γ
Fe ν
p Going further with high-energy multi-messenger astronomy
Kumiko Kotera - Institut d’Astrophysique de Paris
EuCAPT Symposium - 07/05/2021 Multi-messengers!
Gravitational waves
Cosmic rays
p Fe ν Gamma rays ν
Neutrinos
2 Multi-messengers! cosmic rays + others —> temporal coincidence impossible (deflections) but studies of di!use fluxes
Gravitational waves GW + electromagnetic (+gamma) GW170817
neutrinos + gammas? TXS0506+056 (blazar flare) Cosmic rays
p Fe ν Gamma rays ν
Neutrinos Neutrinos
3 RESEARCH | RESEARCH ARTICLE
as a fitted parameter. The model parameters are the RA coordinates within each data-taking period. contributing events at boundary times). For the correlated and are expressed as a pair, (F100, g), The resultant P value is defined as the fraction of box-shaped time window, the uncertainties are where F100 is the flux normalization at 100 TeV. randomized trials yielding a value of TS greater than discontinuous and not well defined, but the un- The time-dependent analysis uses the same for- or equal to the one obtained for the actual data. certainties for the Gaussian window show that it mulation of the likelihood but searches for Because the detector configuration and event is consistent with the box-shaped time window Neutrino flares andclustering TXS in time as well0506+56 as space by introducing selections changed as shown in Table 1, the time- fit. Despite the different window shapes, which an additional time profile. It is performed sep- dependent analysis is performed by operating on lead to different weightings of the events as a IC-170922A – a 290arately TeV for two different Neutrino generic profile shapes: a each data-taking period separately. (A flare that function of time, both windows identify the same Gaussian-shaped time window and a box-shaped spans a boundaryRESEARCH between two periods could be time interval as significant. For the box-shaped time window. Each analysis varies the central partially detected in either period, but with re- time window, the best-fitting parameters are sim- ◥ trinos, IceCube provides real-time triggers for time of the window, T0,andthedurationTW duced significance.) An additional look-elsewhere ilar to those of the Gaussian window, with fluence RESEARCH ARTICLE SUMMARY observatories2 around the world measuring (from seconds to years) of the potential signal to correction then needs to be applied for a result in at 100 TeV and spectral index giveng-rays, by x-rays,E J100 optical,= radio, and gravitational T T 1:0 4 –2 waves, allowing for the potential identification find the four parameters (F100, g, 0, W)that an individual dataNEUTRINO segment, ASTROPHYSICS given by the ratio of 2:2þ0:8 10À TeV cm and g =2.2±0.2.This maximize the likelihood ratio, which is defined the total 9.5-year observation time to the obser- fluenceÀ  corresponds to an averageof even flux rapidly over fading sources. 0:7 15 –1 –2 –1 as the test statistic TS. (For the Gaussian time vation time of that data segment (30). 158 days of F100 =1:6þ0:6 10À TeVRESU LTS:cm A high-energys . neutrino-induced Multimessenger observations ofÀ a  muon track was detected on 22 September 2017, window, TW represents twice the standard de- When we estimate the significance of the time- Neutrinos from the direction of automatically generating an alert that was viation.) The test statistic includes a factor that flaring blazar coincidentdependent with result by performing the analysis at distributed worldwide ◥ corrects for the look-elsewhere effect arising TXS 0506+056 the coordinates of TXS 0506+056 onON randomized OUR WEBSITE within 1 min of detection and prompted follow-up from all of the possible time windows that could The results of thehigh-energy time-dependent analysis neutrino per- datasets, IceCube-170922A we allow in each trial a newRead the fit full for article all at http://dx.doi. searches by telescopes over RESEARCH ARTICLE RESEARCH | be chosen (30). formed at the coordinatesThe IceCube of Collaboration, TXS 0506+056Fermi-LAT, are MAGIC,theAGILE parameters:, ASAS-SN, HAWC,F100, g H.E.S.S.,, T0, TW. Weorg/10.1126/ find that the a broad range of wave- INTEGRAL, K anata, K iso, K apteyn, Liverpool Telescope, Subaru, Swift/NuSTAR, science.aat1378 lengths. On 28 September For each analysis method (time-integrated and shown in Fig. 1 for each of the six data periods. fraction of randomized trials that result...... in a more 2017, the Fermi Large Area VERITAS, and VLA/17B-403 teams*† –5 Downloaded from lower limit of 183 TeV, dependingtime-dependent), only weakly on a robusttrophysical significance spectrum, together estimate with is simulatedOne ofcannot the data be excluded. periods, Elect IC86bromagnetic from observations 2012 to 2015, significant excess than the real dataTelescope is 7 × 10 Collaborationfor reported that the di- the assumed astrophysicalobtained energy spectrum by performing (25). data, was the used identical to calculate analysis the probability on thatcontains a are avaluable significant to assess excess, the possible which association is identified of the box-shaped time window and 3 rection× 10–5 offor the neutrino the was coincident with a The vast majority of neutrinos detected by neutrino at the observed track energy and zenith a single neutrinoINTRO to an DU astrophysical CTIO N: Neutrinos source. are tracers of mic rays. The discovery of an extraterrestrial cataloged g-ray source, 0.1° from the neutrino IceCube arise from cosmic-raytrials interactions with randomized within angle datasets. in IceCube These is of are astrophysical produced origin.by This both time-windowFollowing thecosmic-ray alert, shapes. IceCube acceleration: The performed excess electrically consists a neutral Gaussiandiffuse flux of high-energy time window. neutrinos, Thisannounced fraction,direction. once The cor- source, a blazar known as TXS Downloaded from Earth’s atmosphere. Althoughby randomizingatmospheric neu- theprobability, event times the so-called and recalculating signalness of the eventof 13 ±complete 5 events analysis aboveand traveling of the relevant expectation at nearly data the prior speed from ofto light, the they rectedby IceCube for in the 2013, ratio has characteristic of the total prop- observation0506+056 at time a measured redshift of 0.34, was trinos are dominant at energies below 100 TeV, (14), was reported to be 56.5% (17). Although 31 October 2017.can Although escape the no densest additional environments excess and may erties that hint at contributions from extra- in a flaring state at the time with enhanced atmospheric background.be tracedThe back to significance their source of depends origin. High- togalactic the sources, IC86b although observation the individual time sources (9.5 years/3g-ray activity years), in the GeV range. Follow-up ob- their spectrum falls steeply with energy, allowing IceCube can robustly identify astrophysical neu- of neutrinos was found from the direction of TXS –4 –4 astrophysical neutrinos to be moreTable easily 1. identi-IceCubetrinos neutrino at PeV data energies, samples. for individual neutrinoson the0506+056 energies near ofenergy the the time events, neutrinos of the theirare alert, expected proximitythere to are be produced to resultsremain as inyet unidentified.P values Continuously of 2 × 10 mon-andservations 10 ,respec- by imaging atmospheric Cherenkov in blazars: intense extragalactic radio, optical, itoring the entire sky for astrophysical neu- telescopes, notably theMajorAtmospheric
fied at higher energies. The muon-neutrino as- at several hundred TeV, an atmospheric originthe coordinatesindications at the of 3s TXSlevel of 0506+056, high-energy neutrino and their tively, corresponding to 3.5s and 3.7s.Because http://science.sciencemag.org/ Six data-taking periods make up the full x-ray, and, in some cases, g-ray sources Gamma Imaging Cherenkov (MAGIC) Candidate9.5-year data sample. Sample Neutrino numbers Source:clustering in time.characterized This TXS is illustrated by relativistic 0506+056 jets in of Fig. 2, there is no a priori reason to prefer onetelescopes, of the revealed periods where http://science.sciencemag.org/ plasma pointing close to our line of the detected g-ray flux from the blazar Fig. 1. Event display for which shows the time-independent weight of generic time windows over the other, we take the correspond to the number of detector sight. Blazars are among the most reached energies up to 400 GeV. Mea- neutrino event IceCube- strings that were operational. During the individual events inpowerful the likelihood objects in the analysis Universe and during more significant one and include a trial factorsurements of of the source have also 170922A. The time at which a the IC86b data period.are widely speculated to be sources 2forthefinalsignificance,whichisthen3.5beens completed. at x-ray, optical, and DOM observed a signal is first three periods, the detector was still of high-energy cosmic rays. These cos- radio wavelengths. We have inves- reflected in the color of the hit,under construction. The last three periods The Gaussian timemic rays window generate high-energy is centered neutri- at 13 Outside the 2012–2015 time period, thetigated next models associating neutrino with dark blues for earliest hitscorrespond to different data-taking December 2014 [modifiednos and g-rays, Julian which day are (MJD) produced 57004] most significant excess is found using the Gauss-and g-ray production and find that and yellow for latest. Times with an uncertaintywhen of the ±21 cosmic days rays andaccelerated a duration in ian window in 2017 and includes the IceCube-correlation of the neutrino with the conditions and/or event selections with the the jet interact with nearby gas or shown are relative to the first 35 flare of TXS 0506+056 is statistically
T =110 days. The best-fitting parameters for 170922A event. This time window is centered DOM hit according to the trackfull 86-string detector. W þ24 photons. On 22 September 2017, the IC170922A:significant at the level of 3 standard on July 12, 2018 À cubic-kilometer IceCube Neutrino deviations (sigma). On the basis of the reconstruction, and earlier and the fluence J100 = ∫F100(t)dt and the spectral at 22 September 2017 with duration TW =19days, later times are shown with the Observatory2 detected0:9 a ~290-TeV4 –Downloaded from 2 2 0:4 redshift4 of TXS 0506+056, we derive index are given by EneutrinoJ100 =2 from:1þ a dir0:7ection10 consistentÀ TeV cm g =1.7±0.6,andfluenceE J100 =0:2þ0:2 constraints10À for the muon-neutrino same colors as the first and À Â –2 À Â
with the flaring g-ray blazar TXS luminosity for this source and find
290 TeV neutrino on July 12, 2018 last times, respectively. The Sample Start End at 100 TeV and g = 2.1 ± 0.2, respectively. The TeV cm at 100 TeV. No other event besides the 0506+056. We report the details of them to be similar to the luminosity total time the event took to joint uncertainty on these parameters is shown IceCube-170922A event contributes significantly IC40 5 April 2008 20 May 2009 this observation and the results of a observed in g-rays. cross the detector is ~3000 ns...... in Fig. 3 along withmultiwavelength a skymap follow-up showing campaign. the result to the best fit. As a consequence,IceCube the Coll. uncertainty The size of a colored sphere isIC59 20 May 2009 31 May 2010 CO NCLU SIO N: The energies of the ...... proportional to the logarithm of the time-dependentRATIO NALE: analysisMultimessenger performed astron- at the on the best-fitting window location andg width-rays and the neutrino indicate that IC79 31 May 2010 13 May 2011 http://science.sciencemag.org/ Science (2018) of the amount of light ...... location of TXS 0506+056omy aims for globally and in coordinated its vicinity spans the entire IC86c period, because anyblazar win- jets may accelerate cosmic rays observed at the DOM, with IC86a 13 May 2011 16 May 2012 observations of cosmic rays, neutri- to at least several PeV. The observed ...... during the IC86bnos, data gravitational period. waves, and electro- dow containing IceCube-170922A yields a similarassociation of a high-energy neutrino Signalness: 56.5% larger spheres corresponding IC86b 16 May 2012 18 May 2015 to larger signals. The total ...... The box-shapedmagnetic time radiation window across is a broad centered value of the test statistic. Following the trialwith cor- a blazar during a period of en- range of wavelengths. The combi- hanced g-ray emission suggests that charge recorded is ~5800 photoelectrons.IC86c Inset 18 May is an 2015 overhead perspective 31 October view 2017 of the event. The13 daysbest-fitting later track with direction duration is shownT as=158days(from an arrow, rection procedure for different observation periods ...... nation is expectedW to yield crucial Multimessenger observations of blazar TXS 0506+056. The blazars may indeed be one of the long- 0:50 consistent with a zenith angle 5:7þ0:30 degrees below the horizon. MJD 56937.81 toinformation MJD 57096.21, on the mechanisms inclusive50% of and 90%as described containment above, regions for the the significance neutrino IceCube- of thissought excess sources of very-high-energy IceCube, Fermi-LAT, MAGIC, AGILE, ASAS-SN,À HAWC, H.E.S.S, INTEGRAL, Kapteyn, EM13±5 flare above the backgroundenergizing of the[Science mostatmospheric powerfulPage astro- 361 8 170922A (2018) (dashed neutrinos, red no.6398, and solid gray contours, eaat1378] respectively),3.5σ cosmic rays, and hence responsible for physical sources. That the produc- overlain on a V-band optical image of the sky. Gamma-ray sources a sizable fraction of the cosmic neu- Kanata, Kiso, Liverpool, Subaru, Swift, VERITAS, VLA, Science 2018 tion of neutrinos is accompanied by in this region previously detected with the Fermi spacecraft are trino flux observed by IceCube. electromagnetic radiation from the shown as blue circles, with sizes representing their 95% positional ▪ The list of author affiliations is available in the full source favors the chances of a multi- uncertainty and labeled with the source names. The IceCube on July 12, 2018 wavelength identification. In par- neutrino is coincident with the blazar TXS 0506+056, whose article online. *The full lists of participating members for each ticular, a measured association of optical position is shown by the pink square. The yellow circle team and their affiliations are provided in the high-energy neutrinos with a flaring shows the 95% positional uncertainty of very-high-energy g-rays supplementary materials. source of g-rays would elucidate the detected by the MAGIC telescopes during the follow-up campaign. †Email: [email protected] Cite this article as IceCube Collaboration et al., mechanisms and conditions for ac- The inset shows a magnified view of the region around TXS 0506+056 Science 361, eaat1378 (2018). DOI: 10.1126/
celeration of the highest-energy cos- on an R-band optical image of the sky. science.aat1378 IMAGES: PHOASAS-SN FOR THE V-BAND OPTICAL; KANATA FOR THE R-BAND IN MAGNIFIED VIEW
The IceCube Collaboration et al., Science 361, 146 (2018) 13 July 2018 1 of 1
4
Fig. 1. Time-dependent analysis results. The orange curve corresponds fitting time windows (durations TW) over all times T0, with the height to the analysisneutrino using the Gaussian-shaped flare time profile. The central time T indicating[Science the significance 361 (2018) of that window.no.6398, In each 147-151] period, the most 0
and width TW are plotted for the most significant excess found in each significant time window forms a plateau, shaded in blue. The large blue period, with the P value of that result indicated by the height of the peak. band centered near 2015 represents the best-fitting 158-day time window Fig. 2. Fermi-LATand MAGICThe observations blue curveMarkus of IceCube-170922A corresponds Ahlers (NBI)’ tos the analysispixel, using using detected theNeutrinos box-shaped photons with and energyγ time-rays of 1from to 300found Extragalactic GeV in using a 2° by the 2° Sources box-shaped timeAugust profile. The28, 2018 vertical dotted slide line 19 in IC86c location. Sky position of IceCube-170922A in J2000 equatorial coordinates region around TXS0506+056. The map has a pixel size of 0.02° and was overlaying the g-ray countsprofile. from Fermi The-LAT curve above 1traces GeV (A) the and theouter signal edgesmoothed of the with superposition a 0.02°-wide Gaussian of the kernel. best- MAGICindicates data are shown the as time of the IceCube-170922A event. significance as observed by MAGIC (B) in this region. The tan square signal significance for g-rays above 90 GeV. Also shown are the locations of indicates the position reported in the initial alert, and the green square a g-ray source observed by Fermi-LAT as given in the Fermi-LAT Third indicates the final best-fittingIceCube position Collaboration, from follow-up reconstructionsScience 361, (18 147). –151Source (2018) Catalog 13 July (3FGL) 2018 (23) and the Third Catalog of Hard Fermi-LAT 2 of 5 Gray and red curves show the 50% and 90% neutrino containment regions, Sources (3FHL) (24) source catalogs, including the identified positionally respectively, including statistical and systematic errors. Fermi-LAT data are coincident 3FGL object TXS 0506+056. For Fermi-LAT catalog objects, shown as a photon counts map in 9.5 years of data in units of counts per marker sizes indicate the 95% CL positional uncertainty of the source.
The IceCube Collaboration et al., Science 361, eaat1378 (2018) 13 July 2018 2 of 8 RESEARCH | RESEARCH ARTICLE
as a fitted parameter. The model parameters are the RA coordinates within each data-taking period. contributing events at boundary times). For the correlated and are expressed as a pair, (F100, g), The resultant P value is defined as the fraction of box-shaped time window, the uncertainties are where F100 is the flux normalization at 100 TeV. randomized trials yielding a value of TS greater than discontinuous and not well defined, but the un- NeutrinoThe time-dependent flares analysis and uses the TXS same for- 0506+56or equal to the one obtainedPost-trials for the actual p-value data. certainties for association: for the Gaussian window 3.0σ show that it mulation of the likelihood but searches for Because the detector configuration and event is consistent with the box-shaped time window Theclustering Multi-Messenger in time as well as space by introducing selections Light changed as shownCurve in Table 1, the time- fit. Despite the different window shapes, which an additional time profile. It is performed sep- dependent analysis is performed by operating on lead to different weightings of the events as a IceCubearately Collaboration, for two different Fermi-LAT, generic profile MAGIC, shapes:a AGILE,each ASAS-SN, data-taking periodHAWC, separately. H.E.S.S., (A flareINTEGRAL, that function et al., ofScience time, both (2018) windows identify the same Gaussian-shaped time window and a box-shaped spans a boundaryRESEARCH between two periods could be time interval as significant. For the box-shaped time window. Each analysis varies the central partially detected in either period, but with re- time window, the best-fitting parameters are sim- ◥ trinos, IceCube provides real-time triggers for time of the window, T0,andthedurationTW duced significance.) An additional look-elsewhere ilar to those of the Gaussian window, with fluence RESEARCH ARTICLE SUMMARY observatories2 around the world measuring VHE(from gamma-rays seconds to years) of the potential signal to correction then needs to be applied for a result in at 100 TeV and spectral index giveng-rays, by x-rays,E J100 optical,= radio, and gravitational T T 1:0 4 –2 waves, allowing for the potential identification find the four parameters (F100, g, 0, W)that an individual dataNEUTRINO segment, ASTROPHYSICS given by the ratio of 2:2þ0:8 10À TeV cm and g =2.2±0.2.This maximize the likelihood ratio, which is defined the total 9.5-year observation time to the obser- fluenceÀ  corresponds to an averageof even flux rapidly over fading sources. 0:7 15 –1 –2 –1 as the test statistic TS. (For the Gaussian time vation time of that data segment (30). 158 days of F100 =1:6þ0:6 10À TeVRESU LTS:cm A high-energys . neutrino-induced Multimessenger observations ofÀ a  muon track was detected on 22 September 2017, GeVwindow, gamma-raysTW represents twice the standard de- When we estimate the significance of the time- Neutrinos from the direction of automatically generating an alert that was viation.) The test statistic includes a factor that flaring blazar coincidentdependent with result by performing the analysis at distributed worldwide ◥ corrects for the look-elsewhere effect arising TXS 0506+056 the coordinates of TXS 0506+056 onON randomized OUR WEBSITE within 1 min of detection and prompted follow-up from all of the possible time windows that could The results of thehigh-energy time-dependent analysis neutrino per- datasets, IceCube-170922A we allow in each trial a newRead the fit full for article all at http://dx.doi. searches by telescopes over RESEARCH ARTICLE RESEARCH | X-raysbe chosen (30). formed at the coordinatesThe IceCube of Collaboration, TXS 0506+056Fermi-LAT, are MAGIC,theAGILE parameters:, ASAS-SN, HAWC,F100, g H.E.S.S.,, T0, TW. Weorg/10.1126/ find that the a broad range of wave- INTEGRAL, K anata, K iso, K apteyn, Liverpool Telescope, Subaru, Swift/NuSTAR, science.aat1378 lengths. On 28 September For each analysis method (time-integrated and shown in Fig. 1 for each of the six data periods. fraction of randomized trials that result...... in a more 2017, the Fermi Large Area VERITAS, and VLA/17B-403 teams*† –5 Downloaded from lower limit of 183 TeV, dependingtime-dependent), only weakly on a robusttrophysical significance spectrum, together estimate with is simulatedOne ofcannot the data be excluded. periods, Elect IC86bromagnetic from observations 2012 to 2015, significant excess than the real dataTelescope is 7 × 10 Collaborationfor reported that the di- the assumed astrophysicalobtained energy spectrum by performing (25). data, was the used identical to calculate analysis the probability on thatcontains a are avaluable significant to assess excess, the possible which association is identified of the box-shaped time window and 3 rection× 10–5 offor the neutrino the was coincident with a The vast majority of neutrinos detected by neutrino at the observed track energy and zenith a single neutrinoINTRO to an DU astrophysical CTIO N: Neutrinos source. are tracers of mic rays. The discovery of an extraterrestrial cataloged g-ray source, 0.1° from the neutrino IceCube arise from cosmic-raytrials interactions with randomized within angle datasets. in IceCube These is of are astrophysical produced origin.by This both time-windowFollowing thecosmic-ray alert, shapes. IceCube acceleration: The performed excess electrically consists a neutral Gaussiandiffuse flux of high-energy time window. neutrinos, Thisannounced fraction,direction. once The cor- source, a blazar known as TXS X-rays spectral index Downloaded from Earth’s atmosphere. Althoughby randomizingatmospheric neu- theprobability, event times the so-called and recalculating signalness of the eventof 13 ±complete 5 events analysis aboveand traveling of the relevant expectation at nearly data the prior speed from ofto light, the they rectedby IceCube for in the 2013, ratio has characteristic of the total prop- observation0506+056 at time a measured redshift of 0.34, was trinos are dominant at energies below 100 TeV, (14), was reported to be 56.5% (17). Although 31 October 2017.can Although escape the no densest additional environments excess and may erties that hint at contributions from extra- in a flaring state at the time with enhanced atmospheric background.be tracedThe back to significance their source of depends origin. High- togalactic the sources, IC86b although observation the individual time sources (9.5 years/3g-ray activity years), in the GeV range. Follow-up ob- their spectrum falls steeply with energy, allowing IceCube can robustly identify astrophysical neu- of neutrinos was found from the direction of TXS –4 –4 astrophysical neutrinos to be moreTable easily 1. identi-IceCubetrinos neutrino at PeV data energies, samples. for individual neutrinoson the0506+056 energies near ofenergy the the time events, neutrinos of the theirare alert, expected proximitythere to are be produced to resultsremain as inyet unidentified.P values Continuously of 2 × 10 mon-andservations 10 ,respec- by imaging atmospheric Cherenkov in blazars: intense extragalactic radio, optical, itoring the entire sky for astrophysical neu- telescopes, notably theMajorAtmospheric
