≡ Menu

The SN 1987A Experiment

If neutrinos really do travel at a velocity slightly higher than the speed of light, we have a measurement that challenges Einstein, a fact that explains the intense interest in explaining the results at CERN that we discussed on Friday. I think CERN is taking exactly the right approach in dealing with the matter with caution, as in this statement from a Saturday news release:

…many searches have been made for deviations from Einstein’s theory of relativity, so far not finding any such evidence. The strong constraints arising from these observations make an interpretation of the OPERA measurement in terms of modification of Einstein’s theory unlikely, and give further strong reason to seek new independent measurements.

And this is followed up by a statement from CERN research director Sergio Bertolucci:

“When an experiment finds an apparently unbelievable result and can find no artifact of the measurement to account for it, it’s normal procedure to invite broader scrutiny, and this is exactly what the OPERA collaboration is doing, it’s good scientific practice. If this measurement is confirmed, it might change our view of physics, but we need to be sure that there are no other, more mundane, explanations. That will require independent measurements.”

All this is part of the scientific process, as data are sifted, results are published, and subsequent experiments either confirm or question the original results. I’m glad to see that the supernova SN 1987A has turned up here in comments to the original post. The supernova, which exploded in February of 1987 in the Large Magellanic Cloud, was detected by the “Kamiokande II” neutrino detector in the Kamioka mine in Japan. It was also noted by the IMB detector located in the Morton-Thiokol salt mine near Fairport, Ohio and the ‘Baksan’ telescope in the North Caucasus Mountains of Russia.

Neutrinos scarcely interact with matter, which means they escape an exploding star more quickly than photons, something the SN 1987A measurements confirmed. But SN 1987A is 170,000 light years away. If neutrinos moved slightly faster than the speed of light, they would have arrived at the Earth years — not hours — before the detected photons from the supernova. The 25 detected neutrinos were a tiny fraction of the total produced by the explosion, but their timing matched what physicists believed about their speed. The OPERA result, in other words, is contradicted by an experiment in the sky, and we have a puzzle on our hands, one made still more intriguing by Friday’s seminar at CERN, where scientists like Nobel laureate Samuel Ting (MIT) congratulated the team on what he called an ‘extremely beautiful experiment,’ one in which systematic error had been carefully checked.

Image: In February 1987, light from the brightest stellar explosion seen in modern times reached Earth — supernova SN1987A. This Hubble Space Telescope image from the sharp Advanced Camera for Surveys taken in November 2003 shows the explosion site over 16 years later. Supernova SN1987A lies in the Large Magellanic Cloud, a neighboring galaxy some 170,000 light-years away. That means that the explosive event – the core collapse and detonation of a star about 20 times as massive as the Sun – actually occurred 170,000 years before February 1987. Credit: P. Challis, R. Kirshner (CfA), and B. Sugerman (STScI), NASA.

It’s true that OPERA was working with a large sample — some 16000 neutrino interaction events — but skepticism remains the order of the day, because as this New Scientist story points out, there is potential uncertainty in the neutrinos’ departure time, there being no neutrino detector at the CERN end. As for the GPS measurements, New Scientist labels them so accurate that they could detect the drift of the Earth’s tectonic plates. Can we still tease out a systematic error from the highly detailed presentation and paper produced by the CERN researchers? They themselves are cautious, as the paper makes clear:

Despite the large significance of the measurement reported here and the stability of the analysis, the potentially great impact of the result motivates the continuation of our studies in order to investigate possible still unknown systematic effects that could explain the observed anomaly. We deliberately do not attempt any theoretical or phenomenological interpretation of the results.

A prudent policy. Let’s see what subsequent experiments can tell us about neutrinos and their speed. The paper is The OPERA Collaboration, “Measurement of the neutrino velocity with the OPERA detector in the CNGS beam,” available as a preprint.


Comments on this entry are closed.

  • David September 24, 2011, 10:16


    Same possible error source as New Scientist. He watched the whole press conference and read the paper. I cant say any better.

  • Dan Ibekwe September 24, 2011, 16:32

    To play Devil’s Advocate… If neutrinos from SN1987A had exceeded c by the same fraction as those in the OPERA experiment apparently did, they would have arrived in the late sixteenth century.

    Would they have carried significant information about the supernova, and if not, would they have violated special relativity at all?

