well [27]. These BHs weigh up to several tensof solar masses, and have low spins. However, despite intensive searches, noother messengers associated with BBH mergers have been detected so far,except for a possible γ-ray burst [28] in GW150914.

• Electromagnetic Detections.- Except for the binary black holes, all theother sources detected with other messengers had been previously exten-sively studied through their EM emissions at various wavelengths. Of majorrecent relevance are the observations in the optical, X-ray and up to 150keV γ-rays with the Swift satellite, and between 10 keV X-rays to <∼ TeVγ-rays with the Fermi satellite (Fig. 1, upper right) [29], which detecteda large number of Gamma-Ray Burst (GRB) sources, active galactic nuclei(AGNs) including blazars, supernovae, etc., as well as a diffuse cosmic γ-background. Of increasing importance for such sources are the ground-based

7

air Cherenkov imaging telescopes, e.g. MAGIC (Fig. 1, upper middle),HESS, VERITAS [30] and the High Altitude Water Cherenkov observatoryHAWC [31, 32], which measure gamma-rays in the 100 GeV to multi-TeVrange. These have been amply supported by ground and space observationswith multiple radio, infra-red, optical and UV telescopes.

2.2 Developments in Joint Multi-Messenger Astrophysics

• Solar and Supernova Neutrinos and Photons.- The two earliest multi-messenger detections involved neutrinos in the MeV range. Davis and col-laborators, starting in the 1960’s, detected the electron neutrinos producedby the nuclear reactions that are the energy source for the Sun’s light, usinga 600 ton perchlorethylene (cleaning fluid) tank located deep undergroundin the Homestake mine in South Dakota, US. This neutrino flux was con-firmed by various other experiments including the one in the Kamioka minein Japan, by Koshiba and collaborators. The other early multi-messengerdetection was that of neutrinos from a core-collapse supernova, SN 1987a,resulting from inverse beta decay as protons are converted into neutrons.This was detected by three different underground detectors, Kamiokande inJapan, Baksan in the Soviet Union, and Irvine-Michigan-Brookhaven in theUS [33, 34, 35]. The neutrino detection preceded significantly the spectacu-lar optical brightening characterizing supernovae. These discoveries earnedDavis and Koshiba the Physics Nobel Prize in 2002 [36, 37].

• Cosmic ray, Gamma-ray and Neutrino Background Interdependences.- Themeasurements of the diffuse UHECR energy spectrum by the Pierre AugerObservatory starting in 2008 put on a firm ground the detection of a spectralcutoff above 1019.5 eV, compatible with the GZK energy losses due to the cos-mic microwave background photons [2], after earlier work by HiRes. Then,starting in 2008, the Fermi satellite (following on previous work by COS-Band other missions) measured a diffuse gamma-ray background extendinginto the sub-TeV range [39]. And starting in 2012-2013 IceCube discovered,with increasing detail, a diffuse high energy neutrino (HEN) background ofastrophysical origin at multi-TeV to PeV energies [14, 15]. There is so far nofirm identification of the sources of either the UHECR, HEN or gamma-raydiffuse backgrounds, although the extragalactic gamma-ray background isknown to be dominated by blazars [40]. However, theoretical relationshipsand mutual constraints are expected from the basic physics of these three

8

f

Figure 2: Some recent cosmic multi-messengers advances involving the electromagnetic,weak, gravitational and strong forces. (a) The first GW detection from LIGO/VIRGOin the O1-O2 observing run, of the GW150914 binary black hole merger [26], showingfor the inspiral, merger and ringdown phases the theoretical and measured waveform, theseparation and the relative velocity. (b) Interrelation expected between (from left to right)the energy spectrum of the diffuse backgrounds in gamma-rays, high energy neutrinos andultra-high energy cosmic rays, based on a black hole jet source model [38]; (c) LIGO,VIRGO and Fermi simultaneous multi-messenger discovery of the binary neutron starmerger GW/GRB 170817; (d) Light track of the muon produced by a 290 TeV muonneutrino coming from the direction of the blazar TXS 0506+056, detected on 22 September2017 by IceCube.

radiations, e.g. Fig. 2 (b). The HENs are produced when UHECRs collidewith low energy target photons and nuclei resulting in charged and neutralpions, which decay in a predictable fraction of high energy neutrinos andgamma rays. The resulting energy spectra of neutrinos and photons implycorresponding diffuse backgrounds which must fit the observed results, in-cluding also the constraint provided by the observed UHECR background.The fact that the energetics of these three messengers is comparable has ledto the idea of a unification of high-energy cosmic particles, e.g., [41, 38]. Onthe other hand, significant constraints are also placed on generic pp hadronu-clear production models of HENs and gamma-rays when when one comparesthem to the Fermi diffuse γ-ray flux, especially accounting for the γγ cas-

9

cades initiated by γ-rays scattering off cosmic radiation backgrounds [42, 43].The constraints are more stringent for Galactic sources [44]. HAWC [31] isexpected to uniquely contribute to measurements of the γ-ray background inthe 10 to 100 TeV energy range, which could strongly constrain the fractionof IceCube neutrinos from Galactic origin. Among pγ photomeson produc-tion models of HENs, valuable constraints have been put on the contributionof the simpler classical GRB neutrino emission models [45, 46, 47], whileleaving open the possibility of contribution by choked GRBs or supernovaedriven by choked jets [48, 49, 47].

• Gravitational Waves and Photons from Binary Neutron Star (BNS) Mergers.-As the culmination of a long series of previous BNS GW/multi-messengersearches, e.g. [50, 51, 52, 53], the joint GW/EM detection of the transientGW/GRB 170817 was the first high significance proof of the strength ofthe joint multi-messenger technique in the GW realm [54], e.g. see Fig. 2(c). The GWs in GW/GRB 170817 showed that this was a neutron starbinary merger, providing a measurement of their masses, the distance [55]and gave constraints on the neutron star equation of state [56], while γ-rayand X-ray measurements by Fermi and Swift showed it was an off-axis shortGRB, e.g. [57]. The near simultaneous observation of EM and GW signalsfrom GW170817 showed that they both travel at the speed of light to betterthan 1 part in 1015, thereby ruling out many alternative theories of gravity.Optical observations with various telescopes showed that it also manifesteditself as a Kilonova, which is an outflow rich in the so-called r-process highatomic number nuclear elements, providing a rich interlocking picture, e.g.[58, 59, 60, 61]. For their role in the discovery of binary gravitational wavesources Barish, Thorne and Weiss received the 2017 Nobel Prize in Physics[62, 63, 64].

• High Energy Neutrinos and Gamma-rays from Blazars.- The joint neutrino[65] and electromagnetic detection [66, 67, 68] of the flaring blazar TXS0506+056 was an extremely exciting result, being the first time that a knownsource was shown to be associated (albeit at the ∼ 3σ level) with a highsignificance astrophysical high energy neutrino (Fig. 2(d). Blazars areactive galactic nuclei (AGNs), which are galaxies with a massive central blackhole powering a relativistic jet outflow pointing close to the observer line ofsight; they are classified into BL Lac objects and flat spectrum radio-quasars

10

(FSRQs), TXS 0506+056 appearing to be of the BL Lac type 2. Blazars arenotorious for exhibiting sporadic and intense gamma-ray flaring episodes, oneof which was in progress at the time the track-type neutrino was observed.Further analysis indicated that in previous years other neutrinos may havebeen associated with this source [?]. This provided valuable constraints onthe radiation mechanisms and the sources of the diffuse HEN background.Based on simple one-zone emission models where both HENs and gamma-rays originate from the same region, the neutrino is a low probability event[70, 67, 71] and based on a stacking analysis of HENs and blazars it appearsthat the blazar population as a whole may account for <∼ 10 − 30% of theentire IceCube neutrino background [72], so other sources may in any caseneed to be appealed to.

3 Emerging Questions and Challenges

• The Lack of EM or HEN counterparts of binary black hole mergers is frus-trating, with ten binary black hole mergers detected in GWs so far (as ofMarch 2019). Such emissions are expected to be faint at best in BBHs, e.g.[73, 74], but they would be very useful for a better understanding of thebinary origin and environment, as well as to get a far better localizationthan provided by the GWs, e.g. [75, 76, 77, 78]. A much larger sample ofBBHs will be needed extending to both smaller and larger masses to testthe hypothesis that BBHs provide a cosmologically important dark mattercomponent, e.g. [79, 80, 81].

• Detection of HEN from GW/EM-detected binary neutron stars would pro-vide an example of a “triple-messenger” source, and would clarify major openquestions in our understanding of these objects. Possible signals and ob-serving strategies have been discussed in, e.g., [82, 83, 78]. Expected HENfluxes are low, especially for off-axis jet viewing [84, 85], but in the bestcase they may be marginally detectable by IceCube (or more plausibly, bya future IceCube-Gen 2), and would greatly aid in clarifying the physics ofthe relativistic jet and the larger-angle slower outflows which give rise to theGRB, the afterglow, and the kilonova emission of these events.

2Recently, however, arguments have been presented [69] indicating that TXS 0506+056may be an FSRQ instead of a BL Lac object as thought previously.

11

• Confirmation or refutation of the occurrence of HEN flares in blazars throughadditional observations of TXS 0506+056 and other AGNs is urgently neededto address the origin of the IceCube background and illuminate any possibleconnection between the HEN and UHECR backgrounds. Progress in thesestudies will also require more targeted calculation of AGN neutrino produc-tion models, yielding detailed predictions for X-ray and other EM constraints.The stacking analyses of blazar EM flares against observed HENs [72, 86] aswell as theoretical arguments [87] indicate that sources other than blazarsmust provide the dominant contribution to the HEN background, and obser-vational correlation studies involving alternative source candidates may needto be undertaken, e.g. [41].

