Interestingly, about the time I first started writing about near Earth supernovae, this fascinating paper was already published. Though by some quirk of fate, I’ve only just been able to get a copy of it to read!
So as I’ve said previously, there’s some pretty compelling evidence that supernovae have gone off in our cosmic back yard in the past. We know, because they’ve left a few telltale signs. Isotopes with half lives shorter than the age of the solar system are a good clue. 60Fe atoms have been found, with over 100 times the expected natural abundance. To cut a long story short, the most likely reason for these isotopes are a supernova occuring near Earth around 2.8 million years ago (give or take 400,000 years) at an estimated 15-100 parsecs away (this concept goes by the delightful name of ‘supernova archaeology’). That’s quite easily consistent with our old friend the Scorpius-Centaurus cloud (our local producer of such stellar popcorn).
But the big question is, how much would the Sun protect us from a nearby supernova? The wind of particles given off by the Sun is pretty powerful. Powerful enough to blow a bubble in the interstellar medium (the heliosphere) and help shield us from interstellar gas and cosmic rays. So how would this interact with a nearby supernova? How much protection does the Sun give us?
Well Fields et al, based at the University of Illinois, have put together the first ever hydrodynamic simulation to study the interaction of the Sun’s heliosphere with an expanding supernova remnant. Systemic, and in plenty of detail, they use observational data on the solar wind to create an idealised model (tailored to recreate the conditions around Earth’s orbit), together with a Sedov-Taylor model of an expanding supernova remnant. Then they put the two models together and see what happens! As such, they can recreate a huge array of possible scenarios, varying solar wind speed, supernova distance, ISM density and other parameters.
The image to the right there is a snapshot of the Sun’s heliosphere as it might look in the middle of a supernova shockwave. This is what could happen with an average solar wind speed, if a supernova occurred 10 parsecs away. The interesting thing is that the overall shape isn’t too different to the heliosphere normally. Certainly, it’s smaller, more compressed, longer and thinner… (actually, a lot smaller — the termination shock is normally around 75-90AU away, while here it’s only 1.1AU!) Mind you, the side where the supernova blast is impacting still has a bow shock, the inside still has a termination shock encircling the Sun (the inner limit of where the supernova blast could reach). While there’s a ragged looking trail of gas (caused by Kelvin-Helmholtz instabilities), Earth would be safe from the actual supernova blast wave itself. The black circle shows Earth’s orbit, safely within the Sun’s influence.
What’s interesting is that the effect can be directly scaled. A stronger blast (say, from a closer supernova) or a weaker solar wind causes the heliosphere to shrink accordingly. If the solar wind were half its average (and incidentally, it’s at a very low point at the moment), the termination shock would be just inside Earth’s orbit. Scarily, this would mean the Earth might take the direct force of the supernova blast.
Still, it’s heartening to know that a supernova 10 parsecs (that’s around 33 light years) away probably wouldn’t destroy the Earth! Ditto, it has some implications for astrobiology. Planets orbiting near to their stars would have a fair amount of protection, even from a supernova, which bodes rather well for the development of life on such planets. A planet within a star’s habitable zone would likely be fairly well protected by its parent star.
What this means is that it’s highly unlikely that the supernova 2.8 million years ago actually hit us directly. The Sun would have given us plenty of shielding from the actual blast wave. More likely, all of that 60Fe got to Earth as dust (supernova condensates… I’ve read about them somewhere before!), peppering the atmosphere like a shotgun blast. Even at the moment, much slower interstellar dust makes it as far into the heliosphere as Earth. We must’ve received a good dusting 2.8 million years ago!
The authors fully acknowledge that their model still needs work, reminding us that this is still a first ever model:
“Of course, the precise quantitative “cutoff” distance for terrestrial exposure will depend on the details of the problem, some of which we have simplified…”
Their plans are to build more detailed simulations to investigate the scenario even further, and their work shows a lot of promise. I’ll bet they have some more interesting results already!
Brian D. Fields, Themis Athanassiadou and Scott R. Johnson (2008). Supernova Collisions with the Heliosphere The Astrophysical Journal, 678, 549-562 DOI: 10.1086/523622