# Would going underground into a planet offer protection from a supernova? [duplicate]

I am considering a scenario where people are anticipating the star of their system going supernova, which would be somehow triggered by an advanced, hostile alien race. However, the people cannot predict when it will happen. Their best option is to prepare some temporary shelter they can stay in until people from other systems can come extract them after the brunt of the explosion has passed.

I considered them building bunkers deep underground that could shield them from radiation for a few days or weeks until help arrives.

Is this a realistic way at all to temporarily protect yourself from your own star’s explosion? Or, would the planet heat up all the way through, making the underground unsafe? Or, would the planet just be completely destroyed? Taking into account planet size, distance from the star, size of the star, and radiation shielding technologies.

In addition, as the supernova would be triggered by a hostile alien force, there is the possibility that its power may not be equal to that of a natural supernova.

If not, any ideas for more realistic options of escape/defense in this situation? (Or some other stellar phenomenon that an underground bunker might be reasonable for?)

• Comments are not for extended discussion; this conversation has been moved to chat. – L.Dutch - Reinstate Monica Jul 8 '20 at 7:35
• It might be possible to survive by going underground, if the star is not the sun of their own system, but a star somewhere far enough that adding some 1km of solid earth between you and incoming radiation helps. that would also make it possible that the hostile aliens are far away, too and can't come destroy planet directly. how far should supernova be that a 1km underground distance makes a difference, is another question though. – Gnudiff Jul 8 '20 at 20:47

# No.

A typical supernova will perhaps not destroy the planet, and "just" leave it as a burnt-out cinder. If that were all, then a refrigerated, shielded, very deep underground hideout could survive the onslaught. Indeed, such a hideout would survive the onslaught.

But the supernova explosion will also lash out with a shock wave of neutrinos. While mostly harmless, neutrinos cannot be shielded by any ordinary matter, because their interaction probability is just too low. By the same token, a huge enough number of them will interact with matter; the whole planet will be flashed inside and out, and a significant portion of its constituent elements will undergo inverse beta decay.

The necessary calculations are found here and yield a fluence of 8.4E+22 events per square meter at a distance of one parsec. Assuming a neutrino energy range of 5 to 15 MeV, the average dose at that distance would be between 1.4E-9 Sv and 1.6E-10 Sv. Randall Munroe derives an estimate of 0.5E-9 Sv, giving a LD at 2.3 AU; from the original figures, we can re-derive an absorbed LD (for a human-mass organism) between 1.3 and 4 AU depending on neutrino energy spectrum.

While not too worrisome for mechanical systems and most electronics, that level of inverse beta decay is potentially lethal to anything organic.

The only chance to avoid destruction is to escape - which will be a considerable endeavour, because you'll need to be at least 300 million kilometer distant to be barely out of the lethality envelope. In all likelihood you will want to achieve a distance at least two orders of magnitude greater, which means completely escaping the doomed solar system.

(This heavily relies on human absorbed doses. Even for humans, 5Sv is only a ballpark estimate of lethality; it is the single-exposure dose that kills 50% of exposed humans within one month. Even a 1Sv dose, which would happen at a distance of approximately 5 AU, has a death rate of around 5% for humans. On the other hand, other organisms would probably tolerate it nicely).

