High-Energy Neutrinos from Magnetized Coronae of Active Galactic Nuclei and Prospects for Identification of Seyfert Galaxies and Quasars in Neutrino Telescopes
DDraft version February 10, 2021
Typeset using L A TEX twocolumn style in AASTeX62
High-Energy Neutrinos from Magnetized Coronae of Active Galactic Nuclei andProspects for Identification of Seyfert Galaxies and Quasars in Neutrino Telescopes
Ali Kheirandish, ∗ Kohta Murase, † and Shigeo S. Kimura ‡ Department of Physics; Department of Astronomy & Astrophysics; Center for Multimessenger Astrophysics, Institute for Gravitationand the Cosmos, The Pennsylvania State University, University Park, PA 16802, USA Department of Physics; Department of Astronomy & Astrophysics; Center for Multimessenger Astrophysics, Institute for Gravitationand the Cosmos, The Pennsylvania State University, University Park, PA 16802, USA&Center for Gravitational Physics, Yukawa Institute for Theoretical Physics, Kyoto University, Kyoto, Kyoto 606-8502, Japan Frontier Research Institute for Interdisciplinary Sciences, Tohoku University, Sendai 980-8578, Japan&Astronomical Institute, Tohoku University, Sendai 980-8578, Japan
ABSTRACTParticles may be accelerated in magnetized coronae via magnetic reconnections and/or plasma turbu-lence, leading to high-energy neutrinos and soft gamma rays. We evaluate the detectability of neutrinosfrom nearby bright Seyfert galaxies identified by X-ray measurements. In the disk-corona model, wefind that NGC 1068 is the most promising Seyfert galaxy in the Northern sky, where IceCube is themost sensitive, and show prospects for the identification of aggregated neutrino signals from Seyfertgalaxies bright in X-rays. Moreover, we demonstrate that nearby Seyfert galaxies are promising tar-gets for the next generation of neutrino telescopes such as KM3NeT and IceCube-Gen2. For KM3NeT,Cen A can be the most promising source in the Southern sky if a significant fraction of the observedX-rays come from the corona, and it could be identified in few years of KM3NeT operation. Ourresults reinforce the idea that hidden cores of supermassive black holes are the dominant sources ofthe high-energy neutrino emission and underlines the necessity of better sensitivity to medium-energyranges in future neutrino detectors for identifying the origin of high-energy cosmic neutrinos.
Keywords: astroparticle physics – galaxies: active – galaxies: jets – gamma rays: galaxies – neutrinos– radiation mechanisms: non-thermal INTRODUCTIONThe observation of high-energy cosmic neutrinos inIceCube (Aartsen et al. 2013a,b) has revealed thathadronic interactions play a major role in the non-thermal emission in the high-energy universe. While themagnitude of the measured spectrum at high energies, (cid:38)
100 TeV, was found to be consistent with some the-oretical predictions for astrophysical neutrinos (Loeb &Waxman 2006; Murase et al. 2008; Kotera et al. 2009;Murase et al. 2013), the emergence of an order of mag-nitude higher flux at medium energies in the 10 TeVrange (Aartsen et al. 2015a,b, 2020a) pointed towardssources with efficient neutrino emission. While the dom-inant sources of high-energy cosmic neutrinos are yet ∗ [email protected] † [email protected] ‡ [email protected] to be identified, the multimessenger data indicate thattheir sources should be opaque to high-energy γ -raysin the GeV-TeV range, i.e., hidden to the γ -ray tele-scopes operating at these energies (Murase et al. 2016;Capanema et al. 2020a,b). Otherwise, the contributionof the γ -ray flux leaving transparent sources to the dif-fuse isotropic γ -ray background (IGRB) would create anexcess that would overshoot the measured diffuse flux by Fermi -LAT (Ackermann et al. 2015).Cores of the active galactic nuclei (AGN), which areoptically thick for GeV-TeV γ -rays are one of the bestcandidates as the source of the high-energy neutrino fluxat medium energies. The isotropic distribution of thearrival direction of high-energy neutrinos, together withthe constraints on the Galactic component of the neu-trino flux, points at extragalactic sources hidden in γ -rays as the origin of this flux. Moreover, in order toproduce this level of flux, efficient neutrino production a r X i v : . [ a s t r o - ph . H E ] F e b is required. Such conditions can be met with large col-umn densities of target material and radiation, whichform dense environments suppressing the γ -ray flux atthe site of production.In Seyfert galaxies, accretion dynamics and magneticdissipation will form a magnetized corona above thedisk, see e.g., Miller & Stone (2000); Merloni & Fabian(2001); Liu et al. (2002a); Blackman & Pessah (2009); Io& Suzuki (2014); Jiang et al. (2014); Jiang et al. (2019).Thanks to the dense environments near the supermas-sive black holes, the accelerated cosmic rays (CRs) inthe corona interact efficiently with the gas and tar-get photons, producing charged and neutral pions thatpromptly decay to neutrinos and γ -rays. The neutrinoswill escape the environment while GeV-TeV γ -rays arecascaded down inside the source.The disk-corona model for high-energy neutrino emis-sion from the core of AGNs can successfully accommo-date the flux of cosmic neutrinos at medium energies inthe 10 −
100 TeV range (Murase et al. 2020). In ad-dition, modeling of the high-energy emission from thehidden cores of AGNs presents a robust connection to γ -ray emission at MeV energies.The dominant role of AGN cores to the flux of high-energy neutrinos at medium energies has been reinforcedby the IceCube’s most recent search for the sources ofhigh-energy neutrinos using 10 years of data (Aartsenet al. 2020b). The all-sky (untriggered) and the sourcelist (triggered) time-integrated analysis of the data findsthe direction of NGC 1068 as the most significant pointin the sky. The excess in the direction of NGC 1068is found local pre-trial p-value of 1 . × − , yielding a2.9 σ after correcting for trials.NGC 1068, aka Messier 77, is a nearby, bright Seyfert2 galaxy. The observed infrared luminosity of NGC 1068is similar to starburst galaxies. It has been consideredas a potential CR accelerator and motivated study ofneutrino and γ -ray emission, see , e.g., Yoast-Hull et al.(2014); Murase & Waxman (2016); Liu et al. (2018);Lamastra et al. (2019). The best-fit neutrino flux mea-sured by IceCube, however, exceeds the observed γ -rayemission by Fermi -LAT (Ackermann et al. 2012). There-fore, the models that are built upon the measured γ -rayemission in Fermi cannot accommodate the reportedneutrino flux. In the meantime, neutrino flux predic-tions from starburst activity are directly correlated withthe
Fermi γ -rays and face a similar obstacle.In this paper, we present the expected neutrino emis-sion from the brightest Seyfert galaxies in the disk-corona model by incorporating the measured intrinsicX-ray luminosities. We estimate the parameters relevantfor emission from the magnetized coronae and establish the parameter space compatible with the reported neu-trino spectrum of NGC 1068 and the medium-energyexcess in the cascade data. We further employ the pa-rameters that can accommodate the emission from NGC1068 for the brightest Seyfert galaxies and study theirlikelihood of observation, individually and when stackedtogether.While providing a consistent level of neutrino flux, ourmodeling finds NGC 1068 as the most promising sourcein the Northern hemisphere, where IceCube is most sen-sitive. While the majority of the bright sources residein the Southern sky, we show that the stacking analy-sis has a good chance of identifying neutrino emissionfrom this class of sources. In addition, the significanceof the search for neutrino emission from such analysiscan inform the principal mechanism for the high-energyemission.We use the predicted neutrino emission to examine theprospects for observation of the nearby, bright Seyfertgalaxies in future neutrino telescopes, KM3NeT and thenext generation of IceCube at the South Pole: IceCube-Gen2.While indicating that Seyfert galaxies are the dom-inant contributors to the high magnitude flux of cos-mic neutrinos at energies below 100 TeV, our predictionprovides a testable scenario in current and future neu-trino telescopes. The subset of bright Seyfert galaxies islikely to be observed with the continuous operation ofthe IceCube detector and commissioning of the neutrinotelescopes in the Northern hemisphere.In the next section, we discuss details of accelera-tion mechanism and neutrino production in the mag-netized corona model. Then, in Sec. 3 we present theneutrino emission under 3 distinct scenarios from NGC1068 and the brightest Seyfert galaxies identified by theX-ray surveys. Subsequently, in Sec. 4, we will studythe prospects for identification of the neutrino emis-sion in current and future neutrino telescopes. Finally,we will discuss the implications of observing neutrinosfrom these sources and the relation to the total flux inSecs. 5.2. MAGNETIZED CORONA MODELProgress in X-ray observations of AGN led to theestablishment of the magnetized coronal paradigm(Haardt & Maraschi 1991; Liu et al. 2002b; Ricci et al.2018). This is partly confirmed by recent magneto-hydrodynamic (MHD) simulations (e.g., Io & Suzuki2014; Jiang et al. 2019). The magnetorotational insta-bility (MRI; Balbus & Hawley 1991) supports that thecoronae above the optically thick accretion disk are nat-urally expected to be hot, magnetized, and turbulent. igh-Energy Neutrinos from Seyfert Galaxies
Stochastic acceleration scenario
It is widely believed that turbulent magnetic fieldsgenerated by the MRI is responsible for the angular mo-mentum transport in accretion flows, which has beenconfirmed by MHD simulations (Stone & Pringle 2001;Machida & Matsumoto 2003; Ohsuga & Mineshige 2011;Narayan et al. 2012; Porth et al. 2019). Turbulence canalso be generated through magnetic reconnections re-sulting from magnetic fields amplified by the MRI. CRsin the MHD turbulence randomly change their energiesthrough interactions with MHD waves. If the acceler-ation in momentum space homogeneously occurs with-out significant trapping (Lemoine & Malkov 2020), thephenomenon can be described by the following diffusionequation in energy space: ∂ F p ∂t = 1 ε p ∂∂ε p (cid:32) ε p D ε p ∂ F p ∂ε p + ε p t cool F p (cid:33) − F p t esc + ˙ F p, inj , (1)where F p is the CR distribution function ( dN/dε p =4 πp F p /c ), D ε p is the diffusion coefficient in energyspace, t cool is the cooling timescale, and t esc is theescape timescale. The injection function is given by˙ F p, inj = f inj L X δ ( ε p − ε inj ) / [4 π ( ε inj /c ) V ] , where ε inj isthe injection energy and f inj is the injection fraction at ε inj , L X is the X-ray luminosity, V is the volume of thecorona, and δ ( x ) is the delta function. The diffusion co-efficient is assumed to scale as D ε p ∝ ε qp , where q is thepower-law index of the diffusion coefficient in momen-tum space. If the gyro-resonant scattering is dominant,which is not necessarily the case, it corresponds to thespectral index of the turbulence power spectrum. Fordemonstrative purposes, we use q = 5 / q = 3 / q = 2 (that is a hard-spheretype; see, e.g., Kimura et al. 2019b) may be possible. Ascooling processes, we take into account Bethe-Heitler, pp inelastic collisions, and photomeson ( pγ ) productionprocesses. We consider advective (infall to a centralblack hole) and diffusive escapes. In this scenario, weintroduce two parameters, the CR pressure to the ther-mal pressure P CR /P th and turbulent strength η − . The former determines the normalization of CRs, whereasthe latter is related to the maximum energy of CRs.Note that we use the thermal pressure at the virial tem-perature, i.e., P th = n p kT vir , where n p is the thermalproton density and T vir is the virial temperature, andthe realistic thermal pressure can be reduced if the iontemperature is lower than T vir . See Supplemental Mate-rial of Murase et al. (2020) for details (see also SectionIII-B of Kimura et al. 2019a, in the context of radiativelyinefficient accretion flows).2.2. Magnetic reconnection scenario
Recent Particle-In-Cell (PIC) Simulations revealedthat relativistic magnetic reconnections in the ion-electron plasma, where the ion magnetization param-eter σ i = B / (8 πn p m p c ) (cid:38)
1, can accelerate non-thermal relativistic particles very efficiently, given thatthe plasma β , which is defined by β ≡ πn p kT p /B , issmaller than ∼ .
01 (Guo et al. 2016; Werner et al. 2018;Ball et al. 2018). For AGN coronae, β can be sufficientlysmall (e.g., β (cid:46) −
3) but we typically expect that theplasma is semi-relativistic, σ i = 2 kT p / ( m p c β ) < ∼ m p c , and further ac-celeration via turbulence would be necessary (Comisso& Sironi 2018; Zhdankin et al. 2019; Wong et al. 2020;Lemoine & Malkov 2020). Not only the stochastic accel-eration but also the first-order Fermi acceleration mayoperate due to scatterings with reconnection outflows(e.g., Pisokas et al. 2018; Hoshino 2012). In this work,for the purpose of phenomenological studies, we sim-ply assume that the injection spectrum of protons isgiven by a single power-law with an exponential cutoff,providing the power-law index, s , as a parameter.To calculate neutrino emission, we solve the transportequation for the non-thermal protons with single-zoneand steady-state assumptions: − ddε p (cid:18) ε p N ε p t cool (cid:19) = ˙ N ε p , inj − N ε p t esc , (2)where N ε p = dN/dε p , ˙ N ε p , inj ∝ ε − sp is the injectionterm.We consider the same cooling and escape processesas those in the stochastic acceleration scenario. Theproton cutoff energy of the injection spectrum is givenby the balance between cooling and acceleration. Theacceleration time is phenomenologically given by t acc ≈ η acc r L /c , where η acc is the acceleration efficiency param-eter, r L is the Larmor radius. The normalization of CRsare provided such that (cid:82) ˙ N inj ε p dε p = (cid:15) CR ˙ M c , where (cid:15) CR is the energy fraction carried by CRs and ˙ M is themass accretion rate in the accretion disk. Note that thecorona is connected to the accretion disk through mag-netic fields, and a significant fraction of the accretionenergy may be dissipated in the disk-corona interface.The previous works focused on the stochastic acceler-ation (Murase et al. 2020) or shock acceleration mech-anism (see below), and the magnetic reconnection sce-nario was not considered in detail. In this work, weshow that the diffuse neutrino flux can be explained bythe magnetic reconnection scenario in the AGN coronamodel. In Sec. 6 we discuss the possibility of explainingthe diffuse neutrino flux at medium energies with thisscenario and comment on the feasibility of identifyingthis scenario in the near future.2.3. Remarks on other acceleration scenarios
Alternatively, accretion shocks have been discussed asa CR production site in the accretion flows (Begelmanet al. 1990; Stecker et al. 1991). Originally, this scenariowas proposed to explain X-ray observations of AGNsby CR-induced electromagnetic cascades (Berezinskii &Ginzburg 1981; Zdziarski 1986). However, cutoff andsoftening features were discovered in bright AGNs in1990s (Maisack et al. 1993; Zdziarski et al. 2000), whichruled out the hadron-induced electromagnetic cascadescenario for the observed X-rays. The accretion shockcould still be viable to be a neutrino production site aslong as the CR-induced cascade flux is below the X-rayand gamma-ray data (Stecker 2005, 2013; Inoue et al.2019). However, there are several problems in this sce-nario. First, the accretion shocks are not seen in anyglobal simulations dedicated for accretion flows (e.g.,Narayan et al. 2012; Kimura et al. 2014; Jiang et al.2019). Although there are solutions with an accretionshock in the one-dimensional hydrodynamic equationsystem with a steady state assumption (Becker et al.2011; Chattopadhyay & Kumar 2016), such solutionsare not realized in global simulations. The second pointis the angular momentum of the accreting matter. Weexpect a large specific angular momentum at an outerscale, which prevents the matter from freely falling toa vicinity of the supermassive black hole. An anoma-lously efficient angular momentum transport mechanismis necessary to form an accretion shock, which is con-trary to the free-fall assumption made by Inoue et al.(2019). Lastly, the accretion rate through the accretionshocks is assumed to be extremely high to explain theIceCube data (Inoue et al. 2019). It is comparable toor even higher than the accretion rate expected in thegeometrically thin, optically thick disk, which requirestwo accretion components with totally different angularmomentum distributions. Note that P CR < . P th is required in this scenario, otherwise hot coronae cannotbe maintained.We also note that the Bethe-Heitler process, whichwas often ignored, is relevant. The 10 −
100 TeV neu-trino flux from the photomeson production with coro-nal X-rays is suppressed by Bethe-Heitler interactionswith disk photons (Murase et al. 2020). On the con-trary, given that shock acceleration leads to higher max-imum energies, PeV neutrinos are mainly produced viathe photomeson production with disk photons (Steckeret al. 1991).Note that Kalashev et al. (2015) assumed electric ac-celeration in a gap formed in the magnetosphere. Thegap formation is possible only in low-luminosity AGNssuch as M87, but it is likely to be screened in the caseof Seyferts and quasars because the accretion rate is sohigh that the plasma satisfies the quasi-neutrality con-dition (Levinson & Rieger 2011). NEUTRINO EMISSION FROM BRIGHTSEYFERT GALAXIESThe high-energy neutrino flux modeled in this studyincorporates both hadronuclear ( pp ) and photohadronic( pγ ) interactions. The inclusion of both the processes inthe production of high-energy neutrinos, together withthe cooling processes in the source, results in featuresin the neutrino flux from Seyfert galaxies. Thus, thespectrum of high-energy neutrinos would deviate fromsimplistic single power laws.The stochastic acceleration scenario for the produc-tion of high-energy neutrinos may yield a spectrum thatpeaks at ∼
10 TeV energies and steeply falls around ∼
100 TeV. This feature deviates sharply from the singlepower-law assumption for the spectrum incorporated inthe search for neutrino sources. Meanwhile, the otherscenario considered here, the magnetic reconnection sce-nario, can yield a neutrino spectrum that mimics apower-law with a cutoff at high energies. Neutrino pro-duction in both scenarios may also receive a significantcontribution from the photomeson production process.This often leads to a bumpy feature before the cutoff inthe spectrum.In this section, we first estimate the neutrino fluxfrom NGC 1068 and adopt parameters in each accel-eration scenario that can describe the reported neutrinospectrum from the source while maintaining other con-straints on the parameters in a physical range. The mainparameter in our modeling that constrains the neutrinospectrum is the ratio of the CR to thermal pressures. Af-ter finding the parameters for the neutrino flux that canaccommodate the neutrino flux from NGC 1068, we es-timate the neutrino flux for the brightest Seyfert galax- igh-Energy Neutrinos from Seyfert Galaxies L X (cid:38) erg/s. Less bright sources suchas Sgr A*, with L X < erg / s are better character-ized by measurements at lower-energy bands. Through-out this work, we estimate the distance of the Seyfertgalaxies based on their redshift provided by BASS. Toevaluate the luminosity distance, we employ the follow-ing cosmological parameters: H = 70 km s − Mpc − ,Ω m = 0 .