• The masses and spins of the GW-detected compact mergers offers new puz-zles. One is the origin of the ”heavy” binary stellar black holes (>∼ 30 solarmasses), it is not clear how they form and evolve. Another question is whydo the LIGO BHs have very low spins or spins mis-aligned with the orbitalangular momentum. This is in contrast to X-ray BH candidates, some ofwhich have very large spins. Also the fate of the remnant in the BNS mergerGW 170817 is unknown, e.g. how long did the remnant last before turninginto a black hole, if it finally did. These and related questions are discussedin, e.g., [27, 88]. Future GW observations could resolve this issue.

• UHECR arrival direction uncertainties are large, and UHECR arrival timesare delayed by ∼104 to 105 years relative to any simultaneous EM or neu-trino emission, so direct correlation attempts have been made only againstquasi-steady, non-bursting sources, so far unsuccessfully [2, 89, 90]. At thehighest energies, UHECR positional correlations with muon neutrino tracks,UHE neutrinos, and/or gamma-rays could lead to a better pinpointing ofthe sources. This will require much better instruments and more sensitiveneutrino/EM correlation analyses as well as much more detailed productionmodels for likely source candidates.

• Statistically significant measures of UHECR/EM/HEN/GW correlations(or lack thereof) are urgently needed, and it is also necessary to explain theUHECR spectrum and chemical composition with an appropriate distribu-tion of specific sources, e.g. [91, 92]. Observations must be fitted in statisticaldetail to model predictions of possible candidates, such as AGNs, GRBs, tidaldisruption events, clusters of galaxies, etc., and more sophisticated models

12

must be calculated, and tested against the observed diffuse neutrino andgamma-ray backgrounds.

• Theory and simulations are still in their infancy, as far as UHECR, HENand GW sources. While the HEN/EM inter-relation is in principle straight-forward, aside from non-linearities introduced by EM and hadronic cascades,an understanding of the HEN/EM inter-relations with UHECRs is signifi-cantly complicated by the fact that the latter are charged, and hence travelin complicated paths, dependent on the magnetic fields, e.g. [93]; in addi-tion, at energies above >∼ 1017 − 1018 eV, cross-section uncertainties start toset in. For BBH and BNS mergers, a lot of progress is urgently required tounderstand post-merger dynamics, the final state of the remnant, the physicsof the ejecta and how BH-NS mergers differ from BNS mergers [94]. Su-pernova simulations have also been a challenge [95, 96]. Lack of reliableGW waveforms means that we have to rely on sub-optimal techniques fortheir detection, and it also makes it far more difficult to distinguish betweendifferent collapse models/scenarios.

• The ultra-high energy neutrino range 1017−1020 eV explorations by ANITA[22] and other future experiments need to achieve at least an order of magni-tude greater sensitivity to probe the cosmogenic neutrino background. Thisis due to UHECRs interacting with the cosmic photon backgrounds, and de-generacies are induced by the effect of the UHECR source luminosity func-tion and redshift distribution as well as the CR chemical composition, e.g.[97, 98, 99]. The ANITA anomalous upward-going events [100], if confirmed,are very exciting for what they may tell us about cosmic tau-neutrinos, orpossibly about beyond the standard model physics. Concordance studies be-tween ANITA, IceCube and Auger will need to be carried out, together withsignificantly more detailed theoretical investigations, e.g. [101, 102, 103, 104].

4 Looking Ahead: New Instruments & Results Ex-pected

The spectacular results achieved, mainly in the last decade, by multi-messengerfacilities such as those in the first row of Table 1, has opened wide new vis-tas in high energy astrophysics. This has spurred the building and planningof more sophisticated and more powerful experiments and missions, geared

13

towards the elucidation of the key new questions raised. The second row ofTable 1 shows some of the new experiments currently under constructions,while the third row shows some of the next generation of experiments plannedfor the period between approximately 5 to 15 years from now.

Figure 3: Some major new detectors in the planning stage: (a) SVOM China-FranceGRB multi-wavelength follow-up satellite, exp. 2022 [105]; (b) Schematic of the CTACherenkov Telescope Array gamma-ray ground array, exp. 2024 [106]; (c) IceCube-Gen2,including current IceCube and DeepCore, and the planned high energy array, super-densePINGU sub-array and extended surface array (larger ARA radio array not shown) [107];(d) KM3NeT planned 3-4 km3 neutrino detector planned in the Mediterranean sea, whichwill include also the high-energy ARCA and low energy ORCA sub-arrays [108]; (e)Schematic of the planned EU next generation Einstein gravitational wave interferome-ter [109].

14

Figure 4: TABLE 115

• New Electromagnetic Detectors .- Among the major space-based electro-magnetic facilities coming online within the next 5 to 10 years are theChinese-French Space Variable Object Monitor (SVOM) [105] (Fig. 3(a)),designed for detecting gamma-ray, X-ray and optical transients, scheduled forlaunch in late 2022. Two other Chinese mission in preparation are GECAM[110], with an all-sky coverage of the sky aimed at detecting GW counterpartsin the 6 keV to 5 MeV energy range, scheduled for the mid-2020s; and thetime-domain Explorer-class Einstein Probe (EP) [111], with a 3600 sq.deg.field of view sensitivity at 0.5-5 KeV, scheduled for end of 2022. An Israeli-US mission called ULTRASAT [112] has been proposed, with an ultraviolet(250–280 nm), fast slewing (∼ minutes) imaging detector of 250 deg2 fieldof view, which could detect hundreds of supernovae, ∼10 BNS counterpartsper year, and ∼100 tidal disruption events per year. There is a very strongcase to be made for a US X-ray-gamma-ray satellite for providing real-timetriggers and data, which would be critical for multi-messenger studies. Apossible NASA mission that recently completed Phase A study is the ISS-TAO “Transient Astrophysics Observer” [113], on the International SpaceStation, with a GTM gamma-ray transient monitor and WFI wide field (350sq.deg.) lobster-type X-ray imager, whose prime target would be EM coun-terparts of GW sources, and which might fly by 2032. A significant rolein detecting or confirming transients of multi-messenger importance will beplayed by the ZTF (Zwicky Transient Facility) [114]) and the ASAS-SN fa-cility [115]. Also in the 5 to 10 year timeframe, the multi-national CherenkovTelescope Array (CTA) [116] (Fig. 3(b)) and the Chinese Large High Al-titude Air Shower Observatory (LHAASO) [117] ground-based facilities willsurvey the sky at TeV-PeV gamma-ray energies, e.g. [30]. Both of these usethe air Cherenkov technique, but while CTA includes steerable dishes thatprovide good angular localization, useful for point sources, LHAASO largelyobserves as the sky goes by, as does HAWC, which works better for extendedor diffuse emission. A major optical/IR survey facility instrument is theSpectroscopic Survey Telescope (LSST) [118], while the Square KilometerArray (SKA) [119] will provide milliarcsecond spatial resolution images atradio frequencies. Also in preparation are the 30-meter class TMT, ELTand GMT optical/IR ground-based telescopes [120, 121, 122].

For the early 2030s, the European Space Agency ESA is preparing a ma-jor flagship X-ray mission called ATHENA [123], which will trace the galaxyformation and metallicity evolution of the Universe with its large area de-tectors, and can study Population III GRBs. Due for final ESA selection in

16

2022 for a launch in 2032 is the smaller but nimbler, fast-slewing (∼ minute)satellite THESEUS [124], designed to discover long GRBs at redshifts z >∼ 9and seek BNS counterparts with a soft X-ray imager, an X- and gamma-rayspectroscopic imager and an 0.7-m class infra-red telescope, which will alsoprovide triggers for ATHENA. The NASA AMEGO satellite [125], sensitiveto gamma-rays from 0.2 MeV to >∼10 GeV and the German eROSITA [126]0.3–10 keV space detector will play important roles in in X- and gamma-rayastronomy, as well as in the EM detection of hidden neutrino sources. Tocomplement the above large facilities, it is extremely important to have alsovarious fast, large field-of-view robotic ground-based telescope systems whichcan follow up transient candidates within seconds after an alert is triggered.

• High and Low Energy Neutrino Detector Improvements and Plans.- Highenergy neutrino detector planned upgrades include the IceCube High En-ergy Array and the denser PINGU sub-array, as part of an extended (10Gtons) IceCube Gen2 [107] (Fig. 3(c)). In the Northern hemisphere, thecompletion in the Mediterranean sea of the KM3NeT [108] 3 to 4 Gton EUdetector is expected by ∼ 2026 (Fig. 3(d)); the error box improvementsfor muon tracks are expected to be <∼ 0.3 − 0.5 deg2. The relative advan-tages/disadvantages of ice vs. water as a Cherenkov detector medium arethat the light absorption length in ice is ∼ 100 m vs. ∼ 15 m for clear oceanwater; while the scattering length for ice is ∼ 20 m vs. ∼ 100 m for water.Also ocean water contains radioactive 40K, affecting the energy resolution sig-nal to noise ratio. Thus one gets relative better/worse energy resolution, andworse/better angular resolution in ice/water, e.g. [107, 108]. Another Gi-gaton water-based neutrino detector, Baikal-GVD [127], in the lake Baikal,Russia, is expected by ∼ 2021-22. Goals include determining large scaleanisotropies and individual source identifications by neutrinos alone or intandem with other multi-messengers. They will facilitate the use of doubletsand multiplets for source population studies, and increase the prospects foridentifying galactic sources, reliably identify tau neutrinos, and determinethe flavor composition of the high energy neutrino background.