  • Chris T September 24, 2011, 16:45

    My understanding is that the difference is on the order of 2.5E-5; at 167000 ly this would be a difference of 4.2 years. Kamakinde II was not operational until 1986, but it might be worth checking the other two detectors’ data over that time period and check for anomalies. It seems rather unlikely that anyone would have checked for such data.

    Another blog noted that if this only appears at extremely high energies, we should be able to check lower energies.

    All that said, it’s almost certainly an error, but if it’s not, it’s probably been hiding in plain sight and no one has thought to look for it before now!

  • Dan Ibekwe September 24, 2011, 17:05

    That’s what happens when you can’t tell the difference between 0.0025 an 0.0025 *per cent* (red face).

  • Bob September 24, 2011, 17:32

    These arguments are loaded with assumptions. The SN 1987A data only says that some neutrinos arrived as expected. It says nothing about other possible neutrinos, the circumstances of their creation, the various types and their changes en route. There are just way too many unanswered questions here to claim the SN 1987A proves anything or the new data proves anything yet . I just hope it all gets sorted out without the heavy hand of bias and status quo squashing new ideas for who knows how long. Sometimes I think if there are really advanced species out there they would be laughing at our slow progress.

  • Bob September 24, 2011, 17:40

    Oh, by the way, I fully expect a cultural divide between the establishment conservatives and those who they deem get “a little too excited” about this story. Perhaps I am too cynical but I sense it won’t be long before the mere mentioning of this in some circles will close doors, minds and get one labeled as a scientific non person.

  • Joy September 24, 2011, 19:47

    As to the suggestion that perhaps some of the neutrinos from SN 1987a (which arrived in a tight 13 second burst) traveled at c or very, very close to c, but maybe some other types of neutrinos arrived years earlier …

    Well, that is a novel suggestion, that some neutrinos are superluminal but others are not, and contrary to what CERN is saying, as they claim all of their neutrinos are superluminal …

    In any case IMB, the Irvine-Michigan-Brookhaven detector was operational in 1982 and the Baksan underground scintillation telescope has been operational since June 1980 and no other neutrino bursts were observed. This is significant as it covers the period in which CERN predicts that their superluminal neutrinos from SN 1987a would have arrived …

    As to the idea that the data from the neutrino observatories are soft due to low numbers of neutrinos detected, not true. The background event rate is very low. For Baksan alone “5 events during 9.1 seconds, random background probability……….. 1.7 x 10 – 5” Nothing similar seen before or since. This really was a WOW! event (sorry SETI).

    Source, great PDF lecture notes from a 2007 talk by Alexeyev:

    Add in the very statistically significant observations from the other observatories in Japan and the US, and the SN 1987a neutrino detection is a dead certainty. And this is being challenged by CERN, when they don’t even have a neutrino detector at their emitter to accurately time their actual neutrino emissions? Absurd.

    In any case, the Milky Way is well overdue for a SN event. If we are lucky enough to get one even within 50,000 LY, the neutrino flux will be at least 10x the 1987 LMC event. That will yield another, even more robust, observational result. Hope to see it in my lifetime.

  • Rob Henry September 24, 2011, 21:03

    The more I read the more confused I get. I know that it seems well established that measuring neutrinos over a few hundred kilometers is a very different than measuring their travel over greater distances, since over greater distances they oscillate between forms. I also note that not long ago, whenever one neutrino of the family was postulated to be a tachyon, it was always the assumed lightest, the electron neutrino, but even if it were, all other members of the family were assumed tardyons. I imagine that the answer is that one particle can’t oscillate between tachyon and tardyon forms, but I would like a physicist to confirm this.

  • Joy September 24, 2011, 23:31

    PS to Chris T – in 1987 the alarms went off at all three of the neutrino observatories straightaway. It wasn’t a case of visual astronomers seeing a SN and then waking up the neutrino observers to look at their data, not at all. The neutrino observers knew about the SN before the visual astronomers did. The neutrino observers were ALWAYS looking for a SN collapse signal, in the USSR from June 1980 onwards, in the USA from 1982 – that is like the holy grail for these very large and expensive facilities. No equivalent events occurred in 1982 or 1983 or 1984 or whatever. A massive astronomical non-confirmation of CERN’s hypothesis.

    What is so wrong with high energy physics theorists and their facilities that for the last two generations astronomers (astrophysicists) have had to make all of the interesting observations suggesting new physics? Inflation? Dark energy? Dark matter? We still don’t know why we observe evidence for such things in the sky, and maybe none of the current theories about them are correct, but at least the astronomers are pushing the boundaries of knowledge. The high energy physicists? Epic failure, since their glory days in the 40s 50s and 60s.

  • An Infinitude of Tortoises September 24, 2011, 23:48

    A couple naive points:

    1. Weren’t what was being measured from SN 1987A electron neutrinos, and wasn’t what CERN used muon neutrinos? Apples and oranges, perhaps?

    2. I agree with Chris T: if superluminal neutrinos from SN 1987A would have arrived four years earlier, data from four years earlier should be checked. Likewise with Bob: what’s to say that one set of neutrinos didn’t arrive four years earlier in addition to the observed set that arrived right on the mark?

  • John Freeman September 25, 2011, 8:36

    The neutrinos produced at CERN are made using a different method than would occur in a supernova, at significantly higher energy levels IIRC. For all we know some see-saw mechanism is in place that permits neutrinos to go all tachyonic only above a given enrgy. Who knows, I suggest we take a leaf out of the CERN teams book and don’t try to come up with in depth explanations of something that is not yet confirmed?

  • Tom Baty September 25, 2011, 14:33

    Everything that mankind has done including all of our technology has already been done, in one form or another in nature. If it is possible to go faster then the speed of light then it has already been done in the natural world. All we have to do is find out how nature has done it and build our technology from there. Sorry to get philosphical but that boils it down to one point….. if it is allowed in the universe then we can do it too.


  • Chris T September 25, 2011, 14:43

    Joy – I was not aware until yesterday that OPERA (not technically apart of CERN) had indeed explored the energy dependency possibility and come up negative.

    And this is being challenged by CERN,

    CERN didn’t make the announcement, OPERA did. They also are not challenging astrophysics, so much as asking for help with an anomaly they have so far been unable to resolve. They’ve outright stated they believe it’s almost certainly an error, but after three years and 16,000 events they’ve exhausted everything they could think of an are now asking the community for help.

    Other facilities are now planning similar experiments to attempt to reproduce the anomaly or find the error. In the meantime, it bares asking if the anomaly has shown up in prior data, but was missed because no one was looking for it.

    OPERA includes a large number of respected particle physicists and technical staff who know full well what the potential import of this announcement and are not making it lightly. It’s highly unlikely to be an obvious error (if it is) and will likely be highly informative all by itself when it is resolved. An anomaly such as this one is rare and deserves explanation.

  • lunej le September 25, 2011, 14:45

    I still believe that opera made an error, counting a cable delay twice in a subtle way or something … in any case – exciting times for physics!

  • andy September 25, 2011, 18:46

    As I understand it, under special relativity if you decrease the energy of a tachyonic particle it ends up travelling faster. On the other hand, the detected SN 1987A neutrinos were electron neutrinos, rather than the muon neutrinos used in the OPERA experiment.

    Then again, there have been several results from tritium decay experiments which have given a negative best fit value for mass^2 of the neutrino. This is exactly what you’d get for a tachyonic particle under special relativity. On the other hand, the error bounds on these results did allow for a positive value of the mass^2.

    Personally I don’t yet believe in the superluminal velocities for various reasons. My suspicion lies with the measurement of the time of the creation of the neutrinos. While I’d love to be proven wrong, so far I’m not convinced.

  • David September 25, 2011, 21:43

    I dont know if anyone pointed out that Fermilab did produce somethingsimilar but they had greater error bars and sublight was possible so they just assumed the neutrinos were going under the speed of light. In light of this …….however.

  • Joy September 26, 2011, 4:02

    Tom Baty September 25, 2011 at 14:33
    “Everything that mankind has done including all of our technology has already been done, in one form or another in nature. If it is possible to go faster then the speed of light then it has already been done in the natural world. ”

    +infinity! Nothing more needs to be said.

  • Martin J Sallberg September 26, 2011, 9:17

    Interesting that mainstream community does an investigation. Prestige-obsession was not as severe as I thought after all. The supernova examples seem to prove that not all neutrinos travel superluminal under all circumstances, but they do not rule out that some neutrinos may travel superluminal under some circumstances. A new idea: maybe some neutrinos are small enough to shortcut through hidden dimensions, but they dissipate from normal space with distance and are therefore indetectible at interstellar distances.

  • Ron S September 26, 2011, 9:56

    andy: “My suspicion lies with the measurement of the time of the creation of the neutrinos.”

    That item stands out since there is a large inherent uncertainty as to where in the tube the proton interaction (and subsequent particle chain) occurs. Since they can’t say where this happens they rely on a statistical analysis using the pdf of where the collision is likely to occur.

    I read the paper but I have none of technical knowledge needed to properly understand this particular item. However I would be surprised if the pdf is incorrect to that degree since it is so fundamental to the operation of the collider.

  • FrankH September 26, 2011, 15:27

    I think Tom Baty’s observation is correct. Also, the XKCD cartoon on the issue pretty much sums it up for me: http://xkcd.com/955/

  • Astronist September 26, 2011, 19:18

    Paul, you start by saying: “If neutrinos really do travel at a velocity slightly higher than the speed of light, we have a measurement that challenges Einstein”. Am not sure I quite understand this. Special relativity has no problem with spacelike objects which would appear to us as particles travelling faster than light. A contradiction of relativity would only arise if a particle accelerated up or down *through* the light barrier, as this would require a greater than infinite amount of acceleration.

  • AcesHigh September 27, 2011, 1:25

    Its interesting that people are debating something that is pretty much certain: that nothing can travel faster than light

    instead of assuming the most obvious: neutrinos DO NOT travel faster than light (as the supernova detection shows)… and yet, they DID arrive faster than light at the CERN experiment.

    Thus, if BOTH experiments show true results, what does that mean???

    well, some sort of small “hyperspace jump” or maybe some space-time meddling (we know that space CAN expand faster than light… it did after the Big Bang!)

  • Joy September 27, 2011, 5:22

    re: XKCD, LOL and yes, I would be happy to bet anyone (rapidly depreciating euros, usd. quatloos or whatever) that the superluminal neutrino story will follow polywater into the dustbin of history. The only difficulty is defining a time frame, and what would be construed as confirmation or refutation. (After all some people still cling to belief in cold fusion, although in over 20 years no one has been able to successfully and reproducibly replicate the claims of Pons & Fleischmann. Where is my Mr. Fusion!)

  • ljk September 27, 2011, 9:15

    To quote from this article:

    “…there is potential uncertainty in the neutrinos’ departure time, there being no neutrino detector at the CERN end.”

    So how were they detected then?

    A take on this story from io9.com:


    My question is, if it is a system error and these guys could not find it after two years, what can anyone else do? These scientists don’t sound like hoaxers or fringers, and I am sure they know if they tried to pull a fast one, their professional careers and reputations would be over faster than those neutrinos they detected.

    Perhaps neutrinos do “change” at various velocities and in different conditions. That is what they discovered about the ones coming from the Sun, as I recall. Or maybe they can pop in and out of different dimensions as others have said. And I just want to throw in again that if an ETI wanted to get our attention, or at least the attention of a species able to detect such things, maybe they would mess with the known laws of physics.

  • ljk September 27, 2011, 9:27

    Can Neutrinos Kill Their Own Grandfathers?

    by Sean from Cosmic Variance

    Building in part on my talk at the time conference, Scott Aaronson has a blog post about entropy and complexity that you should go read right now. It’s similar to one I’ve been contemplating myself, but more clever and original.

    Back yet? Scott did foolishly at the end of the post mention the faster-than-light neutrino business. Which of course led to questions, in response to one of which he commented thusly:

    Closed timelike curves seem to me to be a different order of strangeness from anything thus far discovered in physics—like maybe 1000 times stranger than relativity, QM, virtual particles, and black holes put together. And I don’t understand how one could have tachyonic neutrinos without getting CTCs as well—would anyone who accepts that possibility be kind enough to explain it to me?

    The problem Scott is alluding to is that, in relativity, it’s the speed-of-light barrier that prevents particles (or anything) from zipping around and meeting themselves in the past — a closed loop in spacetime. On a diagram in which time stretches vertically and space horizontally, the possible paths of light from any event define light cones, and physical particles have to stay inside these light cones. “Spacelike” trajectories that leave the light cones simply aren’t allowed in the conventional way of doing things.

