Predicting supernovae

[Tom Mazanec]

Will we ever be able to foretell the day a supernova candidate will go "BOOM!" millennia in advance? I would like to know that Eta Carinae will light up on, for example, March 5, 57,412 AD.

[Nereid]

Will we ever be able to foretell the day a supernova candidate will go "BOOM!" millennia in advance? I would like to know that Eta Carinae will light up on, for example, March 5, 57,412 AD.
Well, for sufficiently massive core collapse supernovae, the answer is likely to be "yes", with caveats.

If real-time high energy astrophysics neutrino 'telescopes' become sufficiently sensitive, they may well detect the KABOOM! of such a supernova, in its core, before the shock reaches the star's surface. How much of a lead time would that be? Certainly hours, and possibly as much as days, depending on which part of the electromagnetic spectrum we look for the supernova in, how much extinction (a.k.a. dust) there is between us and the doomed star, and how bright it needs to be before it is detected using 'light'.

In the case of supernovae that result from inspirals (i.e. two stars, one or both quite dense - white dwarf, neutron star, black hole), gravitational wave radiation 'telescopes' could give similar advanced notice. In fact, we already know - with some accuracy - when various binary pulsars will coalesce (by estimating the rate of decay of the orbits); the closer to merger, the greater the accuracy in estimating the date/time. Whether such a merger results in a supernova or not, well, that's a different story.

But I somehow don't think you had these sorts of scenario in mind ...

[Grey]

My guess is probably not. Certainly we're nowhere close to being able to make a prediction like that at this point, and we'd have to have both much more accurate models and much better observations. Betelgeuse is likely to go any day now, but in stellar evolution terms, "any day now" could mean tomorrow, or it could mean a million years from now.

[Romanus]

^
Ditto; for core-collapse SN, the mass and age unknowns are too great. For Ia, the Chandrasekhar Limit gives us a firm boundary at least, so we may be able to narrow down the uncertainty, but even then I doubt we could narrow it down to the millennium, let alone the year.

[Nereid]

It might be interesting to try to describe what observational data would need to be obtained, in order to predict when a particular star will go supernova, to within a century or so.

Start by distinguishing at least four different paths (channels?) to supernova-hood: core collapse of a massive star; deposition of matter (mostly hydrogen?) onto a white dwarf (WD), from a close binary (a non-compact star); merger/collision of two WDs; merger/collision of two neutron stars or a neutron star and a WD*.

Taking the last first, if one neutron star is observed as a pulsar, then sufficiently accurate estimates of the orbital decay should produce a robust estimate of when the merger - and so the supernova* - will happen. The closer to merger, the more accurate the estimate, possibly to within hours in the last year or decade of the system's life.

The third - double WD - is just like the pulsar case, except that estimates of orbital decay may be far less accurate unless the binary's light curve shows sufficient variability.

The second - WD plus non-compact close binary companion - would seem impossible to predict, to within a century or so; the rate of mass transfer would be difficult to predict (especially if it were variable), and the mass of the WD impossible to estimate with sufficient accuracy (to know when it is close enough to the Chandrasekhar limit).

The first - core collapse of a massive star - may not be hopeless. For example, if the onset of oxygen/neon burning could be detected, then the star's KABOOM date would be known to within months, possibly days (if the star's mass could be sufficiently accurately estimated); for silicon burning, days. If the onset of carbon burning could be reliably dated, then the KABOOM date could likely be estimated to within a century or so (some details here (http://arxiv.org/abs/astro-ph/0601261)).

In all cases (well, except for the second), being able to estimate the star's (or stars') mass(es) sufficiently accurately is key. Followed by good long-term monitoring. Of course, if you don't know a star is a close, double WD binary ...

Question: can the onset of carbon, or oxygen/neon, or silicon burning be detected? With sufficiently sensitive neutrino telescopes, the latter two could (I think); but what if you have only light to observe with?

* It may be that this does not result in a supernova

[Tom Mazanec]

What about pair instability supernovae? I believe there is a candidate in the LMC.

[Nereid]

What about pair instability supernovae? I believe there is a candidate in the LMC.
I'm not sure.

Isn't that just a particular kind of core-collapse supernova? If so, then the progenitor will go through the same carbon, then oxygen/neon, then silicon burning stages, but perhaps a bit faster. If so, then detecting the onset of any of these should lead to a pretty good estimate of when it'll go KABOOM.

