One of the theories behind the extinction of the dinosaurs is the eruption of one or a series of volcanoes. Supposedly the ash expelled by these eruptions blocked out the sun, killing plants and thus disrupting the entire food chain. A more recent example of the power of volcanoes on the environment would be the Little Ice Age, where the whole world felt colder for a while because of a volcano.

What I'm wondering, though, is if, some time in the near future, there was a volcano that ended most if not all life on this planet (humans can survive, but only because they're so smart), how big would this volcano have to be, and what would the eruption be like? Is there perhaps a place where a volcano would have an easier time of it? Is there maybe a volcano out there that fulfills all these criteria?

  • 2
    $\begingroup$ For all i know, yellowstone should qualify. Although i don't have nearly enough knowledge to explain why and how, let alone provide numbers. $\endgroup$
    – Burki
    Commented May 27, 2015 at 14:42
  • 2
    $\begingroup$ Yellowstone was a VEI-8 volcano, it would cause mass extinction for many species, but probably wouldn't end the world. $\endgroup$
    – ShemSeger
    Commented May 27, 2015 at 16:09
  • $\begingroup$ Intelligence doesn't necessarily guarantee survival. We are talking the Earth's largest volcanic eruption since its molten stage, after all. $\endgroup$
    – Frostfyre
    Commented May 27, 2015 at 16:21
  • 2
    $\begingroup$ "humans can survive, but only because they're so smart" - So you're writing a comedy then? $\endgroup$
    – user2547
    Commented May 27, 2015 at 22:06
  • 2
    $\begingroup$ @Lego Stormtroopr: No, it's obviously going to be a fantasy :-) $\endgroup$
    – jamesqf
    Commented May 28, 2015 at 4:12

4 Answers 4


If I remember right, the global temperature would only have to drop by 3 degrees Celsius in order to throw the world into a new ice age. In 1991, Mt Pinatubo in the Philippines erupted and lowered global temperatures by 0.5 degrees Celsius for a couple years. So in theory, if 6 volcanos of the same magnitude–a Volcanic Explosivity Index (VEI) of 6–were to erupt in the same year, it would likely plunge the world into a mini ice age, which would be devastating for many species.

A VEI-7 Volcano is ten times more destructive than a VEI-6. In 1815 Mount Tambora in Indonesia erupted, which resulted in the year without summer, during which 90,000 people froze to death.

A VEI-8 is an apocalyptic sized volcano, it would eject ten times as much material into the atmosphere (> 1,000 km³) which would effectively blot out the sun. The last VEI-8 volcano to erupt was during the Late Pleistocene era, and ushered in the Pleistocene or "Ice Age" extinction event that saw the extinction of many mammals that were larger than 40kg.

As far as I know there has never been a VEI-9 volcano, such a volcano would release more than 10,000 cubic kilometres of material into the atmosphere, it is doubtful that anything would survive such a catastrophic event.

So the answer to your question would likely be a volcano as destructive, or more destructive than a VEI-8.

To answer your question of where such a volcano would have to be - The obvious answer is where there is volcanic activity, which occurs over subduction zones (The ring of fire), and likely where there would be sufficient mass to blast into the atmosphere.

enter image description here

The largest volcanos we know about were all in the Andes, some were extremely powerful VEI-8's, close to VEI-9, like Tamarana which ejected an estimated volume of 8,600km$^{3}$ of material. It's Tamarana which is theorized to be a part of a series of eruptions that wiped out the last of the dinosaurs. If you wanted to pick a mountain to be your VEI-9 volcano, I'd pick Nevado Ojos del Salado It's currently the highest Stratovolcano we know about at 6,893 m (22,615 ft) this means there is a lot of earth under it, and gives it the best potential to eject 10,000km$^{3}$ of material high into the atmosphere.

  • $\begingroup$ The Earth would probably survive a VEI-9. ...How large would an eruption have to be obliterate the Earth? $\endgroup$
    – Frostfyre
    Commented May 27, 2015 at 16:23
  • 1
    $\begingroup$ @Frostfyre - Well, the Earth has a total volume of 1.08e+12km cubed. So a VEI-18 eruption would literally mean the Earth has exploded into dust. Is that obliterated enough for you? $\endgroup$
    – ShemSeger
    Commented May 27, 2015 at 16:50
  • 1
    $\begingroup$ 90,000 people died, which was roughly 9% of the human population of the world. The world's human population hasn't been one million since we were all in Africa... $\endgroup$
    – gilgamec
    Commented May 27, 2015 at 17:39
  • 3
    $\begingroup$ @gilgamec - I've never claimed to be good at math... $\endgroup$
    – ShemSeger
    Commented May 27, 2015 at 17:49
  • 1
    $\begingroup$ @LindaJeanne The Siberian Traps aren't rated using the VEI, they use a different scale (Dense-rock equivalent or DRE) because they were a different type of volcano. The Traps were a mafic eruption, basically they were a flood of lava that moved across land rather than an explosion of ash and dust that filled the sky. Mafic volcanos are much much larger in size, but have a much shorter reach and are much slower. The volume of lava that came out of the traps was in the millions of cubic kilometres, compared to the >1000 cubic kms of a VEI-8. $\endgroup$
    – ShemSeger
    Commented May 27, 2015 at 21:24

The best guess for volcanism-related extinctions relates to mantle plume eruptions like those seen at the Deccan Traps and the Siberian Traps - in both cases vast eruptions covered hundreds of square miles in lava releasing millions of tons of gas with devastating effect on the climate.

One interesting hypothesis is that these huge events may have been triggered by major asteroid impacts, the shock wave passing through the Earth's core and causing huge eruptions on the opposite side of the planet. This is far from proven, but you can see the mechanical sense of it. That would give you instant bang for your buck, followed by a lot more ongoing bang, with more general chaos than a Roland Emmerich movie.

The Siberian Traps is interesting because as well as being associated with the Permian Extinction Event - which is the big one as far as species loss is concerned - it also appears to have been worsened by the fact that the eruptions occurred in an area with fossil fuel resources, so in addition to the volcanic gasses released, there was a good deal of additional CO2 going into the atmosphere creating the same climatic effects that we are seeing now. Conveniently for a global extinction event, we are already releasing more of that than the PT event, so you could probably get away with less explosion and still wipe out the majority of life on earth. Although if you are basing your work on the hypothesis that humans are smart, you might want to gloss over that part...


Yellowstone is considered to be a possible future supervolcanic event that could lead to an Extinction Level Event. A past reference is the Toba catastrophy that at least in hypthesis lead to a very significant reduction of the number of human population between 69,000 and 77,000 years ago.

Supervolcanos eject volumes greater than 1,000 km3 (240 cu mi) according to Wikipedia. The referenced article also includes a nice list of late supervolcanos, their sizes and the amount of ejecta they produced.


It all depends on how obliterated you want the Earth to be.

If you want nothing left, you need to completely destroy Earth and accelerate all the remaining fragments so that they clear the area pretty quickly. Using some base numbers from this answer:

The gravitational binding energy of Earth, $E_{GB} = 2.4 \times 10^{32} \text{J} $.

The energy released by a VEI 5 event is 24 megatons. 1 megaton = 4.184PJ = $ 4.184 \times 10^{15} \text{J}$. Going on the exponential VEI scale of "ten times more destructive", we can estimate (this is not exact) the energy of a VEI 18 (as ShemSeger suggests) to be $ 4.184 \times 10^{28} \text{J} $. I shall call this number $E_{V18}$.

Let's see: if $E_{V18} = E_{GB} $, the explosion merely destroys the Earth and separates the fragments enough that they don't rebind gravitationally. If however $ E_{V18} \gg E_{GB} $, the planet explodes and is never seen again.

In this case, $E_{V18} \ll E_{GB} $ - so we need more energy. Approximately a VEI 22 event has enough energy to simply gravitationally unbind the planet; to completely splinter it off into space you want more.

Note that this is theoretical: firstly, it relies on perfect energy efficiency and distribution, which are not going to happen; and secondly, the Earth doesn't hold enough energy in total to cause an eruption this size.

(Looking at some collateral effects of that, fragments of this explosion are going to go shooting off at something like 12,000 metres per second... don't put anything in the way.)

Using the scale of 10 times more or less destructive as you move up or down the VEI scale, you can also work out roughly what level of destruction you'll get for a given VEI event. For comparison, a VEI 10 is pretty much guaranteed to cause mass extinction of almost everything.

  • $\begingroup$ Good math, but as a volcano, it just ain't gonna happen. Remember that the primordial Earth got hit by something roughly the size of Mars, and all it did was knock off a Moon-sized chunk. $\endgroup$
    – jamesqf
    Commented May 27, 2015 at 21:28
  • $\begingroup$ @jamesqf something the size of Mars, but not travelling fast enough to deliver sufficient energy to explode the planet. With perfect energy distribution, this volcano would destroy the planet. Of course, energy distribution is not perfect. I'll put something in about that. $\endgroup$
    – ArtOfCode
    Commented May 27, 2015 at 21:37
  • $\begingroup$ An eruption this size leaves no chance for the planet to so much as exist after the eruption. I think the OP is looking for an event that humans have the potential to survive because of their technology, but would wipe out all other life. $\endgroup$
    – ShemSeger
    Commented May 27, 2015 at 22:05
  • $\begingroup$ @ShemSeger if you want survival, you use the last paragraph. I provide the planet destroying explosion as comparison. $\endgroup$
    – ArtOfCode
    Commented May 27, 2015 at 22:08
  • $\begingroup$ @ArtOfCode: What I meant by "ain't going to happen" is that there isn't enough energy to cause such an eruption, let alone a plausible way to release it all at once. Back of the envelope figures: specific heat of iron = 452 J/kg/C (stone is less), core temp 5400 C, mass of Earth 6.0e24 kg. So total thermal energy is less than 1.5e31J, an order of magnitude less than you have for gravitational binding energy. $\endgroup$
    – jamesqf
    Commented May 28, 2015 at 4:33

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .