22

Thanks to the efforts of the aLIGO team, gravitational wave astronomy is a reality. At the same time, neutrino detectors like Hyperkamiokande are becoming much more sensitive.

My question is: what are the prospects for the pseudo-simultaneous detection of gravitational waves and neutrinos from the same supernovae? What sort of stuff could we learn from such an event, both about supernovae and neutrinos? In particular, what are the prospects for estimating the neutrino mass?

uhoh
  • 31,151
  • 9
  • 89
  • 293
ProfRob
  • 151,483
  • 9
  • 359
  • 566
  • Since neutrinos travel at $ V < c$ but gravitational waves travel at $V = c$, is your question about measuring the time delay of one relative to the other? – Carl Witthoft Oct 03 '16 at 12:34
  • @CarlWitthoft Yes, I suppose so. I guess the SN would have to be far enough away and theory sufficiently precise to predict the non-travel time delay. – ProfRob Oct 03 '16 at 12:40
  • This might be a cop out, but currently, the prospect of such a measurement is pretty much nill. aLIGO was just barely able to detect the gravitational waves from ~30 solar-mass black holes colliding. I don't know this for a fact, but likely a supernova signal is going to be far below the limits of detection with current detectors. Adding more detectors to aLIGO in the next few years will allow pinpointing sources more accurately, but won't improve detection. And eLISA, the only other detector in the works that I know of, will have worse sensitivity than aLIGO does now. – zephyr Oct 03 '16 at 12:49
  • 3
    There might be some interesting observations possible, such as (a): does the shape of the gravitational wave pulse tell us anything about the "kick" in asymmetric core collapse, and (b): the pulse presumably doesn't interact with anything as it leaves the core, while some of the neutrinos do, so there may be some interesting properties of the star's structure that can be measured this way. (Both ideas based on pop science treatments so treat with caution. And of course I am assuming measurements will be possible at high enough sensitivity.) – Andy Oct 03 '16 at 13:08
  • 3
    @Andy Point (a) is especially true. We'd never measure gravitational waves from a purely spherical explosion given that you need a quadrupole moment to produce the waves. As such, any wave detection that did occur would necessarily indicate the supernova was asymmetrical to some degree. With sufficient modeling, one could possibly work out just how the explosion must have happened to produce the wave observed. – zephyr Oct 03 '16 at 13:17
  • 2
    @zephyr AFAIK GWs are expected from supernovae as the explosion is expected to be asymmetric. In terms of detection sensitivity, the merging black holes were >1 billion light years away. I was thinking more in terms of a supernova in M31, which might be expected in the next ~20-30 years. But if you write an answer which shows my optimism is misplaced, I guess I'd upvote it. – ProfRob Oct 03 '16 at 14:03
  • Oh yes, I'm not saying GW are not expected. There's no chance supernova will be spherically symmetric. Just pointing out that seeing gravitational waves necessarily implies an asymmetric core collapse. That's a good point concerning the distances though. I'd have to do some digging to see if the proximity makes for SN GW being detectable. – zephyr Oct 03 '16 at 14:07
  • @zephyr In fact I fancy that the neutrino detection bit is the thing that is sensitivity limited... – ProfRob Oct 03 '16 at 15:21
  • @RobJeffries But we've been detecting SN neutrinos for nearly 30 years now. LIGO's been around for about 20 years now and only just recently improved enough to detect (very strong) gravitational waves. – zephyr Oct 03 '16 at 15:26
  • @zephyr Superkamiokande detected one supernova in the Large Magellanic Cloud with only ~20 neutrinos. – ProfRob Oct 03 '16 at 15:29
  • 2
    @RobJeffries Actually only 24 neutrinos were detected from 3 neutrino observatories around the world combined, Kamiokande 2 only detected 11, but your right, SN 1987A is the only recorded supernova to have observed neutrinos associated with it. – Dean Oct 04 '16 at 11:19
  • 4
    The cause for pessimism of GW detection of supernovae is that if the supernova is 1000 times closer than the black hole mergers, the GW amplitude is up by 1000, which sounds pretty good, but there's an efficiency problem. In the case of BH merger, GW generation is an important energy pathway, it allows the orbits to decay. When it was thought there might be a gamma ray detection with the BH merger, models were created that could put some small energy into light, but even so very little energy goes into anything other than GWs. Not so for supernova-- they put a lot of energy in neutrinos. – Ken G Oct 04 '16 at 14:51