    Full story here:


  • ljk September 27, 2011, 9:52
  • J. R. September 27, 2011, 13:34

    Just a wild hypothesis, assuming that the neutrinos really did cover the distance faster than c. The experiment resulted in the neutrinos initially accelerating away. Doesn’t one of Eistein’s theories predict that an accelerating mass produces gravity waves? While the neutrinos are accelerating, would these gravity waves form something around the mass that resembles the theoretical warp bubble? It would be for a brief period and only while the neutrinos were accelerating.

  • Rob Henry September 27, 2011, 17:51

    It’s strange that no one wants to answer even one of my many bewildered questions, but the following two seem to me to have application to more informed comments that have already been made.

    Firstly, to a simpleton like me, the fact that the detected neutrino burst from SN1987a (whether superluminal or subluminal) must have been at least ten thousand times closer to the speed of light than the detected burst would normally imply that one received four years earlier would be about ten thousand times more spread out. In that regard I can’t make head nor tail of Joy’s claim that such a burst would have already been detected without aid of special search.

    Secondly, can the highly statically significant time difference between the already detected SN1987a bursts at different sites be seen in new light now?

    Please, I need help.

  • Michael Simmons September 27, 2011, 19:21

    My initial thoughts…
    If you fired a beam of neutrinos and a beam of light though a block of glass you would expect that the neutrinos would less affected by the glass.
    i.e. the neutrinos would have a higher velocity than the light while passing though the glass.
    My understanding is that Relativity is based on the speed of light in a perfect classical vacuum not in a quantum vacuum (one filled with self annihilating virtual particles).
    Could it be that the properties of a quantum vacuum (e.g. refractive index) are slightly different in in a gravitation field and that this difference affects light more than neutrinos?

    I assuming that the OPERA people are allowing for the material light is having to travel though in their calculations.

  • Rob Henry October 1, 2011, 19:01

    Perhaps I should make explicit my thinking on the off-chance that I can later claim credit.

    It seems obvious that if neutrinos traveled faster than light in a way that depended on the medium, then its path through the interstellar medium is not going to be as straight as SN1987a’s light. These two effects might cancel out to give approximately similar travel times.

    It also seems obvious that if that is the case we would see these neutrinos through a distorted lens – as if viewing a vista through an uneven pane of glass. If so the highly statistically significant different timing of these detected bursts in different neutrino observatories would be the expected outcome, wouldn’t it?

    Given these two obvious points, it is natural for an outsider like myself to think that these have already been considered and dismissed for some very good reasons by physicists (for example for light a refraction index less than one does not mean that either information or energy have traveled faster than light, but neutrinos might really be tachyons). I have found in the past that assuming that the bloody obvious has already been investigated and discarded is a huge error.

  • Laurin October 3, 2011, 5:39

    As an outsider who has an interest in future development of mankind….I think that both left and right brain should be used when exploring new ground on any future frontier …i.e. reason/logic and intuition. But I don’t think we are at that stage of even considering it “reality based”. Hence the slow progress.

  • Kelly October 4, 2011, 17:41

    The problem with this blog as I see it is too many scientists are talking about the Opera and not spending enough time with their nose in a book!

  • Ron S October 5, 2011, 13:19

    Which book would that be?

  • Bob G February 16, 2012, 19:27

    What if the apparent distance between the detector and the source was less than measured. The big difference between the astronomical observation and the terrestrial is that the latter particles are travelling through a great deal of matter. If some of the matter has an occupancy rate then maybe it “Newtons Cradles” its way through it. Giving an apparently higher speed.

  • ljk February 25, 2012, 17:39

    Twenty-five years after supernova 1987A



    Posted: 23 February 2012

    While primitive humans of the Middle Paleolithic hunted prey and sheltered in caves in Africa, a distant star eighteen times more massive than the Sun, located faraway in the Large Magellanic Cloud (LMC) endured a catastrophic collapse as it reached the end of its life. As the star caved in, its outer layers rebounded off its dense core and blasted outwards, ripping the star apart in a supernova.

    Some 160,000 years later the light of this supernova, travelling at 300 million metres per second, finally reached Earth to shine in Southern Hemisphere skies on 24 February 1987.

    Full article here: