How large would an asteroid have to be to completely sterilize Earth?

[m1omg]

Now, I know such an event cannot occur naturally today (all truly big asteroids being on stable orbits far way), but how big would an asteroid have to be to actually sterilize Earth? The KT impact was around 6-17 km, and there is a possibility that the Permian-Triassic extinction might have been caused by an asteroid as large as 55 km (based on the probable crater found in Antarctica), so where the threshold really is? 100 km, 200 km?

[antoniseb]

I think it would take something huge to kill off all the life around the deep-sea vents.

[Ara Pacis]

I recall reading that some studies suggest that life evolved before the moon was created. So even an impact with a Mars-sized asteroid may not do it.

[PlutonianEmpire]

I recall reading that some studies suggest that life evolved before the moon was created. So even an impact with a Mars-sized asteroid may not do it.

That makes sense, actually, because I recall reading a few years ago about some studies suggesting that the "global magma ocean" may in fact have never existed, or something like that, even with the Mars-sized impactor. Granted, I read that a few years ago, so my memory is a little vague on that.

[mutleyeng]

before the moon was created?
how old does this study say the moon is?
I know opinion is still divided on the precise mechanism for the forming of the moon, but whatever way you look at it the event happened 4.5Ga - within 10s of millions of years of earths formation (prior to the final big smash).
I have heard people suggest life may have started before the end of the late heavy bombardment - but before the moon?

[Ara Pacis]

There appears to be evidence that there was never a magma ocean. If so, then perhaps there were locations where life may have survived. Until it can be ruled out, it may be considered a possibility.

[Rhaedas]

Oh, and never say never. Just very, very improbable. Something big enough in a somewhat regular orbit and close enough we would already know about. So that leaves things coming into the system at a much more direct path, and the probability of such a path intersecting the Earth's is unlikely.

However, we do need to watch for the smaller ones, and learn what we can do to stop them. They won't kill off life or likely even affect someone not in the direct area, but if a large population did get hit, it would still be tragic.

At least we can probably do something about the small ones. One the size the OP is discussing, you aren't going to do much to change where it's going.

[Romanus]

Given that life has been found not only at deep sea vents, but also well beneath the seafloor itself, and deep underground in pore spaces, I don't think anything in the Solar System today could do it. The best way to figure this out would be to model impacts for a range of impactor masses and speeds that would vaporize the Earth's oceans and melt the crust; as the OP said, something like that wouldn't have happened outside of the earliest Solar System.

[Ara Pacis]

However, if we consider extra-solar impactors with high masses and higher impact speeds...

[Cougar]

However, if we consider extra-solar impactors with high masses and higher impact speeds...

Right. It's not just "how large" the thing is, but also its relative velocity. Obviously, if an earth-sized object struck us "head on" at 29.78 km/s, that would stop earth in its orbit, and earth and the object would promptly fall into the sun, where no deep-sea vents are known.

[m1omg]

By the way, a different question but how would a K-T sized impact look from space? And a bigger, 60 km impactor impacting at 25 km/s (roughly equivalent to 1 Petaton of TNT)? And a 100 km impactor at 25 km/s? And how would the Earth look shortly after the impact?

[Ara Pacis]

http://simulator.down2earth.eu/

http://impact.ese.ic.ac.uk/ImpactEffects/

http://www.purdue.edu/impactearth/

[m1omg]

http://simulator.down2earth.eu/

http://impact.ese.ic.ac.uk/ImpactEffects/

http://www.purdue.edu/impactearth/

Thanks, but I already know these 3 calculators. I want to have a visual idea, not just "if you're distance xxxx km from the impact, you're gonna get fried by a firebal xxx km wide and crushed by a pressure wave xxx psi strong".

[AK]

Norm Sleep and others wrote a Nature paper on this topic in 1989, but his calculations neglected the [then unknown] deep biosphere.

[Bynaus]

based on the probable crater found in Antarctica

No one should even think of calling that thing a "probable crater". Its a roundish structure, identified with radar under the ice, with an associated gravity anomaly. But that doesn't mean anything. The most recent claim on this being a crater was not even a peer-reviewed publication, only a conference abstract.

I recall reading that some studies suggest that life evolved before the moon was created.

I strongly doubt that this suggestion is rooted in reliable observations. The Moon was certainly formed within the first 130 Myrs after the first solids condensed in the solar system, as this is the age of the oldest lunar rocks found so far. The oldest reliable fossil remains of life are about 3.5 Ga old, with indications that habitable conditions (i.e., liquid water) had existed much earlier, probably since ~4.4 Ga. So life could have evolved very shortly after the Moon forming event (within ~100 Myrs), but nothing so far clearly demonstrates that it has.

On the OP, whatever melts the entire crust (and the upper mantle) will also quite probably sterilize the planet. A Mars-sized impactor would certainly do. On the other hand, all you need is the post-impact runaway greenhouse-effect to be strong enough (peak temperatures high enough) to decompose carbonates, stabilizing the greenhouse and thus lose the oceans. A Venus-on-Earth-scenario would probably pretty much sterilize the planet.

