IceCube’s 28 events: finding high-energy neutrinos from space

Two unusual events led to a wider search

IceCube had already found two very energetic events when researchers widened the search. Looking through data collected from May 2010 to May 2012, they found 26 more that passed their selection. The resulting sample of 28 was reported in 2013.

The researchers looked for events that appeared to begin inside the detector and left a large amount of measurable energy there. This helped separate them from charged particles that entered the instrument from outside. The selection was designed to find rare, energetic interactions among much more ordinary activity.

Finding them raised the central question: could the particles produced in Earth’s atmosphere explain this sample, or was another source needed?

Sources: [1] · [2]

The atmosphere makes neutrinos too

Cosmic rays striking the atmosphere produce showers of particles. These include muons, which are charged particles related to electrons, and neutrinos. Both can leave signals in IceCube, so an event seen deep in the ice need not come from a distant cosmic source.

The outer part of the detector helps check for entering particles. If sensors near the edge light up before the main signal develops inside, that can be a sign that a particle came in from outside. Researchers call this a veto: a reason to reject the event from this particular sample.

Atmospheric neutrinos are more difficult to exclude, because they can begin an interaction inside the instrument just as other neutrinos do. Sometimes particles from the same atmospheric shower accompany them and trigger the veto. Energy, direction and calculations of atmospheric production supply further information. All of these effects have to be included when predicting the background that remains.

Sources: [1] · [3]

Why the collection mattered more than any one picture

The researchers compared the selected events with a calculation of what atmospheric backgrounds should produce. They considered the number of events as well as their energies, directions and light patterns. Taken together, the observations were difficult to explain with the atmosphere alone.

The paper reported the disagreement at about four standard deviations. In everyday terms, the sample was unusual enough under the atmospheric-only model to provide evidence that an additional contribution was present. A contribution from astrophysical neutrinos gave a plausible account of the observations.

That statistical statement applies to the sample and the model used to test it. It does not give a probability that one particular event came from a named galaxy. Some atmospheric background can remain in a sample that, as a whole, shows evidence for something new.

Sources: [1] · [2]

How to read the colored dots

An IceCube event display marks sensors that recorded light. Colors often show arrival times, while marker sizes can show the recorded charge, a measure related to how much light was collected. Always check the legend, because these are display choices.

A long pattern can come from a muon traveling through the ice. A more compact pattern can come from a shower of particles, called a cascade. The shapes help researchers work out what happened, but there is no simple one-to-one match between each shape and a neutrino flavor.

Timing helps estimate the particle’s direction and where an interaction occurred. Charge helps estimate the light yield. Both need to be interpreted through the sensor response and the way light propagates through ice. Even the energy deposited within the detector may be only part of the incoming particle’s energy.

Sources: [3] · [1]

The next question was where they came from

Evidence for an astrophysical contribution opens a new question: which objects are producing it? A search for sources looks for more events near a direction than the background would predict, sometimes also using observations at other wavelengths or of other messengers.

Searching many directions creates many chances for a random grouping, so researchers must account for how widely they searched. Time windows and the choice of candidate sources matter too. The 2013 sample began a line of investigation; it was not a list of 28 identified cosmic origins.

The reconstruction Lab lets you try one earlier step. Generate an event, fit a track from the recorded sensor times, then reveal the direction used to make the event. The small teaching array assumes a known point on the track and does not locate a real sky source. It shows why a direction estimate has to be earned before an astronomer can ask where to look next.

Sources: [1] · [4] · [3]

Try it in the Lab

Primary sources & revision

1. IceCube Collaboration · Evidence for High-Energy Extraterrestrial Neutrinos (2013)
2. IceCube · Neutrinos to the forefront of astronomy (2013)
3. IceCube Collaboration · Instrumentation and Online Systems (2017; arXiv v3)
4. IceCube · Public data releases

First published 2026-10-09; last revised 2026-10-11. 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.