Neutrino astronomy · aiwiki.page

Neutrino astronomy is the study of astronomical objects and processes through the detection of neutrinos and antineutrinos emitted beyond Earth. These electrically neutral particles interact weakly with matter, allowing them to carry information from stellar interiors and other environments that are difficult to investigate through electromagnetic radiation. Their usefulness as cosmic messengers also makes them difficult to observe: experiments require large target masses, sensitive instrumentation, and careful separation of astronomical signals from background events. (nobelprize.org)

Observations span the relatively low-energy neutrinos produced by nuclear reactions in the Sun and stellar collapse to the high-energy neutrinos associated with cosmic particle acceleration. Solar measurements have tested stellar energy generation and helped establish neutrino flavor conversion, while the neutrino burst from Supernova 1987A provided direct evidence about a collapsing stellar core. At higher energies, the IceCube Neutrino Observatory has measured an astrophysical neutrino flux and found evidence of emission from active galaxies and the Milky Way. (nobelprize.org)

The field connects astronomy with particle physics and forms part of multimessenger astronomy, which combines neutrino measurements with light, cosmic rays, and gravitational waves. Its central tasks include reconstructing neutrino energies and arrival directions, identifying their sources, and using their spectra, timing, and flavors to investigate otherwise inaccessible astrophysical processes. (masterclass.icecube.wisc.edu)

Neutrinos as astronomical messengers

Neutrino astronomy studies astronomical objects and processes through the neutrinos and antineutrinos they emit, rather than through light alone. Its observational domain includes the Sun, stellar explosions, and high-energy particle interactions within and beyond the Milky Way. Solar observations pioneered by Raymond Davis Jr. and the detection of neutrinos from supernova SN 1987A by Masatoshi Koshiba’s Kamiokande experiment established two early applications. The field is defined by the particles’ origin, not simply by the detector used: atmospheric neutrinos arise when cosmic rays strike Earth’s atmosphere, whereas reactor antineutrinos and geoneutrinos originate in terrestrial nuclear reactions and radioactive decay. These signals support related particle-physics and geophysical research and can constitute backgrounds to astronomical measurements. (nobelprize.org)

Two properties make neutrinos distinctive messengers. Their zero electric charge means that interstellar and intergalactic magnetic fields do not bend their trajectories as they bend those of charged cosmic rays. An arriving neutrino therefore retains directional information about its production site, although the precision with which an instrument reconstructs that direction is limited. Their exceptionally weak coupling to matter through the weak interaction also allows them to escape many regions opaque to photons, including stellar interiors and environments surrounding energetic particle accelerators. Neutrinos can consequently reveal otherwise concealed activity. The same low interaction probability makes detection difficult: instruments must monitor large amounts of material to record a small fraction of the passing particles. (icecube.wisc.edu)

The principal observables are energy, arrival direction, arrival time, and flavor. Energy helps distinguish production regimes and characterize a source’s spectrum; direction supports comparisons with astronomical objects or extended sky structures; and timing distinguishes steady emission from variable activity or short bursts. Flavor denotes the electron, muon, or tau type, with corresponding antineutrinos. It supplies information about particle production and propagation, although an individual event does not always permit an unambiguous flavor assignment. These quantities are inferred from interaction products rather than from direct imaging of the neutrino itself. In the IceCube Neutrino Observatory, for example, the light pattern provides estimates of energy and direction, while elongated muon tracks generally offer better directional information than compact particle cascades. (icecube.wisc.edu)

Astronomical searches distinguish several spatial and temporal signal classes. A point source is unresolved at the instrument’s angular resolution; it need not be physically pointlike. Extended emission occupies a resolvable region, such as a band associated with the Galactic plane. A diffuse flux describes emission measured collectively rather than assigned to individual objects, potentially combining numerous unresolved sources and distributed particle interactions. These categories require different searches: a localized excess, a match to a broad spatial template, or a population-level energy and directional distribution. A burst adds a restricted time interval, as in the neutrino signal from SN 1987A on February 23, 1987; temporal classification is independent of whether the source is spatially resolved. (icecube.wisc.edu)

Energies are expressed in electronvolts: $1\,\mathrm{MeV}=10^{6}\,\mathrm{eV}$ , $1\,\mathrm{GeV}=10^{9}\,\mathrm{eV}$ , $1\,\mathrm{TeV}=10^{12}\,\mathrm{eV}$ , and $1\,\mathrm{PeV}=10^{15}\,\mathrm{eV}$ . Stellar neutrino observations predominantly concern the sub-MeV to tens-of-MeV regime, whereas searches for cosmic particle accelerators commonly examine TeV–PeV energies. Ultrahigh-energy searches extend to hundreds of PeV and the EeV scale, where $1\,\mathrm{EeV}=10^{18}\,\mathrm{eV}$ . These are observational regimes rather than strict boundaries between source classes: atmospheric and extraterrestrial spectra overlap, so energy alone cannot establish an event’s astronomical origin. At the highest energies, targets include both emission associated with accelerators and neutrinos produced during the propagation of energetic cosmic rays through background radiation. (masterclass.icecube.wisc.edu)

Production in stars and cosmic particle accelerators

In hydrogen-burning stars, neutrinos arise from the weak-interaction steps embedded in nuclear fusion, rather than from every fusion reaction. The proton–proton chain begins with $p+p\rightarrow d+e^{+}+\nu_e$ , converting one proton into a neutron while forming deuterium. Its net conversion of four protons into helium-4 produces two electron neutrinos and releases approximately $26.7\,\mathrm{MeV}$ , shared between the neutrinos and energy retained by the star. Other branches produce neutrinos through electron capture on beryllium-7 and the beta-plus decay of boron-8. The carbon–nitrogen–oxygen cycle likewise produces electron neutrinos through radioactive intermediates. These nuclear reactions yield spectra set by nuclear energy differences: beta decays produce continuous spectra, whereas capture into a specified nuclear final state produces an approximately discrete neutrino energy. (sns.ias.edu)

Thermal neutrino emission instead draws energy from hot stellar matter without requiring a net conversion of one nuclear species into another. An important channel is electron–positron annihilation, $e^-+e^+\rightarrow\nu_\alpha+\bar{\nu}_\alpha$ , where $\alpha=e,\mu,\tau$ . It produces neutrino–antineutrino pairs of all three flavors, although their production rates need not be equal. Other thermal channels include plasmon decay—the conversion of an electromagnetic excitation of the plasma into a neutrino pair—and the photoneutrino process, $\gamma+e^-\rightarrow e^-+\nu_\alpha+\bar{\nu}_\alpha$ . Their relative importance depends on temperature, density, and electron degeneracy. Calculations of advanced stellar burning therefore combine thermal emission with composition-dependent beta processes; a 2017 calculation by Kelly Patton and colleagues followed 204 isotopes and found that nuclear processes supplied much of the electron-neutrino spectral tail above approximately $3$ – $4\,\mathrm{MeV}$ . (arxiv.org)

During stellar collapse, electron capture on nuclei, $e^-+(A,Z)\rightarrow(A,Z-1)+\nu_e$ , and on free protons, $e^-+p\rightarrow n+\nu_e$ , reduces the electron fraction and makes the matter increasingly neutron-rich. Conversely, beta-minus decay and positron capture can produce electron antineutrinos, with the free-nucleon capture reaction written $e^++n\rightarrow p+\bar{\nu}_e$ . Pair processes supply muon- and tau-flavor neutrinos and antineutrinos as well as electron-flavor pairs. Thus an electron-neutrino-rich collapse signal and emission containing all six neutrino and antineutrino species have distinct microscopic origins; they are not simply products of continued fusion. Chinami Kato and colleagues’ 2017 calculations explicitly followed these different flavor contributions up to core bounce. (arxiv.org)

High-energy production begins when accelerated hadrons, especially protons, collide with ambient gas or photons. Proton–proton collisions produce charged and neutral pions; a representative photohadronic channel is $p+\gamma\rightarrow\Delta^+\rightarrow n+\pi^+$ . The charged-pion decay chain is

\pi^+\rightarrow\mu^++\nu_\mu,\qquad \mu^+\rightarrow e^++\nu_e+\bar{\nu}_\mu,

with charge-conjugate products for negative pions. If the pions and muons decay before losing substantial energy, this chain supplies one electron-flavor and two muon-flavor neutrinos or antineutrinos. Each neutrino typically carries about one-quarter of the pion energy; the frequently used photohadronic estimate $E_\nu\sim0.05E_p$ is a characteristic energy partition, not an exact relation for individual events. Accompanying neutral pions decay through $\pi^0\rightarrow\gamma+\gamma$ , linking hadronic neutrino production to gamma-ray emission, although absorption can prevent those photons from escaping the source. (docushare.icecube.wisc.edu)

The same interaction physics can operate far from an accelerator. Source neutrinos originate within the acceleration region or its surrounding gas and radiation fields; cosmogenic neutrinos are generated during the subsequent propagation of ultrahigh-energy cosmic rays through intergalactic background radiation, including the cosmic microwave background. Their production site therefore need not coincide with the object that accelerated the parent particle. This distinction also prevents an extremely high neutrino energy alone from establishing a cosmogenic origin: both source interactions and propagation interactions can generate ultrahigh-energy neutrinos. (km3net.org)

Flavor conversion and propagation

The electron, muon, and tau flavors of the neutrino, denoted $\nu_e,\nu_\mu,\nu_\tau$ , are defined by their charged-current weak interactions with the corresponding charged leptons. They are not states of definite mass: each flavor state is a superposition of three mass states,

|\nu_\alpha\rangle=\sum_{i=1}^{3}U_{\alpha i}^{*}|\nu_i\rangle,

where $U$ is the Pontecorvo–Maki–Nakagawa–Sakata mixing matrix. Different masses accumulate different quantum phases during propagation, changing the probabilities of detecting each flavor. This is neutrino oscillation, not a loss of neutrinos; ordinary three-flavor mixing preserves the total active-neutrino flux. (pdgweb.lbl.gov)

Inside the Sun, coherent forward scattering on electrons gives electron neutrinos an additional effective potential, $V_e=\sqrt{2}G_F n_e$ , in natural units, where $G_F$ is the Fermi constant and $n_e$ the electron number density. This changes the effective mixing and propagation states. Under sufficiently gradual density changes, a neutrino follows an instantaneous matter eigenstate as it travels outward—the mechanism underlying the Mikheyev–Smirnov–Wolfenstein effect. High-energy solar electron neutrinos consequently emerge predominantly as the second mass state rather than retaining their original flavor. (pdgweb.lbl.gov)

The electron-neutrino survival probability $P_{ee}$ is therefore energy dependent: it is approximately 0.55 at low solar energies, where vacuum averaging dominates, and approximately 0.30 at higher energies, where matter conversion dominates. (pdgweb.lbl.gov) The transition occurs broadly around 1–4 MeV. An electron-neutrino measurement consequently samples only an energy-dependent fraction of the solar flux: at a specified energy, $\Phi_e=P_{ee}\Phi_{\mathrm{total}}$ , assuming conversion among the three active flavors. Measurements sensitive to all active flavors provide a complementary total-flux determination; SNO established that roughly two thirds of the solar electron neutrinos in its high-energy sample arrived as muon or tau neutrinos. (snoplus.phy.queensu.ca)

For distant high-energy sources, rapid oscillation phases are generally averaged over the source extent, energy distribution, and experimental resolution. The resulting vacuum transition probability is

\overline{P}_{\alpha\rightarrow\beta} =\sum_i |U_{\alpha i}|^2|U_{\beta i}|^2, \qquad \Phi_\beta^\oplus =\sum_\alpha\overline{P}_{\alpha\rightarrow\beta}\Phi_\alpha^{\mathrm{source}}.

A complete charged-pion–muon decay chain gives the familiar approximate source ratio $\nu_e:\nu_\mu:\nu_\tau=1:2:0$ , counting neutrinos and antineutrinos together, which becomes approximately $1:1:1$ at Earth. Equal proportions are a benchmark, not a universal outcome. If muons lose energy before decaying, a source can approach $0:1:0$ ; neutron beta decay instead supplies an electron-antineutrino-dominated component, approaching $1:0:0$ . Their propagated mixtures differ from the pion-chain benchmark, while the precise predictions also depend on the mixing parameters and energy-dependent source conditions. (arxiv.org)

Passage through Earth introduces attenuation as well as flavor evolution. At high energies, neutrino–nucleon interaction cross sections increase: charged-current interactions remove incident neutrinos, while neutral-current interactions redistribute them toward lower energies. Long trajectories through the dense interior are therefore more strongly attenuated than near-horizontal paths. In November 2017, the IceCube Neutrino Observatory reported an Earth-absorption measurement using approximately 10,800 upward-going events, predominantly atmospheric neutrinos, with sensitivity over 6.3–980 TeV. The observed suppression agreed with Standard Model expectations. A telescope’s directional acceptance thus depends jointly on neutrino energy and the amount of Earth traversed; neutrinos cannot be treated as passing unimpeded through all matter. (icecube.wisc.edu)

