Searching for neutrino emission from hard X-ray sources with IceCube
SSearching for neutrino emission from hard X-raysources with IceCube
The IceCube Collaboration ∗ http://icecube.wisc.edu/collaboration/authors/icrc19_icecubeE-mail: [email protected] The IceCube neutrino observatory, a cubic-kilometer particle detector at the South Pole, first an-nounced the discovery of an astrophysical flux of high-energy neutrinos in the TeV-PeV range in2013, followed in 2017 by the detection of a high-energy neutrino event in temporal and direc-tional correlation with the flaring gamma-ray blazar TXS 0506+056. This observation, combinedwith archival neutrino detections in 2014-2015, has provided compelling evidence for the detec-tion of the first high-energy astrophysical neutrino source. A promising way of detecting addi-tional sources is to correlate neutrino detections with sources where a hadronic electromagneticsignature is observed. If blazars are a significant source of neutrinos, the high-energy gamma raysproduced in pionic decays in coincidence with the neutrinos may cascade in the strong photonsfields present in blazar jets, leading to strong emission in the hard X-ray to MeV gamma-ray en-ergy range. We here present plans for a search for neutrino emission from a large sample of hardX-ray sources from the BAT AGN Spectroscopic Survey (BASS).
Corresponding authors:
Marcos Santander Department of Physics and Astronomy, University of Alabama, Tuscaloosa, AL 35487-0324,USA36th International Cosmic Ray Conference -ICRC2019-July 24th - August 1st, 2019Madison, WI, U.S.A. ∗ For collaboration list, see PoS(ICRC2019) 1177. c (cid:13) Copyright owned by the author(s) under the terms of the Creative CommonsAttribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0). http://pos.sissa.it/ a r X i v : . [ a s t r o - ph . H E ] A ug earching for neutrino emission from hard X-ray sources with IceCube
1. Introduction
High-energy neutrinos are unique tracers for hadronic processes occurring in astrophysicalobjects. Neutrinos in the TeV-PeV energy range are produced in the interactions of cosmic rays withambient photon or matter fields at the source or during propagation. The charged pions producedin these interactions eventually decay into neutrinos which oscillate to flavor equipartition as theytravel over cosmic distances. Their neutral charge and low interaction cross section implies that,unlike their parent cosmic ray particle, they can propagate in straight lines and suffer no absorption.The neutral pions produced in the same interactions decay into gamma rays in a similar energyrange that accompany the neutrino emission [1, 2]. These gamma rays can also point back to theirsource, but may be absorbed at the source or during propagation through interactions with theextragalactic background light (EBL) [3].Hadronic gamma rays produced in photon-rich environments, such as Active Galactic Nuclei(AGN), may lose energy through electromagnetic cascading, which results in a high flux emittedin the hard X-ray to MeV range [4, 5, 6]. While these sources and the propagation medium may beopaque to TeV-PeV gamma rays, the X-ray to MeV emission may be observed largely unattenuated,and therefore represent an important probe for hadronic emission (see Fig.1 for an example usingTXS 0506+056). a
10 15 20 25 30 - - - - - log ( Frequency / Hertz ) l og ( E d N / d E / e r g c m - s - ) LeptonicOptical X - raysoft hard GeV - γ TeV - γ No neutrinos b
10 15 20 25 30 - - - - - log ( Frequency / Hertz ) l og ( E d N / d E / e r g c m - s - ) LeptonicHadronic PhotonsMuon NeutrinosOptical X - raysoft hard GeV - γ TeV - γπ cascade p + γ→ p + e ± pair production Figure 1:
Modeling of the spectral energy distribution of the blazar TXS 0506+056 during the 2017 flare po-tentially associated with a high-energy neutrino for purely leptonic ( a ) and purely hadronic ( b ) models. Thehadronic models tend to overproduce flux in the hard X-ray to MeV range due to hadronic pair productionand cascading effects (from [7]). The angular distribution of the astrophysical neutrinos discovered by IceCube in the TeV-PeVrange [8] is consistent with isotropy, which favors an extragalactic origin for the signal. AGN aretherefore a prime candidate neutrino source class, as they dominate the high-energy extragalacticsky. While no significant correlation has been observed between IceCube neutrinos and gamma-rayAGN from the
Fermi earching for neutrino emission from hard X-ray sources with IceCube from TXS 0506+056 with no associated ∼ GeV gamma-ray flare [12] points to a more complexscenario, although the lack of sensitive broad multiwavelength coverage, specially in the X-rayband, makes an interpretation challenging (e.g. [13]).While previous neutrino searches from AGN have concentrated on the correlation with gamma-ray emission, we here propose a search for neutrinos from a catalog of hard X-ray AGN detectedwith Burst Alert Telescope (BAT) [14] onboard the Neil Gehrels
Swift space telescope.