• from the XKCD mentioned in L.Dutch's answer: "The idea of neutrino radiation damage reinforces just how big supernovae are. If you observed a supernova from 1 AU away—and you somehow avoided being being incinerated, vaporized, and converted to some type of exotic plasma—even the flood of ghostly neutrinos would be dense enough to kill you." The problem - also mentioned by that XKCD - is that you'd be inside the exploding star to be within 1 AU from its center. – John Dvorak Jul 5 '20 at 14:04
• “If it’s going fast enough you can absolutely knock someone down with a feather” - Randall Munroe, What If. – Joe Bloggs Jul 5 '20 at 15:26
• @skyprezzel I'm afraid so. The problem being that if the supernova was caused by an alien race (see e.g. The Dark Forest by Liu Cixin, or Calculating God by Robert J. Sawyer), then the same race could make short work of an escape fleet... – LSerni Jul 5 '20 at 21:14
• Sedna has an aphelion of 140 Tm or 140 billion km. Deep underground in Sedna during its aphelion may be far enough? OP didn't specify how far away the planet was orbiting. – gerrit Jul 6 '20 at 7:09
• @gerrit absolutely, yes. Sedna would likely survive the supernova. It is basically the plot of Arthur C. Clarke's The Star, ( en.wikipedia.org/wiki/The_Star_(Clarke_short_story) ), with a redoubt instead of a time vault. The problem remains that I referred to in my previous comment - anyone wielding supernovas could probably make short work of Sedna. – LSerni Jul 6 '20 at 11:33

A supernova is an hazardous object to be within in a radius of several light years because it's very, very, very energetic.

To quote Randall Munroe

his rule of thumb for estimating supernova-related numbers: However big you think supernovae are, they're bigger than that.

Here's a question to give you a sense of scale: Which of the following would be brighter, in terms of the amount of energy delivered to your retina:

• A supernova, seen from as far away as the Sun is from the Earth, or

• The detonation of a hydrogen bomb pressed against your eyeball?

Applying the physicist rule of thumb suggests that the supernova is brighter. And indeed, it is ... by nine orders of magnitude.

The radiation shower following a supernova explosion is dangerous not only because it will exceed the tanning dose by several orders of magnitude, but also because it will strip the planet from its atmosphere and maybe even from a considerable amount of whatever it's below that atmosphere.

Hiding underground to get shelter from a supernova looks like when Wile E Coyote opens an umbrella to get repair from the rocks about to drop on him.

• Might it partially do a nuke-in-space? Spreading the force to the empty space (getting filled by the explosion but still) and get deflected enough not to kill people underground? I'll not contest the atmosphere being unliveable afterwards (if it's even there) but anyone in closed-system bunkers deep underground could be safe? – Demigan Jul 5 '20 at 17:23
• @Demigan - nope. The amount of energy involved here is ridiculously staggering. – jdunlop Jul 5 '20 at 22:32
• that comparison makes me think that supernovas are universe's way of saying "you have made things explode? that's cute" – conquistador Jul 6 '20 at 13:58
• @Demigan: The problems with bunkers deep underground is that they still need to vent the heat upwards. And when waht remains of the surface of the earth is now also boiling lava, you're sandwiched with no place left for the heat to go. – MSalters Jul 6 '20 at 14:59
• @MSalters That depends on how deep you are. If your shelter isn't also swimming in boiling lava, then there's a buffer of relatively cooler rock between you and the surface. – chepner Jul 6 '20 at 21:09

This is an attempt to elaborate on L. Dutch’s answer with a simplified mathematical model and some calculations.

To take a simplified view of a supernova, let’s assume that all of the gravitational binding energy of the star is converted into kinetic energy, radiation, etc during the blast. Assuming that the star has radius $$R$$, mass $$M$$, and uniform density $$\delta=3M/4\pi R^3$$, its gravitational binding energy is equal to $$\frac{3GM^2}{5R}$$ For our sun, this is about $$2.3\cdot 10^{41}$$ joules. Wow!

Not all of this energy would be directed straight towards Earth. Let’s assume the supernova explodes symmetrically in all directions. Then, by the time this energy reaches Earth, it will be spread out over the surface of a sphere with radius equal to the distance between the Earth and the sun. The fraction of this energy that hits the Earth is approximately equal to $$\frac{\pi R_E^2}{4\pi \text{AU}^2}\approx 4.5\cdot 10^{-10}$$ where $$R_E$$ is the radius of the Earth and $$\text{AU}$$ is the Astronomical Unit (distance between the Earth and sun). Thus, the amount of energy received by the Earth is about $$(2.3\cdot 10^{41}\space\text{J})(4.5\cdot 10^{-10})\approx 1.04\cdot 10^{32}\space\text{J}$$ To estimate how much damage that would inflict, let’s calculate the binding energy of the Earth: $$\frac{3GM_E^2}{5R_E}\approx 2.24\cdot 10^{32}\space\text{J}$$ Surprisingly, that’s actually greater than the amount of supernova energy that would reach Earth! This means that the Earth wouldn’t be completely obliterated.

However, as mentioned in L. Dutch’s answer, it would be very nearly obliterated. Although there wouldn’t be enough energy to completely disintegrate Earth, it would probably rip off the atmosphere and the upper layers of the Earth’s crust. It’s doubtful that your people would be able to dig deep enough underground to protect themselves.

EDIT: According to @GaryWalker in the comments, my estimate for the energy output of a supernova is off by a factor of $$1000$$. So if there were any doubts before about whether a human bunker could survive, they can be safely dismissed.

• "Mr President, shall we get roasted from the supernova blast or from the mantle heat?" – L.Dutch - Reinstate Monica Jul 5 '20 at 15:13
• @L.Dutch-ReinstateMonica An interesting follow-up to the OP’s question might be “is it practically possible to create a strong-box capable of protecting people inside from a supernova (even if it is flung out into space by the blast)?” – Franklin Pezzuti Dyer Jul 5 '20 at 15:21
• You underestimate the power of a supernova, minimum size is about 1.5E44 Joules, so nearly another factor of 1000 more powerful than your guess. – Gary Walker Jul 5 '20 at 17:06
• @skyprezzel My simplified model would suggest that energy drops off as the inverse square of the distance from the star $\sim 1/D^2$. (For example, if you double the distance, the amount of energy reaching the planet is cut by a factor of $4$.) Unfortunately, the farther the planet is away from the star, the less solar energy it enjoys before the supernova blast, and the less likely it is to host life. – Franklin Pezzuti Dyer Jul 5 '20 at 21:29
• @SteveJessop fair point. Star explosion != supernova – Gensys LTD Jul 7 '20 at 6:58

Everybody isnthinking only about radiation. You probably did too, since you are considering whether people can shelter from it.

When a star explodes as supernova it sheds mass. From the Wikipedia article about supernova remnants:

A supernova remnant (SNR) is the structure resulting from the explosion of a star in a supernova. The supernova remnant is bounded by an expanding shock wave, and consists of ejected material expanding from the explosion, and the interstellar material it sweeps up and shocks along the way.

(...) the resulting supernova explosion expels much or all of the stellar material with velocities as much as 10% the speed of light (or approximately 30,000 km/s). These speeds are highly supersonic, so a strong shock wave forms ahead of the ejecta. That heats the upstream plasma up to temperatures well above millions of K. The shock continuously slows down over time as it sweeps up the ambient medium, but it can expand over hundreds or thousands of years and over tens of parsecs before its speed falls below the local sound speed.

Even if you could shield yourself from the radiation in a planet (which you can't, as everybody else is showing you), the planet will be vaporized. The oncoming plasma that hits it will be millions of degrees hot in any scale, with each particle packing a hulk punch due to the immoral oncoming speed. And remember that supernova come from stars way more massive than the sun. The oncoming stream of stellar material will be more massive than any planet, so even if the planet could stay whole and solid after the initial blast it would still be surrounded by plasma hotter than a star; Hot enough to melt the planet.

Just in case I haven't made myself clear enough, let's put this all in scale:

• The dino killing asteroid had a mass estimated around at most 4.6 × 1017 kg, with impact speed estimated at around 20 km/s.

• For all we know, for the Moon to form the Earth had to be hit by Theia, with a mass of approximately 6.4 × 1023 kg, and an impact speed of around 4 km/s. This impact re-melted the crust of the Earth. By the way, the kinect energy input from this is five orders of magnitude greater than the dino killing one (i.e.: like 100,000 dino asteroids hitting at once).