Detection methods and experimental architectures

Neutrino detectors register secondary particles produced by the weak interaction, rather than neutrinos directly. In charged-current interactions, $\nu_\ell+N\rightarrow\ell^-+X$ , the outgoing charged lepton identifies the neutrino flavor; antineutrinos produce the corresponding positive lepton. Electron production is accessible at MeV energies, whereas muon and tau production requires substantially higher energies because of their masses. In neutral-current interactions, $\nu_\ell+N\rightarrow\nu_\ell+X$ , all three active flavors can transfer energy to the target, but the escaping neutrino prevents full energy measurement and generally leaves the flavor unidentified. Neutrino–electron elastic scattering also accepts all active flavors, with greater sensitivity to electron neutrinos because they interact through both charged and neutral currents. (pdgaws.lbl.gov)

Radiochemical detectors collect the daughter atoms created by electron-neutrino capture and count their subsequent radioactive decays. Raymond Davis Jr.’s chlorine experiment used tetrachloroethylene, in which $\nu_e+{}^{37}\mathrm{Cl}\rightarrow{}^{37}\mathrm{Ar}+e^-$ ; gallium experiments used ${}^{71}\mathrm{Ga}$ , producing ${}^{71}\mathrm{Ge}$ , with a threshold of 233 keV that admitted low-energy solar neutrinos. Chemical extraction yields an integrated capture rate over each exposure, not an individual event’s arrival time, direction, or energy. Water detectors instead record Cherenkov radiation from charged particles moving faster than light propagates through water. Photomultiplier arrays provide event-by-event timing, energy estimates, and directional information. SNO’s heavy water added a distinctive capability: electron-neutrino charged-current breakup, $\nu_e+d\rightarrow p+p+e^-$ , could be compared with all-flavor neutral-current breakup, $\nu_\ell+d\rightarrow p+n+\nu_\ell$ , detected through the liberated neutron. (nobelprize.org)

Organic liquid scintillators produce abundant, predominantly isotropic light when charged particles excite their molecules, enabling lower-energy measurements but providing less direct directional information than Cherenkov imaging. Borexino used recoil electrons from solar-neutrino scattering. Hydrogen-rich scintillators also detect electron antineutrinos through inverse beta decay, $\bar{\nu}_e+p\rightarrow e^++n$ , above approximately 1.8 MeV: a prompt positron signal followed by delayed neutron capture supplies a characteristic coincidence. Liquid-argon detectors offer a different information yield. In a time-projection chamber, an electric field drifts ionization electrons toward segmented readout planes, permitting spatial imaging of deposited energy. DUNE’s supernova program exploits electron-neutrino absorption on ${}^{40}\mathrm{Ar}$ , complementing the electron-antineutrino sensitivity of inverse-beta detectors. (lngs.infn.it)

High-energy optical telescopes use natural ice or deep water as both target and transparent detection medium. IceCube’s original cubic-kilometer array contains 5,160 optical modules on 86 strings, at depths of 1,450–2,450 m; its main-array geometry uses approximately 125 m horizontal and 17 m vertical spacing. Denser instrumentation in DeepCore extends sensitivity toward approximately 10 GeV. Larger spacing economically encloses more target material, but favors brighter, higher-energy events; closer spacing samples the weaker light signals of lower-energy interactions. ANTARES, operated in the Mediterranean from 2006 to 2022, demonstrated deep-sea detection with 12 instrumented lines covering approximately $0.1\ \mathrm{km}^2$ . (doi.org)

KM3NeT makes this density trade-off explicit: ARCA uses approximately 90 m between strings and 36 m between optical modules vertically for high-energy astronomy, whereas ORCA uses approximately 20 m and 9 m, respectively, for lower-energy atmospheric-neutrino measurements. Each module houses 31 small photomultipliers rather than one large tube. Baikal-GVD uses independently functioning eight-string clusters in freshwater; its first construction phase, completed in 2021, comprised eight clusters with an instrumented volume of approximately $0.4\ \mathrm{km}^3$ . Modular construction allows additional target volume to be instrumented while existing clusters continue collecting data. (link.springer.com)

At ultrahigh energies, radio arrays seek coherent pulses generated by the excess negative charge in neutrino-induced particle showers—the Askaryan effect. Radio attenuation lengths in ice permit antennas to monitor much larger volumes than densely spaced optical sensors, although higher shower energies are needed. Earth-skimming searches instead target tau neutrinos interacting in rock: an emerging tau can decay in the atmosphere and initiate an air shower. Instruments may detect that shower through radio emission, atmospheric fluorescence, optical Cherenkov light, or ground particles. The atmosphere and intervening rock therefore become components of the detection architecture, rather than an instrumented underwater or in-ice volume. (pdgaws.lbl.gov)

Event reconstruction, backgrounds, and statistical evidence

Event reconstruction treats the recorded light as an inverse problem: the positions of optical sensors, photon arrival times, and measured charges constrain the interaction vertex, particle direction, and energy deposition. Predictions must include photon absorption and scattering in the detector medium, sensor efficiencies, and timing calibration. A fit compares the observations with alternative particle trajectories or shower configurations. In large optical arrays, a muon produces an elongated track, whereas a compact electromagnetic or hadronic particle shower produces a cascade. Tracks generally provide stronger directional constraints because light is sampled along a long trajectory; cascades concentrate the energy measurement within a smaller region. (arxiv.org)

Topology does not uniquely identify neutrino flavor. Electron-neutrino charged-current interactions normally produce cascades, but neutral-current interactions of every flavor do so as well. Tau-neutrino charged-current interactions can produce two cascades: one at the initial hadronic interaction and another when the tau lepton decays. The mean separation scales approximately as $\langle L_\tau\rangle \simeq 50\,\mathrm{m}\,(E_\tau/\mathrm{PeV})$ . Whether these are distinguishable depends on their separation, relative brightness, location, and the optical response. An IceCube analysis published in 2022 required a reconstructed separation of at least 10 m; shorter tau events usually remained classified as single cascades. Non-tau events can also imitate this signature through reconstruction errors or localized muon energy losses. (masterclass.icecube.wisc.edu)

Deposited energy is not the same quantity as the parent neutrino’s energy. A contained electron-neutrino charged-current event can deposit nearly all its energy locally, whereas a neutral-current event leaves an outgoing neutrino carrying unobserved energy. Tau decays also release invisible neutrinos. A through-going muon may originate outside the instrumented volume and lose energy both before entering and after leaving it; its measured energy loss therefore provides only partial information about the initiating neutrino. Parent-energy estimates consequently require interaction and propagation models. IceCube’s 2014 reconstruction-methods paper reported approximately 15% uncertainty in electromagnetic-equivalent deposited energy above 10 TeV—a detector-response measure, not a universal neutrino-energy resolution. (link.springer.com)

Background rejection depends on energy and detector architecture. Cosmic-ray showers generate atmospheric muons and atmospheric neutrinos; the latter are genuine neutrino events but backgrounds to extraterrestrial searches. Underground, underwater, or under-ice shielding suppresses muons, while upward-going selections use Earth to exclude penetrating atmospheric muons. Contained-event selections reject early light in outer veto layers, also suppressing downward atmospheric neutrinos accompanied by shower muons. At lower energies, radioactive contaminants can mimic signals; purification, shielding, and fiducial-volume cuts reduce them. Sensor dark counts and electronic noise require coincidence and data-quality cuts. Control samples—such as tagged muons, off-source data, and rejected-event sidebands—test background predictions without relying entirely on simulation. (pdgaws.lbl.gov)

Flux inference requires the effective area $A_{\mathrm{eff}}(E,\Omega)$ , which incorporates interaction probability, propagation, detection efficiency, and selection cuts rather than simply describing geometrical size. For a steady differential intensity $\Phi$ , the expected count over live time $T$ is

N_{\mathrm{exp}}=T\int \Phi(E,\Omega)\,A_{\mathrm{eff}}(E,\Omega)\,\mathrm{d}E\,\mathrm{d}\Omega .

The product of live time and effective area defines exposure. Monte Carlo simulations supply response distributions mapping true energies and directions into reconstructed quantities. Sensor sensitivity, optical-medium properties, interaction cross sections, and atmospheric-flux models contribute systematic uncertainties. IceCube’s 2021 release of its 2008–2018 point-source data supplied effective areas and energy–direction response information for this purpose. (pdgaws.lbl.gov)

Source searches commonly maximize a likelihood combining signal and background probabilities for each event, using angular separation, directional uncertainty, reconstructed energy, and sometimes arrival time. A likelihood-ratio statistic measures improvement over the background-only hypothesis. Background ensembles establish its significance; scanning many sky positions, sources, or time windows requires trial corrections, so a local p-value is not the probability for the whole search. These tests distinguish three claims: a detected event establishes an interaction candidate; a population excess over atmospheric expectations supports an astrophysical flux; and statistically significant directional or temporal clustering supports association with a particular source. An astrophysical flux can therefore be established before its individual sources are resolved. (arxiv.org)

Solar neutrinos and the development of the field

Raymond Davis Jr.’s chlorine experiments and John Bahcall’s solar calculations made the Sun a testable source of astronomical neutrinos. Neutrinos emerge directly from the reactions sustaining solar nuclear fusion, whereas electromagnetic energy undergoes repeated absorption and re-emission before reaching the surface. Davis built an underground prototype in Ohio in 1961, followed by the Homestake experiment in South Dakota. Its radiochemical method exploited $\nu_e+{}^{37}\mathrm{Cl}\rightarrow{}^{37}\mathrm{Ar}+e^-$ : radioactive argon atoms were extracted from a large chlorine-bearing target and counted through their decay. Bahcall calculated the expected capture rate by combining solar structure, nuclear-reaction rates, and neutrino-capture probabilities. Their experimental and theoretical proposals appeared in paired papers in 1964. (nature.com)

Homestake’s first published results in 1968 revealed a deficit that persisted through subsequent measurements: the capture rate was approximately one-third of the standard prediction. Different detector materials tested whether this reflected experimental error, an inadequate solar model, or neutrino physics. Chlorine principally sampled higher-energy components rather than the abundant low-energy pp neutrinos. The SAGE and GALLEX gallium experiments extended sensitivity to pp neutrinos through capture on gallium-71, again finding a deficit. Kamiokande and Super-Kamiokande instead observed neutrino–electron scattering in water, recording individual events and their recoil directions; their solar-pointing signal independently established the source. Unlike radiochemical measurements, these observations also supplied energy and time information. (journals.aps.org)

The Sudbury Neutrino Observatory (SNO) separated missing electron neutrinos from missing total flux. In heavy water, charged-current breakup of deuterium measured electron neutrinos, while neutral-current breakup measured all three active flavors with equal sensitivity. Following initial results in 2001, SNO’s 2002 analysis obtained an electron-neutrino flux of approximately $1.76\times10^6\ \mathrm{cm^{-2}\,s^{-1}}$ , but a total active-flavor flux of $5.09\times10^6\ \mathrm{cm^{-2}\,s^{-1}}$ , consistent with solar-model predictions for boron-8 neutrinos. The non-electron component differed from zero by $5.3$ standard deviations. The central discrepancy therefore arose from flavor conversion, described by neutrino mixing and oscillations, rather than a comparable failure of the predicted solar energy-generation rate. (nobelprize.org)

Solar neutrino spectroscopy distinguishes reactions within the proton–proton chain: the continuous pp spectrum from the initial fusion reaction, monoenergetic pep neutrinos, electron-capture lines from beryllium-7, and continuous spectra from boron-8 decay and the rare hep reaction. The separate carbon–nitrogen–oxygen (CNO) cycle uses carbon, nitrogen, and oxygen nuclei as catalysts; beta decays within it generate additional neutrino spectra. Borexino, a highly purified liquid-scintillator detector at Italy’s Gran Sasso laboratory, extracted these components from electron-recoil spectra. It reported direct pp-neutrino measurements in 2014 and a comprehensive pp-chain analysis in 2018, measuring pp, beryllium-7, pep, and boron-8 contributions within one experimental framework. (nobelprize.org)

In November 2020, Borexino reported experimental evidence for solar CNO neutrinos, rejecting the no-CNO hypothesis at greater than $5\sigma$ . This tested the secondary fusion mechanism expected to supply roughly 1% of solar energy. The astrophysical interpretation of such measurements differs from determining oscillation parameters: after correcting for flavor conversion, component fluxes constrain nuclear-reaction branching, core conditions, and energy production. Boron-8 emission is particularly sensitive to core temperature, while CNO emission also depends on the core’s carbon and nitrogen abundances. Consequently, CNO measurements probe solar composition independently of surface spectroscopy. Comparisons with models must nevertheless include uncertainties in nuclear rates, opacity, and flavor conversion; a measured interaction rate is not itself a direct abundance determination. Combining pp-chain and CNO information helps separate temperature-dependent effects from composition-dependent effects. (nature.com)