3, Ω Λ = 0 . Neutrino emission from NGC 1068
NGC 1068 is a Seyfert 2 galaxy at z = 0 . (cid:38) cm − ,implying a Compton thick environment. The high den-sity environment at the core of NGC 1068 and the highlevel of mass accretion rate provide favorable environ-ments for efficient production of high-energy neutrinos.In order to estimate the neutrino spectrum in theAGN corona model for NGC 1068, we use the intrin-sic X-ray luminosity, L X , suggested by Swift -BAT AGNSpectroscopic Survey (BASS) (Ricci et al. 2017). TheBASS catalog provides X-ray fluxes and luminosities ina wide range, covering both soft (below 2 keV) and hard(above 10 keV) X-rays. Here, we use measured spectrumand luminosity in the 2-10 keV band, which provides amedium range. This choice suits the disk-corona model-ing, as the abundance of data in this energy band enablesus to use empirical correlation.The intrinsic X-ray luminosity, L X measured from thedirection of NGC 1068 by BAT is ∼ erg / s for the2-10 keV band, although it could be larger depending onestimates of the column density. We use this luminosityto model the neutrino spectrum from NGC 1068 in threedifferent scenarios.Assuming a single power-law spectrum, the Ice-Cube Collaboration has reported a best-fit flux of3 × − TeV − cm − s − at 1 TeV with an index of (cid:39) stochastic acceleration scenario. As mentioned earlier, E ν [GeV]10 − − − − − − E Φ ν µ + ¯ ν µ [ G e V c m − s − ] Stochastic(high CR pressure)MagneticReconnection
IceCube 10 yr
NGC 1068 ( L X = 10 erg s − ) Stochastic(modest CR pressure)
Figure 1.
Modeled neutrino spectrum for NGC 1068 com-pared to the best fit flux (yellow band) reported by the Ice-Cube Collaboration 10 yr point source study (Aartsen et al.2020b). The red line shows the expected flux in the stochas-tic acceleration scenario matching IceCube’s best fit at TeVs.The purple line depicts the flux that would give the medium-energy neutrino flux, compatible with the total neutrino fluxreported in the cascade analysis (Aartsen et al. 2020a). Theblue line presents the flux expected for the magnetic recon-nection scenario. in this scenario, the neutrino spectrum has a more com-plicated shape than a single power-law. Accommodat-ing the IceCube flux at TeV energies requires a relativelyhigh normalization while the spectrum has to cutoff fastenough that the spectrum drops around 100 TeV. Suchconditions would result in a high level of CR pressure inthe corona model.In order to maintain realistic scenarios, we restrictourselves to the range of parameters for which the ra-tio of the CR pressure ( P CR ) to the thermal pressure( P th ) is bound to less than 0.5. In this limit, the non-thermal energy is equal to half of the gravitational bind-ing energy at the coronal radius without leaving roomfor thermal particles. Although the coronal plasma maybe heated more through magnetic fields connected to theinner disk, we assume 0.5 as the maximal case in thiswork, and the neutrino spectrum peaks at ∼ E ν [GeV] − − e v e n t s / b i n / y r NGC 1068
Stochastic (High CR pressure)Stochastic (Modest CR pressure)Magnetic ReconnectionAtmospheric
Figure 2.
Expected event distribution per energy and yearin IceCube is shown for the stochastic and reconnection sce-narios presented for NGC 1068 in Figure 1. The atmo-spheric neutrino background event distribution is also shownfor comparison (grey). trino flux at tens of TeV, motivated by the medium-energy excess in the neutrino spectrum. In this case, asshown previously (Murase et al. 2020) we adopt param-eters that can explain the high-energy neutrino flux ex-cess observed at medium energies (Aartsen et al. 2020a).In this case, the P CR /P th is set to (cid:39) .
01. Here, the neu-trino spectrum peaks at ∼
40 TeV, which corresponds toa lower level of neutrino flux compared to the previousscenario. We refer to this case as “Modest CR pressure”hereafter.These results are compatible with the spectra pre-sented previously by Murase et al. (2020) where theCR pressure considered to explain the medium-energyneutrino flux and NGC 1068 are found at the levelof ∼ ∼
30 percent of the thermal pressure, re-spectively. Here, we allow the pressure ratio to be ashigh as 50% to explain the soft spectrum reported forNGC 1068 by the IceCube Collaboration (Aartsen et al.2020b). Note that, in principle, both the High CR pres-sure and Modest CR pressure cases can be viable withinthe same stochastic acceleration scenario. For exam-ple, the Modest CR pressure may be realized in averageAGN, whereas some sources such as NGC 1068 may havea large CR pressure.Finally, we consider the magnetic reconnection sce-nario for particle acceleration. In this case, the neu-trino flux approximately follows mainly the initial CR t [yr]10 − − − − − − − − − p - v a l u e σ σ NGC 1068
Stochastic (High CR pressure)Stochastic (Modest CR pressure)Magnetic Reconnection
Figure 3.
Prospects for identification of neutrino emissionfrom NGC 1068. Local p-values are shown for stochastic withhigh CR pressure (red), stochastic with modest CR pressure(violet), and reconnection (blue) scenarios considered in thisstudy. spectrum until the pγ process becomes the dominantchannel for the production of pions. Therefore, this sce-nario leads to the spectrum close to a power-law spec-trum with a cutoff at high energies. For the injectedCR spectrum, we assume a spectral index of 2. Thenormalization and cutoff parameters are set by the Ice-Cube steep spectrum reported for NGC 1068 as well asthe P CR /P th , bound to be smaller than 0.5.Figure 1 shows the three modeled neutrino fluxes fromNGC 1068. We also projected the best-fit spectrumreported by the IceCube Collaboration. The best-fitpower-law spectrum corresponds to the ∼
51 excess neu-trinos found from the direction of NGC 1068. Theshaded area shows the uncertainty on the fitted spec-trum as reported by IceCube. As shown, all modeledneutrino spectra are within the 68% uncertainty of themeasured spectrum. The parameters that we adapt ineach scenario for particle acceleration and interaction ef-ficiency are presented in Table 1. The common param-eters among different scenarios are the same as Muraseet al. (2020). The injected CR, i.e., proton, differen-tial luminosity for the three scenarios shown in Fig. 1 ispresented in the Appendix (see Fig. 13).We should note that a single power-law spectrum isnot a realistic spectral energy distribution for neutrinoemission from individual astrophysical objects. Whileneutrino and γ -ray spectra may, in general, reflect the igh-Energy Neutrinos from Seyfert Galaxies γ -rayfluxes depends on the nature of the interaction, thresh-olds, and the opacity of the source. The neutrino spectraprovided in this study take all these into account. Onthe other hand, the diffuse flux of high-energy neutri-nos (or γ -rays) over a specific range of energies may beexplained by a power-law since the superposition of theindividual sources would wash out the features.We use the modeled neutrino spectra for NGC 1068to compare with the findings of the IceCube 10 yr pointsource study. In addition, we investigate the prospectsfor identification of each neutrino emission scenario inthe next decade of IceCube operation.In order to find the p-value for observation of neutri-nos from NGC 1068 over the background of atmosphericneutrinos, we calculate the number of signal neutrinosusing the publicly available effective area for the Ice-Cube point source selection (Aartsen et al. 2017). Wealso estimate the expected number of background at-mospheric neutrinos using the zenith dependent atmo-spheric neutrino flux reported by Honda et al. (2007),assuming a resolution of 0.7 ◦ . This is indeed larger thanIceCube’s nominal angular uncertainty of 0.5 ◦ at highenergies ( (cid:38)
100 TeV). We chose this value because mostneutrinos in our predicted spectra are found at the 1-10TeV range. The energy distribution of the events peryear corresponding to the three neutrino scenarios weconsider here is shown in Fig. 2.The energy distribution of the events is a convolu-tion of the spectrum and the IceCube effective area inthe direction of NGC 1068. The stochastic accelerationscenario with the high CR pressure case results in anevent distribution that peaks around 10 TeV. The mod-est CR pressure case for the stochastic acceleration yielda lower rate, and the majority of the events are foundwith energies around 30 TeV. On the other hand, theharder spectrum at higher energies in the magnetic re-connection scenario leads to a relatively flat distributionof events and creates an excess beyond 100 TeV whereatmospheric neutrinos are scarce.We present the p-value for the observation of eachscenario in 5 to 15 years of operation of IceCube inFig. 3. We use the method described in ATLAS (2011)to estimate the statistical significance for identificationof the neutrino emission; for details, see Appendix. Thestochastic acceleration scenario with the high CR pres-sure, compatible with the IceCube 10 year flux measure-ments, provides the most likely scenario for observation Note that the p-values presented here are pre-trial local p-values and do not account for the trials associated with pointsource studies. E ν [GeV] − − − − − E Φ ν µ + ¯ ν µ [ G e V c m − s − ] NGC 1068
Stochastic (High CR pressure)IceCube 10 yr
Figure 4.
