PaperlayerView exact revision on arXiv

Source-linked AI summary

Observation of high-energy neutrinos from the Galactic plane

R. Abbasi, M. Ackermann, J. Adams, J. A. Aguilar, M. Ahlers, M. Ahrens, J. M. Alameddine, A. A. Alves, N. M. Amin, K. Andeen, T. Anderson, G. Anton, C. Argüelles, Y. Ashida, S. Athanasiadou, S. Axani, X. Bai, A. Balagopal V., S. W. Barwick, V. Basu, S. Baur, R. Bay, J. J. Beatty, K. -H. Becker, J. Becker Tjus, J. Beise, C. Bellenghi, S. Benda, S. BenZvi, D. Berley, E. Bernardini, D. Z. Besson, G. Binder, D. Bindig, E. Blaufuss, S. Blot, M. Boddenberg, F. Bontempo, J. Y. Book, J. Borowka, S. Böser, O. Botner, J. Böttcher, E. Bourbeau, F. Bradascio, J. Braun, B. Brinson, S. Bron, J. Brostean-Kaiser, R. T. Burley, R. S. Busse, M. A. Campana, E. G. Carnie-Bronca, C. Chen, Z. Chen, D. Chirkin, K. Choi, B. A. Clark, K. Clark, L. Classen, A. Coleman, G. H. Collin, A. Connolly, J. M. Conrad, P. Coppin, P. Correa, D. F. Cowen, R. Cross, C. Dappen, P. Dave, C. De Clercq, J. J. DeLaunay, D. Delgado López, H. Dembinski, K. Deoskar, A. Desai, P. Desiati, K. D. de Vries, G. de Wasseige, T. DeYoung, A. Diaz, J. C. Díaz-Vélez, M. Dittmer, H. Dujmovic, M. Dunkman, M. A. DuVernois, T. Ehrhardt, P. Eller, R. Engel, H. Erpenbeck, J. Evans, P. A. Evenson, K. L. Fan, A. R. Fazely, A. Fedynitch, N. Feigl, S. Fiedlschuster, A. T. Fienberg, C. Finley, L. Fischer, D. Fox, A. Franckowiak, E. Friedman, A. Fritz, P. Fürst, T. K. Gaisser, J. Gallagher, E. Ganster, A. Garcia, S. Garrappa, L. Gerhardt, A. Ghadimi, C. Glaser, T. Glauch, T. Glüsenkamp, N. Goehlke, A. Goldschmidt, J. G. Gonzalez, S. Goswami, D. Grant, T. Grégoire, S. Griswold, C. Günther, P. Gutjahr, C. Haack, A. Hallgren, R. Halliday, L. Halve, F. Halzen, M. Ha Minh, K. Hanson, J. Hardin, A. A. Harnisch, A. Haungs, K. Helbing, F. Henningsen, E. C. Hettinger, S. Hickford, J. Hignight, C. Hill, G. C. Hill, K. D. Hoffman, K. Hoshina, W. Hou, F. Huang, M. Huber, T. Huber, K. Hultqvist, M. Hünnefeld, R. Hussain, K. Hymon, S. In, N. Iovine, A. Ishihara, M. Jansson, G. S. Japaridze, M. Jeong, M. Jin, B. J. P. Jones, D. Kang, W. Kang, X. Kang, A. Kappes, D. Kappesser, L. Kardum, T. Karg, M. Karl, A. Karle, U. Katz, M. Kauer, M. Kellermann, J. L. Kelley, A. Kheirandish, K. Kin, J. Kiryluk, S. R. Klein, A. Kochocki, R. Koirala, H. Kolanoski, T. Kontrimas, L. Köpke, C. Kopper, S. Kopper, D. J. Koskinen, P. Koundal, M. Kovacevich, M. Kowalski, T. Kozynets, E. Krupczak, E. Kun, N. Kurahashi, N. Lad, C. Lagunas