How many stars we see are already gone?

August 18, 2012 12:34 AM

How many of the stars whose light we see at night are actually already dead/gone/disintegrated?

(Typical-kid's-question-to-which-you-find-you-don't-know-the-answer-Filter.) Considering only the ones visible to the naked eye, on a beautiful starry night in the countryside, my wild guess was about a third. How off was I?

posted by progosk to Science & Nature (7 answers total)

Any actual star you can see unaided is in our galaxy and so the light would be on average a few tens of thousands of years old (the Milky Way being ~100,000 light years across but most visible stars are much closer than that). Nearly all of them would not have gone nova during such a short period of time (for a star). Only a blue giant like Betelgeuse would have any reasonable chance of having done so.

There are only a handful of galaxies you can see with the naked eye even in a very dark sky. Only the large Magelianic cloud is reasonably bright but is only visible from the southern hemisphere (as is the small Magellanic cloud) The rest are super dim and very hard to see.

According to wikipedia, "it has been reported" that someone has seen the Messier 83 galaxy unaided. That would be about the most distant thing you could see in perfect conditions at a bit over 14 million light years.

That's a long time for the very largest blue giant stars but for the grand majority of stars that is hardly any time at all.
posted by Riemann at 12:47 AM on August 18, 2012 [11 favorites]

The last naked eye supernova visible from the Earth with the naked eye was in 1604, with a previous visible ones in 1572, 1181, 1054 and 1006. Other Milky Way stars have gone supernova, but even during the explosion, were not visible as naked eye objects. So there's a supernova visible from earth every 200 years or so.

There are about 2000 naked eye stars, the farthest visible one being (under the best conditions) about 4075 LY away. If the average visible star is 2000 LY away and there are 2000 of them visible in the sky, that 10 of them have gone pop.

But wait - none of the stars I list were, as far as we know, naked eye stars prior to going supernova. This handy list of supernova remnants suggest that only 3 in the last 40,000 years were close enough to possibly be visible prior to going supernova (and one of them is debatable) which makes the odds something like a 1 in 3 chance that one of them is no longer with us.

So while the stars are tremendously far away, they also have a tremendously long shelf life.
posted by Kid Charlemagne at 3:29 AM on August 18, 2012 [8 favorites]

Best answer: Oddly enough, this question appeared on Reddit here just three days ago... many good answers.
posted by crapmatic at 3:39 AM on August 18, 2012 [1 favorite]

Define 'already'. at those distances, concepts like 'now' and 'then' get more than a little bit fuzzy.
posted by empath at 7:30 AM on August 18, 2012

It's complicated further by how long it takes a star to die. Most aren't going to go pop in a supernova (Eta Carina is probably the only visible star that has even a small chance of doing it/having done it already). If it goes through a planetary nebula phase, that phase is going to last on the order of ten thousand years anyway. There might not be a sufficiently distinct moment at which you term most stars 'dead', especially if you don't consider a white dwarf to be dead - that phase lasts longer than the current age of the universe.
posted by edd at 7:48 AM on August 18, 2012 [1 favorite]

Response by poster: Now I know why I didn't know: it's not all that simple to put. Thanks for the oddly coincidant reddit post, crapmatic - between that and the answers here, I guess I don't need to try Quora ;-)
So, in a nutshell: probably none of the between 2000 and 9000 stars we would be seeing has already popped, with the possible exception of Betelgeuse and Eta Carina. So I was very off indeed. Thanks!
posted by progosk at 8:33 AM on August 18, 2012

Define 'already'. at those distances, concepts like 'now' and 'then' get more than a little bit fuzzy.

This is true, but it is fuzzy in a very bounded way. The obvious/naïve/useful way to answer this question is to consider simultaneity in the galactic or solar-system barycenters' frames, but the most you can do by choosing other frames is to roughly double the amount of time available for a star to have "already" popped— and that extra few thousand years for a star in our galaxy isn't going to substantially change the answer.
posted by hattifattener at 11:46 AM on August 18, 2012