People are excited about the announcement from CERN scientists on the OPERA collaboration eliminating one source of potential error in their finding that neutrinos can travel faster than light.
The widespread interest in this experimental result is justified, because if true, it would mean a revolution in our understanding of fundamental physics, space, and time.
However, the CERN collaboration measurement is in conflict with another prominent constraint on neutrino speeds that comes from looking at the sky. In this diary I will briefly describe that result, and why it means, most likely, that they both can't be right.
Very briefly, the CERN experiment consists of making neutrinos via collisions in the particle accelerator at CERN, and sending them through the Earth 450 miles to an underground detector in a complex called Gran Sasso, in the Italian Alps. The time it takes the neutrinos to traverse the distance is measured, and the measurements show that the neutrinos can make the trip in some 60 nanoseconds (60 billionths of a second) less than it would take light, indicating that these neutrinos can travel faster than light.
Einstein's relativity, which has been a fundamental pillar of our understanding of physics and the nature of space and time, states that nothing can travel faster than light. Given that 60 nanoseconds is a tiny difference, requiring extremely precise determinations of time intervals and lengths - and that faster-than-light neutrinos would require a revolution in physics - people are rightly searching for a small source of experimental error that could explain the CERN team's result.
----------------------------
However, there is another experimental constraint on neutrino speeds, which comes to us from the sky.
The picture above shows an optical image of the Large Magellenic Cloud, which is a satellite galaxy of our Milky way Galaxy, a kind of mini-galaxy containing millions of stars. In 1987, the LMC hosted a supernova - the massive explosion of a star at the end of its life - which was the nearest supernova to Earth since the invention of the telescope 400 years ago.
The picture above shows a closer view of the site of the supernova explosion that occurred in the LMC in 1987, known as "SN 1987A". Superimposed on the background is an image from the Hubble Space Telescope from several years later showing the expanding shell of material from the original site of the explosion.
Supernovae explosions involve the rapid expansion outward of much of the mass of the star, as well as enormous quantities of light. The brightest supernovae can even briefly outshine the other billions of other stars in their galaxy.
Supernovae are the source of every atom in the Universe heavier than iron, and those expanding shells are the source of most of the Galactic cosmic rays that lead to the radio and gamma-ray light we see from our own and nearby galaxies.
--------------------------
So what does this have to do with how fast neutrinos go?
Supernovae put out an enormous quantity of neutrinos. In fact more than 90% of the mass of the star actually comes out as neutrinos. If a supernova happens close enough to Earth, we would expect to see some of the neutrinos from it in detectors here on Earth. Neutrinos being very small and neutral, so they are hard to detect, but we would still expect to see a few. That is exactly what happened with SN 1987A, which remains the only distinct celestial object from which we have confirmed neutrinos, other than the Sun.
In the case of SN 1987A, the light was first seen on February 24, 1987. If neutrinos travelled at the speeds necessitated by the CERN result, then, given the large distance to the LMC, we would have seen the neutrinos from SN 1987A four years earlier.
But that's not what happened. The neutrinos arrived three hours earlier. Does this still mean that neutrinos travel a little faster than light? No, because the neutrinos from the explosion are made immediately, while the light we get is actually emitted sometime slightly later by the expanding shell of material.
So unless our understanding of supernova kinematics is way off, the neutrinos from SN 1987A could not have travelled at the speed implied by the CERN result.
----------------------
These two measurements - the CERN to Gan Sasso beam involving a short distance on the Earth and the other, neutrinos from SN1987 A involving an enormous distance in space - are in conflict, and it is unlikely that both can be right. SN1987 A argues for slower-than-light neutrinos, while the recent CERN result argues for faster-then-light neutrinos.
While it is possible to come up with a convoluted model in which neutrino speed is highly energy dependent, most people agree that these two results can't both be correct. Given my irrational bias toward astrophysics, I personally come down on the side that the SN 1987A observations are right and neutrinos are slower-than-light. But that's just my hunch and further observations are needed to settle this.