The Hyper-Kamiokande (Hyper-K) [128] next generation megaton waterCherenkov detector, operating at MeV to GeV energies, is located in theKamioka mine (Japan) and is scheduled to begin construction in 2020. Itwill be an order of magnitude larger than its predecessor instrument Super-K,where the addition of Gadolinium to the water is providing significantlygreater sensitivity. It will be able to detect individual supernova explosions

17

out to ∼4 Mpc, roughly one every 3 to 4 years, and in 10 to 20 years it couldmeasure the relict supernova diffuse neutrino flux in the 16–30 MeV energyrange [129].

At the highest energies, 1017 eV to 1021 eV, the ANITA balloon exper-iment [22] will over the next several years undergo further sensitivity im-provements. On a longer timescale of 2022-2032, there is ongoing work andplans for much larger ground-based detectors using the Askaryan radio tech-nique for detecting neutrinos and UHECRs in the same 1017−1021 eV range,such as ARIANNA [130] and ARA [131], or a possible combination of partsof these efforts (RNO/ARA). These are aimed at detecting the cosmogenicneutrino background component produced by GZK UHECRs, as well as forprobing more deeply the nature of the decline of the UHECR spectrum be-yond 1020 eV. Another large-scale detector proposal aimed at this goal is theChinese-led GRAND 10,000 km2 array being planned for the 2025 to 2030s[132], as well as the POEMMA [133] and Trinity [134] projects.

• UHECR Detector Improvements and Extensions.- UHECR facilities un-dergoing major upgrades include the Auger-Prime addition of 1600 surfacescintillation detectors on top of the existing water Cherenkov tanks as wellas updated electronics [?]; and the Telescope Array upgrade to the TAx4configuration four-fold surface enlargement [135]. In the next 5-10 years theplanned K-EUSO experiment [136] on the International Space Station (ISS)could achieve uniform exposure across the Southern and Northern hemi-spheres of 4 × 104 km2 sr yr per year, an order of magnitude larger thanAuger or TAx4. Radio observations with LOFAR (Low Frequency RadioArray) [137] may also contribute substantially to an understanding of UHE-CRs. In the lower energy range of 1012 to >1015 eV the ISS-CREAM [138]on the International Space Station, building on earlier work of the CosmicRay Energetics And Mass (CREAM) balloon flights, as well as the futureChinese mission HERD [139], could greatly increase knowledge about thespectrum and compositions of CR nuclei with charges in the Z = 1 − 26range. In the next 10 to 15 years the planned POEMMA [133] is expectedto achieve an increase in the exposure by 100×, while the planned FASTground array [140] would provide 10x the exposure with high quality events.These advances will address the chemical composition and anisotropy issuesof UHECR, the interpretation of the surmised three CR components makingup the entire spectrum (e.g. see Fig. 4 of [141]) , the maximum energy ofgalactic CRs, and will probe in much more detail the nature of the spectral

18

cutoff, the transition from galactic to extragalactic components, the strengthof the galactic and intergalactic magnetic fields, etc.

• Gravitational Wave Detectors Improvements & Planned New Facilities.-The upgrades of LIGO and VIRGO are continuing [142], and this will reducethe 90% median credibility error box size for source identifications downto 120 − 180 deg2 (with 12-21% with ≤ 20 deg2) by the 2019+ O3 run.The Japanese KAGRA detector being built in the Kamioka mine in Japanis expected to reach a sensitivity comparable to aLIGO/aVIRGO by 2024.A new LIGO observatory is under construction in India to house the thirddetector (LIGO-India). With these additional facilities, the expected median90% localization error box sizes will be 9-12 deg2. The number of expectedbinary black hole and neutron star detections and the limiting distances areshown in Table 2 (Fig. 5).

Figure 5: TABLE 2. Plausible target detector sensitivities, giving the average detectiondistance (range) at which a 2 × 1.4M� BNS and 2 × 30M� BBH may be detected withLIGO, VIRGO and KAGRA [142].

Among the next generation of ground-based GW detectors planned for the2020-30s [143], in the US, the “A+” upgrade to the LIGO facilities has beenapproved, which should provide a further factor of two increase in detectionrange beyond the detectors’ advanced sensitivity. A further upgrade, termedLIGO Voyager, would have a new detector operating in the existing LIGO,with cryogenic mirrors in the existing LIGO vacuum envelope. This couldbring a further factor of 3 increase in the BNS range (to 1100 Mpc), with alow frequency cut-off down to 10 Hz. In the EU, the planned undergroundEinstein Telescope [109] (Fig. 3(e)), with three arms of 10 km length, willbe able to measure the GW polarization of BBHs and BNS sources from

19

distances 3 to 10 times farther than with the current design sensitivity of theLIGO/VIRGO designs. Further in the future, the Cosmic Explorer ground-based observatory [144] would use 40 km arms to achieve a further order ofmagnitude improvement in sensitivity over the 10 to 104 Hz frequency rangeand detect compact binary in-spirals throughout our Hubble volume.

In order to study the merger of the much larger (106 to 108M�) “super-massive” black holes located at the center of galaxies, the EU is planninga large space-based GW detector called eLISA [145], consisting of an inter-ferometer using three small satellites in Solar orbit. A different techniquefor the exploration of supermassive BH mergers is provided by the PulsarTiming Arrays (PTAs) [146], such as NANOGrav, PPTA, and EPTA, e.g.[147, 148]. This technique relies on measuring the time delays in the EM ra-dio signals from distant pulsars caused by the space-time variations inducedby the GW field of merging BH binaries, and is expected to yield the firststochastic (population-integrated) detections in the near future.

• Exploiting the Multi-Messenger Synergy.- The Astrophysical Multi-messengerObservatory Network (AMON) [149] is a multi-institution consortium whichhas signed MoUs with a number of observatories using different messen-gers. One major purpose is to combine disparate rare signals appearing incoincidence, so that even sub-threshold detections in one messenger, whencombined with other sub-threshold signals in other messengers, can yield areliable above-threshold detection. The other purpose is to rapidly distributeinteresting transient alerts arising from any observatory to all the other obser-vatories and the community, to facilitate rapid follow-up. The architectureof AMON consists of a central hub with radial spokes connecting to indi-vidual observatories, from which it receives individual sub-threshold (andalso above threshold) alerts, which are then subjected to analysis and/orredistributed to other observatories. This greatly increases the speed andeffectiveness of reaction to a trigger, compared to the large number of tradi-tional individual observatory-to-observatory connections. Observatories thathave signed up to the AMON network include, so far, ANTARES, Auger,Fermi LAT, Fermi GBM, HAWC, IceCube, Swift-BAT, VERITAS, and oth-ers. Live data streams from IceCube, HAWC, Fermi and Swift are being re-ceived by AMON, triggering alerts as in the TXS 0506+056 blazar neutrinoand gamma-ray flare discovery [67, 150], and other coincident sub-thresholdanalyses are being carried out, e.g. for LIGO + Swift BAT, HAWC, and oth-ers. The unique and most promising aspect of AMON is its massive emphasis

20

on leveraging multiple live sub-threshold alerts. In addition, sub-thresholdanalyses are also being carried out using archival data from different indi-vidual observatories, e.g. [151]. Other groups are also developing algorithmsand strategies for multi-messenger searches, e.g. [152].

• Theory and Simulations.- The high-quality data provided by the facilitiesoutlined above will only yield fruit insofar as it is thoroughly analyzed andinterpreted through realistic source models satisfying state-of-the-art physics(while keeping in mind the ultimate possibility of beyond the standard modelphysics). Such models must be considered at three levels. The basic levelis based on an overall conceptual picture, using analytical or semi-analyticalsource models, with their constitution, dynamics and multi-messenger radia-tion or micro-physical properties, e.g. [42, 44, 87], see [18, 153, 41, 154, 142]for reviews. The second level involves detailed numerical simulations of thedynamics of the formation of the sources and their evolution leading to thestate at which the various types of multi-messenger emission occurs, e.g.[155, 156, 157, 158]. The third level involves detailed calculations and simu-lations of the radiation physics, using large-scale numerical codes to describethe emission of multi-messengers, followed by their possible changes duringpropagation from the source to the observer, and their detailed effects onparticular types of detectors, e.g. [38, 159, 149, 160, 161, 92]. For low sourcenumbers or low signal rates, diffuse backgrounds must be calculated usingsimulated source signals convolved over cosmological luminosity and redshiftdistributions, e.g. [92].

All three of these types of calculations will have to be considerably ex-panded and refined to address and exploit the potential of the much moredetailed data expected from the above new facilities. Even for the semi-analytical studies, the increasing sensitivity and range of the detectors willmake it necessary to make use of farther and fainter reaching source lumi-nosity functions, more extensive redshift and mass distributions, interveningplasma and radiation field spectral densities, etc. The source formation anddynamics studies, which are increasingly incorporating general relativisticand magnetohydrodynamic (MHD) effects, will need to be extended to the3-dimensional regime much more commonly than before, and the use of adap-tive mesh and shock capture methods will have to be exploited and developedfurther. There will be an increasing need for a better cross-calibration of thevarious Monte Carlo methods used in simulating high energy particle interac-tions and cascades, incorporating the errors due to theoretical uncertainties,

21

especially in the extrapolation to energies beyond those of accelerator data.As another example, the atomic and nuclear physics of very heavy elementsis poorly understood and sparsely studied in the lab, yet to reliably elucidatethe sources of r-process elements in the Universe (lately ascribed largely toBNS mergers), the error estimates arising from these theoretical and exper-imental lab uncertainties will need to be quantified and taken into account.We must emphasize that nuclear and atomic physics lab experiments thatcan shed light on the r-process details are crucial for our understanding ofthis important phenomenon.