Neutrino spectrum for NGC 1068, assuming dif-ferent values of the X-ray luminosity. Here, we study how theneutrino spectrum depends on the uncertainties associatedwith X-ray observations. The line shows expected neutrinoemission for the stochastic acceleration scenario, for L X be-tween 10 . erg s − and 10 . erg s − . of NGC 1068 in IceCube, reaching to 5 σ level in lessthan 12 years of data. This is in accordance with thereported local p-value in IceCube 10 year point sourcestudy. The magnetic reconnection scenario is expectedto be identified at better than 3 σ within the 10 years,while near a decade more observation might be requiredto establish a significance at the discovery level. Finally,the stochastic acceleration scenario with the modest CRpressure seems difficult to identify with enough signifi-cance. However, the prospects for this scenario may beconservative given uncertainties on the X-ray luminos-ity. As demonstrated in Fig. 4, the neutrino flux wouldbe higher for larger intrinsic X-ray luminosities.The key parameter in modeling the neutrino emis-sion from NGC 1068 is the intrinsic X-ray luminosity.The column density ( N H ) for NGC 1068 is very high.Therefore, it is difficult to estimate the intrinsic X-rayluminosity, and measurements often carry large uncer-tainties. The dominant source of uncertainty in mea-suring the X-ray luminosity from NGC 1068 is the frac-tion of X-rays that is scattered into our line of sight(Janssen et al. 2015). X-ray luminosities as high as10 erg/s have been considered for NGC 1068. NuS-TAR and
XMM–Newton monitoring campaigns (Mar-inucci et al. 2016) found L X (cid:39) +7 − × erg/s with N H (cid:39) cm − . Here, we used a central value of 10 erg/s, which is a conservative choice. We further in- Model P CR /P th ( (cid:15) CR ) q η tur s η acc Stochastic acceleration with high CR pressure 0.5 (0.03) 5/3 50 - -Stochastic acceleration with modest CR pressure 0.008 (0.0009) 5/3 10 - -Magnetic Reconnection 0.5 (0.01) - - 2.0 10 Table 1.
Model-dependent parameters and quantities adapted for estimating the neutrino flux from NGC 1068 in each scenarioof particle acceleration shown in Fig. 1. vestigate the variation of the measured intrinsic X-rayluminosity for the first scenario. Figure 4 shows how thespectrum changes when the X-ray luminosity is variedbetween 10 . to 10 . erg/s. Here, we set the CRs tothermal pressure to the maximal value of 0.5. The largervalues of X-ray luminosity will increase the flux by al-most an order of magnitude, while the peak of the spec-trum is slightly shifted towards higher energies. With alarger value of L X , smaller values of η tur and P CR /P th are demanded to match the IceCube data, which mayhelp relaxing the extreme CR production efficiency forthe high CR pressure case.3.2. Neutrino emission from bright Seyfert galaxies
We now use the parameters of the modeled neutrinoflux from NGC 1068 to estimate neutrino spectra andprospects for observations of other bright Seyfert galax-ies. We first focus on IceCube as it is currently themajor operating neutrino telescope. We will extend ourstudy into KM3NeT and the next generation of IceCube,IceCube-Gen2, in the next section.For this purpose, we rank Seyfert galaxies by theirintrinsic X-ray fluxes measured by BAT AGN Spec-troscopic Survey (Ricci et al. 2017). We select the 10brightest sources for our study and extract their intrin-sic X-ray luminosities from the BASS catalog. Table 2presents the list of sources and their declination, dis-tance, redshift, intrinsic X-ray flux, and luminosity in2-10 keV band. In order to estimate the luminosity dis-tance for source, we incorporate the redshift reportedin BASS together with the cosmological parameters in-troduced in the Introduction, with the exception of CenA, Circinus Galaxy, and NGC 4945, for which we em-ployed astronomical measurements for evaluation of thedistance as reported by Harris et al. (2010); Karachent-sev et al. (2013); Tully et al. (2013). We should notethat the neutrino flux is insensitive to the distance, be-cause the neutrino flux is roughly proportional to theX-ray flux.We implement the parameters that we adopted inthe previous section for modeling the neutrino spectrumfrom NGC 1068, listed in Table 1. Again, we consider 3scenarios: the stochastic acceleration scenario compati-ble with the 10-year IceCube observation of NGC 1068(high CR pressure), the stochastic acceleration scenario compatible with the medium-energy excess in the diffuseneutrino flux (modest CR pressure), and the magneticreconnection scenario. Similar to NGC 1068, we imposethe physical constraint on the ratio of CR pressure to thethermal pressure to 0.5. Therefore, the normalizationof the CR spectrum is adjusted to maintain this boundwhenever this ratio exceeds 0.5 with NGC 1068 parame-ters. Figure 5 demonstrates the spectrum for the sourcelist considered in this study. For the stochastic acceler-ation scenario with the high CR pressure, we have ad-justed flux to match the ratio of CR pressure to the ther-mal pressure to 0.5 by rescaling f inj , which are used inour analyses hereafter. The measured intrinsic X-ray lu-minosity and the distance of the source define the shapeand magnitude of the spectra. In the stochastic accel-eration scenario, sources with L X (cid:38) . erg s − peakat relatively higher energies with a narrow width whilesources with smaller intrinsic X-ray luminosity demon-strate a broader spectrum peaking at lower energies. Inthis scenario, Cen A has the highest level of flux in thelist of bright Seyfert galaxies. For the stochastic accel-eration with modest CR pressure, we employ f inj valuethat yields the CR to thermal pressure of ∼ f inj (cid:39) − (5 × − )at 2 m p c (1000 m p c ). Finally, in the magnetic recon-nection scenario, the larger X-ray luminosity leads to alarger contribution of photohadronic processes.The majority of bright nearby Seyfert galaxies in Ta-ble 2 are located in the Southern hemisphere, as pointedout by Murase et al. (2020). IceCube’s event selectionis optimal for the Northern sky, where the Earth actsas a shield for the atmospheric muons. In the Southernhemisphere, the event selection imposes a higher-energythreshold on the energy of the neutrinos to suppress theatmospheric muon background. This feature suppressesthe event rate for the majority of the luminous Seyfertgalaxies. We show the expected events from the sourcesin this list in Fig. 14 in Appendix. Except for the sourcesNGC 1275, UGC 11910, and CGCG 164-019 that are inthe Northern hemisphere, the event rates for the rest ofthe sources are low, weakening the likelihood of identi-fying individual sources in IceCube.Using the expected signal and background rates, weestimate the likelihood for observations of these sources igh-Energy Neutrinos from Seyfert Galaxies Source Declination z d L Intrinsic flux log(Intrinsic luminosity)[deg] [Mpc] [10 − erg cm − s − ] [erg s − ]Circinus Galaxy -65.34 0.0014 4.2 (cid:63) (cid:63) (cid:63) Table 2.
Seyfert galaxies with the highest intrinsic X-ray fluxes in the range of 2-10 keV in BAT AGN Spectroscopic Survey(BASS) (Ricci et al. 2017). Here, we show the declination, redshift, and the intrinsic X-ray fluxes and luminosities reportedby BASS in 2-10 keV band. For sources closer than 10 Mpc, specified by (cid:63) , we employ distances reported by astronomicalmeasurements. For other sources, the luminosity distances are evaluated from the reported redshift. in IceCube. Table 3 summarizes the expected p-valuesunder each emission scenario for 10 years of IceCubeoperation. The listed p-values show that for all threeacceleration scenarios, NGC 1068 is the brightest sourcein IceCube. While the prospects for the identification ofmost sources are not much promising due to the suppres-sion of events in the Southern hemisphere, with contin-ued collection data CGCG 164-019 and NGC 1275 arelikely to be observed at 3 σ level in 20 years of IceCubeoperation. However, we should note that the likelihoodof observations depends on the neutrino emission sce-nario: the stochastic acceleration scenario with the highCR pressure, compatible with NGC 1068 parameters,would yield ∼ σ .Another source in the list worth discussing is NGC4945. The IceCube 10 year analysis found an excess of ∼ σ in less than 10 years of IceCube operation. The hardspectrum in the magnetic reconnection scenario wouldresult in an excess distinguishable from the backgroundat high energies that can yield a 3 σ evidence in about 7years of IceCube. The significance can reach a discov-ery level of 5 σ within the 20 years of operation for thisscenario.The most conservative scenario here is the stochasticacceleration scenario with the modest CR pressure. Thisscenario would not yield a signal at a significant levelwith the current generation of IceCube. Mainly due tothe fact that most sources are in the Southern hemi-sphere. We investigate the likelihood of identifying thisscenario in the next generation of neutrino telescopes inthe next section.We should note that the significance for the stackinganalysis for stochastic acceleration with high CR pres-sure scenario is dominated by the neutrino flux in thedirection of NGC 1068. However, it is worth mention-ing that for this scenario, a stacking analysis of nearbybright Seyfert galaxies when NGC 1068 is excluded isexpected to identify neutrino emission at the level of 3 σ with 15 years of IceCube data. Such analysis can testthis scenario independently, without relying on the neu-trino emission reported from the direction NGC 1068,which this scenario is based upon.0 E ⌫ µ + ¯ ⌫ µ [ G e V c m s ] NGC 424 NGC 1275
Stochastic (High CR pressure) Stochastic (Modest CR pressure) Magnetic Reconnection NGC 4945 E ⌫ µ + ¯ ⌫ µ [ G e V c m s ] Cen A Circinus Galaxy CGCG 164-019 E ⌫ [GeV]10 E ⌫ µ + ¯ ⌫ µ [ G e V c m s ] ESO 138-1 E ⌫ [GeV]10 UGC 11910 E ⌫ [GeV]10 NGC 7582
Figure 5.
