Nuke Apophis?

[GeorgeK]

I've been watching the History and Discovery channels' shows about asteroids and all the physicists laugh or even cackle at the concept. However, they all seem to have the same assumption, that someone would nuke it when it is on its way TOWARD Earth. What would be wrong with watching it to see if it goes through the "keyhole" and if it does, then nuke it on its way AWAY from Earth, you know, for maximum vector effect? Obviously you'd have to have the nuke platform already out there when Apophis pokes its face around next time.

[efkelley]

Well, you have to remember that an orbit is circular. Nuke it here or there and the pieces are still going along the same path. If the notion is to break it into a lot of smaller chunks that can do no harm to Earth, then a nuclear device is not the way to do that.

Far simpler to move it with thrusters, mirrors, or gravity.

[Lhun]

The idea is to use the nuke as a thruster. Lifting a rocket thruster to an asteroid would be hugely complicated and also nearly ineffectual at moving such a huge object. Nuclear powered thrusters like a NSWR or maybe just a fission heated steam thruster might work, but would only be really efficient if the asteroid has enough ice to serve as reaction mass.

[DaveK]

Hitting a moving object with another moving object takes a lot of getting things just so. So if something is going to hit the earth almost any significant push will cause it to miss. Unfortunately in space it loops around and comes at you again. Fortunately space is big so it usually misses again. It doesn't matter too much when you shove it, hitting it when it approaches may be easier because you don't have a long stern chase to catch up. If your pushing technique involves a gravity tug or landing a rocket on the object then you have to match velocities and coming or going is the same.

The orbits can be calculated fairly accurately but not perfectly. Saying an object passed through a keyhole is saying that given what we know about it's orbit and other masses in space the object will probably impact earth. Since it is easier to measure the orbits of objects when they are closer and many will make a close pass to earth before actually colliding these keyholes are usually near earth.

[GeorgeK]

I realize it would take some tech and training to do the job, but the question is what's wrong with the job? A nuke is obviously not as precise as a thruster, but can be used to cause a change in vector, right? The further you are from a target, the less change of vector is needed to miss the target. Ergo, as the rock passes Earth is the point of maximum vector effect with regard to altering its path. If it is such a hard thing for an asteroid to do, to hit Earth, then why would messing with its trajectory be so difficult, particularly if you are dealing with something that shows up every 7 years or whatever it is? We (Meaning scientists) have shot asteroids and at least once landed a spacecraft on a comet, so we know that it is possible.


Some of the mass of the rock will be cast off out of the same orbital path and a mass ov x-1 is still less than x.

Or is the problem that the margins of error are too great to predict anything? But still, even if it was that, should it get to some theoretical limit of a 1 in four chance of Apophis or some other hunk of rock in the shooting gallery called the Sol System, hitting the earth and causing an extinction level event, What would be wrong with having the back up plan in place? At some calculated odds the science committee approaches the Sec General of the UN and says, "It's time."

[Lhun]

Most likely the objective is simply money. Nukes are expensive, and so's sending stuff into interplanetary space. The space shuttle couldn't do that.
As for precision, that's not really necessary. Since we're not launching the asteroid from earth, the fact that orbits are circular doesn't really matter. We could just nuke the asteroid and see if it's enough to move it out of the way, there's hardly any chance of making things worse, but that would cost billions of dollars to try even once.

[DaveK]

It's not that getting a nuke to an asteroid is hard it is that the asteroid is big and the amount of energy needed to change it's path even a little is large. I don't think it is how close in miles to the earth that matters but how long in time your change of velocity has to work.

If the rock is 10 years away to move it an earth diameter away in that time would be 1/10 mph. (Mental math and heavy rounding.) But a rock one mile in diameter is massive so I would bet it takes more than one nuke.

Moving it vs. blowing it to pieces is a question. If it is iron then the nukes would probably move it. But if it is an accumulation of rocky bits then the nuke would probably fragment it. Depending on how close you explode the nuke.

Do you want to get hit by fragments or a solid piece? If you blow it up far away then many of the pieces would probably miss earth so that is good. If it's close then all the pieces would hit and the effect would be similar. I make an analogy of getting shot at with a shotgun. If the shooter is far away then you would like for him to have a slug, there is a better chance of him missing then if he had buckshot. If you can blow it up really good then it is like birdshot and that may cause a little damage but not fatal. If he's close than it really doesn't matter.

[Lhun]

Originally Posted by DaveK

It's not that getting a nuke to an asteroid is hard it is that the asteroid is big and the amount of energy needed to change it's path even a little is large. I don't think it is how close in miles to the earth that matters but how long in time your change of velocity has to work.

In the end it's just a question of total dV you can impart on it. Slow acceleration over time might be more efficient, but nukes are the tool of choice because there's simply nothing else that comes even close to producing that amount of energy.

Moving it vs. blowing it to pieces is a question. If it is iron then the nukes would probably move it. But if it is an accumulation of rocky bits then the nuke would probably fragment it. Depending on how close you explode the nuke.

To get any effect, you'd detonate the nuke inside the rock. Detonating it at a distance will have pretty much no effect. The idea is to vaporize part of the asteroid and use that as reaction mass to move the rest.
Fragmentation is close to an ideal situation actually. The fragments will spread out at quite a speed, so unless it's done extremely close to earth there's no chance of more than one fragment hitting, and a couple of dozen fragments have no significantly higher chance of scoring a hit than a single one, given the size of the target. If the asteroid fractures into a lot of small parts, it'd also be comparatively easy to deflect one of the small fragments with a similar nuke to the one that fractured the original one.

Most of that is theoretical of course, since stuff like this is far to chaotic to be figured out completely even with modern simulations. We should really test the principle on some unrelated asteroid to have some empirical data in case we actually have to move one in the future, but again it's a question of money.

[veinglory]

The model I saw involved attaching a devive to the asteroid that pounded bits off the rock and then shot them into space to created propulsion. That would seem an efficient way to do it.

[Lhun]

Originally Posted by veinglory

The model I saw involved attaching a devive to the asteroid that pounded bits off the rock and then shot them into space to created propulsion. That would seem an efficient way to do it.

Yes, it's using the asteroid itself as reaction mass. But the fuel is still a problem, and there's nothing that comes even close to fusion or even fission in energy production. And so far, we can only utilize those in bombs.

[FOTSGreg]

Maybe one of the most efficient means of deflecting an asteroid appears to be using a "gravity tractor". Essentially you park a space craft near the asteroid and let the spacecraft's own gravity deflect the rock over a span of time. The drawback seems to be that the spacecraft's mass would be minuscule by comparison and thus the actual deflection would also be minuscule unless you could keep it parked for what looks to be years at a stretch. In addition there's the matter of keeping the spacecraft in position to deflect the asteroid the way you want it to go. That's going to require reaction mass or very, very precise placement of the craft near the object, matching speeds, etc. Either way, a gravity tractor looks like it would need a long time to work.

Of course, all you really need is a slight deflection at range to keep the thing from smacking Earth.

[whacko]

I think discovery is the big word. It's hard to nuke something that nobody knows is on a collision course.

And that's before we even consider its size.

[Astronomer]

Without an atmosphere to create a blast wave, a nuke in space wouldn't have much impact on an asteroid's trajectory. Although the energy is impressive, the momentum-transferring mass is extremely tiny (when compared, say, to a rocket booster). Moreover, since the explosion is spherical, only half the energy is directed in the asteroid's direction, and much of that is tangential to the direction you want to push it.

If you bury the nuke and try to blow the asteroid apart, then you're trading the devil you know for the devil you don't know. It's impossible to determine how many pieces will result from such an exercise and what their trajectories would be. You could make things far worse.

It's been argued that blasting an asteroid into smaller chunks would "soften" its blow. And to some extent, this is certainly true. Large thunderstorms release many times the energy of the first atom bombs, but because this energy release is spread over a large area over a large amount of time, the damage is most often negligible.

But everything is a matter of degree. Our planet will happily endure the energy of a puny atom bomb spread over a large area and not even notice. But if an impactor has the kinetic energy sufficient to end civilization, then civilization will end, no matter how that energy is distributed.

My preferred method of deflecting Apophis is to splatter it with buckets of white paint. Painting it white will make it subject to photon pressure from the sun, slightly altering its course. Over time, this could be enough to save us. Of course, this method works only if you have lots of time (as we do with Apophis), and only if pushing it further from the sun is the direction it needs to go to prevent disaster.

[Lhun]

Originally Posted by Astronomer

If you bury the nuke and try to blow the asteroid apart, then you're trading the devil you know for the devil you don't know. It's impossible to determine how many pieces will result from such an exercise and what their trajectories would be. You could make things far worse.

You don't have to either detonate at the centre of the asteroid or at a distance. The idea is to detonate the nuke only a few meters below the surface, creating an improvised rocket thruster.

My preferred method of deflecting Apophis is to splatter it with buckets of white paint. Painting it white will make it subject to photon pressure from the sun, slightly altering its course. Over time, this could be enough to save us. Of course, this method works only if you have lots of time (as we do with Apophis), and only if pushing it further from the sun is the direction it needs to go to prevent disaster.

Even if you mirror the whole surface, you could at most double the current pressure caused by the sun.

[Manuel Royal]

Originally Posted by Lhun

Yes, it's using the asteroid itself as reaction mass. But the fuel is still a problem, and there's nothing that comes even close to fusion or even fission in energy production. And so far, we can only utilize those in bombs.

Well, we use fission in reactors, and avoiding an asteroid strike might justify putting a friendly little pile into space.

(We currently have a number of offworld robots powered by radioisotope thermoelectric generators, but I don't know if you could get enough juice for our needs that way.)