5 Conclusions and Perspectives

Some of the most important questions that will be addressed in the next 5 to10 years with upgraded GW detectors such as LIGO, VIRGO and KAGRA,as they improve sensitivity at frequencies above 0.1 kHz, are to explore indetail the lower mass range of binary black hole mergers, to test whether thefinal outcome of neutron star mergers is a massive neutron star or a blackhole, to probe the final ringdown of space-time around a newly formed blackhole, to determine the maximum mass of neutron stars, to test whether Gen-eral Relativity remains valid under extreme density and pressure conditions,and to explore the nature of the central engine of GRBs. As they push to-wards lower frequencies below 10 Hz it will be possible to probe intermediatemass BHs of 100 to 500 solar masses, important for understanding how themost massive BHs at the center of galaxies form. To get more reliable re-sults, larger interferometer arm lengths such as the ∼ 10 km of the Einsteinor Cosmic Explorer experiments will be needed. On the 10-20 year timescale,even longer arms, such as the 2.5 million km in the space-based eLISA in-terferometer, will measure frequencies ∼ 0.1 Hz which can probe the mergerof >∼ 106M� black hole mergers, important for understanding the growth ofgalaxies and clusters of galaxies in the Universe, and the existence of a pri-mordial GW background left over from the inflationary era of the early BingBang.

The neutrino detector upgrades such as PINGU in IceCube and ORCAin KM3NeT will get better statistics in the 1-10 GeV energy range to probefundamental neutrino physics issues such as the hierarchy of the mass order-ing, as well as the mixing angles between flavors, constraining the neutrinomasses and testing the existence of sterile neutrinos. These issues are likely to

22

be resolved in the next 5-10 years with the help of these and other detectors,while the next generation IceCube Gen-2 is likely to identify the sources ofthe origin of high energy neutrino, e.g. [41]. In the Northern hemisphere, the3 to 4 Gton KM3NeT detector will be able to observe HENs from our galac-tic center, where most of the (so far undetected) galactic neutrino sourcesreside. The upgraded ANITA balloons and, if funded, the ARIANNA andRNO/ARA experiments in Antarctica will make progress towards detect-ing the 1017 − 1019 eV cosmogenic neutrinos produced by UHECR protonsinteracting with the cosmic microwave background, or by UHECR nuclei in-teracting with the diffuse starlight. These experiments will also complementIceCube-Gen2 in the pursuit of the identification of tau-neutrinos. More reli-able determinations may need to wait for larger experiments such as GRANDand POEMMA, which will also address the chemical composition of the high-est energy UHECRs. The next Galactic supernova should be an ideal andimportant event for multi-messenger astrophysics, which can be exquisitelystudied by the Hyper-K [128] and JUNO [162] experiments to address fun-damental issues of neutrino oscillation and supernova physics, e.g. [163, 164].

The next upgrades of the Auger and TA UHECR arrays are expectedto answer the chemical composition question independently of the answersobtained through the above neutrino detectors, providing a much neededconsistency check. When the TAX4 array is completed its area will be com-parable to Auger’s, and being in the Northern hemisphere while Auger is inthe Southern, will prove or disprove possible anisotropies of arrival, also test-ing whether the UHECR production is dominated by a few nearby sourcesor more numerous distant ones. Together, both of them will probe the de-tails of the properties of hadronic interactions at energies three orders ofmagnitude higher than what is achievable in laboratory accelerators. Thesequestions can be more thoroughly investigated from space with K-EUSO andPOEMMA in the 10-15 years.

Future all-sky monitors such as the Large Spectroscopic Survey Telescope(LSST) in the optical and the Square Kilometer Array (SKA) in the radio,as well as the Fermi, Swift and expected (2022) SVOM satellites at keV toGeV energies will provide rapid EM triggers and accurate sky localizationas well as follow-up capability for studying cataclysmic events such as BHor NS mergers, supernovae, gamma-ray bursts, AGN flares, etc., where CRs,neutrinos and GWs are also expected in varying amounts. The strength andmix of these different messengers is model-dependent, and multiple triggers indifferent messengers, as well as model development and extensive simulations,

23

will be the key for understanding the physics of these energetic sources.The key to fully exploit the power of these new facilities is the multi-

messenger approach, due to the complementary advantages and limitationsof the different messenger particles. Cosmic rays provide unique informationabout particle accelerators well beyond terrestrial laboratory capabilities, andabout source magnetic fields and total energetics, but they do not reach usfrom beyond ∼100 Mpc and have at best poor angular resolution. Neutrinosreach us from the most distant reaches of the Universe, and probe the inner,denser regions of the most energetic and cataclysmic events, and ultra-highenergy neutrinos, being intimately linked to UHECRs, can provide uniqueclues as to how the latter reach their enormous energies. Gravitational waveobservations probe the most compact regions of high energy sources; the GWwave strain amplitude at earth is directly proportional to the source compact-ness, measured in terms of GM/c2R, and the GW luminosity goes as the fifthpower of the compactness. GWs provide excellent information about centralobject masses, angular momenta, and distances, and they will eventually bedetected from the farthest distances and earliest epochs in the Universe, butthey have modest angular angular resolution at best, and do not probe thebulk of the stellar or baryonic mass of their sources. Electromagnetic wavesprovide excellent angular resolution, velocity and redshift determination ca-pabilities, but the opacity of matter prevents them from probing the inner,denser regions of astronomical sources, while at higher gamma-ray frequen-cies they cannot reach us directly from the much grater distances probed byneutrinos or gravitational waves. By using several messengers in conjunc-tion, astronomers and physicists can hedge the foibles of each against theadvantages of the others, forging them into a formidable toolkit for probingthe highest energies and densest, most violent corners of the Universe, andin this fashion, putting our physical theories of the universe to their mostextreme and exacting possible tests.

6 Explanatory Boxes

BOX 1: Neutrino Production and OscillationsCosmic ray protons pcr interacting with target protons pt and target pho-tons γt lead to a reduced energy proton p or neutron n, and to intermediatecharged or neutral short-lived unstable particles, such as pions π±,0, muonsµ±, neutrons n and (at higher energies) Kaons K±, whose decay results in

24

neutrinos νi of different flavors i, γ-rays and electrons or positrons e±. CRnucleons, on the other hand, are primarily subject to spallation against tar-get protons and photo-disintegration against target photons, both resultingin smaller nucleons including protons, the latter subsequently undergoing thesame above mentioned interactions.

pcr + pt/γt → p/n+ π± + π0 +K± + · · · (1)

π+ → µ+ + νµ,µ+ → e+ + νe + νµ

π− → µ− + νµ,µ− → e− + νe + νµ (2)

π0 → γ + γK+/K− → µ+/µ− + νµ/νµ

n→ p+ e− + νe (3)

The primary CR proton’s mean relative energy loss per interaction, calledthe inelasticity, is κpp ∼ 0.5 for pp (or pn) and κpγ ∼ 0.2 for pγ interactions.For pcr interactions with target protons (or target neutrons) the mean ratioof charged to neutral pion secondaries is r±/0 ∼ 2, and for interactions withtarget photons it is r±/0 ∼ 1. The electrons and positrons e± quickly lose theirenergy via synchrotron or inverse Compton scattering, resulting in furtherγ-rays, so the final result of the pcr + pt/γt interactions are high energyneutrinos νi and γ-rays. The mean energy of the resulting neutrinos andγ-rays is ∼ 0.05 and ∼ 0.1 of the initial CR proton’s energy.

Once neutrinos of any flavor are produced, during their travel from thesource to the observer the neutrinos of any flavor can change into neutrinosof any of the three flavors, in the so-called neutrino oscillation phenomenon.As a result, independently of the ratio of neutrino flavors initially produceda the source where the CR interactions took place, after traveling over as-tronomical distances typically all three neutrino flavors are present at theobserver. For the neutrino sources considered here, the observable neutrinoflavors are expected to be oscillation-averages due to the large source dis-tances and the finite energy resolution of the neutrino observatories.

25

BOX 2: Multi-messengers and their inter-relation

Figure 6: Multiple messen-ger particles possibly em-anating from (A) a blazarflare; (B) a tidal disruptionevent (gravitational wavesdetectable by eLISA forsome events); (C) a longgamma-ray burst; (D) anengine-driven supernova, or(D) a supernova (gravita-tional waves detectable forGalactic events); (F) a dou-ble black hole merger, or (G)a double neutron star mergerleading to a short gamma-ray burst.

A multi-messenger source might emit two, three, or even all four differenttypes of messengers. For instance, panel (G) of Fig.6 shows a binary neutronstar merger such as the GW/GRB 170817 event, from which two types ofmulti-messengers, gravitational waves (GW) and photons (γ), were observed[54, 57, 59]. Such sources may also emit high energy neutrinos (HENs) andcosmic rays (CRs) e.g. [84, 85, 165], although for this particular sourcetheories predict fluxes too low for current detectors; if so, even closer eventsor next-generation HEN facilities will be required to observe HEN from thesesources. Another panel, (B), shows a tidal disruption event (TDE) of astar by a massive black hole; in this case shocks in the disrupted gas canaccelerate particles and lead to CRs and HENs, e.g. [166, 167, 168, 169].TDEs involving white dwarf stars and ∼1000M� black holes lead to stronglow-frequency (∼1 mHz) gravitational wave emission that could be observedby the forthcoming eLISA mission.

A solitary supermassive black hole with a jet may emit gamma-rays, HEN,and cosmic rays (panel B), as we suspect occurred during the 2017 flaringepisode of the BL Lac-type blazar TXS 0506+056 [66, 67, 68, 65, 70, 71].Here and in related sources, the coproduction of CRs, HEN, and high-energygamma-rays is anticipated, as the physics of these three messengers areclosely connected – high-energy particle acceleration and shocks lead to the

26

interaction of highly-relativistic protons (or nuclei) with ambient gas or in-tense radiation fields, resulting in neutrinos, gamma-rays, and e±. For singleobjects, even those of extreme mass and undergoing substantial accretion,relatively weak gravitational wave emission is expected as the time-varyingquadrupole moment (which requires the breaking of azimuthal symmetry) inthese cases are thought to be minimal. The sole exception would be a Galac-tic engine-driven supernova or a Galactic supernova (panels D and E), whichwould be sufficiently nearby that detection of coherent or incoherent gravi-tational waves by current and future ground-based detectors is anticipated.As far as thermal (∼ 10 MeV) neutrino detection from such supernovae, Ice-Cube is well matched, having for such emission also a roughly galactic reach.A challenge for theory is to predict the amplitude and spectrum of GW andneutrinos from different types of supernovae.