Neutrino flux for bright Seyfert galaxies considered in this study. Here, we show the high (red) and modest (purple)CR pressure stochastic acceleration scenarios as well as the magnetic recoonection scenario (blue), which provide compatiblefluxes with the best-fit flux for NGC 1068 or the total neutrino spectrum measurement for parameters presented in Table 1.
Our predicted neutrino emission was built upon thedisk-corona model of AGN. Two of the sources in thelist of bright Seyfert galaxies, Cen A and NGC 1275, areseen with a high jet activity and it is likely that the X-ray emission arises from the jet rather than the disk. Wewill further investigate the prospects for observations ofbright Seyfert galaxies if the disk-corona emission wouldnot be dominant. The dashed lines in Fig. 6 show thep-value when Cen A and NGC 1275 are not consideredin the source list. While the likelihood of observation is decreased, the cumulative neutrino emission is strongand the stacking analysis of the rest of the bright Seyfertgalaxies can still reach a significant level in the lifetimeof the IceCube detector.For Cen A, we should note that although one zonemodels typically attribute the high-energy to thejet Abdo et al. (2010), it may be difficult to explain someof the X-ray properties with such scenario (Tachibanaet al. 2015). The observed soft lags in X-rays from CenA may indicate their coronal origin. We should note igh-Energy Neutrinos from Seyfert Galaxies p-valueSource Stochastic (High CR pressure) Stochastic (Modest CR pressure) Magnetic reconnectionNGC 1068 10 − × − NGC 1275 0.03 0.3 0.3CGCG 164-019 0.04 0.3 0.4UGC 11910 0.1 0.4 0.4Cen A 0.5 0.2 0.06Circinus Galaxy 0.5 0.3 0.1NGC 7582 0.5 0.5 0.3ESO 138-1 0.5 0.5 0.5NGC 424 0.5 0.5 0.5NGC 4945 0.5 0.5 0.5
Table 3.
Prospects for observations of bright nearby Seyfert galaxies in 10 years of IceCube operations. . . . . . . . t [yr]10 − − − − − − − p - v a l u e Stochastic(high CR pressure)Stochastic(modest CR pressure) MagneticReconnection 3 σ σ IceCube Stacking
Figure 6.
Projected p-values for stacking of 10 brightestSeyfert galaxies in 5 to 20 years of IceCube operation, forthe three acceleration scenario considered in this study. Thesolid lines show the p-values for the 10 bright sources listedin Table 2. The dashed line presents the prospects when theemission from Cen A and NGC 1275 is excluded. that despite the large magnitude of the flux predictedunder both stochastic acceleration and magnetic recon-nection scenarios for Cen A in our study, our predictionis compatible with current upper limits imposed by theANTARES and IceCube Collaboration. In Fig. 7 wecompare the predicted neutrino flux with the currentlimits from the IceCube cascade source search in theSouthern sky (Albert et al. 2020). The upper limit isobtained for a power-law with exponential cutoff, whichis the closest shape to the predicted spectra in our study.In summary, observations of stacked neutrino emis-sion from nearby, bright Seyfert galaxies is promisingwith IceCube, which could reveal the dominant sources E ν [GeV]10 − − − − − − − − E Φ ν µ + ¯ ν µ [ G e V c m − s − ] P CR /P th ’ . P CR /P th ’ . Cen A
Stochastic (High CR pressure)Stochastic (Modest CR pressure)Magnetic Reconnection
Figure 7.
Neutrino spectra from Cen A compared with thediscovery potential of a joint IceCube cascade source searchfor the sources in the Southern sky (Albert et al. 2020). responsible for the medium-energy excess in the spec-trum of high-energy cosmic neutrinos. The likelihoodfor observations of these sources will be enhanced bythe improvements in the event selection in the Southernhemisphere. We will address this in Sec. 6. FUTURE NEUTRINO TELESCOPESIn this section, we explore the prospects for identify-ing bright nearby Seyfert galaxies in the next generationof neutrino telescopes. The major development in thehigh-energy neutrino astrophysics would be driven byKM3NeT, currently under construction in the Mediter-2 E ν [GeV]10 − − e v e n t s / b i n / y r NGC 1068 – KM3NeT
Stochastic (High CR pressure)Stochastic (Modest CR pressure)Magnetic ReconnectionAtmospheric
Figure 8.
Expected signal and background events inKM3NeT neutrino spectra presented in Fig. 1. ranean, and the next generation of IceCube: IceCube-Gen2.KM3NeT is a cubic km-scale water Cherenkov detec-tor, which is designed to enhance the exposure in theSouthern hemisphere. The larger effective area and im-proved angular resolution, compared to the current neu-trino detector in the Mediterranean, ANTARES, are ex-pected to provide better sensitivity to the sources inthe Southern sky (Adrian-Martinez et al. 2016). Thisadvancement would enhance the potential of identify-ing neutrino emission from bright Seyfert galaxies asthe majority of them are located in the Southern hemi-sphere.In order to estimate the likelihood of identifying neu-trino emission, we incorporate the effective area for up-going muon neutrinos in KM3NeT. We use the effectivearea reported in (Adrian-Martinez et al. 2016) and in-clude the trigger efficiency for background rejection inpoint source analyses. The publicly available effectivearea for the two blocks of ARCA is averaged over thezenith covering the upgoing events. Therefore, in our es-timation of the atmospheric background events, we usethe average atmospheric neutrino flux instead of zenithdependent ones that we used for IceCube. We estimatethe signal and background events from the sources listedin Table 2. Unlike IceCube, KM3NeT is not locatedat the geographic pole and sources’ zenith angle vary.Thus, we need to take into account the duration thatthe source is positioned below the horizon. As such, we t [yr]10 − − − − − − − p - v a l u e σ σ IceCube-Gen2 KM3NeT δψ = 0 . δψ = 0 . Figure 9.
Prospects for observation of the brightSeyfert galaxies in the next-generation neutrino telescopes:KM3NeT and IceCube-Gen2. The solid (dashed) lines showexpectations for 0.3 ◦ (0.7 ◦ ) angular resolution for the mod-est CR pressure scenario. The tick lines show the prospectsfor identification of the 10 nearby bright sources in Table 2in a stacking analysis. The thin lines show the prospects foridentification of the sources in the absence of a signal fromdisk-corona model from Cen A and NGC 1275. take into account the visibility of the source when cal-culating the event rate and evaluating the prospects fortheir identification.We first examine the likelihood for observation ofNGC 1068 in KM3NeT. Figure 8 shows the expectedsignal and background rate. Given that the source vis-ibility is 50%, it would not be identified in KM3NeT,even for the most optimistic scenario, i.e, the stochas-tic acceleration with the high CR pressure which wouldyield a p-value of few percent after 5 years.We present the p-values for observations of all sourcesin 1 and 3 years of operation of KM3NeT in Table 4.The brightest source in KM3NeT in the list of brightSeyfert galaxies is found to be Cen A given that its X-rays come from the coronal region (that may not bethe case). Thanks to the higher level of neutrino flux,the stochastic acceleration scenario with the high CRpressure is going to be found at 3 σ in the first yearof observation, and the significance can reach 5 σ withadditional two years of observation.Circinus Galaxy is found to be the second brightestsource in the Southern sky. In addition to its high levelof neutrino flux at TeV energies, the source benefits from igh-Energy Neutrinos from Seyfert Galaxies p-value 1 yr (3 yr)Source Visibility Stochastic (high CR pressure) Stochastic (Modest CR pressure) Magnetic ReconnectionCen A 0.7 0.001 (9.3 × − ) 0.2 (0.07) 0.1 (0.01)Circinus Galaxy 1.0 0.008 (1.9 × − ) 0.2 (0.09) 0.1 (0.02)ESO 138-1 1 0.1 (0.02) 0.4 (0.3) 0.4 (0.3)NGC 7582 0.7 0.2 (0.04) 0.4 (0.3) 0.4 (0.2)NGC 1068 0.5 0.2 (0.05) 0.4 (0.4) 0.4 (0.2)NGC 4945 0.8 0.5 (0.2) 0.5 (0.4) 0.5 (0.4)NGC 424 0.7 0.4 (0.2) 0.5 (0.4) 0.5 (0.4)UGC 11910 0.5 0.4 (0.4) 0.5 (0.5) 0.5 (0.5)CGCG 164-019 0.4 0.4 (0.3) 0.5 (0.5) 0.5 (0.5)NGC 1275 0.3 0.4 (0.4) 0.5 (0.5) 0.5 (0.5) Table 4.
