Results 1 to 23 of 23

Thread: How fast would we look at a supernova?

  1. #1
    Join Date
    May 2004
    Posts
    1,743

    How fast would we look at a supernova?

    Suppose tonight a supernova brighter than Venus lights up the sky.
    How long would it take to instrumentally observe it?
    Ten seconds?
    Five minutes?
    An hour?
    The next night?
    SHARKS (crossed out) MONGEESE (sic) WITH FRICKIN' LASER BEAMS ATTACHED TO THEIR HEADS

  2. #2
    Join Date
    Aug 2006
    Posts
    3,140
    Quote Originally Posted by Tom Mazanec View Post
    Suppose tonight a supernova brighter than Venus lights up the sky.
    How long would it take to instrumentally observe it?
    Ten seconds?
    Five minutes?
    An hour?
    The next night?
    What do you mean "how long"?

    Do you mean how long before its light reaches us? That wold kind of depend on how far away it is.

    Or do you mean how long do they shine before dimming? This is from WikiMedia:



    https://en.wikipedia.org/wiki/File%3...ght_curves.png
    Attached Images Attached Images
    Last edited by DaveC426913; 2018-Sep-22 at 06:16 PM.

  3. #3
    Join Date
    May 2004
    Posts
    1,743
    I mean how long, after the supernova was first seen, would it take for observatory and space instruments to start studying it?
    SHARKS (crossed out) MONGEESE (sic) WITH FRICKIN' LASER BEAMS ATTACHED TO THEIR HEADS

  4. #4
    Join Date
    Aug 2006
    Posts
    3,140
    Quote Originally Posted by Tom Mazanec View Post
    I mean how long, after the supernova was first seen, would it take for observatory and space instruments to start studying it?
    Oh. A few minutes to a few hours.

    These days, there are so many instruments scanning the skies, it might already be observed by the time someone noticed it visually. Astronomers are aware that some events unfold rapidly, and have steps to cover this.

  5. #5
    Join Date
    May 2008
    Location
    The Netherlands
    Posts
    14,848
    How quickly can you setup your telescope? The SuperNova Early Warning System is aimed to provide a warning potentially hours before visible light is emitted from a supernova, and allows public subscription. Up to you to find out if a supernova of your suggested brightness would trigger an alert from this system.

    https://snews.bnl.gov/alert.html
    ____________
    "Dumb all over, a little ugly on the side." -- Frank Zappa
    "Your right to hold an opinion is not being contested. Your expectation that it be taken seriously is." -- Jason Thompson
    "This is really very simple, but unfortunately it's very complicated." -- publius

    Moderator comments in this color | Get moderator attention using the lower left icon:
    Recommended reading: Forum Rules * Forum FAQs * Conspiracy Theory Advice * Alternate Theory Advocates Advice

  6. #6
    Join Date
    Sep 2003
    Posts
    12,363
    For a SN to appear that bright, it would have to be relatively close to us (roughly 10,000 lyrs to 20,000 lyrs. depending on SN type; closer if their are some gas/dust extinctions). Would this not be close enough to see some pre-activity enough to cause more regular monitoring?
    We know time flies, we just can't see its wings.

  7. #7
    Join Date
    Jul 2005
    Location
    Massachusetts, USA
    Posts
    21,783
    Quote Originally Posted by George View Post
    ... Would this not be close enough to see some pre-activity enough to cause more regular monitoring?
    Depending on the type of SN, we might see a giant neutrino spike, or a gravitational wave, or a gamma-ray spike right at the beginning. The gamma ray spike would probably be used to identify the source within a minute (days before the SN gets to peak brightness). I suspect that the GW, or Neutrino spikes would take longer than a few minutes to process.
    Forming opinions as we speak

  8. #8
    Join Date
    Mar 2004
    Posts
    3,112
    From listening to the Ice Cube neutrino people, SN neutrinos are of such low energy (by Ice Cube standards) that they wouldn't detect them individually but would see a dramatic increase in their "noise" level. So they would know essentially that a core-collapse SN event had occurred whose signature reached us at c (close enough to not matter) button where in the sky. So people run out, take a look at Betelgeuse, Eta Carinae, etc. and at least know to keep a wide-field search going at high cadence. For a really nearby SN, we might not have to wait for sunset to get useful spectra (although being able to do low-dispersion spectra of stars that bright is sort of a niche thing). If one needs to wait for sunset it's more of a problem - even ignoring weather, there are time, date, and coordinate combinations where it could be a couple of hours before any site on land has a shot. (Example: the NGC 4993 GW event just missed being identified several hours earlier because a refined position came available just after it effectively set from South Africa while it was still daylight in Chile).

    There have been successful observations of GRB optical afterglows within a couple of minutes of the triggering detection, so when everything goes right the gap can be quite short.