Strong GW emissions have been observed from the mergers of compact bi-nary systems, either from two merging stellar-mass black holes (panel F) [27],two merging neutron stars (panel G) [54], or (in the future) BH-NS mergers,because the final in-spiral to coalescence yields a strong gravitational wavesignal in the “sweet spot” frequency range for ground-based gravitationalwave detectors. In the case of 30M� + 30M� black hole binary systems,such coalescence events can already be observed out to ∼500 Mpc distances[142]. However, in the case of BH-BH mergers little EM flux is expected,because the ambient matter density (protons, electrons) in the vicinity ofthe binary, at the time of the merger, is typically very low. A key exceptionis the accreting supermassive black holes at the centers of massive galaxies,which are expected to merge in the wake of the coalescence of their compo-nent galaxies. These SMBH mergers are key targets for the eLISA mission,and may well exhibit accompanying EM, CR, and HEN emission [170].

Acknowledgments: We are grateful to Stephane Coutu, Douglas Cowen,Miguel Mostafa and Bangalore Sathyaprakash and the referees for usefuldiscussions and comments.

References

[1] Pierre Auger Collaboration, Nuclear Instruments and Methods in Physics ResearchA 798, 172 (Oct. 2015).

[2] Auger collaboration, ArXiv e-prints 1710.09478 (Oct. 2017), 1710.09478.

27

[3] R. Abbasi et al. (HiRes), Phys. Rev. Lett. 100, 101101 (2008), astro-ph/0703099.

[4] K. Greisen, Physical Review Letters 16, 748 (Apr. 1966).

[5] G. T. Zatsepin and V. A. Kuz’min, Soviet Journal of Experimental and TheoreticalPhysics Letters 4, 78 (Aug. 1966).

[6] N. M. Gerasimova and I. L. Rozental, JETP 14, 350 (Feb 1962).

[7] A. Aab, P. Abreu, M. Aglietta, E. J. Ahn, I. Al Samarai, I. F. M. Albuquerque,I. Allekotte, P. Allison, A. Almela, J. Alvarez Castillo, et al., Physics Letters B 762,288 (Nov. 2016), 1609.08567.

[8] D. Gora and for the Pierre Auger Collaboration, ArXiv e-prints (Nov. 2018),1811.00343.

[9] S. Petrera, arXiv e-prints (Mar. 2019), 1903.00529.

[10] H. Kawai, S. Yoshida, H. Yoshii, K. Tanaka, F. Cohen, M. Fukushima, N. Hayashida,K. Hiyama, D. Ikeda, E. Kido, et al., Nuclear Physics B Proceedings Supplements175, 221 (Jan. 2008).

[11] Telescope Array Collaboration, R. U. Abbasi, M. Abe, T. Abu-Zayyad, M. Allen,R. Azuma, and et al., ArXiv e-prints (Aug. 2018), 1808.03680.

[12] T. AbuZayyad et al., JPS Conf. Proc. 19, 011003 (2018).

[13] IceCube Collaboration, A. Achterberg, M. Ackermann, J. Adams, J. Ahrens, K. An-deen, D. W. Atlee, J. Baccus, J. N. Bahcall, X. Bai, et al., Astroparticle Physics 26,155 (Oct. 2006), astro-ph/0604450.

[14] M. G. Aartsen, R. Abbasi, Y. Abdou, M. Ackermann, J. Adams, J. A. Aguilar,M. Ahlers, D. Altmann, J. Auffenberg, X. Bai, et al., Physical Review Letters111(2), 021103, 021103 (Jul. 2013).

[15] IceCube Collaboration, Science 342 (Nov. 2013), 1311.5238.

[16] M. D. Kistler and R. Laha, Phys. Rev. Lett. 120(24), 241105, 241105 (Jun 2018),1605.08781.

[17] F. Halzen, Nature Physics 13, 232 (Mar. 2017).

[18] M. Ahlers and F. Halzen, Progress in Particle and Nuclear Physics 102, 73 (Sep.2018), 1805.11112.

[19] IceCube Collaboration, M. G. Aartsen, M. Ackermann, J. Adams, J. A. Aguilar,M. Ahlers, M. Ahrens, I. A. Samarai, D. Altmann, K. Andeen, et al., ArXiv e-prints(Oct. 2017), 1710.01179.

[20] G. Illuminati, Nuovo Cimento C Geophysics Space Physics C 41(4), 134, 134 (Jul2019).

[21] Baikal-GVD Collaboration, :, A. D. Avrorin, A. V. Avrorin, V. M. Aynutdinov,R. Bannash, I. A. Belolaptikov, V. B. Brudanin, N. M. Budnev, A. N. Dyachok,et al., arXiv e-prints (Aug. 2018), 1808.10353.

28

[22] P. Allison, O. Banerjee, L. Batten, J. J. Beatty, K. Bechtol, K. Belov, D. Z. Besson,W. R. Binns, V. Bugaev, P. Cao, et al., ArXiv e-prints (Mar. 2018), 1803.02719.

[23] A. Aab, P. Abreu, M. Aglietta, E. J. Ahn, I. Al Samarai, I. F. M. Albuquerque,I. Allekotte, P. Allison, A. Almela, and the Auger collab., Phys.Rev. D 91, 092008,092008 (May 2015), 1504.05397.

[24] B. P. Abbott, R. Abbott, R. Adhikari, P. Ajith, B. Allen, G. Allen, R. S. Amin,S. B. Anderson, W. G. Anderson, M. A. Arain, et al., Reports on Progress in Physics72(7), 076901, 076901 (Jul. 2009), 0711.3041.

[25] F. Acernese, M. Agathos, K. Agatsuma, D. Aisa, N. Allemandou, A. Allocca,J. Amarni, P. Astone, G. Balestri, G. Ballardin, et al., Classical and QuantumGravity 32(2), 024001, 024001 (Jan. 2015), 1408.3978.

[26] B. P. Abbott, R. Abbott, T. D. Abbott, M. R. Abernathy, F. Acernese, K. Ackley,C. Adams, T. Adams, P. Addesso, R. X. Adhikari, et al., Physical Review Letters116(6), 061102, 061102 (Feb. 2016), 1602.03837.

[27] The LIGO Scientific Collaboration and the Virgo Collaboration, ArXiv e-prints(Nov. 2018), 1811.12907.

[28] V. Connaughton, E. Burns, A. Goldstein, L. Blackburn, M. S. Briggs, B.-B. Zhang,C. M. Hui, P. Jenke, J. Racusin, C. A. Wilson-Hodge, et al., ArXiv e-prints1602.03920 (Feb. 2016), 1602.03920.

[29] D. J. Thompson, Comptes Rendus Physique 16, 600 (Aug. 2015), 1506.07733.

[30] T. Montaruli, arXiv e-prints (Feb. 2019), 1902.10484.

[31] F. Salesa Greus, in A. Rozanska, ed., XXXVIII Polish Astronomical Society Meeting(Aug. 2018), vol. 7, pp. 316–321.

[32] S. Casanova, in M. Bianchi, R. T. Jansen, and R. Ruffini, eds., Fourteenth MarcelGrossmann Meeting - MG14 (2018), pp. 3303–3306.

[33] K. Hirata, T. Kajita, M. Koshiba, M. Nakahata, Y. Oyama, N. Sato, A. Suzuki,M. Takita, Y. Totsuka, T. Kifune, et al., Physical Review Letters 58, 1490 (Apr.1987).

[34] E. N. Alexeyev, L. N. Alexeyeva, I. V. Krivosheina, and V. I. Volchenko, PhysicsLetters B 205, 209 (Apr. 1988).

[35] T. Haines, C. B. Bratton, D. Casper, A. Ciocio, R. Claus, M. Crouch, S. T. Dye,S. Errede, W. Gajewski, M. Goldhaber, et al., Nuclear Instruments and Methods inPhysics Research A 264, 28 (Feb. 1988).

[36] R. Davis, Rev. Mod. Phys. 75, 985 (Aug 2003), https://link.aps.org/doi/10.1103/RevModPhys.75.985.

[37] M. Koshiba, Rev. Mod. Phys. 75, 1011 (Aug 2003), https://link.aps.org/doi/10.1103/RevModPhys.75.1011.

[38] K. Fang and K. Murase, Nature Physics 14, 396 (Apr. 2018), 1704.00015.

29

[39] M. Ackermann, M. Ajello, A. Albert, W. B. Atwood, L. Baldini, J. Ballet, G. Bar-biellini, D. Bastieri, and the Fermi-LAT collab., The Astrophysical Journal 799(1),86 (jan 2015), https://doi.org/10.1088%2F0004-637x%2F799%2F1%2F86.

[40] M. Ackermann and the Fermi collab. , Physical Review Letters 116(15), 151105,151105 (Apr. 2016), 1511.00693.

[41] K. Murase and E. Waxman, Phys.Rev. D 94(10), 103006, 103006 (Nov. 2016),1607.01601.

[42] K. Murase, M. Ahlers, and B. C. Lacki, Phys.Rev. D 88(12), 121301, 121301 (Dec.2013), 1306.3417.