2. The hard X-ray AGN sample
The first data release (DR-1) of the BAT AGN Spectroscopic Survey (BASS) [15] constitutesthe most complete all-sky AGN catalog in the hard X-ray range (14-195 keV). The sources includedin BASS were selected from the 70-month BAT catalog, which lists 1210 objects in the >
10 keVrange. Given the large positional uncertainty of BAT sources,
Swift
XRT observations were usedto identify counterparts which resulted in 836 BAT-detected AGN. Redshift values are listed for828 AGN, which we use in this study. BASS includes dedicated and archival optical spectroscopicobservations of the identified AGN for redshift determination and spectral line measurements. Askymap of all BASS sources used in this study is shown in Fig. 2. l o g ( F k e V [ e r g c m s ]) Figure 2:
Skymap in equatorial coordinates of the BASS sources that will be used in the proposed search.The dashed line indicates the Galactic plane, with a black marker showing the Galactic Center position.Marker colors show the intrinsic (i.e. deabsorbed) energy flux of the source in the 14-195 keV band inlogarithmic scale.
Most BASS sources are nearby, with a median of z ∼ .
04 and 90% of the sources locatedwithin a redshift of 0.2. Approximately 11% of the sources are potentially beamed (either BL Lacobjects or quasars) where Doppler boosting increases the observed emission from the source.
Fermi
As several sources in BASS may also be listed in the
Fermi catalogs that have been used in earching for neutrino emission from hard X-ray sources with IceCube z Redshift z X - r a y l u m i n o s i t y ( - k e V ) [ e r g / s ] Non-blazar AGNBlazar (BL Lac)Blazar (FSRQ)Blazar (unclassified)Gamma-ray AGN 11.010.510.09.59.0 l o g ( F k e V [ e r g c m s ]) Figure 3:
Left:
Redshift distribution of the BASS catalog sources.
Right:
Hard X-ray luminosity (isotropic-equivalent) for the BASS sources as a function of redshift. Different markers indicate AGN classes, while thecolors indicate the derived intrinsic luminosity in the 14-195 keV band. The highest luminosities are mostlyassociated with blazars thanks to the contribution of Doppler boosting. AGN with a potential gamma-raycounterpart from the
Fermi -LAT 4FGL catalog are marked with a gray circle. the past to search for neutrino correlations, we evaluate the overlap between BASS and the 8-year
Fermi -LAT catalog (4FGL).Sources in 4FGL are searched for around the position of each BASS AGN within a 0.2 ◦ radius.This is a conservative estimate of the position uncertainty of 4FGL sources as more than 96%have better localizations (at 95% CL). Of the 836 BASS sources, 84 have a 4FGL source withina 0.2 ◦ distance (10%) and 61 (7%) are consistent with the 4FGL position taking into accountits localization uncertainty. As expected, the overlap (illustrated in Fig. 3) increases for high-luminosity blazars.To evaluate the overlap of the proposed study with those neutrino- Fermi
AGN correlationstudies that use a gamma-ray flux weighting, we compare the distribution of energy fluxes for4FGL sources included in BASS to the entire catalog of high Galactic latitude sources ( | b | > ◦ ).The 4FGL sources in BASS have a slightly higher median energy flux than the catalog median (seeFig.4) but given their small overall representation in the sample (7%) we estimate that the overlapwith previous studies is not significant and therefore the BASS sample represents a novel sourceclass selection to be evaluated as potential neutrino emitters.The search for neutrino emission from the BASS AGN will use a time-integrated method thatstacks neutrino candidate events from the positions of the BASS sources searching for an excess ofevents when compared to a background expectation. As the relation between the potential neutrinoemission and the hard X-ray observables is not straightforward, different weighting schemes willbe used, as well as a scheme that weights all sources equally regardless of their X-ray emission.
3. Data Sample and Analysis Method
IceCube is a cubic-kilometer neutrino detector deployed deep within the glacier at the SouthPole [16]. It detects neutrinos using a volumetric array of 5160 photomultiplier tubes that record3 earching for neutrino emission from hard X-ray sources with IceCube Energy flux above 100 MeV [erg cm s ]10 Median (all 4FGL)Median (in BASS)All 4FGL sourcesIn BASS catalog
Figure 4:
Gamma-ray energy fluxes above 100 MeV for all 4FGL sources (light blue) and those included inBASS (purple). The BASS gamma-ray sources have a slightly higher energy flux. the Cherenkov light emitted by relativistic charged particles produced in high-energy neutrino in-teractions in the ice or the bedrock underneath the detector. The energy, incoming direction, andflavor of the original neutrino can be inferred from the optical module signals.Charged-current interactions of muon neutrinos and antineutrinos lead to the production ofenergetic muons that can propagate over kilometers. The incoming direction of these muon trackscan be reconstructed to within 1 ◦ of the original neutrino at energies above 10 TeV. In this searchwe plan to use muon neutrino candidate events from the entire sky collected in 8 years of IceCubeoperating in its final configuration. The search for neutrino emission will be performed using a time-integrated unbinned maximum-likelihood approach. The likelihood function is of the form L ( n s , γ ) = N ∏ i = (cid:104) n s N S ( (cid:126) x i , γ ) + (cid:16) − n s N (cid:17) B ( (cid:126) x i ) (cid:105) , (3.1)where n s is the number of signal events, N is the total number of events, (cid:126) x i is a vector thatcontains spectral and positional information for each event, and S and B represent the source andbackground probability distributions, respectively. The parameter γ is the spectral index of theassumed neutrino energy spectrum for all sources, modeled as a power law of the form F ( E / E ) − γ .The use of neutrino energy information can help further distinguish an astrophysical signal, whichis expected to have a harder spectrum than the background ( γ bkg ∼ . ω k and the sensitivity of the detector given thesource declination δ k and global spectral index γ . The source PDF for the entire catalog of M = earching for neutrino emission from hard X-ray sources with IceCube S i = ∑ Mk = ω k R k ( δ k , γ ) S ( (cid:126) x i ,(cid:126) x k , γ ) ∑ Mk = ω k R k ( δ k , γ ) . (3.2)A test statistic Λ is constructed from the likelihood ratio Λ = (cid:20) L ( ˆ n s , ˆ γ ) L ( n s = ) (cid:21) (3.3)where ( ˆ n s , ˆ γ ) are the number of signal events and the spectral index for which the likelihoodratio with respect to the background-only hypotheses ( n s =
0) is maximized. The significanceof any neutrino signal inferred from this analysis will be determined by comparing the Λ valueobtained by applying this method to the actual neutrino data to a high-statistics distribution of Λ values obtained by scrambling multiple times the coordinates of the neutrino data set. The weights ω k assigned to each of the 828 BASS sources will be determined using fiveschemes:1. Flux weighting : ω k is proportional to the intrinsic flux of the AGN in the 14-195 keV band,which selects for near, higher flux sources.2. Luminosity weighting : ω k is proportional to the isotropic-equivalent intrinsic luminosity ofthe AGN in the 14-195 keV band, which selects for more powerful objects. The luminosityis calculated using a cosmological model with Ω Λ = . Ω M = . H = 70 km s − Mpc − .3. Spectral index weighting : ω k is inversely proportional to the X-ray power-law spectralindex ( E − Γ ). This selects for harder sources that may have significant flux at higher energies,specially towards the MeV range.4. Column density weighting : ω k is proportional to the total column density of the source n H ,which selects for obscured sources with substantial target material along the line of sight.5. Equal weighting : ω k is constant across the catalog as an unbiased weight.The distributions of flux, luminosity, spectral index and column density weights are shown inFig. 5.
4. Conclusions and future work
We have discussed a proposed search for neutrino emission from hard X-ray AGN using 8years of IceCube data and enumerated a weighting strategy for this analysis. Next steps includethe generation of sensitivities and discovery potentials for this analysis and, in case no significantsignal is identified, estimates of the constraints that can be set on this class of objects as contributorsto the all-sky astrophysical neutrino flux. 5 earching for neutrino emission from hard X-ray sources with IceCube
41 42 43 44 45 46 47 48log(L
14 195keV [erg/s])0100200300400500600700800 N A G N ( < L k e V )
14 195keV [erg cm s ])0100200300400500600700800 N A G N ( < F k e V ) N A G N ( < )
20 21 22 23 24 25log n H N A G N ( < l o g n H ) Figure 5:
Luminosity (top left), intrinsic flux (top right), spectral index (bottom left) and column density(bottom right) weights based on the BASS catalog values.
References [1] S. R. Kelner, F. A. Aharonian, and V. V. Bugayov,
Phys. Rev.
D74 (2006) 034018. [Erratum: Phys.Rev.D79,039901(2009)].[2] S. R. Kelner and F. A. Aharonian,
Phys. Rev.
D78 (2008) 034013. [Erratum: Phys.Rev.D82,099901(2010)].[3] A. Cooray,
Royal Society Open Science (Mar, 2016) 150555.[4] M. Petropoulou, S. Dimitrakoudis, P. Padovani, A. Mastichiadis, and E. Resconi, Mon. Not. Roy.Astron. Soc. (2015) 2412–2429.[5] M. Petropoulou and A. Mastichiadis, "Mon. Not. Roy. Astron. Soc." (Feb, 2015) 36–48.[6] K. Murase, F. Oikonomou, and M. Petropoulou,
Astrophys. J. (2018) 124.[7] S. Gao, A. Fedynitch, W. Winter, and M. Pohl,
Nat. Astron. (2019) 88–92.[8] IceCube
Collaboration, M. G. Aartsen et al.,
Science (2013) 1242856.[9]
IceCube
Collaboration, M. G. Aartsen et al.,
Astrophys. J. (2017) 45.[10]
IceCube
Collaboration, M. Huber,
These proceedings (2019). earching for neutrino emission from hard X-ray sources with IceCube [11] IceCube, Fermi-LAT, MAGIC, AGILE, ASAS-SN, HAWC, H.E.S.S., INTEGRAL, Kanata,Kiso, Kapteyn, Liverpool Telescope, Subaru, Swift NuSTAR, VERITAS, VLA/17B-403
Collaboration, M. G. Aartsen et al.,
Science (2018) eaat1378.[12]
IceCube
Collaboration, M. G. Aartsen et al.,
Science (2018) 147–151.[13] X. Rodrigues, S. Gao, A. Fedynitch, A. Palladino, and W. Winter,
Astrophys. J. (2019) L29.[14] S. D. Barthelmy et al.,
Space Sci. Rev. (2005) 143.[15] M. Koss et al.,
Astrophys. J. (2017) 74.[16]
IceCube
Collaboration, M. G. Aartsen et al.,
JINST (2017) P03012.(2017) P03012.