Expert: James Gort - 1/6/2010

Question
There was some mention in the media about a supernova, T Pyxidis, being about to blow up any time soon. They also said that even though it was c.3,260 lightyears away, that it would wipe out all life on Earth. What do you think re the possibility that it might happen soon in astronomical terms(ie within a thousand years or so)? And what is the rough distance at which Earth starts getting affected by supernovae in a mild way, and what is the rough distance at which life on Earth is likely to perish from a supernova? I would have thought 3,000 light-years away was far enough for safety.

Answer
Hi Geoff,

That's a really interesting question!  I'm going to go through a "back of the envelope" calculation, using some worst case numbers, and see what I come up with.  I'll then compare my numbers with a somewhat more detailed calculation.

Let's assume that the most violent SN has a total light output (all wavelengths) of 100 billion times that of the sun.  That's about equivalent to the combined light output of all the stars in the galaxy.  It's actually between 10 to 100 billion, but we'll go with the larger number.

Since the sun is about 1.6 x 10^(-5) l-y's from earth, you can do an inverse square calculation to determine the distance our SN has to be in order to be the SAME visual brightness as the sun.  That turns out to be 5 l-y's (feel free to check my arithmetic!).  So if a SN occurred at a distance of 5 l-y from earth, we'd have "two suns".  No question that the earth and its inhabitants couldn't stand that.  Talk about global warming!!  And please keep in mind that so far, we've made the (bad) assumption that the light spectrum of the SN is the same as the sun.  Much more of the SN's energy is actually in the gamma ray and x-ray part of the spectrum (with high energy electrons and neutrinos thrown in).  But let's ignore that for a moment.

So if having two suns is lethal, what might be a safe distance?  Well, I'll make the assumption (subject to argument!) that a 1% increase in the sun's output would be enough to upset the ecology of earth and be very harmful to life.  With that assumption, we'll move our SN 10 times farther out (giving us 1% of its radiation), or 50 l-y.  So the conclusion we can make here is that any SN closer than 50 l-y would be detrimental to life on earth (if we still assumed its spectrum was similar to the sun's).

Now, we'll modify our results to account for the heavy influx of gamma rays and x-rays.  This requires more of a detailed analysis of what this energy actually does to earth (especially the ozone layer), but time for more assumptions.  An educated guess would be we could move our SN out 2 to 4 times farther before we're out of danger.  That means that between 100 and 200 l-y is the crucial distance where a SN could threaten life on earth.  I hope I'm not too far wrong in this, because Betelgeuse (a supergiant star in Orion) is only 640 l-y away, and has been found to be shrinking rapidly (possibly indicating an imminent SN).  Since its mass is about 20 times the mass of the sun, the core collapse will be spectacular, likely resulting in a neutron star, but possibly a black hole.

So how do my quick calculations compare to a more detailed scrutiny?  See http://stupendous.rit.edu/richmond/answers/snrisks.txt

That study concluded that a Type II SN (like Betelgeuse) would need to be closer than about 33 l-y's to cause damage, but a Type 1a (like T Pyxidis) could be several times farther away.  This seems to be consistent with my thoughts.  If you read Appendix A of that study, he also concludes that radiation could be lethal at distances closer than 300 pc (1000 l-y).  I'm a little skeptical of that, but it could be.

So what about the T Pyxidis scare?  I don't believe it.  It's too far away.  That's my opinion, and I hope I'm right!  Hope this helps (and sets you mind at ease).

Cheers,

Prof. James Gort