fied at higher energies. The muon-neutrino as- at several hundred TeV, an atmospheric originthe coordinatesindications at the of 3s TXSlevel of 0506+056, high-energy neutrino and their tively, corresponding to 3.5s and 3.7s.Because http://science.sciencemag.org/ Six data-taking periods make up the full x-ray, and, in some cases, g-ray sources Gamma Imaging Cherenkov (MAGIC)
optical http://science.sciencemag.org/ Candidate9.5-year data sample. Sample Neutrino numbers Source:clustering in time.characterized This TXS is illustrated by relativistic 0506+056 jets in of Fig. 2, there is no a priori reason to prefer onetelescopes, of the revealed periods where plasma pointing close to our line of the detected g-ray flux from the blazar Fig. 1. Event display for which shows the time-independent weight of generic time windows over the other, we take the correspond to the number of detector sight. Blazars are among the most reached energies up to 400 GeV. Mea- neutrino event IceCube- strings that were operational. During the individual events inpowerful the likelihood objects in the analysis Universe and during more significant one and include a trial factorsurements of of the source have also 170922A. The time at which a the IC86b data period.are widely speculated to be sources 2forthefinalsignificance,whichisthen3.5beens completed. at x-ray, optical, and DOM observed a signal is first three periods, the detector was still The Gaussian timeof high-energy window cosmic is rays. centered These cos- at 13 Outside the 2012–2015 time period, theradio next wavelengths. We have inves- reflected in the colorradio of the hit, under construction. The last three periods mic rays generate high-energy neutri- tigated models associating neutrino with dark blues for earliest hitscorrespond to different data-taking December 2014 [modifiednos and g-rays, Julian which day are (MJD) produced 57004] most significant excess is found using the Gauss-and g-ray production and find that and yellow for latest. Times with an uncertaintywhen of the ±21 cosmic days rays andaccelerated a duration in ian window in 2017 and includes the IceCube-correlation of the neutrino with the conditions and/or event selections with the the jet interact with nearby gas or shown are relative to the first 35 flare of TXS 0506+056 is statistically
T =110 days. The best-fitting parameters for 170922A event. This time window is centered DOM hit according to the trackfull 86-string detector. W þ24 photons. On 22 September 2017, the significant at the level of 3 standard on July 12, 2018 À cubic-kilometer IceCube Neutrino deviations (sigma). On the basis of the reconstruction, and earlier and the fluence J100 = ∫F100(t)dt and the spectral at 22 September 2017 with duration TW =19days, later times are shown with the Observatory2 detected0:9 a ~290-TeV4 –Downloaded from 2 2 0:4 redshift4 of TXS 0506+056, we derive index are given by EneutrinoJ100 =2 from:1þ a dir0:7ection10 consistentÀ TeV cm g =1.7±0.6,andfluenceE J100 =0:2þ0:2 constraints10À for the muon-neutrino same colors as the first and À Â –2 À Â
with the flaring g-ray blazar TXS luminosity for this source and find last times, respectively. The Sample Start End at 100 TeV and g = 2.1 ± 0.2, respectively. The TeV cm at 100 TeV. No other event besides the on July 12, 2018 0506+056. We report the details of them to be similar to the luminosity total time the event took to joint uncertainty onthis observation these parameters and the results is of shown a IceCube-170922A event contributes significantlyobserved in g-rays. cross the detector is ~3000IceCube ns.IC40, Fermi-LAT, 5 April 2008 MAGIC, 20 May 2009 AGILE, ASAS-SN, HAWC, H.E.S.S, INTEGRAL, Kapteyn, ...... in Fig. 3 along withmultiwavelength a skymap follow-up showing campaign. the result to the best fit. As a consequence,Page 16 the uncertainty The size of a colored sphere isIC59 20 May 2009 31 May 2010 CO NCLU SIO N: The energies of the Kanata,...... Kiso, Liverpool, Subaru, Swift, VERITAS, VLA, Science 2018 proportional to the logarithm of the time-dependentRATIO NALE: analysisMultimessenger performed astron- at the on the best-fitting window location andg width-rays and the neutrino indicate that IC79 31 May 2010 13 May 2011 http://science.sciencemag.org/ of the amount of light ...... location of TXS 0506+056omy aims for globally and in coordinated its vicinity spans the entire IC86c period, because anyblazar win- jets may accelerate cosmic rays observed at the DOM, with IC86a 13 May 2011 16 May 2012 observations of cosmic rays, neutri- to at least several PeV. The observed ...... during the IC86bnos, data gravitational period. waves, and electro- dow containing IceCube-170922A yields a similarassociation of a high-energy neutrino larger spheres corresponding IC86b 16 May 2012 18 May 2015 to larger signals. The total ...... The box-shapedmagnetic time radiation window across is a broad centered value of the test statistic. Following the trialwith cor- a blazar during a period of en- range of wavelengths. The combi- hanced g-ray emission suggests that charge recorded is ~5800 photoelectrons.IC86c Inset 18 May is an 2015 overhead perspective 31 October view 2017 of the event. The13 daysbest-fitting later track with direction duration is shownT as=158days(from an arrow, rection procedure for different observation periods z~0.337 Paiano...... et al. 2018 nation is expectedW to yield crucial Multimessenger observations of blazar TXS 0506+056. The blazars5 may indeed be one of the long- 0:50 consistent with a zenith angle 5:7þ0:30 degrees below the horizon. MJD 56937.81 toinformation MJD 57096.21, on the mechanisms inclusive50% of and 90%as described containment above, regions for the the significance neutrino IceCube- of thissought excess sources of very-high-energy À EM flare energizing the[Science most powerful astro- 361170922A (2018) (dashed red no.6398, and solid gray contours, eaat1378] respectively), cosmic rays, and hence responsible for physical sources. That the produc- overlain on a V-band optical image of the sky. Gamma-ray sources a sizable fraction of the cosmic neu- tion of neutrinos is accompanied by in this region previously detected with the Fermi spacecraft are trino flux observed by IceCube. electromagnetic radiation from the shown as blue circles, with sizes representing their 95% positional ▪ The list of author affiliations is available in the full source favors the chances of a multi- uncertainty and labeled with the source names. The IceCube on July 12, 2018 wavelength identification. In par- neutrino is coincident with the blazar TXS 0506+056, whose article online. *The full lists of participating members for each ticular, a measured association of optical position is shown by the pink square. The yellow circle team and their affiliations are provided in the high-energy neutrinos with a flaring shows the 95% positional uncertainty of very-high-energy g-rays supplementary materials. source of g-rays would elucidate the detected by the MAGIC telescopes during the follow-up campaign. †Email: [email protected] Cite this article as IceCube Collaboration et al., mechanisms and conditions for ac- The inset shows a magnified view of the region around TXS 0506+056 Science 361, eaat1378 (2018). DOI: 10.1126/
celeration of the highest-energy cos- on an R-band optical image of the sky. science.aat1378 IMAGES: PHOASAS-SN FOR THE V-BAND OPTICAL; KANATA FOR THE R-BAND IN MAGNIFIED VIEW
The IceCube Collaboration et al., Science 361, 146 (2018) 13 July 2018 1 of 1
Fig. 1. Time-dependent analysis results. The orange curve corresponds fitting time windows (durations TW) over all times T0, with the height to the analysisneutrino using the Gaussian-shaped flare time profile. The central time T indicating[Science the significance 361 (2018) of that window.no.6398, In each 147-151] period, the most 0
and width TW are plotted for the most significant excess found in each significant time window forms a plateau, shaded in blue. The large blue period, with the P value of that result indicated by the height of the peak. band centered near 2015 represents the best-fitting 158-day time window Fig. 2. Fermi-LATand MAGICThe observations blue curveMarkus of IceCube-170922A corresponds Ahlers (NBI)’ tos the analysispixel, using using detected theNeutrinos box-shaped photons with and energyγ time-rays of 1from to 300found Extragalactic GeV in using a 2° by the 2° Sources box-shaped timeAugust profile. The28, 2018 vertical dotted slide line 19 in IC86c location. Sky position of IceCube-170922A in J2000 equatorial coordinates region around TXS0506+056. The map has a pixel size of 0.02° and was overlaying the g-ray countsprofile. from Fermi The-LAT curve above 1 traces GeV (A) the and theouter signal edgesmoothed of the with superposition a 0.02°-wide Gaussian of the kernel. best- MAGICindicates data are shown the as time of the IceCube-170922A event. significance as observed by MAGIC (B) in this region. The tan square signal significance for g-rays above 90 GeV. Also shown are the locations of indicates the position reported in the initial alert, and the green square a g-ray source observed by Fermi-LAT as given in the Fermi-LAT Third indicates the final best-fittingIceCube position Collaboration, from follow-up reconstructionsScience 361, (18 147). –151Source (2018) Catalog 13 July (3FGL) 2018 (23) and the Third Catalog of Hard Fermi-LAT 2 of 5 Gray and red curves show the 50% and 90% neutrino containment regions, Sources (3FHL) (24) source catalogs, including the identified positionally respectively, including statistical and systematic errors. Fermi-LAT data are coincident 3FGL object TXS 0506+056. For Fermi-LAT catalog objects, shown as a photon counts map in 9.5 years of data in units of counts per marker sizes indicate the 95% CL positional uncertainty of the source.
The IceCube Collaboration et al., Science 361, eaat1378 (2018) 13 July 2018 2 of 8 16
expectation of an associated HE neutrino detection by Table 8. Parameter values for hadronic models (HMs) for IceCube. TXS 0506+056 discussed in the text and presented in Fig. 6.