[43] K. Murase, D. Guetta, and M. Ahlers, Physical Review Letters 116(7), 071101,071101 (Feb. 2016), 1509.00805.

[44] M. Ahlers and K. Murase, Phys.Rev. D 90(2), 023010, 023010 (Jul. 2014), 1309.4077.

[45] R. Abbasi, Y. Abdou, T. Abu-Zayyad, M. Ackermann, J. Adams, J. A. Aguilar,M. Ahlers, D. Altmann, K. Andeen, J. Auffenberg, et al., Nature 484, 351 (Apr.2012), 1204.4219.

[46] M. G. Aartsen, M. Ackermann, J. Adams, J. A. Aguilar, M. Ahlers, M. Ahrens,D. Altmann, T. Anderson, C. Arguelles, T. C. Arlen, et al., Astrophys.J.(Lett.)805, L5, L5 (May 2015), 1412.6510.

[47] M. G. Aartsen, M. Ackermann, J. Adams, J. A. Aguilar, M. Ahlers, M. Ahrens,I. A. Samarai, D. Altmann, K. Andeen, T. Anderson, et al., ArXiv e-prints (Jul.2018), 1807.11492.

[48] P. Meszaros and E. Waxman, Physical Review Letters 87(17), 171102 (Oct. 2001),arXiv:astro-ph/0103275.

[49] K. Murase and K. Ioka, Physical Review Letters 111(12), 121102, 121102 (Sep.2013), 1306.2274.

[50] B. Abbott, R. Abbott, R. Adhikari, A. Ageev, B. Allen, R. Amin, S. B. Anderson,W. G. Anderson, and LIGO collaboration, Phys.Rev. D 72(4), 042002, 042002 (Aug2005), gr-qc/0501068.

[51] B. Abbott, R. Abbott, R. Adhikari, P. Ajith, B. Allen, G. Allen, R. Amin, S. B. An-derson, and LIGO/VIRGO collaboration, Classical and Quantum Gravity 25(11),114051, 114051 (Jun 2008), 0802.4320.

[52] J. Kanner, T. L. Huard, S. Marka, D. C. Murphy, J. Piscionere, M. Reed, andP. Shawhan, Classical and Quantum Gravity 25(18), 184034, 184034 (Sep 2008),0803.0312.

[53] B. Abbott, R. Abbott, R. Adhikari, J. Agresti, P. Ajith, B. Allen, R. Amin, S. B.Anderson, K. C. Hurley, and LIGO collaboration, Astrophys.J. 681(2), 1419 (Jul2008), 0711.1163.

30

[54] B. P. Abbott, R. Abbott, T. D. Abbott, F. Acernese, K. Ackley, C. Adams,T. Adams, P. Addesso, R. X. Adhikari, V. B. Adya, et al., Astrophys.J.(Lett.)848, L13, L13 (Oct. 2017), 1710.05834.

[55] The LIGO Scientific Collaboration, the Virgo Collaboration, B. P. Abbott, R. Ab-bott, T. D. Abbott, F. Acernese, K. Ackley, C. Adams, T. Adams, P. Addesso, et al.,Physical Review X 9, 011001, 011001 (Jan. 2019), 1805.11579.

[56] The LIGO Scientific Collaboration, the Virgo Collaboration, B. P. Abbott, R. Ab-bott, T. D. Abbott, F. Acernese, K. Ackley, C. Adams, T. Adams, P. Addesso, et al.,Phys. Rev. Lett. 121, 161101, 161101 (Oct. 2018), 1805.11581.

[57] E. Troja, L. Piro, H. van Eerten, R. T. Wollaeger, M. Im, O. D. Fox, N. R. Butler,S. B. Cenko, T. Sakamoto, C. L. Fryer, et al., ArXiv e-prints (Oct. 2017), 1710.05433.

[58] D. A. Coulter, R. J. Foley, C. D. Kilpatrick, M. R. Drout, A. L. Piro, B. J. Shappee,M. R. Siebert, J. D. Simon, N. Ulloa, D. Kasen, et al., Science 358, 1556 (Dec.2017), 1710.05452.

[59] M. M. Kasliwal, E. Nakar, L. P. Singer, D. L. Kaplan, D. O. Cook, A. Van Sistine,R. M. Lau, C. Fremling, O. Gottlieb, J. E. Jencson, et al., Science 358, 1559 (Dec.2017), 1710.05436.

[60] B. P. Abbott, R. Abbott, T. D. Abbott, F. Acernese, K. Ackley, C. Adams,T. Adams, P. Addesso, R. X. Adhikari, and LIGO-VIRGO collaboration,The Astrophysical Journal 848(2), L12 (oct 2017), https://doi.org/10.3847%

2F2041-8213%2Faa91c9.

[61] R. Margutti, E. Berger, W. Fong, C. Guidorzi, K. D. Alexander, B. D. Metzger, P. K.Blanchard, P. S. Cowperthwaite, R. Chornock, T. Eftekhari, et al., The Astrophys-ical Journal 848(2), L20 (oct 2017), https://doi.org/10.3847%2F2041-8213%

2Faa9057.

[62] R. Weiss, Annalen der Physik 531(1), 1800349 (2019),https://onlinelibrary.wiley.com/doi/pdf/10.1002/andp.201800349, https:

//onlinelibrary.wiley.com/doi/abs/10.1002/andp.201800349.

[63] B. C. Barish, Annalen der Physik 531(1), 1800357 (2019),https://onlinelibrary.wiley.com/doi/pdf/10.1002/andp.201800357, https:

//onlinelibrary.wiley.com/doi/abs/10.1002/andp.201800357.

[64] K. S. Thorne, Annalen der Physik 531(1), 1800350 (2019),https://onlinelibrary.wiley.com/doi/pdf/10.1002/andp.201800350, https:

//onlinelibrary.wiley.com/doi/abs/10.1002/andp.201800350.

[65] IceCube, and other Collaborations, Science 361(6398), eaat1378 (2018).

[66] R. Mirzoyan, The Astronomer’s Telegram 10817 (Oct. 2017).

[67] A. Keivani, K. Murase, M. Petropoulou, D. B. Fox, S. B. Cenko, S. Chaty, A. Coleiro,J. J. DeLaunay, S. Dimitrakoudis, P. A. Evans, et al., Astrophys.J. 864, 84, 84 (Sep.2018), 1807.04537.

31

[68] D. B. Fox, J. J. DeLaunay, A. Keivani, P. A. Evans, C. F. Turley, J. A. Kennea,D. F. Cowen, J. P. Osborne, M. Santander, and F. E. Marshall, The Astronomer’sTelegram 10845 (Oct. 2017).

[69] P. Padovani, F. Oikonomou, M. Petropoulou, P. Giommi, and E. Resconi,M.N.R.A.S. 484, L104 (Jan. 2019), 1901.06998.

[70] S. Gao, A. Fedynitch, W. Winter, and M. Pohl, Nature Astronomy 3, 88 (Jan. 2019),1807.04275.

[71] M. Cerruti, A. Zech, C. Boisson, G. Emery, S. Inoue, and J.-P. Lenain, M.N.R.A.S.483, L12 (Feb. 2019), 1807.04335.

[72] M. G. Aartsen, K. Abraham, M. Ackermann, J. Adams, J. A. Aguilar, M. Ahlers,M. Ahrens, D. Altmann, K. Andeen, T. Anderson, et al., Astrophys.J. 835, 45, 45(Jan. 2017), 1611.03874.

[73] R. Perna, D. Lazzati, and B. Giacomazzo, Astrophys.J.(Lett.) 821, L18, L18 (Apr.2016), 1602.05140.

[74] K. Murase, K. Kashiyama, P. Meszaros, I. Shoemaker, and N. Senno, Astro-phys.J.(Lett.) 822, L9, L9 (May 2016), 1602.06938.

[75] K. Murase, K. Kashiyama, P. Meszaros, I. Shoemaker, and N. Senno, Astrophys.J.822(1), L9, L9 (May 2016), 1602.06938.

[76] I. Bartos, Z. Haiman, Z. Marka, B. D. Metzger, N. C. Stone, and S. Marka, NatureCommunications 8, 831, 831 (Oct 2017), 1701.02328.

[77] K. E. S. Ford, I. Bartos, B. McKernan, Z. Haiman, A. Corsi, A. Keivani, S. Marka,R. Perna, M. Graham, N. P. Ross, et al., arXiv e-prints p. arXiv:1903.09529,arXiv:1903.09529 (Mar 2019), 1903.09529.

[78] A. Albert, M. Andre, M. Anghinolfi, M. Ardid, J. J. Aubert, J. Aublin, T. Avgitas,B. Baret, and LIGO-VIRGO collaboration, Astrophys.J. 870(2), 134, 134 (Jan2019).

[79] S. Bird, I. Cholis, J. B. Munoz, Y. Ali-Haımoud, M. Kamionkowski, E. D. Kovetz,A. Raccanelli, and A. G. Riess, ArXiv e-prints 1603.00464 (Mar. 2016), 1603.00464.

[80] R. Magee and C. Hanna, ArXiv e-prints 1706.04947 (Jun. 2017), 1706.04947.

[81] B. Carr, arXiv e-prints (Jan. 2019), 1901.07803.

[82] I. Bartos, C. Finley, A. Corsi, and S. Marka, Phys. Rev. Lett. 107(25), 251101,251101 (Dec 2011), 1108.3001.

[83] S. Ando, B. Baret, I. Bartos, B. Bouhou, E. Chassande-Mottin, A. r. Corsi, I. DiPalma, A. Dietz, C. Donzaud, D. Eichler, et al., Reviews of Modern Physics 85(4),1401 (Oct 2013), 1203.5192.

[84] S. S. Kimura, K. Murase, P. Meszaros, and K. Kiuchi, Astrophys.J.(Lett.) 848, L4,L4 (Oct. 2017), 1708.07075.