• To blow up as supernova, a star needs to have at least eight solar masses (so the exploding star will have at least ~1.6 × 1031 kg). And while the explosion is omnidirectional, potentially most of the mass will be blown across the star's plane of rotation (I believe clouds such as the Veil Nebula provide evidence for this), which is where planet orbits will tend to be. We are therefore talking about multiple planetary masses coming onto your planet at 30,000 km/s.

How many planetary masses will depend on your planet's size and distance from the star, with such data we might be able to calculate. But the mass and speeds involved don't bode well for anyone trying to escape the apocalypse in a vault. For a single Earth mass (already factoring in relativistic mass) in plasma hitting your planet at 10% the speed of light, the impact is nine orders of magnitude greater than the impact that created the Moon - or, in approximate terms, the same kinetic energy of a billion planets the size of Mars falling onto Earth.

And that's still less energy than what you'll get in the form of gamma radiation from the supernova.

• Back of envelope: the radius of the earth is 6300 km, so the area of its disk is 1.25E8 km². Distance to Sun is 150 million km, so the area of the shell at Earth's orbit is 2.8E17 km². Therefore even in the optimistic assumption that it's ejected spherically we'd get 4.4E-10 of whatever mass is coming. If only 50% of a Solar mass is ejected (1E30 kg), that means we take 4.4E20kg to the face. 1/1000 of the mass of Theia, but going 7000 times as fast, so 49000 times the kinetic energy using the non-relativistic equation. – Steve Jessop Jul 7 '20 at 2:48

If the planet is Earth-like, e.g. it has a hot molten core, then you can't burrow very deeply without it getting too hot. That's why human's haven't managed to dig through the Earth's crust. The Kola Superdeep Borehole is 12,226 m deep and it's 180 C down there.

• Good call-out. At the point where you're talking a strong enough bunker to (potentially) withstand a supernova, a little thing like a planet's molten core probably won't be a problem. But if it's just a misguided attempt that can't work, this would be a valid problem during construction. – Bobson Jul 7 '20 at 20:58

An issue as well is your "until help arrives" comment. It suggest you think someone could shield themselves from a supernova, but there exists a spaceship which could whitstand said supernova for long enough to evacuate a colony during the worst part of it. according to this time line you have hours, if that, until your solar system is screwed, and it will get worse over the next few weeks. If somehow you could withstand said supernova through all that by bunkering, you're not looking at a few weeks until help can arrive, but more likely a good few years. During which the surface of your planet will instantly go ozonless, then boil and liquify, then slowly freeze solid. Survival during that time with absolutely no access to natural biological resources will be enough trouble, but the inability to replenish your air makes this even worse.

Practically, you'd need to build the equivalent of a space station underground close to the core of your planet. If you're going through all that effort, it would be much better to build an actual space ship and leave. And I mean leave now, because even as soon as you notice the start of a supernova, you're already too late to launch a space ship and escape.

Your basic problem from step one is 'anticipating' that a hostile Alien race 'cause' a Supernova to occur in the first place. That and the fact not every star is capable of becoming a supernova (aliens igniting it or not). Only the most massive stars can do it this. And there aren't many within reasonable range.

It also implies contact with the race in question, so either we have traveled out into interstellar space and met them or they have traveled here. If its the latter case they literally don't need a supernova to kill us. By definition if they are that advanced (compared to us) and hostile we are DOA Supernova or not. A race with that kind of power would simply not have to bother with Supernova to deal with a species that wasn't at least near their (God Like) tech level.

That leaves us meeting them out there. And if that the case we have not only have advance warning we have also probably spread out in the opposite stellar direction anyway. So taking out Earth, while a tremendous blow doesn't kill off all the annoying apes anyway - we're still out there in the stars somewhere.

In terms of the effect on Earth. Bunkers would protect humanity and any species we selected from the worst effects of a Supernova which primarily involve the destruction of the ozone layer. Humanity could remain underground for a few decades/generations until the worst effects have past. So if the aliens are intending to kill us all off they have gone to an awful lot of effort for very little effect.