3.3. Hadronic Models (HMs) HM1 HM2 HM3 In hadronic scenarios, while the low-energy peak in the B! [G] 85 16 blazar’s SED is explained by synchrotron radiation from R! [in 10 cm] 2 3 4.5 14 relativistic primary electrons, the HE peak is explained δ 5.2 10 15 by EM cascades induced by pions and muons as de- 43 1 Table 7. Model-specific parameter values for leptonic models (LMs)L! for[in TXS 10 0506+056erg s− ]9.3 discussed 0.6 in 0.06 the text cay products of the photomeson production (Mannheim e 1993; M¨ucke et al. 2003), or synchrotron radiation from se,1 1.8 LMBB1a LMBB1b LMBB1c LMBB2a LMBB2b LMBB2c LMPL1a LMPL1b LMPL2a LMPL2b relativistic protons in the ultrahigh-energy range (Aha- s 4.2 3.6 3.6 (max) 44 1 e,2 Lp! [10 erg s− ]0.540.270.341 5.4100.540.541010 ronian 2000; M¨ucke et al. 2003). We coin this scenario 2 ! sp 22.53 2 2γ 2e,min [in 2 10 ]6.311 2 2 2 “HM”, which stands for Hadronic Model6 ,inreference6 2 γ! 1310 3 10 1111111 p,min × × γe,! br [in 10 ]7.96.35 to the hadronic origin8 of the γ-rays. The synchrotron γp,! max [10 ]3030301.60.160.01630300.0160.0164 γ! 10 and IC emission of3 secondary pairs may have an im- e,max uext! [erg cm− ]0.0330.0330.0670.040.08 46 1 5 portant contributionT ! [K] to the bolometric radiation 3 of10 the Lp! [in 10 erg s− ]2.70.10.01n/a source. In contrast to the leptonic scenario (Sec.× 3.2), α n/asp 3 2 32.1 2
the parametersεmin! [keV] describing the proton distribution n/a can be 0.05 γp,! min 1 directly constrainedεmax! [keV] from the NuSTAR and Fermi n/aLAT 5 9 γ! 2 10 data. For the TXS 0506+056 flare, in the hadronic sce- p,max × Notenario,—See the Table SED5 for can parameter be fully definitions, explained and without Table 6 invokingfor parameter values common to all LMs. In LMBB models, the external photon fieldexternal is blackbody-like radiation with fields. comoving temperature T !,whileinLMPLmodels,itisapower-lawbetweencomovingenergiesNote—Parameter definitions are providedε inmin! Tableand 5ε.max! , withThere photon are index diffαerent.Inallcases, combinationsuext! is the of comovingparameters energy that density of the external photon field. Note that the isotropic-equivalent 4 cosmic-ray proton luminosity is Lp = δ Lp! . can successfullyCandidate explain Neutrino the SED in Source: the HM sce- TXS 0506+056 Keivani et al. 2018 nario (B¨ottcher et al. 2013; Cerruti et al. 2015). As a starting point,leptonic we search model for combinations of δ and hadronic model 1047 −11 1047 10 δ=δ ∼ 5 B! that lead to(max) rough energy equipartition between eq Lp δ=10 LMBB1a the10− magnetic10 field× (max) and protons, since the primary elec- 10−10 LMBB1b 2 Lp LM δ=15 HM tron energy density is negligible in this scenario. With LMBB1c LMBB2a analytical calculations we derive rough estimates10 of46 the cascade 1046 LMBB2b ] 1 ] parameter values for equipartition: δeq 5, Beq! 80 G, LMBB2c ] 1 1 − −12 − −11 −11 − ] 16 9 ∼ ∼ ] 10 1 10 1 10 s LMPL1a − R 10 cm, and ε 10 GeV (Petropoulou & − eq! p,! max ! 2 − s s 2 ∼ ∼ 2 LMPL1b − Dermer 2016). − ! LMPL2a 1045 1045 The parameter values obtained by numerically mod- LMPL2b [erg cm [erg cm HM3 ε elingε −12 the SED (see Fig. 6) are summarized in Table 8 −12 [erg cm
10 ν 10 F F at z=0.33 [erg s at z=0.33 [erg s ε ε ε ε ε
and are similar to the estimates provided above. The L L
F −13 ε ε ν
10 44 ε 44 jet power computed for this parameter set (HM1)10 is 10 close to the minimum value expected in the hadronic −13 X-rays −13 X-rays scenarios.10 10−12 More specifically, the absolute power of a 10 10−12 two-sided jet inferred for these parameters is Lj 103 104 105 103 104 105 2 2 47 101 43 ≈ 1043 2πcR! (δ/2) (up! + ue! + uB! ) 4 10 erg s− ,with −14 0 5 3 ∼10 × 15 10 0 5 10 15 up! 2uB! 10500 erg cm10 − ,where10 up! , u10e! , uB! are comov- 102 10 4 10 6 10 8 ≈ ∼ ε [eV] 10 10 ε [eV] 10 10 ing energy densities of relativistic protons, electrons, and ε leptonic with radiatively subdominant hadronic component cascade implies high X-rayν [TeV] level to match magnetic fields, respectively. As demonstrated in Fig. 6, detection proba. with IC in real time during 6-month flare = 1-2% observed neutrino flux Figurethe emission 4. fromLeptonic the EM Model cascade (LMBB2b) forms a “bridge” for the be- Figure 6. Hadronic Model (HM3) for the SED of TXS 0506+056 flare (Ep. 1). Two SED cases (gray TXS 0506+056 flare (Ep. 1), as computed for different values tween the low-energy and high-energy peaks of the SEDlucky?Figure 5. Upper limits on the all-flavorGao neutrinoet al. 2018 (ν +¯ν) lines) are plotted against the observations (colored points, of the Doppler factor (gray curves), together with resulting for δ = δ (graybut dottedthe 2014-2015 line). Despite neutrino minimizing flares require the higherfluxes rates predicted for our modeling of theCerruti SED in et the al. leptonic2018 showing allowedeq ranges at 90% confidence), one with all-flavor neutrino fluxes (red curves) and electromagnetic (L~ 1047 erg/s over 158 days ~ 4 x average gamma-ray(LMx) andluminosity) hadronic (HMx) models. Zhang, Fang & Li 2018 hadronicpower of component the jet, the set adopted to the set maximum of parameters allowed for proton HM1 observations (colored points, showing allowed ranges at 90% (max) 50 1 Gokus et al.11 2018 luminositycannot explainLp the2 SED10 dueerg to s the− (solid associated gray), significant and the confidence). Photon attenuation at εγ > 3 10 eV due to ≈ × Sahakyan× 2018 otherEM set cascade to twice component. this maximal value (dashed gray line). interactionsProton maximum with the energy extragalactic — Motivated background∼ by the light hypoth- is not Multi-zone or more complicated models? CorrespondingThe EM cascade all-flavor emission neutrino can fluxes be forsuppressed the maximal if the esisincluded that blazars here. are UHECR accelerators, i.e., at ener- - Additional photomeson production by external radiation fields Murase, 18 (solidsource red) becomes and “twice less opaque maximal” to (dashed the intra-source line) casesγγ areab- gies above 3 10 eV (Murase et al. 2012), we ex- - hadronuclear production (e.g., jet-cloud interaction)11 Oikonomou,× also shown. Photon attenuation at εγ > 3 10 eV due to plore the effect of the proton maximum energy on the More parameters introduced, the setup is× ad-hoc Petropoulou 2018 interactions with the extragalactic background∼ light is not neutrino flux upper limits. We thus explore cases with included here. 8 9 9 6 γ! =1.6 10 , 1.6 10 , and 3 10 – see Ta- p,max × × × ble 7. Our results on the neutrino fluxes are presented In what follows, we show that our neutrino flux limits in Fig. 5. are fairly insensitive to the exact parameter values that Neutrino spectra in the LMBB1x models are more may affect the photomeson production optical depth. extended in energy compared to the default case (LMBB2b). They peak around 10 PeV (100 PeV) for Multi-messengers! cosmic rays + others —> temporal coincidence impossible (deflections) but studies of di!use fluxes
Gravitational waves GW + electromagnetic (+gamma) GW170817
neutrinos + gammas? TXS0506+056 (blazar flare) Cosmic rays
p Fe ν Gamma rays
ν GW + neutrinos ???
Neutrinos
7 15
700 70 7 700 70 7 1 1
0.9 0.9 0.8 0.8 0.7 0.7 0.6 0.6 0.5 0.5 0.4 0.4 0.3 0.3
0.2 0.2
0.1 0.1 0 0 10 -23 10 -22 10 -21 10 -23 10 -22 10 -21
Figure 1. Fraction of simulated astrophysical GW+neutrino events whose significance exceeds a threshold as a function of the GW hrss, assuming a sine-Gaussian gravitational waveform described in Section 3.1. Separate curves are shown for the cases Populationof detections constraints by IceCube+LIGO from GW+neutrino (left) and Antaresnon-detection+LIGO (right). Results are shown for different significance thresholds, 1 with thresholds set at the most significant event [GW+νAlbert(obs.)], et al. 2019 as ApJ well 870 134 as thresholds corresponding to FARs 1/10yr− and 1 Advanced LIGO IceCube+ANTARES 1 1 1/50 yr− . For comparison, we further show results for GW-only searches, also for FARs 1/10 yr− and 1/50 yr− .Onthetop 2 2 23 of the figures we also show the source distance corresponding to hrss,assumingEGW =10− M c .Below5 10− ,wefind " × Upper limits thaton rate the density GW of search is unable to detect events (shaded area). GW+neutrino sources Short GRBs were found to have comparable beam- 109 ing factors based on their observed jet breaks and rate (Berger 2014). Nevertheless, the detection of 108 GRB 170817A at a higher observing angle of 30◦ 15◦ ∼ ± 107 (Abbott et al. 2018a) implied weaker effective beaming. Radio observations of the GRB’s afterglow indicate that
106 the outflow had a narrowly collimated relativistic jet with θj < 5◦ as well as a broader, less energetic compo- 105 nent (Mooley et al. 2018a; Ghirlanda et al. 2018). The origin of this structured outflow remains the subject of 104 active debate (Alexander et al. 2018; Lazzati et al. 2018; Mooley et al. 2018b; Ioka & Nakamura 2018; Gottlieb 103 et al. 2018; Haggard et al. 2017; Alexander et al. 2018; -5 -4 -3 -2 -1 10 10 10 10 10 Veres et al. 2018). It is instructive to compare the present limits to pre- Figure 2. Upper limits for the rate density of vious results. Here we look at the latest estimates that 8 GW+neutrino sources as functions of EGW, for different val- used Initial LIGO-Virgo and the partially completed ues of Eiso,ν (see numerical values of Eiso,ν in the figure), for IceCube detector (Aartsen et al. 2014a). Considering a sine-Gaussian gravitational waveform described in Section 2 2 a fiducial source emission of EGW = 10− M c and 3.1. We assume a beaming factor fb = 10. For comparison, 51 # we show the rate density of local core-collapse supernovae Eν,iso = 10 erg, assuming a beaming factor of fb = 10, (CCSNe; dashed line, rate error region shown in blue), and this previous search obtained a joint source rate upper 7 3 1 that of BNS mergers (dotted line, rate error region shown in limit of 1.1 10 Gpc− yr− . The present search updates × 4 3 1 red). this constraint to 4 10 Gpc− yr− , an improvement × of more than 2 orders of magnitude.
gamma-ray bursts (GRBs), it can be as low as fb ! 14 (Liang et al. 2007). For long GRBs, typical jet opening 3.3. Discussion angles are θ =3◦ 10◦, with some extending up to j − Here we briefly review the expected emission parame- 20◦ (Berger 2014), corresponding to a beaming factor ters of sources of interest, and compare the our rate den- ≈ 1 3 f =(1 cos θ )− = 10 10 . b − j − sity constraints to expectations. While our constraints 104
103 Number of events
102
10
1 54800 55000 55200 55400 55600 55800 56000 56200 Modified Julian Day
Figure 7. Time distribution of the reconstructed events. Upper histogram (black): distribution of well-reconstructed events (including downgoing muons). Bottom histogram (red): distribution of the Why focus on transient sources?events selected by this analysis.
