The most energetic neutrino ever detected may have come from a Blazar

In February 2023, the KM3NeT/ARCA neutrino detector, submerged in the depths of the Mediterranean Sea, registered an event with no precedent in particle astrophysics. A cosmic neutrino crossed the detector carrying an energy of approximately 220 petaelectronvolts (PeV), exceeding all previous observations by more than an order of magnitude. The event, catalogued as KM3-230213A, has gone without a confirmed origin for three years. A new study published in the Journal of Cosmology and Astroparticle Physics (JCAP) by the KM3NeT collaboration now proposes that this neutrino may have been produced within a population of blazars — active galactic nuclei among the most extreme particle accelerators in the known universe.

The Detector and the Event

KM3NeT/ARCA is a neutrino observatory installed off the coast of Sicily at significant depth on the Mediterranean seafloor. Its detection principle relies on Cherenkov radiation: when a neutrino interacts with water, it produces secondary particles traveling faster than light in that medium, emitting a blue flash captured by the detector’s optical modules. Reconstructing those light tracks allows physicists to infer both the energy and the arrival direction of the incoming neutrino.

At the time of event KM3-230213A, the detector was operating with only 21 active detection lines — roughly 10% of its planned final volume. Despite that reduced capacity, the event was recorded with sufficient clarity to determine its energy, which proved historically unprecedented. No associated electromagnetic counterpart was detected in any wavelength — radio, optical, X-ray, or gamma-ray — at the time or in the direction of the event.

Why Blazars?

The absence of an electromagnetic counterpart is a key diagnostic. In neutrino events linked to identified point sources — such as the case of blazar TXS 0506+056 associated with an IceCube neutrino in 2017 — activity in other wavelengths is typically observed more or less simultaneously. The absence of that signal in KM3-230213A points in a different direction: the neutrino likely did not originate from a single dramatic event such as an explosion or a flare, but from a diffuse background produced by an extended population of sources.

Blazars are active galactic nuclei powered by a supermassive black hole generating relativistic plasma jets aimed almost directly at Earth. That geometry favors the observation of high-energy emission, and theoretical models have long predicted these objects are capable of accelerating protons to extreme energies. When those protons interact with radiation or gas inside the jet, they produce pions that decay into neutrinos and gamma rays. Crucially, neutrinos escape without further interaction, while very-high-energy photons can be absorbed along the way — which would explain the absence of a detectable gamma-ray counterpart.

Methodology: Simulating a Blazar Population

To test this hypothesis, the team used AM3 (Astro-Multimessenger Modeling), an open-source code that simulates the emissions of an individual blazar jet through a one-zone model. Physical jet parameters — magnetic field, velocity, emission-region size, electron luminosity — were fixed to average values drawn from prior independent observations. The two parameters varied freely in the analysis were the baryonic loading, which quantifies the energy carried by protons relative to electrons, and the proton spectral index, which determines the energy distribution across the jet’s proton population.

The best fit to the combined KM3NeT/ARCA and IceCube data was obtained at a baryonic loading of approximately 10 and a proton spectral index of 1.8. Baryonic loading values above 100 were ruled out above the 3-sigma confidence level. Harder proton spectra were similarly disfavored. The model was also applied to blazar PKS 0605-085, positionally consistent with event KM3-230213A, showing compatibility with available multimessenger data.

Two Alternative Scenarios

The KM3NeT collaboration acknowledges that the blazar hypothesis is one of several circulating in the literature since the event’s detection. The two other main candidates are: first, the production of cosmogenic neutrinos, generated when ultra-high-energy cosmic rays interact with the cosmic microwave background during intergalactic propagation — the so-called GZK process; and second, the existence of a sufficiently powerful Galactic accelerator. The latter option is strongly disfavored: the required energy reaches 4 × 10¹⁸ eV and no supporting gamma-ray counterpart exists.

Cosmogenic neutrinos remain a viable alternative in principle, though they require specific conditions on the composition and density of cosmic-ray sources. The diffuse blazar background hypothesis is currently the most parsimonious explanation relative to the full body of available observations.

Implications for Multimessenger Astrophysics

If the blazar attribution is confirmed with future data, the implications are far-reaching. It would mean these objects are capable of accelerating protons well beyond what previous models considered their operational ceiling, substantially revising particle acceleration models in relativistic jets. It would also establish blazars as the dominant source of the diffuse ultra-high-energy neutrino background, partially closing one of the oldest open questions in cosmic-ray astronomy.

The collaboration anticipates that once KM3NeT/ARCA reaches its full detection volume — 115 lines instead of the 21 operational in 2023 — accumulated statistics will enable far more powerful correlation analyses between neutrino events and blazar catalogs. In the meantime, KM3-230213A remains the highest-energy neutrino event ever recorded, and its precise origin remains an open question that the astrophysical community is, for the first time, properly equipped to answer.

The original paper, authored by the KM3NeT collaboration, is published under the title «Blazars as a Potential Origin of the KM3-230213A Event» in the Journal of Cosmology and Astroparticle Physics (March 9, 2026).

More information in arXiv