  • 2
    So to detect the GW from something that is happening due to all the energy going into neutrinos, you need an efficiency for GW generation that is something like 1/1000 to make it equally detectable as BH mergers 1000 times farther away. These are rough numbers of course, but to get that high efficiency of GW generation in something that is very good at making neutrinos, I think it would require an extreme asymmetry. In other words, I think BH mergers make GWs because they don't have anywhere else to put the energy, but supernovae are very good at putting energy elsewhere. – Ken G Oct 04 '16 at 14:53
  • Still, 1/1000 kinds of efficiency in GW generation might be possible if there are various asymmetries going on, so it ultimately becomes an observer question. An effort to answer it is described in this talk: https://www.physics.ncsu.edu/FOE2015/PRESENTATIONS/FOE15_Szczepanczyk.pdf So far, only null results, but who knows-- might be quite a headline in the works there. – Ken G Oct 04 '16 at 15:15
  • @Ken G All good points. My understanding was that with aLIGO we might be limited to the local group. I was hoping not to have to write my own answer... – ProfRob Oct 04 '16 at 15:19
  • 1
    I think the answer is that the prospects are unknown, because it depends on what is happening inside the supernovae that we don't know yet. So you have optimists actually doing the measurements, and pessimists saying they are wasting resources, and we won't know who is right until they see something, or don't. But it sounds like your interest is in whether or not we can learn more about neutrinos that way than we can in our accelerators. You get a long distance to work with, but poor controls on the emission times. Another optimist/pessimist divide I presume! – Ken G Oct 04 '16 at 15:43
  • We shouldn't even rule out a supernova in our own galaxy, giving another factor of 1000 in GW amplitude to work with. That will take more patience, but who knows. So another way to frame your question is, with our current detectors, what is the ideal distance for a supernova, in the tradeoff between sensitivity and timing discrimination? You don't win that many stars going outside our galaxy if you will stay in the local group! – Ken G Oct 04 '16 at 15:46
  • @KenG It would be an acceptable to provide an answer which gave both the optimist and pessimist point of view. – called2voyage Oct 04 '16 at 17:45
  • Basically, we only see neutrinos from supernovae in the local group. That only gives us 1-2 trillion stars to work with, or a supernova every few decades. So the optimist says, be ready to get lucky. The pessimist never does anything based on getting lucky! A separate issue is if we will see the GWs from any supernovae. The pessimist says we haven't heard anything in a year or so that aLIGO has been operational, the optimist says with more LIGO instruments we can triangulate and beat down the noise. That one depends on how asymmetric the explosion is, which we won't know until we see it. – Ken G Oct 05 '16 at 00:58

1 Answers1

6

This article basically seems to answer the question. They quote from an earlier study:

"Although no CCSNe have currently been detected by gravitational-wave detectors, previous studies indicate that an advanced detector network may be sensitive to these sources out to the Large Magellanic Cloud (LMC). A CCSN would be an ideal multi-messenger source for aLIGO and AdV, as neutrino and electromagnetic counterparts to the signal would be expected. The gravitational waves are emitted from deep inside the core of CCSNe, which may allow astrophysical parameters, such as the equation of state (EOS), to be measured from the reconstruction of the gravitational-wave signal."

Since we know from SN1987A that neutrinos from a supernova can be detected at that range, that seems to be a "yes". The biggest uncertainty seems to be how much gravitational wave energy would be emitted by the supernova, and at what frequencies, which depends on a relatively detailed understanding of exactly how the matter moves around in the explosion, one simulation of which is illustrated in the (rather awesome) video in the article.

Steve Linton
  • 10,273
  • 1
  • 28
  • 49