Average number of events required for a 5σ discovery for source located at a declination of -40 deg, E−2 spectrum time-dependent neutrino searches reduce the background for sub-PeV energies (atmospheric neutrinos+muons)
Less events needed for shorter flares
ANTARES Coll. JCAP12(2015)014
Figure 8.Averagenumberofeventsrequiredfora5σ discovery (50% probability) for a source located o 2 at a declination of -40 and following an E− energy spectrum as a function of the total width of the flaring periods (solid line). These numbers are compared to those obtained without using the timing information (dashed line).
Real-time analysis + multi-messenger follow-up on alerts–10– increase statistical significance of signals
Meszaros et al. 2019 9 The new high-energy transient zoo
AGN/Blazars flares, time-variabilities
Young pulsars
Neutron star Magnetars mergers (AXP/SGR)
Long Gamma Ray Bursts
Black hole Superluminous mergers Tidal disruption Supernovae events 10 n, g n, g blazar flare tidal disruption event
star
dust torus
disk n, g n n, g
jet progenitor
wind n, (g) n, g choked jet long gamma-ray burst engine-driven circumstellar material supernova supernova
gas n, g neutron star n, g n, (g) black hole
wind double black hole merger short gamma-ray burst neutron star merger ejecta The new high-energy transient zoo Murase & Bartos 2019 Figure 5 Schematic picture of various high-energy multi-messenger transients.
Table 1 List of multi-messenger transients that can be promising emitters of high- energy neutrinos and/or gravitational waves. AGN/Blazars flares, time-variabilities Source Rate density EM Luminosity Duration Typical Counterpart 3 1 1 [Gpc− yr− ] [erg s− ] [s] Blazar flarea 10 100 1046Young1048 pulsars106 107 broadband − − − Tidal disruption event 0.01 0.1 1047 1048 106 107 jetted (X) − − − 100 1000 1043.5 1044.5 > 106 107 tidal disruption event (optical,UV) − − − Long GRB 0.1 1 1051 1052 10 100 prompt (X, gamma) − − − Short GRB 10 100 1051 1052 0.1 1 prompt (X, gamma) Neutron star − − − Low-luminosity GRB 100 1000 1046 1047 1000 10000 prompt (X, gamma) Magnetars− − − mergersGRB afterglow < 1046 1051, > 1 10000 afterglow (broadband) (AXP/SGR) − − Supernova (II) 105 1041 1042 > 105 supernova (optical) − Supernova (Ibc) 3 104 1041 1042 > 105 supernova (optical) × − Hypernova 3000 1042 1043 > 106 supernova (optical) − NS merger 300 3000 1041 1042 Long> 105 kilonova (optical/IR) − − 1043 > 107 108 radio flare (broadband) Gamma− Ray BH merger 10 100 ? ? ? − WD merger 104 105 1041 1042 Bursts> 105 merger nova (optical) − − aBlazar flares such as the 2017 flare of TXS 0506+056 are assumed for the demonstration. Abbreviations: BH, black hole; EM, electromagnetic; GRB, gamma-ray burst; NS, neutron star; WD, white dwarf.
Black4.1. holeBlazar Flares Superluminous Tidal disruption Supernovae mergersIn general, blazars are highly variable objects that show broadband spectra from radio, op- tical, X-ray, and gamma-rays.events In the standard leptonic scenario for SEDs, the low-energy and high-energy humps are explained by synchrotron emission and inverse-Compton radia- 11 tion from non-thermal electrons, respectively. For BL Lac objects that typically belong to a low-luminous class of blazars, seed photons for the inverse-Compton scattering are mainly supplied by the electron synchrotron process. In contrast, flat-spectrum radio quasars (FS-
14 Murase and Bartos 30 Chapter 2. Comparing explosive transients
30 Chapter 2. Comparing explosive transients
A "Hillas diagram" for high-energy neutrino transients Guépin & KK 2017
AGN/Blazars Neutrino maximal energy Eν (Γ = 1) Neutrino maximal energy Eflares,ν (Γ = time-variabilities 10) 1056 1056 1021 tvar 9 tvar L 10 L synchrotron losses tot > Young pulsars11 tot > 52 10 54 52 13 10 54 19 10 NS mergers 10 10 10 9 erg 10 15 erg 10 11 10 ) ) 48 10 13 48 Blazar flares 1 1 10 10 LL 17 −
− 10 10 BH mergers GRBs SLSNe Neutron44 star 44 Magnetar GF 10 Magnetar IB 10 15 (eV)
Magnetars (erg s ν (erg s PS16cgx SNe TDEs BH mergers 10 mergers E
bol 15 bol 40 (AXP/SGR) 40 10 L L 10 Novae 10 1013 Magnetar SB 1015 13 1036 1036 10 11 13 Crab flares 10 10 11 32 32 Gamma10 Ray 10 luminosity bounds 10 Bursts (GRBs) 109 4 2 0 2 4 6 8 4 2 0 2 4 6 8 10− 10− 10 10 10 10 10 10− 10− 10 10 10 10 10 tvar (s) tvar (s)
Black hole Superluminous mergers Tidal disruption Supernovae events 12
Figure 2.2: Maximum accessible proton energy Ep,max (left column) and corresponding maximum accessible neutrino energy Eν,max (right column) as a function of the variability timescale tvar and the bolometric luminosity Lbol of a flaring source, with bulk Lorentz factor Γ =1, 10, 100 (from top to bottom). Overlaid are examples of the location of benchmark explosive transients in the Lbol tvar parameter space (see Section 2.4). The beige region indicates the domain where no source is expected− to be found because of the excessive energy budget. The dots locate recently discovered categories of transients (Kasliwal, 2011), superluminous supernovae (SLSNe), peculiar supernovae, and luminous red novae. The small square box (labeled SNe) and the short diagonal line on its upper left indicate core- collapse and thermonuclear supernovae, respectively. Low-luminosity GRBs and type Ibc supernovae should be treated with care (see Section 2.5.2).
Figure 2.2: Maximum accessible proton energy Ep,max (left column) and corresponding maximum accessible neutrino energy Eν,max (right column) as a function of the variability timescale tvar and the bolometric luminosity Lbol of a flaring source, with bulk Lorentz factor Γ =1, 10, 100 (from top to bottom). Overlaid are examples of the location of benchmark explosive transients in the Lbol tvar parameter space (see Section 2.4). The beige region indicates the domain where no source is expected− to be found because of the excessive energy budget. The dots locate recently discovered categories of transients (Kasliwal, 2011), superluminous supernovae (SLSNe), peculiar supernovae, and luminous red novae. The small square box (labeled SNe) and the short diagonal line on its upper left indicate core- collapse and thermonuclear supernovae, respectively. Low-luminosity GRBs and type Ibc supernovae should be treated with care (see Section 2.5.2). PointIceCube Source neutrinos: soonvs. to beDiffuse probed transients Flux
6 7 10 Detectiondiscovery of multiplets potential depends for 10 discovery potential for LL AGN on number density of sources 105 steady source candidates 106 transient source candidates ] 3 in 10 years ] in 10 years starburst & GC/GG-int 1 − 104 − 105 yr 3
[Gpc SNe & newborn pulsars 3 RQ AGN GC-acc − 4 eff 10 10 ρ
[Gpc hypernovae 2 Could be probed in the 3 ˙ 10 next decade ρ 10 RL AGN 10 blazar 102 LL GRB γ-ray burst BL Lac observed IC 1 10 TeV-PeV flux 0.1 IceCube 1 IceCube
local rate density HL GRB effective local density 2 5 IceCube 5 IceCube 10− × 0.1 × jetted TDE 20 IceCube FSRQ 20 IceCube 3 × 2 × 10− 10− 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 effective neutrinoRare powerful luminosity sourcesL [ erg/sare excluded] effective bolometric neutrino energy [erg] ν Eν [Murase & Waxman'16; Ackermann et al.'19]
Murase & Waxman 2016, M. Ahlers talk NBI Copenhagen Dec. 2010 13 Rare sources, like blazars or gamma-ray bursts, can not be the dominant sources of TeV-PeV neutrino emission (magenta band).