32

[85] S. S. Kimura, K. Murase, I. Bartos, K. Ioka, I. S. Heng, and P. Meszaros, Phys.Rev. D98(4), 043020, 043020 (Aug. 2018), 1805.11613.

[86] D. Hooper, T. Linden, and A. Vieregg, ArXiv e-prints (Oct. 2018), 1810.02823.

[87] K. Murase, F. Oikonomou, and M. Petropoulou, Astrophys.J. 865, 124, 124 (Oct.2018), 1807.04748.

[88] The LIGO Scientific Collaboration and The Virgo Collaboration, ArXiv e-prints(Nov. 2018), 1811.12940.

[89] IceCube Collaboration, Pierre Auger Collaboration, and Telescope Array Collabora-tion, Jour. Cosmology and Astro-Particle Phys. 1, 037, 037 (Jan. 2016), 1511.09408.

[90] R. Moharana and S. Razzaque, Jour. Cosmology and Astro-Particle Phys. 8, 014,014 (Aug. 2015), 1501.05158.

[91] R. Aloisio, V. Berezinsky, and P. Blasi, Journal of Cosmology and Astro-ParticlePhysics 2014, 020, 020 (Oct. 2014), 1312.7459.

[92] R. Alves Batista, R. M. de Almeida, B. Lago, and K. Kotera, Jour. Cosmology andAstro-Particle Phys. 1, 002, 002 (Jan. 2019), 1806.10879.

[93] K. Murase and M. Fukugita, Phys. Rev. D99(6), 063012 (2019), 1806.04194.

[94] D. Radice, A. Perego, K. Hotokezaka, S. A. Fromm, S. Bernuzzi, and L. F. Roberts,Astrophys.J. 869(2), 130, 130 (Dec 2018), 1809.11161.

[95] R. Glas, O. Just, H. T. Janka, and M. Obergaulinger, Astrophys.J. 873(1), 45, 45(Mar 2019), 1809.10146.

[96] D. Radice, V. Morozova, A. Burrows, D. Vartanyan, and H. Nagakura, Astrophys.J.876(1), L9, L9 (May 2019), 1812.07703.

[97] D. Seckel and T. Stanev, Phys. Rev. Lett. 95, 141101 (2005), astro-ph/0502244.

[98] K. Kotera and A. V. Olinto, Annu.Rev.Astron.Astrophys. 49, 119 (Sep. 2011),1101.4256.

[99] N. Globus, D. Allard, E. Parizot, and T. Piran, Astrophys.J. 839, L22, L22 (Apr.2017), 1703.04158.

[100] P. W. Gorham, B. Rotter, P. Allison, O. Banerjee, L. Batten, J. J. Beatty, K. Bech-tol, K. Belov, D. Z. Besson, W. R. Binns, et al., ArXiv e-prints (Mar. 2018),1803.05088.

[101] J. Alvarez-Muniz, W. R. Carvalho, K. Payet, A. Romero-Wolf, H. Schoorlemmer,and E. Zas, Phys.Rev. D 97(2), 023021, 023021 (Jan. 2018), 1707.00334.

[102] A. Romero-Wolf, P. W. Gorham, J. Nam, S. Hoover, P. Allison, O. Banerjee,L. Batten, J. J. Beatty, K. Belov, D. Z. Besson, et al., ArXiv e-prints (Sep. 2018),1810.00439.

[103] A. Connolly, P. Allison, and O. Banerjee, ArXiv e-prints (Jul. 2018), 1807.08892.

33

[104] D. B. Fox, S. Sigurdsson, S. Shandera, P. Meszaros, K. Murase, M. Mostafa, andS. Coutu, ArXiv e-prints (Sep. 2018), 1809.09615.

[105] N. Dagoneau, B. Cordier, S. Schanne, and A. Gros, in 42nd COSPAR ScientificAssembly (Jul 2018), vol. 42, pp. E1.17–46–18, 1810.12052.

[106] W. Wild, in Ground-based and Airborne Telescopes VII (Jul. 2018), vol. 10700 ofSociety of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, p.107000X.

[107] IceCube-Gen2 Collaboration, :, M. G. Aartsen, M. Ackermann, J. Adams, J. A.Aguilar, M. Ahlers, M. Ahrens, I. A. Samarai, D. Altmann, et al., ArXiv e-prints(Oct. 2017), 1710.01207.

[108] The KM3NeT Collaboration, S. Aiello, S. E. Akrame, F. Ameli, E. G. Anassontzis,M. Andre, G. Androulakis, M. Anghinolfi, G. Anton, M. Ardid, et al., ArXiv e-prints(Oct. 2018), 1810.08499.

[109] B. Sathyaprakash, M. Abernathy, F. Acernese, P. Ajith, B. Allen, P. Amaro-Seoane,N. Andersson, S. Aoudia, K. Arun, P. Astone, et al., Classical and Quantum Gravity29(12), 124013, 124013 (Jun. 2012), 1206.0331.

[110] D. Zhang, X. Li, S. Xiong, Y. Li, X. Sun, Z. An, Y. Xu, Y. Zhu, W. Peng,H. Wang, et al., Nuclear Instruments and Methods in Physics Research Section A:Accelerators, Spectrometers, Detectors and Associated Equipment 921, 8 (2019),http://www.sciencedirect.com/science/article/pii/S0168900218318291.

[111] W. Yuan, C. Zhang, H. Feng, S. N. Zhang, Z. X. Ling, D. Zhao, J. Deng, Y. Qiu, J. P.Osborne, P. O’Brien, et al., arXiv e-prints p. arXiv:1506.07735, arXiv:1506.07735(Jun 2015), 1506.07735.

[112] I. Sagiv, A. Gal-Yam, E. O. Ofek, E. Waxman, O. Aharonson, S. R. Kulkarni,E. Nakar, D. Maoz, B. Trakhtenbrot, E. S. Phinney, et al., Astron.J. 147, 79, 79(Apr 2014), 1303.6194.

[113] L. Yacobi, R. Abramov, N. Lupu, A. Vdovin, A. Kaidar, R. Rahin, A. Feigenboim,B. M. Levine, A. Osovizky, J. Camp, et al., in Space Telescopes and Instrumentation2018: Ultraviolet to Gamma Ray (Jul. 2018), vol. 10699 of Society of Photo-OpticalInstrumentation Engineers (SPIE) Conference Series, p. 106995U.

[114] M. T. Patterson, E. C. Bellm, B. Rusholme, F. J. Masci, M. Juric, K. S. Krughoff,V. Z. Golkhou, M. J. Graham, S. R. Kulkarni, G. Helou, et al., PASP 131(1), 018001(Jan. 2019), 1902.02227.

[115] C. S. Kochanek, B. J. Shappee, K. Z. Stanek, T. W.-S. Holoien, T. A. Thompson,J. L. Prieto, S. Dong, J. V. Shields, D. Will, C. Britt, et al., PASP 129(10), 104502(Oct. 2017), 1706.07060.

[116] T. Cherenkov Telescope Array Consortium, :, B. S. Acharya, I. Agudo, I. A. Samarai,R. Alfaro, J. Alfaro, C. Alispach, R. Alves Batista, J.-P. Amans, et al., ArXiv e-prints (Sep. 2017), 1709.07997.

34

[117] G. Di Sciascio and LHAASO Collaboration, Nuclear and Particle Physics Proceed-ings 279, 166 (Oct. 2016), 1602.07600.

[118] LSST Science Collaboration, P. Marshall, T. Anguita, F. B. Bianco, E. C. Bellm,N. Brandt, W. Clarkson, A. Connolly, E. Gawiser, Z. Ivezic, et al., arXiv e-prints(Aug. 2017), 1708.04058.

[119] A. M. McPherson, J. McMullin, T. Stevenson, P. Dewdney, A. Casson,L. Stringhetti, M. Deegan, P. Hekman, M. Austin, M. Harman, et al., in Ground-based and Airborne Telescopes VII (Jul. 2018), vol. 10700 of Society of Photo-OpticalInstrumentation Engineers (SPIE) Conference Series, p. 107000Y.

[120] G. H. Sanders, Journal of Astrophysics and Astronomy 34(2), 81 (Jun 2013).

[121] A. M. Varela, H. V. Ramio, J. Vernin, C. Munoz-Tunon, M. Sarazin, H. Trinquet,J. M. Delgado, J. J. Fuensalida, M. Reyes, A. Benhida, et al., Publications of theAstronomical Society of the Pacific 126(938), 412 (apr 2014), https://doi.org/10.1086%2F676135.

[122] M. Johns, P. McCarthy, K. Raybould, A. Bouchez, A. Farahani, J. Filgueira,G. Jacoby, S. Shectman, and M. Sheehan, in SPIE (2012), vol. 8444, https:

//doi.org/10.1117/12.926716.

[123] K. Nandra, in J.-U. Ness and M. Ehle, eds., The X-ray Universe 2011 (Aug. 2011),p. 022.

[124] L. Amati, P. O’Brien, D. Gotz, E. Bozzo, C. Tenzer, F. Frontera, G. Ghirlanda, andet al., Advances in Space Research 62, 191 (Jul. 2018), 1710.04638.

[125] A. Moiseev and Amego Team, International Cosmic Ray Conference 35, 798, 798(Jan. 2017).

[126] P. Predehl, R. Andritschke, V. Babyshkin, W. Becker, W. Bornemann,H. Brauninger, H. Brunner, T. Boller, V. Burwitz, W. Burkert, et al., in SpaceTelescopes and Instrumentation 2016: Ultraviolet to Gamma Ray (Jul. 2016), vol.9905 of Procs. SPIE, p. 99051K.

