TXS 0506+056: one alert, two separate lines of evidence

An alert sent in time for others to look

On 22 September 2017, IceCube detected the high-energy event IceCube-170922A and sent out an alert. Speed mattered: a source’s activity changes, and a telescope pointing there months later could not recover all of it.

The reconstructed direction was consistent with TXS 0506+056, a known gamma-ray blazar that was active at the time. A blazar is an active galactic nucleus whose jet points roughly toward us, which can make its changing radiation especially prominent.

A matching direction raised new questions. How unlikely was the match if only background were at work? And what did observations across the electromagnetic spectrum say about the candidate source? Both had to be checked.

Sources: [1] · [2]

Telescopes across the spectrum followed up

Follow-up observations ran from radio to gamma rays. Fermi-LAT measured the gamma-ray activity, and MAGIC detected very-high-energy gamma rays from the source. Other telescopes helped pin down its spectrum and how it varied.

A useful comparison uses both position and time and takes each instrument’s sensitivity into account. The question is whether a shared physical origin explains the observed activity and the neutrino better than the alternatives do; finding a bright object nearby on a map is not enough.

For the association scenarios it tested, the 2018 multimessenger paper reported evidence at about three standard deviations against a chance coincidence. That number belongs to the coincidence analysis and does not carry over to later TXS studies.

Sources: [1] · [3]

Older data answer a different question

A companion analysis searched 9.5 years of earlier IceCube data at the blazar’s position. It found an excess from September 2014 to March 2015, before the 2017 alert. This was a search over time in the neutrino data, separate from the alert event itself.

The reported significance accounts for the search over observing time and for trying two time-window shapes. This matters, because picking a favorable interval after looking at the data gives background fluctuations extra chances.

As the paper states, the earlier episode is statistically independent of the 2017 one. Still, the direction was chosen because of the later association, and both analyses use the same observatory. Independent event samples can still share assumptions.

Sources: [4] · [5]

A link to a source is only the start of a model

Evidence from the same direction at two different times makes the source more interesting. It still leaves the physics open: which accelerated particles interacted, with what target, and where in the active galactic nucleus?

Gamma rays and neutrinos can escape in different ways. How they relate can depend on absorption, particle cooling and the structure of the emitting region. A model that fits one period may need different conditions to fit another.

This one case does not show that every blazar is an equally important neutrino source, or that blazars produce the whole diffuse flux. A claim about the population needs other sources, an account of selection effects and limits from objects that gave no significant signal.

Sources: [4] · [6] · [3]

Two papers, two questions

In the multimessenger paper, follow the steps from the event direction to the source’s activity and the coincidence test. Ask how an alert’s uncertainty region differs from a precise optical position, and which observations came soon enough to tell us about the active episode.

In the archival paper, follow how the search over time and its background trials were built. Ask which interval was selected, what else was scanned, and how the reported significance accounts for those extra chances.

The two papers complement each other. Adding their significances, or quoting the larger one as the certainty of everything said about TXS, misses the point. What they achieved is a link between specific observations that can keep being tested.

Sources: [1] · [4] · [5]

Try it in the Lab

Primary sources & revision

1. IceCube, Fermi-LAT, MAGIC and partner teams · Multimessenger observations of TXS 0506+056 (2018)
2. IceCube · Real-Time Alerts
3. IceCube · Neutrinos point to a long-sought cosmic ray accelerator (12 July 2018)
4. IceCube Collaboration · Neutrino emission before the IceCube-170922A alert (2018)
5. Particle Data Group · Statistics (2025)
6. Francis Halzen · Astroparticle Physics with High Energy Neutrinos: from AMANDA to IceCube (2006)

First published 2026-10-10; last revised 2026-10-13. Original explanatory text and diagrams by Sandbox Physics. Illustrations are schematic; no experimental event records are reproduced here. This is an independent educational publication, not an official Nobel or experiment collaboration publication.