  9. #9
    Join Date
    Feb 2010
    Posts
    587
    If it were “behind the Sun”, maybe several days? Or even a couple of weeks?

    If it were in the field of an observation that was underway, at a large observatory, seconds to minutes (depends on how vigilant the telescope operators were, how bright, etc).

    If bright enough, and not behind the Sun (or Moon), likely seconds to minutes, for one of the robotic all-sky monitoring outfits (bad weather may wreak havoc, of course).

    For a SNe that peaks at -4, say, are there any which take less than a minute/hour to get that bright? And not all GRBs - which have really fast rise times - are SNes, certainly not immediately...

  10. #10
    Join Date
    Sep 2006
    Posts
    1,784
    Quote Originally Posted by Jean Tate View Post
    For a SNe that peaks at -4, say, are there any which take less than a minute/hour to get that bright?
    Well, when the shock wave of a core-collapse SN reaches the photosphere of its host star, the temperature shoots up by quite a bit in a very short period of time -- maybe from 3000 K to, uh, 100,000 K, at a guess. The luminosity of the star will then increase by something like, hmm, (100/3)^4 ~ 81 x 10^4 ~ 10^5 ~ 12.5 magnitudes. To reach a visual magnitude of -4, the progenitor would have to be around mag 8. How any core-collapse candidates are there which are so bright? I dunno.

    After this brief spike in luminosity, the object cools off quickly, perhaps in a matter of hours, before VERY much more gradually rising to a higher peak luminosity over a period of a few weeks. So this brief burst in brightness might not be noticed by casual stargazers.

  11. #11
    Join Date
    Sep 2003
    Posts
    12,363
    There's a lot of interesting points so far but, as your Average Joe, I would like to see how close I am with the following small plethora of assumptions, given the OP scenario:

    1) From Dave's post (graphs), it looks like the star would be roughly 0 to -1 (my math) in apparent magnitude (-14 absolute) for a period of 10 to 20 days prior to peak luminosity. [This should catch some attention if the Sun isn't hiding it, I assume.]

    2) Based on what I recall, and found, on 1987A, the neutrino count should be between ~ 60x to 250x, roughly, greater than 1987A due to the distance difference alone. Is this about right? [This assumption assumes sensitivity levels for those facilities have not improved since 1987, which I hope isn't the case.]

    2b) But are the neutrino energies important to get that bump in rate if the S/N type is different than 1987A, which I think was a Type II?

    3) From ngc3314, however, ICE Cube might have an advantage (my assumption) though the neutrinos are wimpier for them. Is this right? And are neutrinos from SN weaker than "normal"?

    4) The time difference explains why the neutrinos arrived sooner than light itself in the 1987A event. This time difference was a maximum of about 3 hours, I think I read. This would make the OP S/N time lag to be about 10 to 20 minutes, at best. Is this about right?
    Last edited by George; 2018-Sep-26 at 06:07 PM.
    We know time flies, we just can't see its wings.

  12. #12
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,232
    Quote Originally Posted by George View Post
    There's a lot of interesting points so far but, as your Average Joe, I would like to see how close I am with the following assumptions, given the OP scenario:

    1) From Dave's post (graphs), it looks like the star would be roughly 0 to -1 (my math) in apparent magnitude (-14 absolute) for a period of 10 to 20 days prior to peak luminosity.

    2) Based on what I recall, and found, on 1987A, the neutrino count should be between ~ 60x to 250x, roughly, greater than 1987A due to the distance difference alone. Is this about right?

    2b) But is the neutrino energies important to get that bump in rate if the S/N type is different than 1987A, which I think was a Type II?

    3) From ngc3314, however, ICE Cube might have an advantage (my assumption) though the neutrinos are wimpier for them. Is this right? And are neutrinos from SN weaker than "normal"?

    4) The time difference explains why the neutrinos arrived sooner than light itself in the 1987A event. This time difference was a maximum of about 3 hours, I think I read. This would make the OP S/N time lag to be about 10 to 20 minutes, at best. Is this about right?


    My bold. You may have something backward here. The neutrinos got here three hours earlier, while they are a smidgen slower than the speed of light considering their non-zero mass. They got out of the envelope almost uninhibited, while it would appear to me that it took about three hours for the photosphere to start brightening noticeably. Thus I would expect the same lag for a similar supernova that is much closer.

  13. #13
    Join Date
    Sep 2003
    Posts
    12,363
    Quote Originally Posted by Hornblower View Post
    [/B]
    My bold. You may have something backward here. The neutrinos got here three hours earlier, while they are a smidgen slower than the speed of light considering their non-zero mass.
    Hmmm, I did say "sooner", right?