[127] Baikal-GVD Collaboration, :, A. D. Avrorin, A. V. Avrorin, V. M. Aynutdinov,R. Bannash, I. A. Belolaptikov, V. B. Brudanin, N. M. Budnev, A. A. Doroshenko,et al., arXiv e-prints (Aug. 2018), 1808.10353.

[128] Hyper-Kamiokande Proto-Collaboration, :, K. Abe, K. Abe, H. Aihara, A. Aimi,R. Akutsu, C. Andreopoulos, I. Anghel, L. H. V. Anthony, et al., arXiv e-prints(May 2018), 1805.04163.

[129] J. Migenda and Hyper-Kamiokande Proto-Collaboration, in Proceedings of the Eu-ropean Physical Society Conference on High Energy Physics. 5-12 July (Jul 2017),p. 20, 1710.08345.

[130] S. W. Barwick, D. Z. Besson, A. Burgman, E. Chiem, A. Hallgren, J. C. Hanson,S. R. Klein, S. A. Kleinfelder, A. Nelles, C. Persichilli, et al., Astroparticle Physics90, 50 (Apr. 2017), 1612.04473.

35

[131] P. Allison, R. Bard, J. J. Beatty, D. Z. Besson, C. Bora, C.-C. Chen, C.-H. Chen,P. Chen, A. Christenson, A. Connolly, et al. (The ARA collaboration), Phys. Rev.D 93, 082003 (Apr 2016), https://link.aps.org/doi/10.1103/PhysRevD.93.

082003.

[132] GRAND Collaboration, J. Alvarez-Muniz, R. Alves Batista, A. Balagopal V., J. Bol-mont, M. Bustamante, W. R. Carvalho, D. Charrier, I. Cognard, V. Decoene, et al.,ArXiv e-prints (Oct. 2018), 1810.09994.

[133] A. V. Olinto, J. H. Adams, R. Aloisio, L. A. Anchordoqui, D. R. Bergman, M. E.Bertaina, P. Bertone, M. J. Christl, S. E. Csorna, J. B. Eser, et al., InternationalCosmic Ray Conference 301, 542, 542 (Jan 2017), 1708.07599.

[134] A. Nepomuk Otte, arXiv e-prints (Nov. 2018), 1811.09287.

[135] H. Sagawa and Telescope Array Collaboration, in 34th International Cosmic RayConference (ICRC2015) (Jul. 2015), vol. 34 of International Cosmic Ray Confer-ence, p. 657.

[136] M. Casolino, A. Belov, M. Bertaina, T. Ebisuzaki, M. Fukushima, P. Klimov, M. I.Panasyuk, P. Picozza, H. Sagawa, and K. Shinozaki (JEM-EUSO), PoS ICRC2017,368 (2018).

[137] T. Winchen, A. Bonardi, S. Buitink, A. Corstanje, H. Falcke, B. Hare, J. Horandel,P. Mitra, K. Mulrey, A. Nelles, et al., arXiv e-prints (Mar. 2019), 1903.08474.

[138] E. S. Seo, T. Anderson, D. Angelaszek, S. J. Baek, J. Baylon, M. Buenerd, M. Cop-ley, S. Coutu, L. Derome, B. Fields, et al., Advances in Space Research 53, 1451(May 2014).

[139] S. N. Zhang, O. Adriani, S. Albergo, G. Ambrosi, Q. An, T. W. Bao, R. Battiston,X. J. Bi, and et al., in Space Telescopes and Instrumentation 2014: Ultravioletto Gamma Ray (Jul 2014), vol. 9144 of Society of Photo-Optical InstrumentationEngineers (SPIE) Conference Series, p. 91440X, 1407.4866.

[140] T. Fujii, M. Malacari, J. A. Bellido, P. Horvath, M. Hrabovsky, J. Jiang, D. Mandat,A. Matalon, J. N. Matthews, P. Motloch, et al., in European Physical Journal Web ofConferences (Mar. 2017), vol. 136 of European Physical Journal Web of Conferences,p. 02015.

[141] T. K. Gaisser, T. Stanev, and S. Tilav, Frontiers of Physics 8(6), 748 (Dec 2013),1303.3565.

[142] B. P. Abbott, R. Abbott, T. D. Abbott, M. R. Abernathy, F. Acernese, K. Ackley,C. Adams, T. Adams, P. Addesso, R. X. Adhikari, et al., Living Reviews in Relativity21, 3, 3 (Apr. 2018), 1304.0670.

[143] B. P. Abbott, R. Abbott, T. D. Abbott, M. R. Abernathy, K. Ackley, C. Adams,P. Addesso, R. X. Adhikari, and et al., Classical and Quantum Gravity 34(4), 044001(jan 2017), https://doi.org/10.1088%2F1361-6382%2Faa51f4.

36

[144] B. P. Abbott, R. Abbott, T. D. Abbott, M. R. Abernathy, K. Ackley, C. Adams,P. Addesso, R. X. Adhikari, V. B. Adya, C. Affeldt, et al., Classical and QuantumGravity 34(4), 044001, 044001 (Feb. 2017), 1607.08697.

[145] A. Klein, E. Barausse, A. Sesana, A. Petiteau, E. Berti, S. Babak, J. Gair,S. Aoudia, I. Hinder, F. Ohme, et al., Phys.Rev. D 93(2), 024003, 024003 (Jan.2016), 1511.05581.

[146] K. Schutz and C.-P. Ma, M.N.R.A.S. 459, 1737 (Jun. 2016), 1510.08472.

[147] G. Hobbs and S. Dai, arXiv e-prints p. arXiv:1707.01615, arXiv:1707.01615 (Jul2017), 1707.01615.

[148] Z. Arzoumanian, P. T. Baker, A. Brazier, S. Burke-Spolaor, S. J. Chamberlin,S. Chatterjee, B. Christy, J. M. Cordes, N. J. Cornish, F. Crawford, et al., The Astro-physical Journal 859(1), 47 (may 2018), https://doi.org/10.3847%2F1538-4357%2Faabd3b.

[149] A. Keivani, H. Ayala, and J. DeLaunay, ArXiv e-prints (Aug. 2017), 1708.04724.

[150] H. A. Ayala Solares, S. Coutu, D. F. Cowen, J. J. DeLaunay, D. B. Fox, A. Keivani,M. Mostafa, K. Murase, F. Oikonomou, M. Seglar-Arroyo, et al., arXiv e-prints(Mar. 2019), 1903.08714.

[151] C. F. Turley, D. B. Fox, A. Keivani, J. J. DeLaunay, D. F. Cowen, M. Mostafa, H. A.Ayala Solares, and K. Murase, Astrophys.J. 863, 64, 64 (Aug. 2018), 1802.08165.

[152] S. Countryman, A. Keivani, I. Bartos, Z. Marka, T. Kintscher, R. Corley, E. Blau-fuss, C. Finley, and S. Marka, arXiv e-prints (Jan. 2019), 1901.05486.

[153] B. F. Schutz, Philosophical Transactions of the Royal Society of London Series A376, 20170279, 20170279 (May 2018), 1804.06308.

[154] P. Meszaros, Annual Review of Nuclear and Particle Science 67, 45, 45 (Nov. 2017),1708.03577.

[155] M. Shibata, K. Kiuchi, and Y.-i. Sekiguchi, Phys.Rev. D 95(8), 083005, 083005(Apr. 2017), 1703.10303.

[156] P. J. Easter, P. D. Lasky, A. R. Casey, L. Rezzolla, and K. Takami, arXiv e-prints(Nov. 2018), 1811.11183.

[157] T. Parsotan, D. Lopez-Camara, and D. Lazzati, Astrophys.J. 869, 103, 103 (Dec.2018), 1805.10327.

[158] H. van Eerten, International Journal of Modern Physics D 27, 1842002, 1842002-314(2018), 1801.01848.

[159] N. Senno, K. Murase, and P. Meszaros, Phys.Rev. D 93(8), 083003, 083003 (Apr.2016), 1512.08513.

[160] K. Hotokezaka, P. Beniamini, and T. Piran, International Journal of Modern PhysicsD 27, 1842005, 1842005 (2018), 1801.01141.

37

[161] D. Biehl, D. Boncioli, A. Fedynitch, and W. Winter, Astron.Astrophys. 611, A101,A101 (Apr. 2018), 1705.08909.

[162] J.-S. Lu, Y.-F. Li, and S. Zhou, Phys. Rev. D 94, 023006 (Jul 2016), https://

link.aps.org/doi/10.1103/PhysRevD.94.023006.

[163] J. F. Beacom, Annual Review of Nuclear and Particle Science 60(1), 439 (2010),https://doi.org/10.1146/annurev.nucl.010909.083331.

[164] I. Tamborra and K. Murase, Space Science Reviews 214, 31, 31 (Feb. 2018).

[165] S. S. Kimura, K. Murase, and P. Meszaros, Astrophys.J. 866, 51, 51 (Oct. 2018),1807.03290.

[166] C. Guepin, K. Kotera, E. Barausse, K. Fang, and K. Murase, ArXiv e-prints (Nov.2017), 1711.11274.

[167] D. Biehl, D. Boncioli, C. Lunardini, and W. Winter, ArXiv e-prints 1711.03555(Nov. 2017), 1711.03555.

[168] N. Senno, K. Murase, and P. Meszaros, ArXiv e-prints 1612.00918 (Dec. 2016),1612.00918.

[169] X.-Y. Wang and R.-Y. Liu, ArXiv e-prints 1512.08596 (Dec. 2015), 1512.08596.

[170] A. Klein, E. Barausse, A. Sesana, A. Petiteau, E. Berti, S. Babak, J. Gair,S. Aoudia, I. Hinder, F. Ohme, et al., Phys.Rev. D 93(2), 024003, 024003 (Jan.2016), 1511.05581.

38