[filrabat]

Let's hope we can!! That way, we can become at least partially prepared for when Betelgeuse literally goes out with a bang - and spew radiation all around its radius. I realize it's far enough away that the upcoming supernova won't be particularly lethal to us, but I can't see how it cannot disrupt space travel (even robotic)!

BTW, do astronomers know enough about SNs to know what kind of SN that particular star will be?

[Hornblower]

Let's hope we can!! That way, we can become at least partially prepared for when Betelgeuse literally goes out with a bang - and spew radiation all around its radius. I realize it's far enough away that the upcoming supernova won't be particularly lethal to us, but I can't see how it cannot disrupt space travel (even robotic)! My bold. Why not? Can you give us some references to predicted radiation intensity at that distance? Can you compare it to what the Sun has slammed us with during a major flare? Do you have information on what the spacecraft in question are designed to withstand?


BTW, do astronomers know enough about SNs to know what kind of SN that particular star will be?It is expected to be a core-collapse event, which happens when all sources of exothermic fusion energy are exhausted.

My question is, why is Betelgeuse getting all of the attention. Why not Antares, Mu Cephei and other noteworthy evolved supergiants?

[antoniseb]

... My question is, why is Betelgeuse getting all of the attention. Why not Antares, Mu Cephei and other noteworthy evolved supergiants?

I agree. I suspect it is mostly a question of sensational story writing. Antares is less massive than Betelgeuse, but other than that, I can't imagine how we could be saying one versus the other is next to go... and why not any any of the many WR stars? OK, we do have stuff about Eta Car, or Rho Cas, but I think more people are familiar with Betelgeuse as a star they recognize in the sky so it makes a better story.

[filrabat]

My bold. Why not? Can you give us some references to predicted radiation intensity at that distance? Can you compare it to what the Sun has slammed us with during a major flare? Do you have information on what the spacecraft in question are designed to withstand?It is expected to be a core-collapse event, which happens when all sources of exothermic fusion energy are exhausted.

I admit I've not seen any particulars as to the radiation intensity. Even so, I understand that, for Crab Nebula supernova at least, it would've killed off any life within 50 Light Years of it. Both Betelgeuse and Antares are at similar distances (as close as 500 LY for the latter, min est. and almost right at 500 for the latter). So even at 10 times the distance from a likely "kill radius" for planetary life, that's still likely to cause all kinds of problems for at least non-hardened satellites and spacecraft (I know very well that radiation intensity decreases as to the inverse square of the distance...in this case 50 divided by 500 is 10, and 10^2 is 100. So radiation at 500 LY is going to be 100 times less intense than at 50 LY.

As for the rest of the questions (shielding, coronal mass ejections, etc.) I confess ignorance.


My question is, why is Betelgeuse getting all of the attention. Why not Antares, Mu Cephei and other noteworthy evolved supergiants?

AND


I agree. I suspect it is mostly a question of sensational story writing. Antares is less massive than Betelgeuse, but other than that, I can't imagine how we could be saying one versus the other is next to go... and why not any any of the many WR stars? OK, we do have stuff about Eta Car, or Rho Cas, but I think more people are familiar with Betelgeuse as a star they recognize in the sky so it makes a better story.

Perhaps because it's the closest supernova candidate to us. Betelgeuse is only 497 LY, Antares is 553. Granted that's not a whole lot of difference, but as mentioned the masses are (within the aforementioned limits) similar. So unless there's some relevant difference among the stars that I'm unaware of, I have a hard time seeing how the predicted collapse/rebound of the two would release substantially more energy than the other.

Eta Car and Rho Cas are both extremely far away, and the latter's poles seem pointed well away from Earth. Rho Cas is one of the most massive (at 11K Ly, it's still visible to the naked eye), and yes it does cause a potential problem even at that distance.

Both of you do make convincing cases for the media to bring attention to the other potentially threatening supernovae though.

[Romanus]

RE Betelgeuse:
Agree with those who say it basically makes good "copy"; saying Mu Cephei is on the edge of a SN will just make people scratch their heads, thinking of something like "Cephalon? Safeway? Moo?"

In my op, the next galactic SN we can see will probably be some faint no-name star buried in a spectroscopic catalog, like our own SN 1987A progenitor. Note that at say, the Crab Nebula's distance, a red supergiant of ~Mv = -6 would have an apparent magnitude of only 5.5, and that's without the interstellar absorption considerable at that distance.