    They got out of the envelope almost uninhibited, while it would appear to me that it took about three hours for the photosphere to start brightening noticeably. Thus I would expect the same lag for a similar supernova that is much closer.
    Ah, I see what you're saying. But I think I just have it backwards. Assuming a 3-hour time lag for the 1987A case (160k lyrs), then wouldn't the time lag be about 48 to 24 hours for the 10k to 20k lyr closer distance?

    Or is the time lag at the star and not at the observatory? Maybe that's what's got me.
    Last edited by George; 2018-Sep-26 at 10:15 PM.
    We know time flies, we just can't see its wings.

  14. #14
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,232
    If we take neutrinos three hours earlier as correct, then the time lag has to be a characteristic of the star, because the neutrinos cannot get here earlier if they were emitted at the same time as the photospheric light. Since the neutrinos are ever so slightly slower, the time lag would decrease with increasing distance, not with decreasing distance.

  15. #15
    Join Date
    Sep 2003
    Posts
    12,363
    Quote Originally Posted by Hornblower View Post
    If we take neutrinos three hours earlier as correct, then the time lag has to be a characteristic of the star, because the neutrinos cannot get here earlier if they were emitted at the same time as the photospheric light. Since the neutrinos are ever so slightly slower, the time lag would decrease with increasing distance, not with decreasing distance.
    Right, so if the 1987A event had happened, say 8x, closer to us, the time lag would be much greater, though not 8x the observed lag time since it's not that simple -- at some distance the light catches the neutrinos and there is no time lag observed so an 8x time lag multiplier is proven useless, though simple math should work.

    The problem, I assume, is in, I'll call it decimal dyslexia. [Perhaps there is a more clever term for it.] The number of decimal places we have to work with are two few to get a handle on it accurately. The number needed is troubling for us.

    Consider again 1987A...
    The observed time lag for light after neutrino arrival was, per Wiki, about 2 to 3 hours.

    Fermi (MINOS) determined the lower speed limit of 0.999976c. But this lower limit would mean, at the star, the time lag would be about 3-1/2 years (ug!). The upper limit was just a hair over c, which isn't plausible either and at several levels. So the required number of decimal places needed to get the time lag at the star must be very, well, significant.

    However, I would assume that another SN (Type II) much closer would greatly help determine the stellar time lag, especially if the observed time lag could be measured in minutes. But to optically observe, in minutes of its start, a new SN might be possible given the proximity as stated in the OP, which I find intriguing. Is this assumption reasonable?
    Last edited by George; 2018-Sep-27 at 02:27 PM.
    We know time flies, we just can't see its wings.

  16. #16
    Join Date
    Feb 2009
    Posts
    2,139
    Since 1987, there have been just 5 supernovae brighter than +12, of which 2 were type II and 3 were type I
    Type II:
    1993J, in M81, 8,5 Mly peak +10,2, progenitor K supergiant
    2004dj, in NGC 2403, estimated at 8...11 Mly, peak +11,2. Found after peak
    Type I:
    2013aa, in NGC5643, peak +11,7
    2013cq, peak +12
    2014J, in M82, 11,5 Mly, peak +10,5, missed for at least 6 days, discovered accidentally in a lesson.

    So - the 3rd brightest supernova in 30 years was missed for a week and then found by chance.

    Suppose that there is a Type I supernova to peak at -4 - 25 times closer than 1987A, at 6600 ly. Then it would be something whose progenitor, at magnitude +5, is a naked eye object - like Eta Carinae.

    Sanduleak was spotted at +4,5. How long after the neutrino burst was it?

  17. #17
    Join Date
    Sep 2006
    Posts
    1,784
    Quote Originally Posted by chornedsnorkack View Post
    Sanduleak was spotted at +4,5. How long after the neutrino burst was it?
    One can find the answers with just a little bit of research.

    IAUC 4316 states that Ian Shelton reported the detection of SN 1987A on a photographic plate at 1987 Feb 24.23, with a visual sighting at Feb 24.2.

    The neutrinos were recorded at 1987 Feb 23 07:35:41 = Feb 23.32 or so.

    That makes a time difference of 0.91 days between the neutrinos arriving at Earth, and humans noticing the light from the explosion.

  18. #18
    Join Date
    Feb 2009
    Posts
    2,139
    Quote Originally Posted by StupendousMan View Post
    IAUC 4316 states that Ian Shelton reported the detection of SN 1987A on a photographic plate at 1987 Feb 24.23, with a visual sighting at Feb 24.2.

    The neutrinos were recorded at 1987 Feb 23 07:35:41 = Feb 23.32 or so.