Core-collapse supernovae and the diffuse supernova background

In a core-collapse supernova, the neutrino light curve traces successive stages of the central engine. When the collapsing core reaches nuclear density and rebounds, the outward-moving shock dissociates nuclei, enabling rapid electron capture, $e^-+p\rightarrow n+\nu_e$ . As the shock crosses the region from which neutrinos can escape, this produces a brief neutronization burst, dominated at emission by electron neutrinos and lasting roughly 10–20 milliseconds. During the subsequent accretion phase, infalling matter releases gravitational energy as it settles onto the newly formed proto-neutron star; electron neutrinos and antineutrinos carry substantial accretion luminosity. Accretion can continue for hundreds of milliseconds or longer, depending on the progenitor and explosion outcome. Cooling then dominates: neutrinos and antineutrinos of all flavors diffuse out as the remnant contracts and loses thermal energy and lepton number over roughly ten seconds. The total neutrino energy is typically a few $10^{53}$ erg, far exceeding the explosion’s kinetic energy. (arxiv.org)

The burst from SN 1987A, in the Large Magellanic Cloud, reached Earth on February 23, 1987. The conventionally quoted samples comprise 12 candidate events in Kamiokande-II in Japan, eight in the Irvine–Michigan–Brookhaven detector in the United States, and five in the Baksan scintillation detector in the Soviet Union. These small samples, distributed over intervals of order ten seconds, were predominantly interpreted as electron-antineutrino interactions through inverse beta decay, $\bar{\nu}_e+p\rightarrow n+e^+$ . Background contamination, differing detector thresholds, and event-selection choices matter when fitting their energies and arrival times. Nevertheless, the burst duration and inferred energetics supported the central prediction that collapse produces a hot compact remnant whose gravitational binding energy escapes mainly in neutrinos. The observations did not individually resolve all three emission phases or establish the detailed mechanism of shock revival. (arxiv.org)

A sufficiently nearby future collapse could provide much richer measurements. Before collapse, increasing thermal emission and nuclear beta processes during advanced burning produce a weaker, lower-energy presupernova signal. Models place its practical detection range at approximately one kiloparsec for suitable low-threshold instruments, with strong dependence on stellar mass, distance, and detector background. During the burst, changes in luminosity and spectrum could constrain accretion, explosion onset, and remnant cooling; rapid variations could reveal hydrodynamic instabilities. If the proto-neutron star instead collapses into a black hole, models predict a sharp termination of its neutrino emission. These are prospective diagnostics, not features established by the sparse SN 1987A record, and their interpretation requires source modeling and allowance for flavor conversion. (arxiv.org)

The diffuse supernova neutrino background (DSNB) is the accumulated flux from stellar core collapses throughout cosmic history, rather than a burst associated with one identifiable star. Its spectrum combines the collapse rate, emitted spectra, cosmological redshift, and contributions from collapses that form black holes without bright optical explosions. Searches therefore probe a population of stellar deaths, including otherwise difficult-to-observe events. (arxiv.org)

Electron-antineutrino searches exploit the prompt positron and delayed neutron from inverse beta decay. Neutron tagging identifies this coincidence and rejects many isolated backgrounds. Super-Kamiokande began adding gadolinium to its water in 2020: neutron capture on gadolinium releases a gamma-ray cascade totaling about 8 MeV, more readily detected than the 2.2 MeV photon from capture on hydrogen. Reactor antineutrinos compete at low energies, while atmospheric-neutrino interactions become increasingly important at higher energies; cosmic-ray-induced radioactive products introduce additional backgrounds. These competing signals motivate searches principally in the tens-of-MeV range. (www-sk.icrr.u-tokyo.ac.jp)

On June 26, 2026, the Super-Kamiokande collaboration reported a $2.6\sigma$ indication of the DSNB using approximately 5,000 days of data: 3,349 days of pure-water operation and 1,653 days with gadolinium. This reported excess represents an indication rather than a discovery-level establishment of the background. (icrr.u-tokyo.ac.jp)

High-energy neutrino observations and source identification

In November 2013, the IceCube Neutrino Observatory reported 28 events recorded between May 2010 and May 2012, with deposited energies of approximately 30 TeV–1.2 PeV. Their energy distribution and other properties were inconsistent with atmospheric muons and neutrinos alone at $4.1\sigma$ , providing evidence for an extraterrestrial high-energy flux. An expanded three-year analysis reported in 2014 rejected a purely atmospheric origin at $5.7\sigma$ ; an independent analysis of through-going muon tracks provided further confirmation in 2015. These results established an astrophysical population, not particular sources: the early samples showed no statistically significant clustering that identified an individual accelerator. (arxiv.org)

The event IceCube-170922A, detected on September 22, 2017, connected source searches with contemporaneous electromagnetic observations. Its reconstructed direction was consistent with TXS 0506+056, a blazar—an active galaxy whose relativistic jet points approximately toward Earth. Fermi-LAT observations showed enhanced gamma-ray activity, and MAGIC subsequently detected very-high-energy gamma rays. The spatial and temporal association disfavored chance coincidence at approximately $3\sigma$ , depending on the statistical prescription; it was evidence for an association rather than proof from a single event. A separate archival search found a neutrino excess during 2014–2015, with a best-fitting signal of $13\pm5$ events and $3.5\sigma$ significance after accounting for the time-dependent search. This earlier episode supplied independent neutrino evidence from the same direction, rather than additional observations of the 2017 flare. (arxiv.org)

In 2022, IceCube reported evidence for emission from NGC 1068, also called Messier 77, using data collected during 2011–2020. The analysis inferred an excess of $79^{+22}_{-20}$ neutrinos over background, with a trial-corrected significance of $4.2\sigma$ . This was a statistically fitted source contribution, not 79 events individually identifiable as originating in the galaxy. NGC 1068 contains an obscured active galactic nucleus powered by accretion onto a supermassive black hole. Its substantial TeV neutrino emission, without comparable escaping gamma-ray emission, supports interpretations involving particle interactions in a gamma-ray-opaque environment. Dense gas and radiation near the nucleus can enable neutrino production while attenuating accompanying gamma rays. A hot accretion-disk corona is one proposed production region, but the observation does not uniquely establish that location or a particular acceleration mechanism. (arxiv.org)