[Ara Pacis]

On the OP, whatever melts the entire crust (and the upper mantle) will also quite probably sterilize the planet. A Mars-sized impactor would certainly do. On the other hand, all you need is the post-impact runaway greenhouse-effect to be strong enough (peak temperatures high enough) to decompose carbonates, stabilizing the greenhouse and thus lose the oceans. A Venus-on-Earth-scenario would probably pretty much sterilize the planet.

That was the point: a Mars-sized impactor might not necessarily do that, according to some.

[Bynaus]

That was the point: a Mars-sized impactor might not necessarily do that, according to some.

But based on what evidence? As said above, the oldest reliable signs of life are much younger than than the oldest Moon rocks. In other words: there is simply no reason to assume that a Mars-sized impactor would NOT sterilize the planet.

[Ara Pacis]

But based on what evidence? As said above, the oldest reliable signs of life are much younger than than the oldest Moon rocks. In other words: there is simply no reason to assume that a Mars-sized impactor would NOT sterilize the planet.

From what I've read there's evidence or lack of evidence that suggests there might not have been a planetary magma ocean.

Is There Any Evidence that the Earth Ever
Had a Magma Ocean? No. Another anticipated
result of the Giant Impact is a terrestrial magma ocean
[3]. This melting event was likely not an opportunity
to homogenize but, rather, an opportunity to differentiate.
And because there are mantle spinel lherzolites
whose compositions closely approximate that of the
bulk silicate Earth, this seems to imply that there was
never a global magma ocean

Perhaps you can sterilize without making a magma ocean. Go ahead and try to make that case.

If we don't have rock on earth with signs of life older than a certain age, we can conclude one of two things: there was no life on those rocks and therefore not on Earth, or there was no life on those rocks but there was life on other rocks on Earth. The latter might hold if there was a systemic bias for life to reside on rocks that did not survive and was recycled. This might happen on oceanic rock that gets subducted and recycled if life tended to only exist at that time near deep hydrothermal vents.

But that's speculation that is not important to answering the question at hand as to what size impactor is necessary and your reframing of the question as melting the crust and mantle: the answer to which seems to be not Theia, and therefore not a Theia sized object, a class that includes Mars.

[Bynaus]

I don't see how the existence of some spinel lherzolites with a composition similar to the BSE would exclude that there ever was a magma ocean. It is correct that it is debated among geologists whether the Earth ever had a magma ocean, but this discussion is far from settled (see, e.g., here, a recent article in Nature advocating the existence of an early magma ocean: http://adsabs.harvard.edu/abs/2006Natur.441..825W).

If there ever was a Giant Impact, then there was also a magma ocean. On that, all the models agree: there's no way around it.

This might happen on oceanic rock that gets subducted and recycled if life tended to only exist at that time near deep hydrothermal vents.

Yes. On the other hand, these subducted rocks might just as well have been completely devoid of life.

Given that there is NO evidence at all for life before the LHB (or even before the Giant Impact), it makes no logical sense at all to take the mere assumed possibility of life existing there to claim that a Mars-sized impactor would NOT sterilize the planet, don't you think? As said above, all the models predict a magma ocean after the Giant Impact, and I think we agree that a magma ocean would sterilize the planet.

Find me some rocks with signs of life that are older than the Giant Impact (or at least a plausible source claiming they have found these rocks) - and I will admit that I was wrong.

With regards to the OP: Every impactor that creates a global magma ocean will completely sterilize the planet. It takes much less than a Theia to do that, I guess a Vesta-sized (500 km) body would do as well.

[m1omg]

Can it be calculated how big an impactor/how much energy would be required to create a global magma ocean?

[Bynaus]

The second link above gives you some indications. For a 2000 km projectile, they indicate that 1.4% of the Earth would melt due to the impact - that is probably enough for a relatively shallow global magma ocean. A Giant Impact-type impact (6000 km projectile) leads to 38% of the Earth melting.

[Ara Pacis]

I don't see how the existence of some spinel lherzolites with a composition similar to the BSE would exclude that there ever was a magma ocean. It is correct that it is debated among geologists whether the Earth ever had a magma ocean, but this discussion is far from settled (see, e.g., here, a recent article in Nature advocating the existence of an early magma ocean: http://adsabs.harvard.edu/abs/2006Natur.441..825W).

If there ever was a Giant Impact, then there was also a magma ocean. On that, all the models agree: there's no way around it.



Yes. On the other hand, these subducted rocks might just as well have been completely devoid of life.

Given that there is NO evidence at all for life before the LHB (or even before the Giant Impact), it makes no logical sense at all to take the mere assumed possibility of life existing there to claim that a Mars-sized impactor would NOT sterilize the planet, don't you think? As said above, all the models predict a magma ocean after the Giant Impact, and I think we agree that a magma ocean would sterilize the planet.

Find me some rocks with signs of life that are older than the Giant Impact (or at least a plausible source claiming they have found these rocks) - and I will admit that I was wrong.

With regards to the OP: Every impactor that creates a global magma ocean will completely sterilize the planet. It takes much less than a Theia to do that, I guess a Vesta-sized (500 km) body would do as well.

Okay then, show me an earth-sized that had life and got smacked by a mars-sized object that developed a magma ocean and sterilized the planet. We can do this all day.

[m1omg]

Ok, so if say, an 80 km asteroid impacted the Earth at 20 km/s, what would be the result? Would all multicellular life go extinct or just a hyper-Permian extinction with multicellular life fully recovering after a few tens of millions of years?

[m1omg]

I just watched a part of an interesting series that includes the discussion of giant imacts during Earth's history:

http://www.youtube.com/watch?v=LJ0jBq-hUQ0
http://www.youtube.com/watch?v=P2-LUuS7gEg
http://www.youtube.com/watch?v=2TXptnwVifk
http://www.youtube.com/watch?v=VV3qsop3qRQ

It was calculated that not even a 500 km impact would create a magma ocean, with the searing heat penetrating only a thin layer on the surface and a life zone would still remain between the red hot top layer and the mantle. It is suggested that life actually survived 6 such impacts, each occuring billions of years ago, but already after life has already firmly taken hold, with moon already formed and all.

[jfribrg]

Remember that even if life did originate before a moon-creating impact, that doesn't mean that it survived the impact. I've seen some speculation that perhaps life originated many times only to be possibly wiped out after the next huge impact. If conditions favorable to the development of life persisted after each of these impacts, there is no reason why life couldn't evolve multiple times, and could perhaps evolve independently in several places at the same time. According to this speculation, only after the bombardment period was over did life become a permanent feature of Earth.

[m1omg]

This series is not about the moon creating impact. That one was a Mars sized object, this one deals with latter, smaller, but still incredibly high energy impacts. The seventh minute of this video http://www.youtube.com/watch?v=2TXptnwVifk deals with the survival of microbes underground. Presumably this talks about the post-Late Heavy Bombardment period, considering the LHB period had apocalyptic impacts every century or so. This is about the Archean, not Hadean.

[Bynaus]

One should not forget that there might be more ways to sterilize the planet through an impactor than through a magma ocean. E.g., if an impactor induces a (lasting) runaway greenhouse, the surface will be sterilized as well (although one might make the case that stratospheric microbes would survive even this).

We can do this all day.

I am not the one making the extraordinary claim that requires the extraordinary evidence...

You can download animations of the Giant Impact from Prof. Robyn Canups website (she is probably the worlds most reknowned expert on that subject). I think its pretty obvious that such an impact would melt not only the crust, but also most of the mantle. Actually, that is also the single most important reason why we cannot know if life had already formed before the Giant Impact - there are simply no rocks left from that time. Life, if it had already formed at that point, would certainly not have survived this.

[Noclevername]

I hate to sound like a panspermist, but could some microbe-bearing fragments have been flung into orbit and the life in them survived long enough to "re-infect" the Earth?

Which leads to the question, how long would it take for a global magma ocean to produce enough solid crust and liquid water for a returning extremophile to survive there?


Edit: How much radiation the meteor was exposed to would depend on whether it orbited below the planet's magnetosphere or into the Van Allen belts.

[Ara Pacis]

I am not the one making the extraordinary claim that requires the extraordinary evidence...

Neither am I. I'm not actually making any claims, I'm merely pointing out that other claims might be either flawed or incomplete or unsupported.

[Trakar]

I hate to sound like a panspermist, but could some microbe-bearing fragments have been flung into orbit and the life in them survived long enough to "re-infect" the Earth?

Which leads to the question, how long would it take for a global magma ocean to produce enough solid crust and liquid water for a returning extremophile to survive there?

Edit: How much radiation the meteor was exposed to would depend on whether it orbited below the planet's magnetosphere or into the Van Allen belts.

Your last edit zips up a timeframe. Orbits below Van Allen belts are generally only stable for periods of centuries at most. Inner VA belt 1000km to 5000km.
http://www.orbitaldebris.jsc.nasa.gov/faqs.html#12

I doubt that there is a high likelihood of ejecta maintaining earth orbit for a sufficient period to allow crustal reformation and liquid surface water presence while not being thoroughly sterilyzed by orbital (earth/solar) radiations. I wouldn't deem it impossible, but would find it improbable and generally implausible as a proposed mechanism. Easier, IMO, to consider that while generalized energy calculations indicate the release of energies sufficient to bring the crust and much of the mantle to a roiling boil, this presumes that such energy is released in a more distributed fashion than is consistent with typical impact senarios. I would be more inclined to look at surviving deep, hot rock niches and their extremophile biota in isolated nursery regions, if I were seeking more plausible "carry-over" life senarios.