Prospects for observation of nearaby bright Seyfert galaxies in one years of KM3NeT observations. a 100% visibility in KM3NeT. Therefore, the likelihoodfor its observation is high, which can exceed 3 σ in 3 yearsof operation for the stochastic acceleration scenario withthe high CR pressure.As the signal events from the rest of the sources inthe list fall short of yielding a statistical significance in3 years, we now turn into the prospects for observationof neutrino emission in a stacking analysis. We onlyconsider the modest CR pressure scenario in stochasticacceleration since emission under either of the other twoscenarios should be identified by IceCube. In additionto KM3NeT, we consider IceCube-Gen2 for the stackingsearch in this scenario. Here, we assume that the effec-tive area for IceCube-Gen2 is ∼ ∼ −
30 TeV is crucial forthe identification of neutrino emission from these sourcesespecially in the modest CR pressure case. We furthershow the growth of significance for a given resolution inSec. 6. DISCUSSION5.1.
Aggregated Fluxes
Highly magnetized and turbulent coronae can be pos-sible sites of particle acceleration. The system is calori-metric in the sense that sufficiently high-energy CRs aredepleted via hadronuclear and photohadronic interac-tions. The high magnitude of the neutrino flux at 10–100 TeV makes this scenario a primary candidate for themedium-energy neutrino flux observed in IceCube at thelevel of E ν Φ ν ∼ − GeV cm − s − sr − (Murase et al.2020). The diffuse flux mainly originates from AGN athigh redshifts (with z ∼ − π in order to compare with the total neutrino flux from the6-yr cascade analysis of IceCube (Aartsen et al. 2020a).Overall, each scenario predicts the contribution of thecatalogued nearby sources to the total neutrino flux at10 TeV to be within 2-10%.The stochastic acceleration scenario with the modestCR pressure would mainly contribute to the 10-100 TeVregion. However, the high CR pressure case would gen-erate a significant excess of the flux below 10 TeV. Thisregion is hard to investigate with the overwhelming fluxof atmospheric neutrinos, and detailed veto techniquesare required to distinguish the flux at TeV energies witha good accuracy. The magnetic reconnection scenariohas the highest contribution to the flux at (cid:38)
100 TeV.Distinguishing this scenario from the one responsible forthe flux above 100 TeV would be difficult because of thescarcity of the data at high energies. While the exten-sion of the flux to higher energies could make it easierto identify this flux from the background, such a pos-sibility is less motivated by the compelling signal from4 E ⌫ [GeV]10 E ⌫ + ¯ ⌫ [ G e V c m s s r ] Stochastic (modest CR pressure) E ⌫ [GeV]10 E ⌫ + ¯ ⌫ [ G e V c m s s r ] Stochastic (high CR pressure) E ⌫ [GeV]10 E ⌫ + ¯ ⌫ [ G e V c m s s r ] Magnetic Reconnection
Figure 10.
Accumulated neutrino emission (thick line) and the individual neutrino flux (thin lines) for the bright nearby Seyfertgalaxies considered in this study. We show the neutrino flux for stochastic acceleration with highest CR pressure (left), stochasticacceleration with modest CR pressure (middle), and magnetic reconnection (right). The data points show the measured cosmicneutrino flux in 6 years of IceCube cascade data (Aartsen et al. 2020a)
NGC 1068. We discuss this further later in the nextsubsection.5.2.
Magnetic reconnection scenario and the diffuseneutrino flux
It has been already shown that Seyfert galaxies with aCR pressure at the level 1-10% of the thermal pressurewith the virial temperature can explain the magnitudeof the diffuse neutrino flux at medium energies (Muraseet al. 2020).One of the novel points in this work is that we sug-gest the magnetic reconnection scenario as an alterna-tive possibility of CR acceleration. In this scenario, weexpect a power-law spectrum for CRs. The diffuse neu-trino flux in this scenario is estimated to be E ν Φ ν ∼ − GeVcm s sr (cid:18) K K (cid:19) (cid:18) R p (cid:19) (cid:18) f mes f BH + f mes (cid:19) × (cid:18) ξ z (cid:19) ξ CR (cid:18) L X ρ X × erg Mpc − yr − (cid:19) , (3)where K = 1(2) for pγ ( pp ) interactions, ξ z ∼ R p is the conversion factor from bolo-metric to differential luminosities, ξ CR = L CR /L X =( (cid:15) CR /(cid:15) rad )( L disk /L X ) is the CR loading factor definedagainst the X-ray luminosity, L X is the X-ray luminos-ity, ρ X is the local density of X-ray selected AGN. There-fore, the total flux is found at 10 − GeV cm − s − sr − .In the magnetic reconnection scenario, we assumedan injected spectral index of 2 for the injected CRs.This assumption yields a relatively flat spectrum of high-energy cosmic neutrinos with a cutoff around 100 TeVfor η acc that could describe the flux reported for NGC 1068. In addition, depending on the intrinsic X-ray lu-minosity of the source, a significant contribution from pγ interactions would emerge before the cutoff. The nor-malization in this scenario is set by P CR /P th ≤ . −
10 TeV energy range, the flux be-comes dominant over the background at higher energies,providing an excess that could be distinguished over thebackground in the sufficient years of detector operation.A higher cutoff energy in the spectrum would be there-fore constrained by the lack of observation for such ahard neutrino flux from these sources. Therefore, bothmagnetic reconnection and shock acceleration scenarios(Inoue et al. 2020) would be disfavored unless η acc isfine-tuned. We further investigate the expected flux inthis scenario by considering a steeper CR spectrum. InFig. 11, we show the neutrino flux corresponding to in-jected CR indices of 2.2, 2.3, and 2.5. As can be seen,when restricting the pressure of CRs to less than 50%,it is going to be difficult to explain the measured fluxNGC 1068 by the IceCube Collaboration. SUMMARY & OUTLOOKIn this study, we presented promising neutrino emis-sion scenarios from nearby, bright Seyfert galaxies andquasars, which are typically classified as radio-quietAGN. Our predictions were built upon the disk-coronamodel of AGN. Murase et al. (2020) already demon-strated that this scenario can explain the medium-energy flux of high-energy cosmic neutrinos observedin IceCube, and provide a consistent multi-messengerpicture because such sources are opaque to very high-energy γ -ray emission. Here, we presented the neu-trino emission from the individual bright Seyfert galax-ies by incorporating their measured intrinsic X-ray lu- igh-Energy Neutrinos from Seyfert Galaxies E ν [GeV]10 − − − − − − E Φ ν µ + ¯ ν µ [ G e V c m − s − ] NGC 1068 s = 2 . s = 2 . s = 2 . s = 2 . Figure 11.