Gualda, J. L. Lanfranchi, M. J. Larson, F. Lauber, J. P. Lazar, J. W. Lee, K. Leonard, A. Leszczyńska, Y. Li, M. Lincetto, Q. R. Liu, M. Liubarska, E. Lohfink, C. J. Lozano Mariscal, L. Lu, F. Lucarelli, A. Ludwig, W. Luszczak, Y. Lyu, W. Y. Ma, J. Madsen, K. B. M. Mahn, Y. Makino, S. Mancina, I. C. Mariş, I. Martinez-Soler, R. Maruyama, S. McCarthy, T. McElroy, F. McNally, J. V. Mead, K. Meagher, S. Mechbal, A. Medina, M. Meier, S. Meighen-Berger, Y. Merckx, J. Micallef, D. Mockler, T. Montaruli, R. W. Moore, K. Morik, R. Morse, M. Moulai, T. Mukherjee, R. Naab, R. Nagai, R. Nahnhauer, U. Naumann, J. Necker, L. V. Nguyen, H. Niederhausen, M. U. Nisa, S. C. Nowicki, D. Nygren, A. Obertacke Pollmann, M. Oehler, B. Oeyen, A. Olivas, E. O'Sullivan, H. Pandya, D. V. Pankova, N. Park, G. K. Parker, E. N. Paudel, L. Paul, C. Pérez de los Heros, L. Peters, J. Peterson, S. Philippen, S. Pieper, A. Pizzuto, M. Plum, Y. Popovych, A. Porcelli, M. Prado Rodriguez, B. Pries, G. T. Przybylski, C. Raab, J. Rack-Helleis, A. Raissi, M. Rameez, K. Rawlins, I. C. Rea, Z. Rechav, A. Rehman, P. Reichherzer, R. Reimann, G. Renzi, E. Resconi, S. Reusch, W. Rhode, M. Richman, B. Riedel, E. J. Roberts, S. Robertson, G. Roellinghoff, M. Rongen, C. Rott, T. Ruhe, D. Ryckbosch, D. Rysewyk Cantu, I. Safa, J. Saffer, D. Salazar-Gallegos, P. Sampathkumar, S. E. Sanchez Herrera, A. Sandrock, M. Santander, S. Sarkar, S. Sarkar, K. Satalecka, M. Schaufel, H. Schieler, S. Schindler, T. Schmidt, A. Schneider, J. Schneider, F. G. Schröder, L. Schumacher, G. Schwefer, S. Sclafani, D. Seckel, S. Seunarine, A. Sharma, S. Shefali, N. Shimizu, M. Silva, B. Skrzypek, B. Smithers, R. Snihur, J. Soedingrekso, A. Sogaard, D. Soldin, C. Spannfellner, G. M. Spiczak, C. Spiering, M. Stamatikos, T. Stanev, R. Stein, J. Stettner, T. Stezelberger, B. Stokstad, T. Stürwald, T. Stuttard, G. W. Sullivan, I. Taboada, S. Ter-Antonyan, J. Thwaites, S. Tilav, F. Tischbein, K. Tollefson, C. Tönnis, S. Toscano, D. Tosi, A. Trettin, M. Tselengidou, C. F. Tung, A. Turcati, R. Turcotte, C. F. Turley, J. P. Twagirayezu, B. Ty, M. A. Unland Elorrieta, N. Valtonen-Mattila, J. Vandenbroucke, N. van Eijndhoven, D. Vannerom, J. van Santen, J. Veitch-Michaelis, S. Verpoest, C. Walck, W. Wang, T. B. Watson, C. Weaver, P. Weigel, A. Weindl, M. J. Weiss, J. Weldert, C. Wendt, J. Werthebach, M. Weyrauch, N. Whitehorn, C. H. Wiebusch, N. Willey, D. R. Williams, M. Wolf, G. Wrede, J. Wulff, X. W. Xu, J. P. Yanez, E. Yildizci, S. Yoshida, S. Yu, T. Yuan, Z. Zhang, P. Zhelnin