    That makes a time difference of 0.91 days between the neutrinos arriving at Earth, and humans noticing the light from the explosion.
    I read 0,88 days here. Making 21 hours.
    The time from neutrino burst to shock arrival at photosphere is variously estimated from 1 to 4 hours. This leaves 17 to 20 hours that the supernova was shining, but remained unseen.
    Now consider a supernova 25 times closer and 7 magnitudes brighter.
    Magnitude -4 at peak, -2,5 20 hours after burst.
    If the neutrino detectors running in 1987 caught 25 neutrinos from the +3 supernova, these should catch 15 000 from a -4 supernova.
    How much neutrino detection capacity is operating now, compared to 1987?

  19. #19
    Join Date
    Sep 2003
    Posts
    12,363
    Quote Originally Posted by chornedsnorkack View Post
    I read 0,88 days here. Making 21 hours.
    The time from neutrino burst to shock arrival at photosphere is variously estimated from 1 to 4 hours. This leaves 17 to 20 hours that the supernova was shining, but remained unseen.
    How can that be? Isn't light faster than neutrinos? Light would have to arrive at Earth sooner than that 1 to 4 hour delay at the star after the neutrino detection, right?

    Now consider a supernova 25 times closer and 7 magnitudes brighter.
    Magnitude -4 at peak, -2,5 20 hours after burst.
    If the neutrino detectors running in 1987 caught 25 neutrinos from the +3 supernova, these should catch 15 000 from a -4 supernova.
    And we would have a better time lag to work with since the light wouldn't have had 160,000 years of faster speed in chasing those neutrinos.
    We know time flies, we just can't see its wings.

  20. #20
    Join Date
    Mar 2004
    Posts
    3,112
    The lowest neutrino energies from the Kamiokande detentions of SN1987A were about 10 MeV, compared to current limits on the electron (anti)neutrino mass <2 eV. That puts (1-v/c)<4x10-14 for these, so that over 150,000 years' flight time, the energy-based delay compared to simultaneously emitted photons would be only about 0.2 second. (Related, for a while the best mass limits came from lack of correlation between the SN1987A neutrino arrival times and energies). For 2 eV mass, a supernova would have to be (150,000 * 3600/0.2 = 2.7 Glyr or 900 Mpc) distant to get an hour delay, which would start to get into the realm of the neutrino-optical time delay from supernova simulations. This is why the discussion has centered on the neutrino masses and not distance-induced delays.

    (He didn't notice until the next day, but an Australian amateur happened to have a series of 35mm images of the LMC showing the initial brightening before Shelton's discovery, but while it was not yet so obvious as to jump out visually). (Note after googling - I think that was Rob McNaught).

  21. #21
    Join Date
    Sep 2003
    Posts
    12,363
    Quote Originally Posted by ngc3314 View Post
    The lowest neutrino energies from the Kamiokande detentions of SN1987A were about 10 MeV, compared to current limits on the electron (anti)neutrino mass <2 eV. That puts (1-v/c)<4x10-14 for these, so that over 150,000 years' flight time, the energy-based delay compared to simultaneously emitted photons would be only about 0.2 second. (Related, for a while the best mass limits came from lack of correlation between the SN1987A neutrino arrival times and energies). For 2 eV mass, a supernova would have to be (150,000 * 3600/0.2 = 2.7 Glyr or 900 Mpc) distant to get an hour delay, which would start to get into the realm of the neutrino-optical time delay from supernova simulations. This is why the discussion has centered on the neutrino masses and not distance-induced delays.

    (He didn't notice until the next day, but an Australian amateur happened to have a series of 35mm images of the LMC showing the initial brightening before Shelton's discovery, but while it was not yet so obvious as to jump out visually). (Note after googling - I think that was Rob McNaught).
    That helps greatly! Those neutrinos were haulin' from the SN! Can't blame 'em.

    Thus, the time lag observed for anything in our galactic cluster will essentially be that of the lag that happened at the star.
    We know time flies, we just can't see its wings.

  22. #22
    Join Date
    Sep 2004
    Location
    South Carolina
    Posts
    1,133
    I had thought that the first notice from a supernova type Ia would be the radiation burst, which would be received before the brightening was noticed as the brightening had to build up.
    There is something fascinating about science. One gets such wholesale returns of conjecture out of such a trifling investment of fact.
    — Mark Twain, Life on the Mississippi (1883)

  23. 2018-Oct-04, 11:22 AM

  24. #23
    Join Date
    Sep 2006
    Posts
    1,784
    What creates the "radiation burst" in your notion of the event?

    Type Ia do produce a good deal of gamma-rays, but many of them are created by the radioactive decay of iron-group nuclei which are generated in the runaway nuclear reactions. Those decays take days to weeks, and in many cases, power the optical light curve.

    So, high-energy radiation would grow roughly on the same timescales as the optical radiation.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •