I need to wipe out a city. Not a huge one (I won't let them grow beyond a certain limit), but big enough to need to do some pretty serious damage. Let's say something similar in size to Pompeii. (yes... that Pompeii...)

My weapon of choice is a meteor. Carefully aimed, let it meander the heavens for millions of years and then ----blam!---- right on target at the appointed hour with no need of any further intervention.

Clean, clinical, and no side effects.

Only there are side effects, aren't there...?

Oh sure, a nice small meteor can make a beautiful explosion and a nice crater for itself without really having any effect on anyone outside the blast radius. But when you scale up, a bigger meteor will not only make a crater but also mess with the local or even global climate, maybe even trigger tsunamis, earthquakes, volcanoes...

Argh! I don't want any of that. All I want to do is wipe out a city with a giant space rock, but leave everything else untouched. No collateral damage. Is that really too much to ask?

Ideally, I'm hoping for a lone survivor to get away because he just happened to be leaving the city immediately before the event (uh... because I'll tell him to. Hope he listens!). The exact distance away isn't important, but he needs to be able to turn around at the right moment and see the impact close enough at hand to know exactly what happened but without getting caught up in it himself.

  • 1
    $\begingroup$ It would help if you put some numbers next to the adjectives. How big is a big enough city? What is significant secondary effect? $\endgroup$ – L.Dutch - Reinstate Monica Nov 18 '19 at 14:05
  • $\begingroup$ @L.Dutch-ReinstateMonica Okay, I've added a few extra details. :-) $\endgroup$ – Spudley Nov 18 '19 at 14:13
  • $\begingroup$ I'm avoiding actual numbers, as I honestly don't know what they'd be, but I placed the city size to be like Pompeii, and I added a paragraph explaining why I'm looking to localise the effects as much as possible. $\endgroup$ – Spudley Nov 18 '19 at 14:18
  • $\begingroup$ Would multiple smaller meteors be acceptable, or does it have to be a single rock? $\endgroup$ – Kat Nov 19 '19 at 0:44
  • 2
    $\begingroup$ Your survivor isn't called Lot by chance? ;-) $\endgroup$ – kutschkem Nov 19 '19 at 8:35

Keep it small

First, we need to know the area that must be destroyed:

Pompeii covered a total of 64 to 67 hectares (170 acres) and was home to 11,000 to 11,500 people -Wikipedia

So you need a radius of total destruction of only about 400 m. That's very small.

Second, let's look at the equivalent kind of nuclear detonation will have a 100 % casualty radius of about 400 m. For this, I used https://nuclearsecrecy.com/nukemap/

At a ground burst of 6 kilotons:

Air blast radius (20 psi): 400 m (0.49 km²)

At 20 psi overpressure, heavily built concrete buildings are severely damaged or demolished; fatalities approach 100%. Often used as a standard benchmark for heavy damage in cities.

(6 kilotons is wee. The Hiroshima bomb was 15 kt)

Finally, we need a bolide collision that generates about 6kt.

This impact calculator doesn't even go that small.

Happily, This one does.

Let's assume the bolide has a density of rock (2 g/cm³), a graze angle of 45 degrees, and a poky velocity of only 12 km/s (you can massage those numbers to get the effect you want). A 7 m (23 ft) diameter bolide will create a 6.1 kt explosion. It will create a 69 m diameter crater, and you can assume that everybody within the city wall would be killed by the blast effects and the buildings falling in on top of them.

For fun, let's bump it up to 10 m (33 ft) and 25 km/s. That's 78 kilotons, the crater is 150 m, and you just blew down those sturdy city walls and incinerated the fields and orchards outside. Be careful of going too big or too fast.

Extra credit: Lots of meteors explode or fragment before reaching ground, so that 7 m lump might be a fragment of something larger.

  • 1
    $\begingroup$ The problem with small meteors is that they tend to break up in mid-air. The Chelyabinsk meteor, for example, was twice the diameter of yours, and broke up high enough to cause almost no surface damage. $\endgroup$ – Mark Nov 18 '19 at 23:36
  • $\begingroup$ Answer currently seems to be coming at the question from the opposite direction of that asked. Rather than "What's the smallest" [and then fiddling with finding stuff with a high enough density to ensure it gets deep enough through the atmosphere to do its job...] the question seems to be asking "How big can I go without being 'too big'?" $\endgroup$ – TheLuckless Nov 18 '19 at 23:47
  • $\begingroup$ @mark this depends more on what the meteor is made out of. In fact that matter a lot more than size. $\endgroup$ – John Nov 19 '19 at 6:10
  • $\begingroup$ So it seems the most tricky part is to get a meteor which will stay in one piece when falling through the earth athmosphere. $\endgroup$ – quarague Nov 21 '19 at 10:46

Let's say something similar in size to Pompeii.