arXiv:2307.04427v1astro-ph.HEastro-ph.GAcs.LG

TL;DR

The paper addresses the challenge of identifying Galactic high-energy neutrino emission against substantial backgrounds. It applies machine-learning-based cascade selection and reconstruction to ten years of IceCube data, finding Galactic-plane emission at trial-corrected 4.48σ. The excess is consistent with diffuse Galactic emission, while unresolved point sources remain possible.

  • Problem

    High-energy cosmic rays are difficult to trace because magnetic deflection obscures their arrival directions, motivating searches for associated Galactic neutrinos amid atmospheric backgrounds.

  • Method

    The study applies deep-learning event selection and hybrid reconstruction to ten years of IceCube cascade data, testing diffuse and point-source emission hypotheses with data-derived backgrounds.

  • Results

    4.48σ trial-corrected significance rejects the background-only hypothesis for Galactic-plane emission.

  • Takeaways & Limitations

    The observed Galactic-plane neutrino excess provides strong evidence for Galactic neutrino emission, although it may arise from unresolved point sources.

  • Takeaways & Limitations

    The analysis lacks sufficient statistical power to distinguish the tested emission models or identify embedded point sources.

Abstract

from arXiv · show

The origin of high-energy cosmic rays, atomic nuclei that continuously impact Earth's atmosphere, has been a mystery for over a century. Due to deflection in interstellar magnetic fields, cosmic rays from the Milky Way arrive at Earth from random directions. However, near their sources and during propagation, cosmic rays interact with matter and produce high-energy neutrinos. We search for neutrino emission using machine learning techniques applied to ten years of data from the IceCube Neutrino Observatory. We identify neutrino emission from the Galactic plane at the 4.5$σ$ level of significance, by comparing diffuse emission models to a background-only hypothesis. The signal is consistent with modeled diffuse emission from the Galactic plane, but could also arise from a population of unresolved point sources.

Cascade events in IceCube

IceCube cascade events provide contained energy measurements and, despite poorer angular resolution than tracks, improve sensitivity to TeV neutrino emission from the Southern Galactic plane.

  • Background and event topology: Cascades are short-ranged showers that appear nearly point-like, whereas tracks are elongated and generally provide better directional resolution.Cascades arise predominantly from electron- and tau-neutrino interactions, while tracks arise mainly from muons produced in atmospheric or muon-neutrino interactions.
  • Background and event topology: Contained cascade energy depositions provide a more complete measure of neutrino energy than tracks.This is enabled by the short range of charged particles produced in cascade interactions.
  • Background and event topology: 100 million muons are recorded for every observed astrophysical neutrino, making atmospheric backgrounds a dominant challenge.Southern-sky analyses therefore require selections that reduce incoming atmospheric muons.
  • Cascade sensitivity: Cascade selections reduce atmospheric-neutrino contamination by about an order of magnitude at TeV energies and lower the analysis threshold to about 1 TeV.These benefits compensate for inferior angular resolution relative to tracks in the Southern sky.
  • Cascade sensitivity: Cascade-based analyses are expected to improve sensitivity to extended Southern-sky neutrino emission because they rely more on signal purity than individual-event angular resolution.This is particularly relevant when the emitting region is comparable to or larger than the angular resolution, as for the Galactic plane.

Application of deep learning to cascade events

The analysis uses deep-learning selection and hybrid reconstruction to retain more cascade events, including difficult boundary events, while improving angular resolution and lowering the energy threshold.

  • Deep-learning selection: Convolutional neural networks perform event selection at millisecond inference speed, enabling complex filtering earlier in the pipeline.Earlier filtering retains more lower-energy astrophysical neutrino events and difficult-to-reconstruct cascades.
  • Hybrid reconstruction: The hybrid reconstruction combines maximum-likelihood estimation with deep learning to model photon yields at each detector module.The neural network approximates computationally expensive Monte Carlo simulations while retaining detector symmetries and domain knowledge.
  • Hybrid reconstruction: The hybrid method is intended to use more available information than previous CNN-based methods applied directly to detector data.The approach parameterizes expected light yield from event hypotheses rather than inferring event properties only from observed data.
  • Analysis performance: More than 20 times as many events are retained as in the previous cascade-based Galactic-plane selection.The increased event rate is attributed mainly to improved efficiency, a lower energy threshold, and boundary-event inclusion.
  • Analysis performance: Angular resolution improves by up to a factor of 2 at TeV energies relative to the previous selection.The ten-year sample contains 59,592 selected events from 500 GeV to multiple PeV, compared with 1,980 events in the earlier seven-year selection.