Expected neutrino fluxes for power-law CRspectra expected in the magnetic reconnection scenario, withspectral indices ranging from 2 to 2.5. minosities. We considered three scenarios for particleacceleration and evaluated the prospects for identifyingthe neutrino emission in both point source and stackingsearches.We showed that NGC 1068 is the brightest source inIceCube and can explain the near 3 σ observation re-ported by the 10 year analysis of IceCube data (Aart-sen et al. 2020b). We found that, with a slightly highervalue for the pressure ratio compared to what was previ-ously considered (Murase et al. 2020), the model coulddescribe the steep spectra observed for neutrino emis-sion from NGC 1068 as well as the reported significancefor the number of signal events. Our projection demon-strates that the identification of high-energy neutrinoemission from NGC 1068 at the level of 5 σ is likely whenthe optimistic scenario of the maximal CR pressure tothe thermal pressure is assumed. For less optimistic sce-narios, where a moderate value of the pressure ratio isconsidered, the significance at the level of 3 σ would beestablished . Note that, even in the modest case, ob-servations at the discovery level could be achieved withnext-generation detectors such as IceCube-Gen2.Besides NGC 1068, the rest of the sources are unlikelyto be identified in IceCube point source searches. How-ever, stacking analyses would have sufficient sensitivitiesfor identifying neutrino emission from the bright sourcesin the next few years in IceCube. Identification at ∼ σ is plausible even when the two sources in our list with ambiguities on the contribution of coronal emission (CenA and NGC 1275) are removed from a stacking analysis.We further showed the prospects for identification ofthese sources in the stacking analysis in IceCube-Gen2,where we assume the modest CR pressure. While theincreased effective area of the IceCube-Gen2 makes itpossible to identify neutrino emission from these sources,the observational significance highly depends on the an-gular resolution of IceCube-Gen2 at ∼ −
30 TeV en-ergies. We showed the dependency of the prospects forobserving neutrino emission from bright Seyferts in ourlist on the angular resolution of IceCube-Gen2 in Fig. 12.5 σ identification could be achieved with better than 0.4 ◦ angular uncertainty. This result further strengthens theimportance of improving the angular resolution to iden-tify the “dominant” origin of IceCube neutrinos (Murase& Waxman 2016).As the majority of the bright sources are in the South-ern sky, the operation of KM3NeT in the near futurewill make it possible to identify neutrino emission fromCen A and Circinus Galaxy, even with the modest CRpressure case. Again, the signal from the rest of thesources is not strong enough to yield a significant excess.However, even the modest scenario should be identifiedwithin the five years of its operation. We should alsonote that in addition to KM3NeT, commissioning of theBaikal underwater neutrino telescope (NT-200) (Belo-laptikov et al. 1997; Aynutdinov et al. 2006) and thePacific-Ocean Neutrino Experiment (P-ONE) (Agostiniet al. 2020) will boost the coverage for the sources inthe Southern hemisphere. This would enhance the likeli-hood for identification of neutrino emission from sourcesconsidered in this study.It is worth mentioning that the recent progress in en-hancing IceCube’s sensitivity for sources in the South-ern sky can increase the likelihood for observation of theneutrino emission from Seyfert galaxies in the near fu-ture (Mancina & Silva 2020). The technique enhancesthe exposure of IceCube in the medium-energy range inthe 10 TeV range, which creates a unique opportunityto examine the stochastic acceleration scenario.In this study, we considered two acceleration mecha-nisms in the disk-corona model of AGN: stochastic accel-eration and magnetic reconnection. They phenomeno-logically represent a hard CR spectrum and a broadpower-law spectrum, respectively. One of the major dif-ferences between the spectra of high-energy neutrinosin these two scenarios is the level of the flux at energiesabove 100 TeV. Compared to the stochastic accelerationscenario with the high CR pressure, the flux normaliza-tion in the magnetic reconnection scenario is, in general,lower. This gives the stochastic acceleration scenario a6 . . . . . . δψ [ ◦ ]10 − − − − − − − p - v a l u e σ σ IceCube-Gen2 Stacking
Figure 12.
Dependence of the sensitivity of stacking anal-yses of bright nearby Seyfert galaxies in 5-year observa-tion of IceCube-Gen2 for different angular resolutions in themedium-energy range. Here, we show the expected p-valuefor the stochastic acceleration scenario with the modest CRpressure. better likelihood of observation in point source studies.However, in a stacking search, the accumulation of high-energy events from all sources leads to a stronger sepa-ration of signal and background events. Therefore, themagnetic reconnection scenario yields better p-values.The better likelihood also means a stronger constraintin the absence of evidence for a neutrino spectrum thatextends to higher energies from these sources. At themoment, the excess of events found in the direction ofNGC 1068 is not statistically significant enough to con-strain any of the scenarios discussed here. However, the confirmation of the steep spectrum reported by the Ice-Cube Collaboration in the future will put constraints onparticle acceleration mechanisms. For example, shockacceleration in the Bohm limit may be strongly con-strained.The coronal neutrino emission model for the nearby,bright sources considered in this study is expected to beidentified by current and future neutrino detectors. Themagnitude of the diffuse neutrino flux observed in Ice-Cube (Aartsen et al. 2020a) demonstrated that hadronu-clear or photohadronic interactions play significant rolesin the high-energy universe. The recent developmentsin the search for the origin of high-energy neutrinoshave revealed the first sign of anisotropies in the flux ofhigh-energy neutrinos (Aartsen et al. 2020b). At veryhigh energies, events beyond 100 TeV have facilitatedtime-dependent searches and follow-ups, delivering coin-cidences with flaring blazars and tidal disruption events(Aartsen et al. 2018; Stein et al. 2020). In the meantime,the higher level of the diffuse neutrino flux at mediumenergies implies the possibility of finding sources with ahigh-level of neutrino flux. Identifying their origin willbestow a unique opportunity to probe extraordinaryenvironments in the non-thermal universe and particleacceleration in the dense environments that cannot bereadily probed by electromagnetic observations.ACKNOWLEDGEMENTSWe would like to thank Chad Finley, Francis Halzen,and Ibrahim Safa for their useful comments on themanuscript. A.K. acknowledges the support from theIGC through the IGC postdoctoral fellowship. The workof K.M. is supported by NSF Grant No. AST-1908689,and KAKENHI No. 20H01901 and No. 20H05852.S.S.K. acknowledges JSPS Research Fellowship andKAKENHI Grant No. 19J00198.REFERENCES
Aartsen, M., et al. 2013a, Phys.Rev.Lett., 111, 021103—. 2018, Science, 361, eaat1378—. 2020a, arXiv:2001.09520—. 2020b, Phys. Rev. Lett., 124, 051103Aartsen, M. G., et al. 2013b, Science, 342, 1242856—. 2015a, Phys. Rev., D91, 022001—. 2015b, Astrophys. J., 809, 98—. 2017, Astrophys. J., 835, 151Abdo, A. A., et al. 2010, Astrophys. J., 719, 1433 Ackermann, M., Ajello, M., Allafort, A., et al. 2012, TheAstrophysical Journal, 755, 164.http://dx.doi.org/10.1088/0004-637X/755/2/164Ackermann, M., et al. 2015, Astrophys. J., 799, 86Adrian-Martinez, S., et al. 2016, J. Phys., G43, 084001Agostini, M., et al. 2020, arXiv:2005.09493Albert, A., et al. 2020, Astrophys. J., 892, 92Alexandreas, D., et al. 1993, Nucl. Instrum. Meth. A, 328,570ATLAS, CMS, L. H. C. G. 2011 igh-Energy Neutrinos from Seyfert Galaxies Aynutdinov, V. M., et al. 2006, Phys. Atom. Nucl., 69,1914, [,449(2006)]Balbus, S. A., & Hawley, J. F. 1991, Astrophys. J., 376, 214Ball, D., Sironi, L., & ¨Ozel, F. 2018, Astrophys. J., 862, 80Becker, P. A., Das, S., & Le, T. 2011, ApJ, 743, 47Begelman, M. C., Rudak, B., & Sikora, M. 1990, ApJ, 362,38Belolaptikov, I. A., et al. 1997, Astropart. Phys., 7, 263Berezinskii, V. S., & Ginzburg, V. L. 1981, MNRAS, 194, 3Blackman, E. G., & Pessah, M. E. 2009, Astrophys. J.Lett., 704, L113Capanema, A., Esmaili, A., & Murase, K. 2020a, Phys.Rev. D, 101, 103012Capanema, A., Esmaili, A., & Serpico, P. D. 2020b,arXiv:2007.07911Chattopadhyay, I., & Kumar, R. 2016, Mon. Not. Roy.Astron. Soc., 459, 3792Comisso, L., & Sironi, L. 2018, PhRvL, 121, 255101Eichmann, B., & Becker Tjus, J. 2016, Astrophys. J., 821,87Gonzalez-Garcia, M., Halzen, F., & Niro, V. 2014,Astropart.Phys., 57-58, 39Guo, F., Li, X., Li, H., et al. 2016, ApJL, 818, L9Haardt, F., & Maraschi, L. M. U. 1991, Astrophys. J. Lett.,380, L51Halzen, F., Kheirandish, A., & Niro, V. 2016,arXiv:1609.03072Harris, G. L., Rejkuba, M., & Harris, W. E. 2010, Publ.Astron. Soc. Austral., 27, 457Honda, M., Kajita, T., Kasahara, K., Midorikawa, S., &Sanuki, T. 2007, Phys.Rev., D75, 043006Hoshino, M. 2012, Phys. Rev. Lett., 108, 135003. https://link.aps.org/doi/10.1103/PhysRevLett.108.135003Hoshino, M. 2015, Physical Review Letters, 114, 061101Inoue, Y., Khangulyan, D., & Doi, A. 2020, Astrophys. J.Lett., 891, L33Inoue, Y., Khangulyan, D., Inoue, S., & Doi, A. 2019, ApJ,880, 40Io, Y., & Suzuki, T. K. 2014, Astrophys. J., 780, 46Janssen, A. W., Bruderer, S., Sturm, E., et al. 2015, TheAstrophysical Journal, 811, 74.https://doi.org/10.1088/0004-637x/811/2/74Jiang, Y.