This sounds like a job for nukemap.

15kt surface blast on Pompeii

Here's an example detonation of a 15kt yield device, equivalent to the "Little Boy" device used on Hiroshima. The effects of a surface blast are shown, without showing thermal effects which will be quite different for a meterorite impact than for a nuclear explosion.

As you can see, the inner grey circle (the 5psi air blast radius lies at about 1.13km from ground zero, a little over two thirds of a mile. Within that radius, you can reasonably expect most residential buildings to be pretty badly smashed up. The outer grey circle is the 1psi overpressure radius where things like glass windows can still be smashed and lies at about 3km. If your observer was outside this radius, they've a good chance of surviving alive and intact.

There's a wealth of information on nuclear blasts of this magnitude on the interwebs, including videos. In all cases you can see that whilst 12kt is a pretty fearsome amount of destructive power, it pales in comparison to larger bombs or natural disasters like volcanic eruptions. You need not expect climatic change or huge chunks of debris being thrown for miles.

What's more difficult though is to work out exactly what kind of impactor would cause this very specific kind of devastation.

Fiddling with the Earth Impact Effects Program (also referenced by L.Dutch above), you may find (as I did) that its trick to find a set of parameters that allows a significant amount of damage to the ground within the desired sort of radius, without having a massive airburst (widely distributing the damage, or expending too much of the energy too high up).

I came up with this set of parameters, which involves an iridium asteroid (unlikely, but not impossible) to concentrate a lot of mass into a small volume. Note that the EIEP authors caution against using densities greater than iron, as they may result in unrealistic object strengths, caveat impactor etc. Iron asteroids (much more plausible) tended to break up, or deliver insufficient oomph. There will also be differences between the calculations used by nukemap and EIEP so the results aren't entirely comparable, but then hitting rocks together really hard isn't the same as a nuclear fireball, so there you go.

The exact details will depend on the effects you want. A low-altitude airburst (like this one using a smaller iron meteorite) would look neat, and provide a shotgun-like effect that distributes the damage across an area destroying everything underneath it but not producing as much collateral damage. An object that arrives intact, on the other hand, will leave a big bad crater behind. You can decide what you'd prefer.


I think you can bail out of this problem with an object similar to the one which created the meteor crater

Meteor Crater lies at an elevation of 5,640 ft (1,719 m) above sea level. It is about 3,900 ft (1,200 m) in diameter, some 560 ft (170 m) deep, and is surrounded by a rim that rises 148 ft (45 m) above the surrounding plains.

Though the crater itself is smallish, the impact had consequences on a larger area:

The object that excavated the crater was a nickel-iron meteorite about 160 feet (50 meters) across. The speed of the impact has been a subject of some debate. Modeling initially suggested that the meteorite struck at up to 45,000 mph (20 km/s) but more recent research suggests the impact was substantially slower, at 29,000 mph (12.8 km/s). It is believed that about half of the impactor's bulk was vaporized during its descent through the atmosphere. Impact energy has been estimated at about 10 megatons. The meteorite was mostly vaporized upon impact, leaving few remains in the crater

The vaporization of the meteor would create a nice heat blast, and the following impact and debris deposition would wipe out a small city like the Pompeii you refer, without any dramatic large scale effect.

The impact created an inverted stratigraphy, so that the layers immediately exterior to the rim are stacked in the reverse order to which they normally occur; the impact overturned and inverted the layers to a distance of one to two kilometers outward from the crater's edge

If you want to get a clue of what the impact felt like at various distances, you can use one of the many online impact calculators. I usually refer to this

  • $\begingroup$ Interesting. Especially that bit about the inverted rocks. How far away from this impact would our lone escapee have to be in order to avoid being... uh... inverted? What kind of view would he get of the destruction? $\endgroup$ – Spudley Nov 18 '19 at 14:36
  • $\begingroup$ @Spudley, see my last edit $\endgroup$ – L.Dutch - Reinstate Monica Nov 18 '19 at 14:42
  • $\begingroup$ Oh sweet... I love that calculator. I shall be playing with that for some time! I think that will give me a lot of ammunition (hehe) to play with. Thank you. $\endgroup$ – Spudley Nov 18 '19 at 14:43

House-sized meteor. City-sized crater.

A good starting point is the nuclear bomb that dropped on Hiroshima. The blast was city-sized as opposed to country-sized. It had about $6 \times 10^{13}$ Joules of total energy.$^1$ So we know that's not enough for significant secondary effects.

We want a meteor with a similar energy total. The kinetic energy of the meteor is $\frac{MV^2}{2}$ for mass $M$ and velocity $V$. We want $\frac{MV^2}{2}=6 \times 10^{13}$. This paper (page 1214 and onwards) says meteors enter the atmosphere at $20$-ish kilometres per second or $V = 2 \times 10^4 m/s$. Squaring that we get $V^2 = 4 \times 10 ^8$. Plugging that in we get

$$\frac{M }{2} (4 \times 10 ^8) = 6 \times 10^{13}$$ $$ M (2 \times 10 ^8) = 6 \times 10^{13}$$ $$ M = 3 \times 10^{5}$$

As usual $M$ is measured in kg so the mass should be $300,000$ kg. In other words $300$ tons. So we're looking at a house-sized meteor.

The second question is about your survivor is more about the shape of the blast. I would imagine the amount of material thrown into the air will be MUCH larger than the city. So "viewing the explosion from the top of a nearby hill" is out of the question. More realistic is "caught on the edge of the blast front." Something like this but on a smaller scale, where a tsunami of dust and soil moves outwards from the impact. Granted our meteor is MUCH smaller. But the survivor is also MUCH closer. (the dinosaurs in the video are on the other side of the planet).

$^1$ Granted only a small portion was directed into the city. I imagine the meteor will be much more efficient if it physically impacts the city. Some energy will be burnt off in the atmosphere but I don't know how much.

Edit: The impact calculator talks about something called an airburst, which is the explosion caused by the meteor smashing into the thickest part of the atmosphere. It seems larger bodies airburst closer to the surface and lost proportionally more of their energy in the burst. That means larger impacts are less focused towards the ground.

  • 1
    $\begingroup$ For small impactors, actually traversing the atmosphere is surprisingly difficult. Even if they do make it down intact, you'll lose a lot of energy to the atmosphere. The various impact effects calculators can help inform you about these problems, though I'm increasingly tempted to try and make my own... $\endgroup$ – Starfish Prime Nov 18 '19 at 15:19
  • $\begingroup$ I'm sure the amount of energy burnt off in the atmosphere is indeed a lot. But I don't know how to compute that or if the impact calculators are any good. $\endgroup$ – Daron Nov 18 '19 at 16:50
  • $\begingroup$ You'll find that things like the Earth Impact Effects Program come with explanations of their function and the academic papers and data sources used for the algorithms, so there's reasonable scope for assessing them yourself. $\endgroup$ – Starfish Prime Nov 18 '19 at 17:04
  • $\begingroup$ Seems to give the same order of predictions as me: impact.ese.ic.ac.uk/ImpactEarth/cgi-bin/… $\endgroup$ – Daron Nov 18 '19 at 17:38
  • 1
    $\begingroup$ The link you gave for your example has the airburst happening so far up that the pressure wave when it reaches the ground won't even be enough to break glass. The usual suspects give that as a 1psi overpressure, and your blast is 20-40 times too weak. $\endgroup$ – Starfish Prime Nov 18 '19 at 17:57

As noted, Barringer (Meteor) Crater provides an easily observable benchmark: a 1.2 km crater would be the result of an explosion large enough that you could assume everything within tens of kilometers would be flattened, and right there is even a very large modern city.

For another comparison, Tunguska. The airburst was somewhere between 10 and 30 megatonnes (estimates vary), but flattened roughly 2100 km2 of forest despite exploding at between 5 and 10 km above the surface; that type of overpressure would devastate the majority of urban structures near the center of the explosion. The Castle Bravo test accidentally went into that range, and it vaporized instruments and destroyed bunkers designed to be hit by a nuclear explosion, so you can image what normal civic infrastructure would look like.


Something like the Tunguska event might be suitable. I believe directly under the blast trees were left standing so someone in the city in a deep basement might well have survived https://en.wikipedia.org/wiki/Tunguska_event

  • $\begingroup$ The Tunguska event was...not small. $\endgroup$ – Mark Nov 18 '19 at 23:39
  • 1
    $\begingroup$ But could be scaled down $\endgroup$ – Slarty Nov 19 '19 at 0:20

You could potentially get away with destroying a larger city with a smaller meteor if you took advantage of a knock on effect from something else being destroyed.

An example of this: A very large dam upstream of the city is destroyed by the meteor, causing a catastrophic flood that washes most of the city away.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.