Now that we know how to find high energy neutrinos, we're seeing lots.
The building that houses the IceCube servers.
However, very distant and highly energetic events should produce neutrinos with a correspondingly high energy. If we can detect them, maybe they can tell us about those high-energy events. This idea is more than 30 years old—until now, the technology has simply not been up to the task.
IceCube consists of some 8000 photomultiplier tubes (light detectors), strung out on strings, buried under the ice of the Antarctic. Each photomultiplier tube contains its own data processing computer that provides some preliminary filtering and enables event signals to be synchronized to within 2ns.
These signals are then sent to a local computing center that sits at the center of the array (yes, in Antarctica), which does more processing before sending it out to the world.
Both the instrument and neutrino physics are exquisitely well understood, so the scientists working with the detector have just a single percent uncertainty in their models and another ten percent uncertainty in the instrumentation. Considering everything, that is a fine piece of hardware. However, even buried deeply in the ice, IceCube has a devil of a time finding the neutrinos it is looking for. The instrument records 2700 cosmic rays per second, and a locally produced neutrino turns up every six minutes. The cosmic signal, in contrast is ten neutrinos... per year.
After a fairly long run and a particular type of analysis, the collaboration running IceCube was rather confused; they hadn't found any neutrinos of interest yet. So they took a look inside their processing and found a particular type of event that was being incorrectly filtered out. The problem was an assumption about the energy range they were after.
At high energies (but not too high), a neutrino will tend to collide with an atom outside the volume occupied by the detectors. The resulting high energy muon streaks off like a meteor through the ice, losing energy through radiation, production of electrons and positrons, and other things. IceCube looks for these tracks and figures out where the neutrino came from. (It should be noted that some of these neutrinos have traveled through the entire Earth before being detected.)
Once you go to even higher energies, however, neutrinos that are detected are a result of collisions from within the detector. Again the muon motors off, causing havoc, but the event track began from within the detector volume—a pattern that the researchers initially excluded. They are now actually spotting neutrinos in the 1000TeV range (The LHC operates around 14TeV).
At these energies, the particles have quite a high probability of interacting with atoms, so they don't make it through the Earth; instead, they're detected quite close to the surface of the ice. Even better, because the tracks begin within the detector volume, the energy of the neutrinos can be calculated with high accuracy
After recognizing the problem, a reanalysis produced around 30 neutrinos, and many more are expected to be reported in May. The big question is whether these neutrinos are evenly distributed across the sky (as low energy neutrinos are), or if they have specific sources. At present, there is not enough data to say. As with any small data set, it has a few blobs that look like they may be specific sources, but we should expect those to disappear as more neutrinos are detected.