A better scenario would be a simper one. Human astronomers a few years from now (pick your date) start to detect preliminary changes in local large star (there is a list of candidates) that indicates it has or will go Supernova shortly. Of course Einstein gets in the way (AGAIN) as it could have gone BOOM already and we wouldn't know sans early signs until the first signs/wave front arrived. Giving humanity a precise period of time to prepare based on how many light years the chosen star is away and what advance warning (changes in the star) there are.

• While this is a hard science question the means by which the aliens trigger is can't be as we have no hard science answers to it. Thus we can't rule out the possibility that the aliens can trigger one in a too-small star. How about introducing a core collapse by means of artificial gravity? Squeeze hard enough, it's going to blow even if it's an itsy-bitsy red dwarf. – Loren Pechtel Jul 5 '20 at 18:43
• Good points, but about 'only most massive stars': according to wikipedia, also binary stars may be capable of a supernova. So, for ignition, the enemy aliens might throw another star on ours :-) – Roland Jul 5 '20 at 23:46
• An alternate might simply be to drop a 'very' tiny black hole (BH) into a local star. I think (need to check) that BH will then do what all BHs do - pull in all the matter available in its immediate locality growing as it does so. Once it has consumed the star it does not however have the mass available to sustain itself. The smaller the BH the shorter its life time. Having exhausted all available local matter it degrades shedding radiation/mass as it does. And the end is highly explosive - involving a series of intense gamma ray bursts. – Mon Jul 6 '20 at 3:48
• An alternate might simply be to drop a 'very' tiny black BH into a tight orbit around the outer edge of a local star. Dropping it 'in' won't help. That BH will then do what all BHs do - pull in all the matter available in its immediate locality growing as it does so. In this case starting with the outer layers of the stars structure and working inwards. Once the it has absorbed all available mass it will start to degrade. The smaller a BH is the shorter its life time. And the end is highly explosive - involving a series of intense gamma ray bursts. – Mon Jul 6 '20 at 3:59
• The problem is matching exactly the right sized BH with exactly the right sized star and exactly the right distance from Earth. All very complex. Plus you then have to wait for the whole system to reach its end point - which could be thousands of years. If the enemy civilization is really that high up the Kardashev Scale all they really need to do is map Earths Orbit around the Sun then send one or more small probes our way at a high % of C all timed to arrive when Earth is in the correct position. Even a at a low % of C we would only have at best a few days/hours warning. At highers speeds? – Mon Jul 6 '20 at 4:14

Of course it's possible. With the right assumptions just about anything is possible.

Assume: the planet is massive, but has a cold core and is made of something rigid, so it doesn't flow like a liquid under its own weight. And it's far from the exploding sun. And the race living on it has advanced technology with access to vast amount of unobtanium, which reflects neutrinos, doesn't conduct heat well, and is entirely shatterproof.

Then they tunnel to the core, build their shelter from unobtanium, install their hyperfusion reactors, load up on food sources, pile in and close the hatch. Dimly, they hear a boom, and then they just wait, because the environment outside their shelter isn't going to be healthy for quite some time. Occasionally they burp out any waste heat so they don't cook themselves. Time passes.

Eventually they open the hatch. What do they find? You're the author, so you get to decide, but it's reasonable to assume there's no planet left there anymore, or at least nothing that can support what we usually think of as life. So they'd do best to rig their reactor to generate propulsion and move their habitat to some other star. Volia, they (well, probably their descendants if they aren't immortal) survived.

If they have all that tech, though, they could drive off the attacking aliens with unobtanium bullets and green death-star rays and keep their planet safe that way.

Realistic? Of course not. But you posited aliens that can make a sun go supernova so you're already off the map, in terms of realism. There are a lot of simpler, entirely feasible ways of taking out life on a planet, if that's the goal. A large number of messy nuclear bombs will likely do it. Why get all high tech?