Markus Ahlers (NBI) Multi-Messenger Interfaces 11 Observed high-energy multi-messenger transients Murase & Bartos 2019 Guépin & KK 2017
e.g., Kimura et al. 2017, 2018 Biehl et al. 2018 Decoene, Guépin, Fang, KK, AGN/Blazars Metzger, 2020 flares, time-variabilities Ahlers & Halser 2020 Young pulsars
Fang & Metzger 2018 Neutron star mergers Magnetars (AXP/SGR)
Long Gamma Ray KK & Silk 2016 Bursts De Wasseige et al. 2019
Black hole Superluminous mergers Tidal disruption Supernovae events 14
neutron star mergers
GW170817Computing high-energy neutrino fluxes tidal disruption events
mechanisms: Cosmic-ray shock acceleration ejecta Cosmic-ray interactions + cooling magnetic reconnection… acceleration Neutrino production
at various locations: inner/external/side jet wind X accretion disk… jet —> max. acceleration energy spectrum cascades synchrotron cooling no HE ν no HE ν accretion disk wind ex: pions in kilonova ejecta High-energy neutrinos from fallback accretion of binary neutron star Radiative + hadronic merger remnants V. D. (IAP), Ke Fang (Stanford U.), Claire Guépin (Maryland U.), Kumiko Kotera (IAP), Brian D. Metzger (Columbia U.) (JCAP 2019) backgrounds transparency time
) 50 fallback 1)Ex: A modelred kilonova for the radiative ejecta background inside the ejecta: 1 10 −
s thermal ejecta . density, spectra, time evolution • Thermodynamical equilibrium 1047 Metzger et al. 2011 nuclear heating in acceleration region and beyond d dR 44 E = E E + Q˙ + Q˙ (thermal) 10 r fb outflow
esc − TXS0506+056≈ 4πR2 c Optimistic
cm 19 ! " . 10 1 2 Fall-back − 10 s • 103 s 2 mass accretion rate 1016 104 s Q˙ fb = !fbM˙ fbc red ejecta Inner regions • Nuclear reaction Barnes et al. 2016 1013 Decoene et al. 2020 Q˙ = MX e˙ (t) M. R. Drout et al, 2017
cosmic rays: Photon SED (eV 1010 nuclear mass energy 10 6 10 4 10 2 100 102 104 106 acceleration − − − interactions injection Photon energy (eV) lanthanides mass fraction 15 V. Decoene PhD 7 12 18 High-energy neutrinos from binary neutron-starNeutrinos mergers from GRBs: Off-Axis View of Structured Jets 7 Implications from GW170817 6
cutoff of the neutrino fluence to lower energies, while combined rate of losses from synchrotron emission, pγ in- to attenuation by the ejecta, we compare our neutrino con- 10 Antares there is only a slight impact on the peak for GRBs. This Auger teractions (Bethe-Heitler and photo-hadronic) and adiabatic straints to neutrino emission expected for typical GRB pa- example has been computed with an initial baryonic load- losses. Our model predictions assume a magnetic energy ra- IceCube ing of ξA = 100, as indicated by the scale on the left side 1 Successful jet rameters. For the prompt and extended emissions, we use the tio compared to γ-rays of ξB = 0.1 and a non-thermal bary- of the plot, it scales directly with this parameter. The Kimura et al. 2017 results of Kimura et al. (2017) and compare these to our con- blue band includes the 1σ-uncertainties on the measured onic loading of ξp 1 (see Appendix B). We calculate the 0.1 $ duration T90, time variability tv,redshiftz, γ-ray fluence Biehlneutrino et al. 2018 emissivity j θ , ! from pγ interactions with the straints for the relevant 500 s time window. For extended ] fixed !peak" GRB 170817A ν!α ∗ ν! 2 ( ) Fγ as well as the spectral index α and peak energy Epeak 2 Ahlers & Halser 2020 ± − photon background in sub-shells based on the Monte-Carlo 10− emission we consider source parameters corresponding to of the SED. Note that we use D =2Γ instead of Γ for fixed !peak∗ the boost compared to what is frequently used in the generator SOPHIA (Mucke¨ et al. 2000), that we modified to both optimistic and moderate scenarios in Table 1 of Kimura 3 literature. 10− Chokedaccount forjet synchrotron losses of all secondary charged parti- et al. (2017). For emission on even longer timescales, we The gray scale indicates which fraction of the total cles before their decay (Lipari et al. 2007). The uncertainties [GeV cm 4 Kimura et al. 2018 mass of the neutron star system has to be dumped into µ 10 regarding the photon target spectrum are estimated in the
¯ − compare our constraints for the 14-day time window with ν
the jet. Assuming that the wholeGRAND mass of the sys- + on-axis following via the two models (a) and (b) of the peak photon +0.04 µ
the relevant results of Fang & Metzger (2017), namely emis- ν tem, which is estimated to be 2.74 0.01M [1], goes 5 − " F 10− energy. 2 into the jet, the maximum achievable baryonic loading is ν Beamed and
sion from approximately 0.3 to 3 days and from 3 to 30 days ! 7.5 off-axis The expected fluence of muon neutrinos (ν + ν ) under ξA = 10 . This is to be interpreted only as a rough guid- 6 µ ¯µ following the merger. Predictions based on fiducial emis- ance, since the actually realeased energy (compared to 10− intensedifferent emission model assumptions is shown in Fig. 4. The off-axis TeVthe isotropic equivalentPeV energy) is smallerEeV by the beam- sion models and neutrino constraints are shown in Fig. 2.We 2 7 off-axis fluence at a viewing angle of θv 15◦ is indicated as thick CHAPTER 1. COSMIC NEUTRINOLIGO, SOURCE VIRGO,ing factor IceCube,1/(2Γ ) coveredAuger, by ANTARES the jet, which 2017 relaxes 1910− $ find that our limits would constrain the optimistic extended- thisLIGO,VIRGO,IceCube,Auger,ANTARES constraint.∼ On the other hand, for the structured 17 Biehl et al.(approximation) 2018 black lines. The off-axis prediction has only a weak depen- Kimura et al. 2017 FIG. 3: Fluence of νµ +¯νµ for SGRB170817A and differ- emission scenario for a typical GRB at 40 Mpc, viewed at jet scenario, the released energy in different directions ent values10 of8 the LorentzAhlers factor Γ&in Halser the structured 2020 jet case. dence on the angular scaling of the co-moving peak of the ∼ may be higher, which makes the constraint stronger. We assume− pure proton injection and the same parameters as photon spectrum, Eqs. (32)or(33), as indicated as solid and zero viewing angle. Blue ejecta • Off-axis flux is lower than the on-axisgiven flux in Fig. 2. Solid curves2 refer to3 a fixed4 baryonic5 6 loading7 8 9 10 11 As an additional constraint, the photospheric radius of ξA = 100, where1 10 thick10 solid10 curves10 correspond10 10 to10 colli-10 10 10 10 dotted lines, respectively. This is expected from the normal- scales with the baryonic loading. According to Eq. (13) sions above the photosphere, and thin curves indicate sub- ization of the model to the observed γ-ray fluence under this • 3 !ν [GeV] Extendedthe maximum emission baryonic loading is isnotξ observed10 for the fromphotospheric this collisions. event For the dashed curves, the baryonic A,max ∼ viewing angle. For comparison, we also show in Fig. 4 an 4. CONCLUSION dissipation radius to be super-photospheric. This means loading has been maximized demanding that Rcoll >Rph. —> neutrinos from EE may not be observedFigure 4. Predicted fluence of muon neutrinos (νµ + ν¯µ )associ- Red ejecta that the shown neutrino fluence can be up-scaled by a fac- approximation (thin green lines) of the off-axis neutrino flu- tor of 10 in this scenario, which represents our maximal ated with the prompt emission in the best-fit structured jet model We searched for high-energy neutrinos from the first bi- •Fallback ence based on theGRAND on-axis top-hat jet calculation with Lorentz Promptpossible emission neutrino fluence is for too this SGRB dim in theto internal detectof byGhirlanda IceCubeB. et O al.ff-axis(2019 fireball). We scenario show the predictions0 IceCube based on a nary neutron star merger detected through GWs, GW170817, ) 10 factor and neutrino emissivity evaluated at θ∗ θv. This ap- ejecta shock scenario. Thus, if indeed neutrinos had been de- fixed photon peak in the shell frame2 (“fixed " ”, solid lines) us- Equatorial emission peak! POEMMA $ 11 20 tected, then one would have concluded that the gamma- − proximation has been used by Biehl et al. (2018) to scale the in the energy band of [ 10 eV, 10 eV] using the ingIn Eq. the o (ff32-axis)andintheengineframe(“fixed fireball scenario, the observation angle" ∗Decoene”, dotted et lines) al. 2020 ray emission comes from the photosphere at a larger ra-Decoene, Guépin, Fang, KK, peak off-axis2 emission of the5 structured jet. Note that this approx- ∼ ∼ θobs enters as an additional parameter influencing neu- 10 s 10 s ANTARES, IceCube, and Pierre Auger Observatories, as well dius than the neutrino production radius. using Eq. (33). The thick black lines showcm the off-axis emission at a Metzger, 2020 . 2 trino production and photospheric radius. 10 imation3 significantly underestimates6 the expected neutrino viewing angle θv = 15◦.Thebluelinesshowthecorrespondingpre-− 10 s 10 s as for MeV neutrinos with IceCube. This marks an unprece- We show the impact of the Lorentz factor on the muon In Fig. 4, the dependence of the neutrino fluence on fluence of GRB 170717A compared to an exact calculation. diction for the on-axis emission, which has a strong dependence on 4 neutrino fluence in Fig. 3. The solid curves refer to a the observation angle is shown. The Lorentz factor is 10 s total dented joint effort of experiments sensitive to high-energy Accretion thefixed internal to Γ = 30, photon which meansspectrum. that the The scaling thin green is given lines by show the result Figure 4 also indicates the predicted neutrino fluence for an
fixed baryonic loading ξA = 100, which illustrate that (GeV disk 90 %Wind IsotropicofEq. an (11). approximation Again, emission the solid based curves on represent the standard the unscaled on-axis4 calculation of on-axis observer of the source located at the same luminosity neutrinos. We have observed no significant neutrino counter- Figure 2. Upperthe limits fluence (at scales withconfidenceΓ according level)to Eq. (11) on the without neutrino ν 10− imposing any additionalshocks constraints. The scaling agrees uniformfluences with jets a (Waxman fixed baryonic & Bahcall loading 1997ξA =)withjetparametersfrom 100, while distance. The extrapolated on-axis fluence shows a strong part within a 500 s window, nor in the subsequent 14 days. spectral fluence from GW170817 during a 500 s window centeredthe dashed curves show the maximum achievable neu- ± very well. However, for large shifts± there is an additional the structured jet model at θ∗ = θv .Theuppersolidlinesindicate dependence on the model of the internal photon spectrum; The three detectors complement each other in the energy on the GW triggerdamping time of (top the high-energy panel), and tail a of 14-day the spectrum window due follow- to trino fluence corresponding to the solid curves re-scaled thewith 90% the maximum C.L. upper possible limit baryonic on the loading fluence demanding from 6Albert et al. (2017). model (33) predicts a strong neutrino peak at the EeV scale ing the GW triggersecondary (bottom cooling, panel). which For was neglected each experiment, in the simple limits an- are dNdE 10− bands in which they are most sensitive (see Fig. 2). alytic estimate Eq. (11). that Rcoll >Rph. From the way the curves2 rescaleν it can that exceeds the prediction of model (32) by two orders of calculated separately for each energy decade, assuming a spectralbe deduced that the collisions become sub-photosphericE This non-detection is consistent with our expectations from 2 magnitude. The relative difference of the neutrino fluence at fluence F (E)=ForFup low values[E/GeV] of Γ, the− collisionin that radiusdecade decreases, only. Also!(thin lines)20 alreadyMeV, for insmall tension observation with angles theθobs peak2◦ distribution in- which implies× efficient neutrino production. On the other forpeak this particular values of Γ and ξ . For large obser-∼8 the EeV scale follows from the ratio of ! ! 0 for the two a typical GRB observed off-axis, or with a low-luminosity shown are predictions by neutrino emission models. In the upper $ A 10− peak( ) hand, the photospheric radius increases, which leads to ferredvation angles, from the GRBs fluence observed will be highly by suppressed. Fermi-GBM10 The4 (Gruber et al.106 models (1032)8 and (32): For10 a10 fixed co-moving energy density GRB. Possible gamma-ray attenuation in the ejecta from the Fallback 5 2 plot, models fromsub-photosphericKimura et al. collisions(2017) for forΓ both! 20 extended – indicated emission by 2014maximum). The neutrino phenomenological fluence is a few model10− GeV(b) cmis− motivated by the Relativistic × 3 of the shell, a lower peak photon energy corresponds to a merger remnant could also account for the low gamma-ray (EE) and promptthin emission solid curves. are The scaledwind dashed to a curves distance indicate of 40 the Mpc, max- anddiscussionfor the on-axis of observerIoka & and NakamuraξA,max 10 (.2019 Compared), who to study the con- Eν (GeV) 16 Jets imal neutrinoDecoene fluence et using al. the2020 photospheric constraint, the structured low luminosity jet, the≈ off-axis observation higher photon density and also a higher threshold for neutrino luminosity, which could mean stronger neutrino emission. shown for the casewhich of means on-axis that theviewing curves angle for Γ < (020◦) are and down-scaled selected off-sistencymakes it even of the less likely on-axis to detect emission a neutrino of GRB from this 170817A with the iso production. One can also notice, that the on-axis neutrino Optimistic scenarios for such on-axis gamma-attenuated axis angles to indicateto match theit, and dependence the curves for onΓ this> 20 parameter. are up-scaled GW ac- dataevent.Eγ -!peak correlation suggested by Amati (2006). Here, the fluence in the TeV range depends only marginally on the cordingly. The expected maximal neutrino fluence is at In order to demonstrate how observation angle θ and the redshift of the host-galaxy constrain the viewing angle toon-axis fluence is expected to peakFigure at !peak 6obs.Neutrinospectraforonesourceatdistance178 keV. 40 Mpc (optimistic scenario), for an injection emission are constrained by the present non-detection. most about four orders of magnitude below the neutrino and Lorentz factor Γ are affected by the photospheric$ viewing angle. This energy scale is dominated by the emission Θ [0◦, 36◦] (seetelescope Section sensitivities,3). In thewhich lower means plot, that models the detection from ofFangconstraint, we show a parameter space scanspectral in Fig. 5. indexFor α =1.5 and baryonof the jet loading at θ 10ηp =020 .and1. Lines reflects with the strong increasing angular thickness represent neutrino While the location of this source was nearly ideal for ∈ ∗ $ ◦ − ◦ & Metzger (2017a neutrino) are scaled coming to from a distance this SGRB of 40 was Mpc. extremely All fluences un- each set of parameters, the maximum possiblefluences baryonic integrated up to thedependence indicated of the (increasing) neutrino emission times in after the rest the frame merger. of the Left: pure proton injection Auger, it was well above the horizon for IceCube and are shown as thelikely per in flavor the structured sum of jet neutrino scenario. and anti-neutrino flu-5.2loading Neutrino is calculated such Fluence that the collision is still super- central engine (cf. Fig. 3). ANTARES for prompt observations. This limited the sensitiv- ence, assuming equal fluence in all flavors, as expected for standard and Right: pure iron injection.The Black upper thin solid solid lines lines represent in Fig. 4 show the the IceCube90% confidence point-source sensitivity for two Figure 1.2: Sketch of the regions of the neutron-star merger remnant at play forAs the we discussed in section 4, the neutrino emissivity of a ity of the latter two detectors, particularly below 100 TeV. neutrino oscillation parameters. structured jet is expected to deviatedeclination from the configurations angular dis- oflevel the (C.L.) source upper in limits the on sky: the neutrino0◦ < δ flux< of30 GRB◦ (best 170817A sensitivity) and 30◦ < δ < 60◦ acceleration∼ and interaction of cosmic rays in our scenario. The red and blue envelopes For source locations near, or below the horizon, a factor of tribution of the observable γ-ray[? emission.]. Dashed For high lines opacity are projectedfrom Antares, point-source Auger and IceCube sensitivities (Albert et for al. future2017). The experiments: POEMMA [? ] indicate the location of the so-called blue and red kilonovae ejecta, that emit thermal(τpγ 1) regions of the shell the angular distribution of the predicted neutrino fluence is orders of magnitude below these 10 increase in fluence sensitivity to prompt emission from didates from identified GW sources. A coincident neutrino, % (blue) and GRAND (green) [? ]. ∼ 2 UV/optical/IR radiation over timescales of hours to days (blue) and a week (red). Modelsneutrino emission is expected to follow the distribution of in- combined limits. However, our neutrino fluence predictions an E neutrino spectrum is expected. 2 − related to the GRB jet havewith been a exploredtypical position in scenarios uncertainty involving of 1 GRBs.deg could In this signifi- work,ternal energy (24) that takes into account the efficiency of are proportional to the non-thermal baryonic loading factor, With the discovery of a nearby binary neutron star merger, ∼ we focus on the interaction ofcantly a fast improve wide-angle the fast outflow localization from the of joint accretion events disk compared powereddissipation in internal collisions. This is shown for our effi- and we assume a moderate value of ξp = 1 for our calcula- the ongoing enhancement of detector sensitivity (Abbott tions. In any case, the predicted neutrino flux at an observa- by late-time fall-back of mergerto the debris, GW-only with case. the In slowly-expanding addition, the first joint red GW kilonova and high- ejecta.ciency model (A6) as the thick green line in Fig. 4.Forlow- 6 et al. 2016) and the growing network of GW detectors (Aso opacity (τpγ 1) regions, however,mismatch the energy around distributionE = 10tionGeV angle in of 15 the◦ is manyneutrino orders ofspectra magnitude (since larger than about the 5% percent of the energy This interaction results in theenergy dissipation neutrino of discovery the accretion might power thereby as shocksbe known or magnetic to the % et al. 2013; Iyer et al. 2011), we can expect that several binary has an additional angular scalingof from the the proton opacity goes (27), as intoexpectation neutrinos). from an In off this-axis observation energy range, of a uniform the jet. conversion of proton energy reconnection, accelerating relativisticwider astronomy particles, community in a nebula within behind minutes the ejecta after theshell. event,indicated by the thin green line. One can notice that a low neutron star mergers will be observed in the near future. Not opening a rich field of multimessenger astronomy with parti-opacity environment has an enhancedinto neutrino emission at jet energy angles is not well reproduced by the analytical estimate. The discrepancies only will this allow stacking analyses of neutrino emission, 10 -20 , which is comparable to our relative viewing angle. cle, electromagnetic, and gravitational waves combined. ◦ ◦ at lower energies and in6CONCLUSIONS the high energy tail of the neutrino spectra is due to the photopion but it will also bring about sources with1.2 favorable Principles orientation of phenomenological modelling Note that the angular distributions in Fig. 3 are normalized and direction. to the value at the jet core andproduction do not indicate model. the absolute Our analyticalIn this paper, estimates we have discussed only considers the emissiona of constant neutrinos interaction cross section emissivity of neutrinos or γ-rays, which depend on jet angle in the internal shock model of γ-ray bursts. The majority of The ANTARES, IceCube, and PierreModelling Auger Collaborations astroparticles emissions from high energy sources require to take into account for photopion production, while the accurate implementation of other channels smooth out various complex processes. The key element inACKNOWLEDGMENTS that respect, is the interplay betweenθ∗ theand co-moving cosmic ray energy !p!. previous predictions are based on the assumption of on-axis are planning to continue the rapid search for neutrino can- At each jet angle θ we estimatethe the secondary maximal cosmic particle ray energiesobservations over of auniform wider jets range. with wide Note opening however angles. Here, that the peak of each spectra particles and the background environment provided by the source. The particles undergoes ∗ energy based on a comparison ofis the accurately acceleration rate reproduced, to the we and have the extended good the agreement standard formalism in the of cosmic-ray neutrino pro- spectra implies that the
MNRAS 000, 1–10 (2019) fraction of proton energy converted into meson (pion) energy is correctly estimated. Regarding the cosmic rays spectra it can be seen that already at t = 105 stheprimary cosmic rays undergo severe interactions leading to a large depletion of 2 orders of magnitude ∼ between the pre and post interaction spectra. At time 104 s, most cosmic rays lose energy via drastic photo-pion interactions, hence the absence of cosmic-ray flux at this time in the right-hand side plot. In the final picture, at early times (> 1 s) no cosmic rays can escape the kilonova as the number of interaction is too large, at longer times (> 104 s) a mixed composition appear and in between a transition from pure proton to mixed composition can be seen. However the diminution of the baryon loading with time result in a negligible cosmic-ray flux. The numerical spectra for one source are shown in Figure ?? for different times after the merger, for pure proton (left) and pure iron (right) injections at the kilonova input, for the optimistic scenario. We can clearly identify an optimum neutrino production time around t = 103 104 s. The optimum time is the result of a combination between i) a high cosmic- − ray luminosity, ii) a high efficiency of cosmic-ray interactions leading to mesons production, and iii) a sufficiently low rate of meson (and muon) cascades leading to neutrino production. Consequently at earlier times, the neutrino flux is low (and with limited neutrino energy) because of the strong meson cascade rates and at later times it is low due to the decrease
