So then, where _do_ they come from?

Cosmic rays are a puzzle. Protons and iron nuclei hit the upper atmosphere and splatter in showers that typically just leave muons by the time they reach the ground. But where do they come from? And why does the spectrum have such a long tail--out into 10^17 eV , or a million ergs--and more? What (not) on earth can generate such high energies?

To accelerate and shape these kinds of things needs huge magnetic and electric fields, and most stellar activity isn't that dramatic. Gamma ray bursts are. These are so far (luckily) extra-galactic explosions that generate more energy in a few seconds than the sun will through its lifetime, sending so many gamma rays that some of them survive the trip to Earth through all the intervening gas and dust. So if you see two similar oddities, maybe they're related. Maybe gamma ray bursts (GRBs) generate the ultra-high energy cosmic rays too.

It isn't easy to pinpoint the direction of a gamma ray. It interacts with a bang too, but with layers of detectors you can mostly make out the direction from the shape of the shower. Gamma rays have no charge, and so aren't bothered by the twisty galactic magnetic fields that bend protons around in random directions.

You can't tell where a cosmic ray originally came from. It might have gone in some great arc and swept around to hit the Earth from the opposite direction. Galactic magnetic fields aren't very strong, but they are huge and a little bend for a long time makes a difference.

So if we can't use the direction of the cosmic rays to verify that GRBs create them, what can we do?

It turns out that cosmic rays can splatter against photons, especially when there are so many of them coming out of a GRB. One of the by-products is pions. Pions decay to muons, and muons to electrons which aren't going to survive the journey to Earth. But along with those decays come neutrinos. They'll be headed in almost exactly the same direction as the original proton (for high enough energy protons--which is what we're interesting in figuring out). So they should point back to their origin, as ordinary cosmic rays do not.

Cue the neutrino detector IceCube. The short answer is no, we don't see nearly enough neutrinos coming from GRBs. (No, my name is not on that paper.)

So, now what? Either cosmic rays don't come from GRBs after all (and they were such a good candidate), or something is badly mistaken about the interaction cross section of neutrinos at ultra-high energy. Low energy neutrinos have a famously small interaction cross section, but unless there's something unexpected interfering that cross section should rise significantly at these ultra-high energies. Or something else intervenes. I'm puzzling about that one.