-F., Blaes, O., Stone, J., & Davis, S. W. 2019,arXiv:1904.01674Jiang, Y.-F., Blaes, O., Stone, J. M., & Davis, S. W. 2019,ApJ, 885, 144Jiang, Y.-F., Stone, J. M., & Davis, S. W. 2014, Astrophys.J., 784, 169 Kalashev, O., Semikoz, D., & Tkachev, I. 2015, SovietJournal of Experimental and Theoretical Physics, 120,541Karachentsev, I. D., Makarov, D. I., & Kaisina, E. I. 2013,AJ, 145, 101Kimura, S. S., Murase, K., & M´esz´aros, P. 2019a, PhRvD,100, 083014Kimura, S. S., Toma, K., Suzuki, T. K., & Inutsuka, S.-i.2016, ApJ, 822, 88Kimura, S. S., Toma, K., & Takahara, F. 2014, ApJ, 791,100Kimura, S. S., Tomida, K., & Murase, K. 2019b, MNRAS,485, 163Kotera, K., Allard, D., Murase, K., et al. 2009,Astrophys.J., 707, 370Lamastra, A., Tavecchio, F., Romano, P., Landoni, M., &Vercellone, S. 2019, Astropart. Phys., 112, 16Lemoine, M., & Malkov, M. A. 2020, Mon. Not. Roy.Astron. Soc., 499, 4972Levinson, A., & Rieger, F. 2011, Astrophys. J., 730, 123Liu, B., Mineshige, S., Meyer, F., Meyer-Hofmeister, E., &Kawaguchi, T. 2002a, Astrophys. J., 575, 117Liu, B., Mineshige, S., & Shibata, K. 2002b, Astrophys. J.Lett., 572, L173Liu, R.-Y., Murase, K., Inoue, S., Ge, C., & Wang, X.-Y.2018, Astrophys. J., 858, 9Loeb, A., & Waxman, E. 2006, JCAP, 0605, 003Lynn, J. W., Quataert, E., Chandran, B. D. G., & Parrish,I. J. 2014, ApJ, 791, 71Machida, M., & Matsumoto, R. 2003, ApJ, 585, 429Maisack, M., Johnson, W. N., Kinzer, R. L., et al. 1993,ApJL, 407, L61Mancina, S., & Silva, M. 2020, PoS, ICRC2019, 954Marinucci, A., et al. 2016, Mon. Not. Roy. Astron. Soc.,456, L94Merloni, A., & Fabian, A. 2001, Mon. Not. Roy. Astron.Soc., 321, 549Miller, K., & Stone, J. 2000, Astrophys. J., 534, 398Murase, K., Ahlers, M., & Lacki, B. C. 2013, Phys.Rev.,D88, 121301Murase, K., Guetta, D., & Ahlers, M. 2016, Phys. Rev.Lett., 116, 071101Murase, K., Inoue, S., & Nagataki, S. 2008, Astrophys.J.,689, L105Murase, K., Kimura, S. S., & Meszaros, P. 2020, Phys. Rev.Lett., 125, 011101Murase, K., & Waxman, E. 2016, Phys. Rev., D94, 103006Narayan, R., S¨adowski, A., Penna, R. F., & Kulkarni, A. K.2012, MNRAS, 426, 3241Ohsuga, K., & Mineshige, S. 2011, ApJ, 736, 2 Pisokas, T., Vlahos, L., & Isliker, H. 2018, Astrophys. J.,852, 64Porth, O., Chatterjee, K., Narayan, R., et al. 2019, ApJS,243, 26Ricci, C., et al. 2017, Astrophys. J. Suppl., 233, 17Ricci, C., Ho, L. C., Fabian, A. C., et al. 2018, MNRAS,480, 1819Sanuki, T., Honda, M., Kajita, T., Kasahara, K., &Midorikawa, S. 2007, Phys. Rev., D75, 043005Stecker, F., Done, C., Salamon, M., & Sommers, P. 1991,Phys.Rev.Lett., 66, 2697Stecker, F. W. 2005, Phys. Rev. D, 72, 107301—. 2013, Phys. Rev. D, 88, 047301Stein, R., et al. 2020, arXiv:2005.05340Stone, J. M., & Pringle, J. E. 2001, MNRAS, 322, 461Tachibana, Y., Kawamuro, T., Ueda, Y., et al. 2015,arXiv:1504.03208 Tully, R. B. 1988, Nearby galaxies catalogTully, R. B., Courtois, H. M., Dolphin, A. E., et al. 2013,AJ, 146, 86Waxman, E., & Bahcall, J. N. 1999, Phys.Rev., D59, 023002Werner, G. R., Uzdensky, D. A., Begelman, M. C., Cerutti,B., & Nalewajko, K. 2018, MNRAS, 473, 4840Wong, K., Zhdankin, V., Uzdensky, D. A., Werner, G. R.,& Begelman, M. C. 2020, Astrophys. J. Lett., 893, L7Yoast-Hull, T. M., III, J., Zweibel, E. G., & Everett, J. E.2014, Astrophys. J., 780, 137Zdziarski, A. A. 1986, ApJ, 305, 45Zdziarski, A. A., Poutanen, J., & Johnson, W. N. 2000,ApJ, 542, 703Zhdankin, V., Uzdensky, D. A., Werner, G. R., &Begelman, M. C. 2019, Phys. Rev. Lett., 122, 055101 igh-Energy Neutrinos from Seyfert Galaxies APPENDIX7.1.
Cosmic-ray luminosity
The differential CR luminosity for the three scenarios for particle acceleration in NGC 1068 are shown in Fig. 13.The CR luminosity is deduced from the measured intrinsic X-ray by incorporating the escape and cooling timescales,injection function for the CRs, and the volume of the coronal region. For more details, see Supplemental Material ofMurase et al. (2020). E p [GeV]10 L p [ e r g s − ] NGC 1068 - injected proton luminosity
Stochastic (High CR pressure)Stochastic (Modest CR pressure)Magnetic Reconnection
Figure 13.
Differential CR proton luminosities for NGC 1068 ( L X = 10 erg s − ) for the three different scenarios shown inFig. 1 and parameters in Table 1. Event distributions
To calculate the number of background atmospheric neutrino events, we have integrated the atmospheric flux Hondaet al. (2007) and Sanuki et al. (2007) over an opening angle of Ω = πσ around the direction of the source, where theangle σ is the angular resolution of the detector. Given that the majority of the events in our models appear below ∼
10 TeV, we set the angular resolution for the IceCube and KM3NeT to 0 . ◦ as specified in Fig. 2 of Aartsen et al.(2017) and Fig. 22 of Adrian-Martinez et al. (2016), respectively.To calculate the number of cosmic neutrino events for the flux predicted in each scenario, we integrate the product ofthe neutrino spectrum and effective area over energy and time. For IceCube, we use the zenith dependent effective areafor the point source analysis (Aartsen et al. 2017). Similarly, we use this effective area, together with the atmosphericneutrino flux of Honda et al. (2007) and Sanuki et al. (2007) to estimate the number of background events in thedirection of each source within the opening angle of Ω = πσ . The signal and background event distributions for NGC1068 is presented in Fig. 2 in the main text, and in Fig. 14 for the rest of bright Seyfert galaxies in Table 2. Given thatthe majority of the sources in Fig. 14 are located in the Southern hemisphere, the corresponding event distributionis limited by the strong event selection cuts imposed to remove the atmospheric muon background penetrating thedetector.The event distribution for KM3NeT is presented in Fig. 15. We incorporate the average effective area for upgoingevents in KM3NeT according to Fig. 19 of Adrian-Martinez et al. (2016) and apply the trigger efficiency for thepoints source searches. To take into account the fraction of the duration for which the source is below the horizon forKM3NeT, we rescale the number of events by the visibility of the sources, given by their declination as is defined inFig. 37 of Adrian-Martinez et al. (2016).0 − − − − e v e n t s / b i n / y r NGC 424 − − − − NGC 1275
Stochastic (High CR pressure) Stochastic (Modest CR pressure) Magnetic Reconnection Atmospheric − − − − NGC 4945 − − − − e v e n t s / b i n / y r Cen A − − − − Circinus Galaxy − − − − CGCG 164-019 E ν [GeV]10 − − − − e v e n t s / b i n / y r ESO 138-1 E ν [GeV]10 − − − − UGC 11910 E ν [GeV]10 − − − − NGC 7582
Figure 14.
Expected event distribution in IceCube per year period for the three scenarios considered in this study. Theexpected background component from conventional atmospheric neutrino flux is shown for each source. For the sources locatedin the southern hemisphere, the event at low energies are suppressed due to the strict cuts in IceCube point source selection.The lines in each panel correspond to the fluxes presented in Fig. 5
Estimation of the Statistical significance
We estimate the statistical significance for observing the sources using the analytic expression introduced by ATLAS(2011). This method has been used extensively to evaluate the prospects for observation of Galactic sources of high-energy neutrinos, see e.g., Gonzalez-Garcia et al. (2014); Halzen et al. (2016). In this method, the p-value of identifyingsignal events from a source is given by p value = 12 (cid:20) − erf (cid:18)(cid:113) q obs / (cid:19)(cid:21) , (4) igh-Energy Neutrinos from Seyfert Galaxies − − − e v e n t s / b i n / y r NGC 424 − − − NGC 1275Stochastic (High CR pressure) Stochastic (Modest CR pressure) Magnetic Reconnection Atmospheric − − − NGC 4945 − − − e v e n t s / b i n / y r Cen A − − − Circinus Galaxy − − − CGCG 164-019 E ν [GeV]10 − − − e v e n t s / b i n / y r ESO 138-1 E ν [GeV]10 − − − UGC 11910 E ν [GeV]10 − − − NGC 7582
Figure 15.