IceCube’s June 2023 Galactic-plane result identified extended emission rather than a single extragalactic point source. The analysis used approximately 60,000 cascade-like events from ten years of observations, with machine-learning methods improving background rejection and directional reconstruction. Comparing their sky distribution with diffuse-emission templates yielded $4.5\sigma$ evidence against the background-only hypothesis. The templates included one extrapolated from Fermi-LAT gamma-ray observations. The signal is consistent with energetic protons and other cosmic-ray nuclei colliding with interstellar gas, producing pions whose decays yield neutrinos and gamma rays. It established a Galactic contribution to the high-energy neutrino sky, but did not resolve the individual accelerators that supplied those cosmic rays. (icecube.wisc.edu)

KM3NeT extended observations to substantially higher inferred energies with KM3-230213A, recorded by its Mediterranean ARCA detector on February 13, 2023, and reported on February 12, 2025. The nearly horizontal muon track had an estimated energy of approximately 120 PeV; the inferred parent-neutrino energy had a median near 220 PeV and a central 90% interval of 72 PeV–2.6 EeV. The broad interval reflects the probabilistic inference from the observed muon to its parent neutrino, including unobserved energy and assumptions about the incident spectrum. The reported 68% directional containment radius was $1.5^\circ$ , dominated by uncertainty in the detector’s absolute orientation. The discovery analysis established no astronomical counterpart: both production near a distant accelerator and production during ultrahigh-energy cosmic-ray propagation remained possible interpretations. (nature.com)

Multimessenger coordination and transient searches

Real-time neutrino alerts distribute an event’s arrival time, reconstructed direction, localization uncertainty, and estimated astrophysical probability to other observatories. For the IceCube Neutrino Observatory, automated notices transmitted through NASA’s General Coordinates Network (GCN) typically arrive within 0.5–1 minute; subsequent circulars provide refined reconstructions and follow-up results. Localization is a region rather than an exact position: GCN lists characteristic radii of 0.25–0.75° for track alerts, compared with 2–20° for cascade alerts. Optical, X-ray, and gamma-ray telescopes use these regions to select targets or tile the sky. The estimated astrophysical probability distinguishes cosmic events statistically from atmospheric backgrounds; it is not the probability that a particular object produced the neutrino. An alert therefore initiates a source search rather than announcing a confirmed discovery. (gcn.nasa.gov)

The SuperNova Early Warning System (SNEWS) coordinates a different signal: the seconds-long, MeV-energy burst expected from a nearby core-collapse supernova. Its original coincidence scheme required signals from at least two independent detectors within 10 seconds, suppressing false alarms caused by local detector backgrounds. Neutrinos escape from the collapsing core before the outgoing shock reaches the stellar surface, potentially providing minutes to hours of warning before optical brightening; the interval depends on the progenitor’s structure. Localization can combine directional information from neutrino–electron scattering with triangulation from arrival-time differences between geographically separated detectors. Timing differences constrain the direction through the projected detector separation, although finite event counts and differing detector responses limit precision. The SNEWS 2.0 program extends this exchange of timing and localization information to support early electromagnetic and gravitational-wave observations. (arxiv.org)

Transient searches operate in both directions: neutrino alerts trigger electromagnetic observations, while independently discovered outbursts define directions and time windows for searches in neutrino data. For gamma-ray bursts, analyses test prompt emission as well as precursor and delayed activity; an IceCube study reported in 2022 considered windows extending up to two weeks before and after the burst. Active-galaxy flares and tidal disruption events, in which a star is disrupted by a massive black hole, require longer monitoring. The September 22, 2017 neutrino alert associated with the flaring galaxy TXS 0506+056 illustrates the former strategy. Robert Stein and collaborators’ 2021 study of AT2019dsg illustrates the latter: systematic optical follow-up identified a tidal disruption event as a candidate counterpart, with radio observations supplying evidence for an outflow. Such associations require statistical assessment rather than positional agreement alone. Electromagnetic spectra and light curves characterize the radiation field and outflow, while neutrinos test whether accelerated hadrons undergo particle-producing interactions. (icecube.wisc.edu)

Gravitational-wave triggers provide an independent clock and sky region for neutrino searches around compact-object mergers. Following the binary neutron-star merger GW170817 on August 17, 2017, ANTARES, IceCube, and the Pierre Auger Observatory searched for high-energy neutrinos within $\pm500$ seconds of the merger and during the following 14 days. No directionally coincident neutrinos were found. The resulting limits were consistent with expectations for a short gamma-ray burst viewed away from its jet axis. Here the gravitational waves identified the merger, electromagnetic observations characterized its luminous aftermath, and neutrino measurements constrained possible high-energy particle production. (arxiv.org)

Nondetections constrain a specified emission model, not every possible neutrino contribution from a source class. Their interpretation depends on the assumed spectrum, emission duration, source distance, detector sensitivity in that direction, and the geometry of any jet. A 2017 IceCube analysis found no significant correlation with the prompt emission of 1,172 gamma-ray bursts, placing limits on models of neutrino production during that phase. It did not exclude emission outside the selected windows or from bursts absent from the electromagnetic sample. Likewise, merger limits constrain combinations of neutrino output and viewing geometry rather than establishing that mergers never produce neutrinos. (icecube.wisc.edu)

Connections to fundamental physics and cosmology

Neutrino telescopes test interaction physics at energies beyond those accessible with accelerator neutrino beams. Earth absorption provides one measurement: the energy-dependent deficit of neutrinos crossing long paths through Earth constrains the neutrino–nucleon interaction cross section. IceCube’s 2017 analysis used approximately 10,800 events and was sensitive to absorption between 6.3 and 980 TeV; its result agreed with the Standard Model. The sample consisted mainly of atmospheric neutrinos, with a smaller astrophysical contribution, so this was a particle-physics application of an astronomical instrument rather than a measurement using exclusively extraterrestrial neutrinos. Comparing different trajectories helps distinguish absorption from the unknown incident flux. (icecube.wisc.edu)

Flavor measurements provide a complementary test of propagation. Tracks, cascades, and identifiable tau-decay signatures statistically constrain the electron, muon, and tau contributions. The approximately $1:1:1$ mixture expected after averaged neutrino oscillations for conventional pion-decay sources is a benchmark, not a universal prediction: source conditions can change the emitted mixture. Consequently, tests of neutrino decay or additional, sterile neutrino states must account for both source composition and detector response. A more selective interaction is the Glashow resonance, proposed by Sheldon Glashow in 1960: an electron antineutrino interacts with an electron through $\bar{\nu}_e+e^-\rightarrow W^-$ , with a resonance near 6.3 PeV for an electron at rest. An IceCube event detected on December 8, 2016, and reported in 2021 had a reconstructed visible energy of $6.05\pm0.72$ PeV, consistent with this process. Unlike most telescope events, it specifically probes the electron-antineutrino component, linking electroweak interaction physics to source production mechanisms. (icecube.wisc.edu)

A supernova burst permits tests using arrival times and spectra. For an ultrarelativistic mass eigenstate of mass $m$ , energy $E$ , and propagation distance $D$ , the delay relative to a massless particle is approximately

\Delta t\simeq\frac{D}{2c}\left(\frac{mc^2}{E}\right)^2.

An energy-dependent spreading of the burst can therefore constrain neutrino mass, but intrinsic emission times, flavor conversion, energy resolution, and event statistics also shape the observed distribution. Such analyses cannot simply treat the optical brightening as simultaneous with neutrino emission. The roughly ten-second SN 1987A signal also underlies limits on additional cooling channels: hypothetical weakly coupled particles, including axions, could remove energy from the newly formed neutron star and alter its neutrino luminosity and cooling duration. Quantitative bounds depend on the stellar model, dense-matter microphysics, particle couplings, and the interpretation of the sparse late-time events. (arxiv.org)

Searches for dark matter instead seek an additional neutrino component from annihilation or decay. In the Sun, halo particles could scatter on nuclei, lose sufficient energy to become gravitationally captured, and accumulate before annihilating. A seven-year IceCube DeepCore analysis reported in 2021 found no significant excess; converting its flux limits into dark-matter–proton scattering limits assumed specified annihilation channels and equilibrium between capture and annihilation. Searches toward the Galactic center and halo test annihilation or decay in extended dark-matter distributions. For direct production of a neutrino–antineutrino pair, the idealized line energy is $E_\nu\simeq m_\chi c^2$ for annihilation and $E_\nu\simeq m_\chi c^2/2$ for decay. A five-year IceCube line search reported in 2023 found no significant excess and set annihilation-rate and lifetime limits. These are model-dependent exclusions, sensitive to the assumed density distribution and final states, not detections of dark matter. (arxiv.org)

Cosmological neutrino constraints concern a different population and measurement method. The cosmic neutrino background consists of relics of the hot early universe, not neutrinos emitted by identifiable stars or accelerators. Standard decoupling calculations predict a present temperature near 1.9 K and an effective relativistic-species parameter $N_{\mathrm{eff}}\simeq3.044$ ; this parameter describes radiation density rather than simply counting particles. Primordial element abundances, cosmic microwave background anisotropies, and large-scale structure provide indirect evidence for this background. Neutrino free streaming suppresses matter clustering on small scales, making these observations sensitive principally to the summed masses $\sum_i m_i$ . The resulting limits depend on cosmological assumptions and selected datasets; they are neither direct detections of relic neutrinos nor event-by-event astronomical mass measurements. (pdg.lbl.gov)

Observational limits and emerging capabilities

Most of the diffuse high-energy neutrino flux remains unassigned to individual sources. The unresolved problem is therefore not simply to find more emitters, but to determine how much different populations contribute at different energies: active galactic nuclei, star-forming galaxies, and transient accelerators need not dominate the same spectral intervals. Connecting this emission to the cosmic-ray energy budget requires separating the power injected into accelerated nuclei from the fraction converted into neutrinos before those nuclei escape. Dense environments can be efficient neutrino emitters without releasing a correspondingly large escaping cosmic-ray flux; consequently, a neutrino-bright population need not dominate the cosmic rays measured at Earth. Population models must also reproduce accompanying gamma-ray emission, including radiation absorbed and reprocessed inside sources. (icecube.wisc.edu)

Sparse samples limit both source localization and spectral discrimination. As an illustration—not a current total—a 7.5-year IceCube starting-event analysis contained 60 events with deposited energies between 60 TeV and 10 PeV. Track events generally provide better directions than compact cascades, whereas cascades better constrain deposited energy and extend sensitivity toward the southern sky. Sky coverage is thus uneven in sensitivity rather than simply divided into observable and unobservable hemispheres. Detector response introduces additional limitations: uncertainties in light propagation through the ice affect reconstruction, while deep-sea arrays must track sensor positions as currents move their supporting strings. Optical calibration and acoustic positioning are therefore part of the astronomical measurement, not merely engineering support. (arxiv.org)

Larger low-energy detectors address different questions from sparsely instrumented TeV–PeV telescopes. Hyper-Kamiokande, under construction in Japan with observations targeted for 2028, is designed to provide a 187-kilotonne fiducial water volume for measurements including solar neutrinos, Galactic supernova bursts, and the diffuse supernova background. At the South Pole, the IceCube Upgrade completed deployment in February 2026: five densely instrumented strings added more than 600 sensors and calibration instruments. Besides improving low-energy neutrino measurements, the new devices are intended to characterize the ice sufficiently well to permit improved reconstruction of approximately 15 years of archived observations. These are construction and deployment milestones, distinct from demonstrated improvements in astronomical sensitivity after commissioning and analysis. (www-sk.icrr.u-tokyo.ac.jp)

Expanded optical and radio instruments pursue complementary energy ranges. KM3NeT/ARCA is collecting data while construction continues, with a design comprising two blocks of 115 detection units and roughly one cubic kilometre of instrumented seawater. IceCube-Gen2 remains a proposed extension: its design envisages an approximately eight-cubic-kilometre optical array and a radio array covering about $500\ \mathrm{km}^{2}$ . The Radio Neutrino Observatory Greenland is already operating a partial array while building toward 35 stations; its June 2026 status report described eight deployed stations. It targets neutrinos above approximately $10^{16}\ \mathrm{eV}$ through coherent radio pulses from particle showers, rather than optical light. These radio programs seek the much larger effective volumes needed for the very low fluxes expected at ultrahigh energies. (km3net.org)

Analysis improvements can increase discovery reach without adding detector volume. In July 2025, IceCube reported its first joint point-source search combining 14 years of tracks with 10 years of cascades, using maximum-likelihood estimation and testing 167 gamma-ray sources. Neither hemisphere’s strongest all-sky excess remained significant after trial corrections. Cross-detector combinations provide another route: a 2026 ANTARES–ARCA study reported approximately 10% better performance than ANTARES alone for an assumed $E^{-2}$ spectrum. Longer exposures accumulate repeat events, while independent detectors test whether an excess survives different media, backgrounds, and response uncertainties. For rare-event interpretations, such comparisons constrain both source hypotheses and inferred fluxes rather than making one unusually energetic event sufficient to establish a source. (icecube.wisc.edu)

References

Neutrino astronomynobelprize.org
Pressmeddelande: Nobelpriset i fysik 2002nobelprize.org
Detecting Neutrinosmasterclass.icecube.wisc.edu
Antineutrino Science in KamLANDarxiv.org
IceCube and Neutrinosicecube.wisc.edu
KM3NeT detects the highest energy neutrino ever observedkm3net.org
What can cascade events tell us about neutrino sources?icecube.wisc.edu
Research Highlightsicecube.wisc.edu
Solar Neutrinossns.ias.edu
Neutrino Mass, Mixing, and Oscillationspdg.lbl.gov
Neutrino spectrum from the pair-annihilation process in the hot stellar plasmaarxiv.org
Presupernova Neutrinosdare.uva.nl
Neutrinos from beta processes in a presupernova: probing the isotopic evolution of a massive stararxiv.org
Neutrino emissions in all flavors up to the pre-bounce of massive stars and the possibility of their detectionsarxiv.org
Christopher Weaver, doctoral thesisdocushare.icecube.wisc.edu
M. F. Baker, doctoral thesisuser-web.icecube.wisc.edu
IceCube Preliminary Design Documenticecube.wisc.edu
Neutrino Masses, Mixing, and Oscillationspdgweb.lbl.gov
Flavor Ratios of Astrophysical Neutrinos: Implications for Precision Measurementsarxiv.org
A first look at how the Earth stops high-energy neutrinos in their tracksicecube.wisc.edu
Particle Detectors for Non-Accelerator Physicspdgaws.lbl.gov
Solving the mystery of the missing neutrinosnobelprize.org
Measurement of the solar neutrino capture rate with gallium metal. III: Results for the 2002–2007 data-taking periodarxiv.org
KamLAND: A Liquid Scintillator Anti-Neutrino Detector at the Kamioka Siteweb.stanford.edu
Deep Underground Neutrino Experiment (DUNE), Far Detector Technical Design Report, Volume IV: Far Detector Single-phase Technologyarxiv.org
Supernova and solar neutrino searches at DUNEarxiv.org
Search for a light sterile neutrino with 7.5 years of IceCube DeepCore datadoi.org
Detectors – IceCube-Gen2icecube-gen2.wisc.edu
Particle Physics in Ice with IceCube DeepCorearxiv.org
About us – ANTARESantares.in2p3.fr
Energy Reconstruction Methods in the IceCube Neutrino Telescopearxiv.org
Detection of astrophysical tau neutrino candidates in IceCubelink.springer.com
IceCube Data for Neutrino Point-Source Searches Years 2008-2018arxiv.org
Time-integrated Neutrino Source Searches with 10 years of IceCube Dataarxiv.org
About Raymond Davis Jr.bnl.gov
Raymond Davis, Jr. — Nobel Lecturenobelprize.org
Present Status of the Theoretical Predictions for the Solar-Neutrino Experimentjournals.aps.org
Advanced information on the Nobel Prize in Physics 2002nobelprize.org
Direct Evidence for Neutrino Flavor Transformation from Neutral-Current Interactions in the Sudbury Neutrino Observatoryarxiv.org
Neutrinos from the primary proton–proton fusion process in the Sunnature.com
Comprehensive measurement of pp-chain solar neutrinosnature.com
Experimental evidence of neutrinos produced in the CNO fusion cycle in the Sunnature.com
Supernova Neutrinos: Production, Oscillations and Detectionarxiv.org
Comparative analysis of SN1987A antineutrino fluencearxiv.org
SNEWS 2.0: A Next-Generation SuperNova Early Warning System for Multi-messenger Astronomyarxiv.org
Evidence for High-Energy Extraterrestrial Neutrinos at the IceCube Detectorarxiv.org
A growing astrophysical neutrino signal in IceCube now features a 2-PeV neutrinoicecube.wisc.edu
Timelineicecube.wisc.edu
Multi-messenger observations of a flaring blazar coincident with high-energy neutrino IceCube-170922Aarxiv.org
Neutrino emission from the direction of the blazar TXS 0506+056 prior to the IceCube-170922A alertarxiv.org
Neutrino emission from the direction of the blazar TXS 0506+056 prior to the IceCube-170922A alerticecube.wisc.edu
Evidence for neutrino emission from the nearby active galaxy NGC 1068arxiv.org
Evidence for neutrino emission from X-ray-emitting galaxiesicecube.wisc.edu
Our galaxy seen through a new lens: neutrinos detected by IceCubeicecube.wisc.edu
Observation of high-energy neutrinos from the Galactic planearxiv.org
Observation of an ultra-high-energy cosmic neutrino with KM3NeTnature.com
Observation of an ultra-high-energy cosmic neutrino with KM3NeTnature.com
GCN - Missions - IceCubegcn.nasa.gov
The hunt for neutrinos from gamma-ray burstsicecube.wisc.edu
IceCube data from 2008 to 2017 related to analysis of TXS 0506+056icecube.wisc.edu
A tidal disruption event coincident with a high-energy neutrinoarxiv.org
Search for High-energy Neutrinos from Binary Neutron Star Merger GW170817 with ANTARES, IceCube, and the Pierre Auger Observatoryarxiv.org
GW170817: Observation of Gravitational Waves from a Binary Neutron Star Inspiralarxiv.org
IceCube expands GRB searches to the southern skyicecube.wisc.edu
Measuring the flavor ratio of astrophysical neutrinosicecube.wisc.edu
IceCube data for the first Glashow resonance candidateicecube.wisc.edu
IceCube detection of a high-energy particle proves 60-year-old theoryicecube.wisc.edu
Analyzing the time spectrum of supernova neutrinos to constrain their effective mass or Lorentz Invariance Violationarxiv.org
Muonic boson limits: Supernova reduxdoi.org
A Fresh Look at Axions and SN 1987Aarxiv.org
Search for GeV-scale Dark Matter Annihilation in the Sun with IceCube DeepCorearxiv.org
First spectral search of monochromatic neutrino lines from dark matter annihilation and decay in IceCubeicecube.wisc.edu
Neutrinos in Cosmologypdg.lbl.gov
IceCube’s first hunt for neutrino sources using two types of signalsicecube.wisc.edu
IceCube: Neutrinos from Active Galaxiesarxiv.org
Contribution of starburst nuclei to the diffuse gamma-ray and neutrino fluxarxiv.org
Dynamic position calibrationsftp.km3net.de
Hyper-Kamiokande Design Reportarxiv.org
The IceCube Neutrino Observatory gets a major upgrade beneath the iceicecube.wisc.edu
The ARCA and ORCA detectorskm3net.org
The Radio Neutrino Observatory in Greenland (RNO-G): Status Update and Recent Resultsindico.kit.edu
Combined KM3NeT/ARCA and ANTARES search for compact neutrino sourcesarxiv.org