Searches for Galactic neutrino emission

The study tests diffuse Galactic-plane, catalog-stacking, and point-source hypotheses using data-derived backgrounds and spatially convolved emission templates. The Galactic-plane tests reject the background-only hypothesis at trial-corrected 4.48σ, while no individual point source is significant after trials.

  • Search methodology: Backgrounds are estimated by scrambling event right ascension, preserving detector-specific artifacts while randomizing isotropic background directions.P-values are obtained by comparing results with mock experiments generated from randomized data.
  • Search methodology: The tests include three diffuse Galactic-plane models, catalogs of known Galactic gamma-ray sources, Fermi Bubbles, GeV emitters, and an all-sky point-source scan.Signal hypotheses are defined a priori and evaluated with a maximum-likelihood-based method.
  • Galactic plane neutrino searches: 4.71σ, 4.37σ, and 3.96σ reject the background-only hypothesis for the π0, KRA5, and KRAγ models, respectively.The models are correlated, so the analysis applies a conservative trial factor of three to the most significant template.
  • Galactic plane neutrino searches: 4.48σ is the trial-corrected significance for the Galactic-plane emission result.The tested models use spatial templates convolved with detector acceptance and event angular uncertainty.
  • Point-source searches: No single point in the all-sky scan is statistically significant after accounting for trial factors.Some excess locations coincide spatially with known gamma-ray emitters, but the result does not isolate an individual source.
  • Implications of Galactic neutrinos: The Galactic-plane excess cannot be attributed to a single unresolved point source, but the data lack power to distinguish emission models or identify embedded sources.Catalog excesses cannot be interpreted as detections because the catalogs overlap regions with the largest predicted diffuse fluxes.

Implications of Galactic neutrinos

The observed Galactic-plane neutrino signal favors diffuse emission, while unresolved sources remain a plausible explanation. The inferred Galactic contribution is estimated at approximately 6–13% of IceCube’s astrophysical flux at 30 TeV, but current data cannot distinguish models or embedded point sources.

  • Model spectra: A factor of ∼5 separates the best-fitting flux from the simpler π0 extrapolation from GeV energies to 100 TeV.The π0 best-fitting flux remains consistent with recent 100 TeV gamma-ray observations by the Tibet Air Shower Array.
  • Source interpretation: The tests favor diffuse neutrino emission from the Galactic plane, but cannot distinguish the tested emission models or identify embedded point sources.Model-injection tests show that different source hypotheses can produce statistically compatible results.
  • Flux contribution: ∼6–13% of the astrophysical flux at 30 TeV is contributed by the inferred Galactic-plane template models.The observed Galactic flux is integrated over the entire sky, with higher local contributions expected along the central Galactic plane.
  • Implications: The Galactic-plane neutrino excess provides evidence that the Milky Way is a source of high-energy neutrinos and complements IceCube’s diffuse extragalactic measurement.Together, these measurements provide a more complete picture of the neutrino sky.

List of Supplementary Materials

The supplementary materials accompany the paper on the observation of high-energy neutrinos from the Galactic Plane.

  • Supplementary materials are provided for the observation of high-energy neutrinos from the Galactic Plane.
  • The supplementary materials concern the Galactic Plane neutrino observation.
  • The supplementary materials are associated with the reported high-energy neutrino study.

IceCube Collaboration∗:

This section lists IceCube Collaboration affiliations.

  • The IceCube Collaboration affiliations include institutions in Anchorage, Clark-Atlanta, Atlanta, Cambridge, Brussels, Bochum, and Chiba.
  • Listed affiliations include Harvard University and the Massachusetts Institute of Technology in Cambridge, Massachusetts.
  • The collaboration list also includes the International Center for Hadron Astrophysics at Chiba University.

20 Department of Astronomy, Ohio State University, Columbus, OH 43210, USA

This section lists affiliations associated with Ohio State University and the Niels Bohr Institute.

  • The listed affiliations include the Department of Physics and Center for Cosmology and Astro-Particle Physics at Ohio State University.
  • The affiliations span institutions in Columbus, Ohio, and Copenhagen, Denmark.
  • The Niels Bohr Institute affiliation is located in Copenhagen, Denmark.

23 Department of Physics, TU Dortmund University, D-44221 Dortmund, Germany

This section lists affiliations at TU Dortmund University.

  • TU Dortmund University is listed as an institutional affiliation in Dortmund, Germany.

27 Physik-department, Technische Universität München, D-85748 Garching, Germany

This section lists an affiliation at Technische Universität München.

  • Technische Universität München is listed as a Physik-department affiliation in Garching, Germany.

30 Department of Physics and Astronomy, University of California, Irvine, CA 92697, USA

This section lists research affiliations across universities and institutes in multiple countries.

  • The listed affiliations include Karlsruhe Institute of Technology and universities in the United States, Canada, Europe, Asia, and Australia.
  • The affiliations include physics, astronomy, astrophysics, computer science, and astroparticle-physics departments and institutes.
  • United States affiliations span universities including Wisconsin–Madison, Yale, Oxford-related listings, Drexel, and South Dakota Mines.

Materials and Methods

The analysis uses machine-learning event selection and hybrid reconstruction to retain a larger, lower-energy cascade sample while controlling background and refining event properties. The resulting data-driven search supports Galactic-plane neutrino emission, with systematic effects affecting sensitivity and normalization but not the identification.

  • Event Selection and Reconstruction: Convolutional neural networks and gradient-boosted decision trees reduce atmospheric background before neutrino-source searches.The CNNs provide early event processing, while BDTs use high-level CNN outputs in the final selection.
  • Event Selection and Reconstruction: More than 20 times as many events are retained as in the previous IceCube cascade selection.The increase reflects improved efficiency and a lower threshold of about 500 GeV, compared with several TeV previously.
  • Event Selection and Reconstruction: The energy-dependent angular resolution improves over the previous selection through a hybrid reconstruction method.The hybrid method uses neural networks to model high-dimensional pulse-arrival dependencies and provides better uncertainty estimation.
  • Spatial distribution of Galactic Models: The observed Galactic diffuse flux is consistent with the flux level inferred by gamma-ray observatories in the TeV–PeV range.Warm spots accumulate along the Galactic plane and near the Galactic Center, while individual excesses are not statistically significant alone.
  • Systematic Uncertainties and their Impact: Systematic uncertainties affect sensitivity by up to 20% and effective area by about 10%, but the data-driven method leaves Galactic-plane identification robust.Flux normalizations are susceptible to systematic uncertainties, whereas the p-value remains robust.

Supplemental Text

The supplemental analyses test diffuse Galactic templates, stacked source catalogs, and individual point sources using likelihood-based searches. No individual point source is statistically significant after trials, while the diffuse and stacking tests are generally consistent with background.

  • All-sky search: The all-sky point-source search evaluates signal counts and spectral index on a 0.45° grid, requiring correction for approximately 500 trials.Results are reported using the post-trial p-value for the hottest point in each hemisphere.
  • All-sky search: Both hottest points in the Northern and Southern Hemispheres are consistent with the background-only hypothesis.The all-sky map contains warm spots near known gamma-ray sources, including the Crab Nebula, but these do not establish significant neutrino emission.
  • Sensitivity and resolution: Cascade-event angular uncertainties of 5° to 20° mean that resolving a point source requires combining many events; under the measured parameters, resolution is expected to reach a few degrees.The sensitivity and upper limits depend on the assumed E−2 or E−3 spectrum and source declination.
  • Fermi Bubbles: The Fermi Bubbles search is consistent with background across tested spectral cutoffs, with a trial-corrected p-value of 6.48 × 10−2 (1.52σ).The result accounts for correlations between the different cutoff tests.
  • Stacking searches: Stacking analyses estimate the total signal events and a shared catalog power-law index, with results summarized for Galactic source catalogs and source-list searches.Catalog sources are weighted to contribute equally before detector effects are considered.
Loading 2307.04427v1…