After 3.9 billion years of hurtling unhindered through the vast reaches of the universe, a ghostly neutrino particle died on 22 September 2017. It was annihilated when it collided with an atom in the frozen darkness 2 kilometers beneath the surface of the south polar ice cap.
But this subatomic particle’s death did not go unmarked. It was announced today in Science that its moment of passing—labelled neutrino event 170922A—triggered a worldwide cascade of astronomical observations using a raft of varied technologies. And these led to the first ever identification of the birthplace of a neutrino from outside our galaxy: in this case, the unimaginably violent cosmic forge of a blazar.
Blazars are incredibly bright natural sources of radio waves. They form when some of the swirling material falling into a supermassive black hole is converted into a hot radiating soup of elementary particles and then gets blasted back out into space in the form of twin jets moving at close to the speed of light. Tracing the 170922A neutrino back to a blazar known as TXS 0506+056, located billions of light years away in the Orion constellation, required the rapid coordinated response of a network of observatories around the world and in orbit above it.
The initial observation that kicked this so-called multimessenger observation campaign off was made by the IceCube detector at the South Pole. IceCube was created by using pressurized hot water to melt 86 shafts into the polar ice over a square kilometer. Before the shafts refroze, cables strung with 60 digital optical modules apiece were lowered down so that the modules sit evenly spaced every 17 meters between 1,450 and 2,450 meters deep. The result is a detector that encompasses a cubic kilometer of solid ice. The optical modules are sealed into basketball-size spheres of borosilicate glass to withstand the crushing pressure and are designed to spot the signature flashes of light that occur when a neutrino smashes into an atom in the clear ice.
Illustration: Emily Cooper
When a neutrino collides with any atom, sometimes a muon is produced—a particle that’s essentially a heavier version of an electron. When this happens in ice, the muon travels faster than the local speed of light. (Nothing can travel faster than light in a vacuum, but in ice the speed of light is about 24 percent slower, so a fast particle can outpace it.) When something goes faster than local light speed, a shockwave of photons forms, much like the way that a sonic boom is produced when a plane breaks through the sound barrier. This shockwave creates an eerie blue light known as Cherenkov radiation, and measuring the direction and intensity of the light reveals how much energy the original neutrino possessed, and where in the sky it came from.
Neutrino collisions are exceedingly rare—trillions of neutrinos from the sun stream through your body every second without so much as wobbling an electron—so IceCube has to be big for it to have a statistical chance of catching a collision before the researchers die of old age (IceCube is even bigger than it seems: As long as the resulting muon passes through the detector array, it can spot neutrino collisions that occur in the surrounding ice cap up to about 10 kilometers away). And IceCube has to be deep beneath the surface because only there is the pressure intense enough—700 times normal atmospheric pressure—to squeeze all the air bubbles out of the ice. The bubbles have to be eliminated because they would otherwise scatter the Cherenkov radiation that the detectors are looking for. For a comprehensive description of IceCube’s design, right down to the FPGAs used in the optical modules, you can read the comprehensive account that team member Spencer Klein wrote for IEEE Spectrum in 2011. That article is still au courant: Since it was published “there haven’t been any major changes,” says Klein, “The optical modules are buried under a mile and a half of ice, so there’s no way to really access the hardware. There have been some very minor firmware updates.”
The biggest change since, says Klein, has been in the creation of an automated alert system. This system broadcast an alert to astronomers working around the world just 43 seconds after the 170922A event occurred. IceCube detects muon shockwaves all the time, but these are generally due to low-energy muons that are produced in the Earth’s atmosphere by cosmic rays. These background muons have a different shockwave signature than those produced by high-energy extragalactic neutrinos, but filtering them out automatically requires a detailed model of the optical properties of the specific ice sheet in which IceCube is embedded.
In particular, the ice is slightly contaminated by dust. The dust comes from two sources, accumulated over tens of thousands of years as the sheet slowly formed from surface snowfalls. One is “volcanic eruptions which produce very, very thin, but relatively dense, layers” that run throughout the ice, says Klein. The other source is regular atmospheric dust which isn’t as dense, but occurs throughout the ice: “If there’s dust in the atmosphere, some of that dust will get dragged down with the snow. That does change with time somewhat, so we have a picture of the dust content in Antarctica over the last 70,000 years.”
With experience gained from several years of operation, the IceCube team has considerably improved their understanding of their patch of ice and how the detector behaves in response. Consequently, they spun up the automated system in April 2016. “The detectors and the computer systems at the South Pole look for interesting events and can automatically send out an alert when it sees that something interesting has happened. It takes a fair amount of confidence to get to that point,” says Klein. “It used to be that [a candidate event for an alert] would go to a human being, who would look at it and then send out the alert. That takes time. This alert went out in under a minute.”
The 170922A event alert, with its estimated coordinates of the neutrino’s origin in the sky, went out to astronomers running instruments which can detect gamma rays, such as those onboard the orbiting Swift Observatory. Swift quickly spotted that the 170922A event coordinates matched with those of known blazar TXS 0506+056, and that the blazar was flaring in brightness. “Thanks to the automated trigger…Swift was observing within four hours of the neutrino detection,” said Jamie Kennea, science operation team lead for Swift in a press release.
As the next 14 days rolled on, more and more instruments were brought to bear on TXS 0506+056, allowing it to be monitored across a range of wavelengths from radio, through optical, all the way to X-ray. The 170922A event coincided with a period of heightened activity of TXS 0506+056, and researchers have concluded that it’s 99.7 percent likely that the detected neutrino originated in the flaring blazar. “The fact that we could tie gamma rays and neutrinos together tells us very exciting things about the particle jet,” said Regina Caputo, analysis coordinator for the satellite-based Fermi-LAT gamma ray telescope, at an NSF press conference today.
With the evidence pointing to TXS 0506+056 in hand, the IceCube researchers also checked through their complete records and found that there were between 8 and 18 previous neutrino events that hadn’t met the threshold required for sending an alert, but which likely were also produced by neutrinos streaming from the blazar.
“It’s a pretty amazing finding,” says Klein. “but more data is needed. The multimessenger campaign was based on one neutrino. It’s great to know about one [extragalactic source] but we have a ways to go before we have a systemic understanding.” The IceCube team hopes to get more data by building a next-generation array by spreading out a similar number of digital optical modules over a larger area, and adding seven more closely spaced strings to the original IceCube array. The seven strings in particular “would allow us to do a much better job of understanding the optical properties of the ice,” says Klein, which would let them pinpoint the sources of neutrinos with even greater accuracy.
Stephen Cass is the special projects editor at IEEE Spectrum. He currently helms Spectrum's Hands On column, and is also responsible for interactive projects such as the Top Programming Languages app. He has a bachelor's degree in experimental physics from Trinity College Dublin.