–16– 5
Table 2.Thedetectionprobabilities,P (Nµ ≥ k)fordL =300Mpc.IC:IceCube,Gen2:IceCube-Gen2,up+hor:upgoing7 + horizontal events, down: downgoing events, all: covering-factor-weighted average over the up+hor and down, Aeff,ave :usingthe declination-averaged effective area. decay is expressed as TABLE II. Detection probability of neutrinos by IceCube and EE-mod-dist-A IC (up+hor) IC (down) IC (all) Gen2 (all) IC (Aeff,ave ) iso iso IceCube-Gen2 dNνπ dN 2 µ 1 1 2 pP (Nµ ≥ 1) 0.07 0.04 0.06 0.21 0.06 E π f + f f E , (17) Number of detected neutrinos from single event at 40 Mpc νµ pγ pp π,sup p dEνπ ≈ !8 6 " dEPp µ (Nµ ≥ 2) model0.01 IceCube (up+hor) 0.00 IceCube 0.00 (down) 0.05 Gen2 (up+hor) 0.00 A2.00.168.7 −1 −1 −1 −1 EE-mod-dist-B IC (up+hor) IC (down) IC (all) Gen2 (all) IC (Aeff,ave ) where fpγ = tpγ /tp,cl and fpp = tpp /tp,cl are the neutrino B0.117.0×10−3 0.46 production efficiency through photomesonP production(Nµ ≥ 1) 0.11 0.04 0.08 0.25 0.08 pp Number of detected neutrinos from single event at 300 Mpc and inelastic collision, respectively, and theP ( subscriptNµ ≥ 2) 0.02 0.00 0.01 0.10 0.01 π νµ indicates the muon neutrinos produced from pions. model IceCube (up+hor) IceCube (down) Gen2 (up+hor) − A The muons decay to neutrinos and electrons/positrons,EE-opt-dist-A ICA0.0352.9 (up+hor)Coincident IC (down) IC detection (all)×10 Gen23 (all) with0.15 IC ( effgravitational,ave ) waves × −3 × −4 × −3 whose spectrum is represented as P (Nµ ≥ 1) B1.90.7410 0.25 0.521.3 10 0.868.1 10 0.59 P − Choked jet iso iso (Nµ ≥ 2) 0.42GW+neutrino 0.04 detection 0.25 rate 0.69 [yr 1] 0.24 iso dN µ dN π Kimura et al. 2018 2 dNνe 2 νµ 2 νµ µ π A Eνe Eν fµ,supEν EE-opt-dist-B, (18) IC (up+hor) IC (down) IC (all) Gen2 (all) IC ( eff,ave ) µ µ µ π model IceCube (up+hor+down) Gen2 (up+hor) dEνe ≈ dEνµ ≈ dEνµ P (Nµ ≥ 1) A0.381.2optimistic0.60 0.19 0.41 0.73 0.47 Optimistic model: −1 −1 P B0.0240.091moderate Coincident detection where fµ,sup =1 exp( tµ,dec/tµ,cl) is the suppression(Nµ ≥ 2) 0.31 0.02 0.18 0.55 0.17 − − −1 −1 −1 possible already with factor by the muon cooling, tµ,cl = tµ,syn + tdyn,andthe subscript νµ indicates the muon neutrinos produced from IceCube µ Table 3 muons. These muon neutrinos and electron neutrinos where Aeff is the effective area..Thedetectionprobabilitieswithinagiven IceCube and IceCube- time interval, P∆T .TheSGRBrateisassumedtobe change their flavor during the propagation to the Earth. Gen2 can also detect νesand−ντ sasshowerevents(or− − − 4Gpc 3 yr 1 − 10 Gpc 3 yr 1 The electron neutrinos and muon neutrino fluences at the cascade events). The angular resolution of shower events Successful jet Earth are estimated to be [e.g., 90] is much worse than that of trackNS-NS events. (∆T Also,=10yr) the effec-IC (all) Gen2 (all) tive area for the shower events is smaller than the upgoing Kimura et al. 2017 10 4 track events. Thus, we focus onEE-mod-dist-A the detectability of 0.11ν - – 0.25 0.37 – 0.69 φ = φ0 + (CHAPTERφ0 + φ 1.0 COSMIC), (19) NEUTRINO SOURCE 19 µ Optimistic model: νe+νe νe+νe νµ+νµ ντ +ντ induced track events, although theEE-mod-dist-B shower events may0.16 be – 0.35 0.44 – 0.77 18 18 Coincident detection important for the merger events inEE-opt-dist-A the southern sky.0.76 – 0.97 0.98 – 1.00 Blue ejecta We use the effective area shown in Ref. [92] for Ice- highly probable 4 7 EE-opt-dist-B 0.65 – 0.93 0.93 – 1.00 φ = φ0 + (φ0 + φ0 ), (20) Cube. For IceCube-Gen2, the effective volume can be 10 already with IceCube νµ +νµ νe+νe νµ+νµ ντ +ντ CHAPTER 1. COSMIC NEUTRINO SOURCE 38 18 18 times larger than that of IceCubeNS-BH [93]. (∆T Hence,=5yr) we useIC (all) Gen2 (all) 2/3 0 iso 2 Red ejecta 10 times larger Aeff than that for IceCube, although where φi =(dNi /dEi)/(4πdL) is the neutrino fluence EE-mod-dist-A 0.12 – 0.28 0.45 – 0.88 Fallback ) it depends on the specific configurations. The thresh- 1 Spectral index = 1.5, Proton primaries
without the oscillation and dL is the luminosity distance. − ejecta EE-mod-dist-B 0.18 – 0.39 0.57 – 0.88 7 ARA
old energy for the neutrino detection is set to 0.1 TeV sr 10− We set dL = 300 Mpc as a reference value, which is IceCube POEMMA EE-opt-dist-A 0.85 – 0.99 1.00 – 1.00 1
for IceCube and 1 TeV for IceCube-Gen2. The down- − the declination-averaged horizon distance for face-on NS- Equatorial emission s 8
2 10 going events suffer from the atmosphericEE-opt-dist-BDecoene, background. Guépin,0.77 –Fang, 0.97 KK, 0.99 – 1.00 − NS merger events for the design sensitivity of the second − generation detectors [91]. Although the downgoing events can be usedMetzger, to discuss 2020 9 Accretion 10− The resultant muon neutrino fluences are shown in Fig- thedisk detectability withWind IceCube, Aeff for the downgoing events with IceCube-Gen2 is quite uncertain.Optimistic Thus, wemodel: 10 GRAND ure 5 for optimistic (model A) and moderate (model B) shocks 10− focus on the upgoing+horizontalcould detect events more that EEs have (Nakamura decli- et al. 2014). Here, 10% of IceCube di!use flux (GeV cm sets of parameters tabulated in Table I. These models are ◦
we simply assume that half of SGRBs have EEs, leadingν 11 different in L and Γ , which mainly affect the normal- nation δ > 5 for IceCube-Gen2. KM3NeT will observe 10− k,iso j − ∼ − Interesting for stacking and ization of the fluence and the cutoff energy, respectively. the events in the southernto skyN [94],2 which5for will∆ helpT = make 10 years. Within the sensi- GW170817-like cross-correlation searches 12 Optimistic coincident detections intivity the near range future. of NS-BH Note that mergers the by aLIGO (600 Mpc),dNdE 10−
For model A, the neutrino spectrum has a cutoff around 2 ν −1 −1 4 6 8 10 atmospheric neutrinos arethe negligible SGRB rate owing is to∼ the3.7yr short − 9.0yr , leadingE to 10 10 10 10 Eν 200 TeV, while for model B, the spectrum break Fallback ∼ duration of tdur 2s. ∼ − Eν (GeV) appears at lower energy, Eν 50 TeV, due to the lower Relativistic ∼ wind 9 22 EEs for a 5-year operation. The estimated 17 ∼ We calculateDecoene the expectedet al. 2020 number of detected neutri- Γj . The pion cooling causes the cutoff and the spectral Jets values of P∆T are tabulated in Table 3.Wefindthat nos for models A and B for a single event located at break. The combination of the muon cooling and the neu- the simultaneous detection of gamma-rays, neutrinos, trino oscillation causes a slightly soft spectrum at 3 TeV 40 Mpc, which are tabulated in the upper part of Table Figure 2.ThedetectionprobabilityP (Nµ ≥ 1) as a function and GWs is possible in the era of IceCube-Gen2 and ! E ! 200 TeV for model A and at 1 TeVd ! E ! 50 II. IceCube is likely to detect a coincident neutrino signal ν of luminosity distance L.Theupperandlowerpanelsareν aLIGO/aVirgo/KAGRA, assuming a cosmic-ray load- TeV for model B. for our model A if the source is located on the northern with IceCube and IceCube-Gen2, respectively. The vertical◦ d skyd (δ > 5 ). For ouring model factor, B, detectionξp ∼ 10. for Thisa source will allow us to probe the thin-dotted lines show L =300Mpcand L =600Mpc.− Figure 1.2: Sketch of the regionsin of the the northern neutron-star sky merger is alsophysical remnantpossible, conditions at but play not for guaranteed. during the EEs, including the cosmic- B. Detectionacceleration rates and interaction of cosmicFor IceCube-Gen2, rays in our scenario. detection Theray redloading is probable and blue factor for envelopes the and northern the Lorentz factor (see Section indicate the location of the so-calledsky blue events. and If red we kilonovae put the4). ejecta, source that at 300 emit Mpc, thermal neutrino These neutrinos(e.g., can beNakarUV/optical/IR detected et al. 2006 by radiation IceCube; Wanderman over or timescales &detection Piran of hours 2015 from), to so a days single (blue) eventIn and the a is week unlikely near (red). future, with Models IceCube, KM3NeT will be in operation. while it is possible with IceCube-Gen2 if the optimistic IceCube-Gen2 as νµthe-induced eventrelated track rate to within events, the GRB the whose jet sensitivity have ex- been range explored of in aLIGO scenarios involvingWhile GRBs. IceCube In this is more work, suitable to observe the north- event (model A) occurs at the northern sky. pected event number(300 is estimated Mpc)we focusis to∼ be0 on.46 the yr interaction−1 − 1.1yr of− a1. fast According wide-angle to outflow the fromern the sky, accretion KM3NeT disk powered will achieve a better sensitivity for by late-time fall-back of merger debris,We with now the calculate slowly-expanding the joint redGW+neutrino kilonova ejecta. detection Swift ∼ results,This interaction25 % of results SGRBs in are the accompaned dissipationrate for of a bypopulationthe EEs accretion of powerthe sources, southern as shocks which sky, or we magnetic assume helping to us be improve the possibility of µ = φν Aeff (δ,Eν )dEν , (21) N (#Sakamotoreconnection, et al. 2011 accelerating), noting relativistic that softeruniformly particles, instruments distributed in a nebula behind indetections. the the local ejecta universe. shell. Using the Figure 1.11: Diffuse neutrinos spectra for injection spectral index α =1.5 (top) and α = 1.2 Principles of phenomenological modelling 2.1 (bottom) of proton primaries (left) and iron primaries (right), with baryon loading ηp =0.1. The GW170817-like scenario follows a flat source evolution with raten ˙ 0 = 3 1 Modelling astroparticles emissions from high energy sources require to take into account 600 Gpc− yr− , while the optimistic scenario follows a SFR source evolution with rate 3 1 various complex processes. The key element in that respect, is the interplay between the n˙ 0 = 3000 Gpc− yr− . particles and the background environment provided by the source. The particles undergoes a flat behaviour of the population with redshift (e.g., [127, 128]). On the other hand, the connection of these systems with short GRBs, and the observation of the latter can indicate a source emissivity following the star formation rate (SFR) [129, 130]. 3 1 In order to be conservative, we adopt a flat evolution model withn ˙ 0 = 600 Gpc− yr− for the GW170817-like scenario. Such a hypothesis represents the simplest model and does not presume of any enhancement of the population at earlier times in the Universe history. For the optimistic scenario, we assume a SFR evolution rate following Ref. [131] and a local 3 1 merger raten ˙ 0 = 3000 Gpc− yr− . The SFR evolution can enhance the diffuse neutrino flux level by a factor of ξ 2 4 [132]. z ∼ − Multi-messengers! cosmic rays + others —> temporal coincidence impossible (deflections) but studies of di!use fluxes
Gravitational waves GW + electromagnetic (+gamma) GW170817
neutrinos + gammas? TXS0506+056 (blazar flare) Cosmic rays
p Fe ν Gamma rays
ν GW + neutrinos ???
Neutrinos
ultra-high energy neutrinos ???
Mauricio's Talk 18 UHE neutrinos: an unchartered territory! Mauricio's Talk
Alves Batista, de Almeida, Lago, KK, 2018 GRAND Science & Design, 2018 KK, Allard, Olinto 2010
GR ND
UHE ν ? RNO PUEO
19