Your robot could settle on the naughty asteroid, then start pushing in one of a few ways. You could use an ion thruster (like a magnetaplasmadynamic thruster carrying solid lithium for propellant).

Or -- I like the Wile E. Coyote directness of this -- your robot could bore out chunks of asteroid material and, through a pneumatic or even spring-loaded mechanism, simply throw them in a desired direction, like a pitching machine. (I just thought of this, which surely means somebody else has already come up with a detailed plan for doing it.)

A mile-wide asteroid has a very low escape velocity. Let me check: mile in average diameter; even assuming it's mostly iron ... whoa. Less than 2 meters per second. You could hock a loogie clear off the asteroid. (Wouldn't recommend it.)

Now, if it was a comet you were concerned about, you'd have the advantage of the nucleus basically being partly made out of propellant suitable for either ionic or chemical rockets. Cometary nuclei are natural spaceships when they're close enough to the Sun for heat to boil off frozen volatiles. (Here's a great picture of this happening with Comet Hartley 2.)

Note: orbits aren't circular.

ETA: It occurs to me that you'd probably want to use a mass driver to hurl your chunks of asteroid away. But for some reason the idea of doing it with something I could build on my porch appealed to me.

[Manuel Royal]

Originally Posted by Lhun

Even if you mirror the whole surface, you could at most double the current pressure caused by the sun.

True -- but I think we're talking about the Yarkovsky effect, not direct radiation pressure.

[Lhun]

Originally Posted by Manuel Royal

Well, we use fission in reactors, and avoiding an asteroid strike might justify putting a friendly little pile into space.

(We currently have a number of offworld robots powered by radioisotope thermoelectric generators, but I don't know if you could get enough juice for our needs that way.)

While the various nuclear battery designs have a theoretically very high energy density (that's why they're being used on mass-critical missions) it takes years and years for them to overtake the absolute energy produced by something such as rocket fuel. Ideal for powering something that requires very low power, but runs constantly for decades, but insufficient for anything that needs to happen now.
Could work out for moving an asteroid if we just start early enough. (Still much less energy than a nuke though)

Or -- I like the Wile E. Coyote directness of this -- your robot could bore out chunks of asteroid material and, through a pneumatic or even spring-loaded mechanism, simply throw them in a desired direction, like a pitching machine. (I just thought of this, which surely means somebody else has already come up with a detailed plan for doing it.)

That's the idea with nuking the thing. Just with a bomb instead of a spring. It's like Wile E. Coyote on speed.
It also a question of the cost of getting something to the asteroid. Putting a railgun or other mass driver on the asteroid requires you to get that machinery up there, to get machinery up there to power it, to get machinery up there to pick up (or dog for) the "ammunition" and get machinery up there to build or assemble all that stuff. A nuke on the other hand requires you to get a warhead up there, including a little casing to make it a bunker buster that buries a few meters before detonating.
While the nuke will not be a very efficient rocket motor, (i.e. we can ideally expect 50% thrust in the intended direction and another 50% laterally) it will have a much much higher energy density than whatever is used to power the railgun, let alone the whole railgun ensemble.
I'd need some numbers on how much of a nuke's energy will be absorbed by a few meters of rock to make a useful estimate, but my gut feeling is that the energy density is simply so much higher that it more than compensates for the lower efficiency (i.e. energy-to-thrust ratio)
And what's most important is the amount of thrust on the asteroid we get for every kilo of machinery we have to lift from earth.

Note: orbits aren't circular.

Not in the geometrical sense of the word, but in the sense of things-returning-to-point-of-origin.

[Astronomer]

The idea is to detonate the nuke only a few meters below the surface, creating an improvised rocket thruster.

Ah, I get it now. That's actually something worth exploring. I still think Bruce Willis should be involved, though.

True -- but I think we're talking about the Yarkovsky effect, not direct radiation pressure.

Actually, we are talking direct radiation pressure, since we're modifying the asteroid to reflect more radiation away and thus diminishing the absorption/re-radiation latency.

Even if you mirror the whole surface, you could at most double the current pressure caused by the sun.

Which is why it would take a long time to have a civilization-sparing effect. Even an added force equal to the weight of a sticky note will alter an asteroid's orbit after a few decades, which, thankfully, is the amount of time we have. (We're still talking about Apophis, right?)

[Manuel Royal]

Originally Posted by Lhun

Not in the geometrical sense of the word, but in the sense of things-returning-to-point-of-origin.

Not necessarily. Depends upon how the object's orbit is changed by its near-encounter with Earth.

(ETA: Let me get back to that. I'm trying to remember what I used to know about gravity-assist changes in orbit.)

I'd like to try a bunch of stuff before abandoning (or altering) the 1963 Limited Nuclear Test Ban Treaty. The ESA is planning a probe (charmingly called Don Quijote) to test a kinetic impact technique.

ETA: Wouldn't it be awesome to capture Apophis into an Earth orbit? I was just reading an article about that. Something like two dozen megatons of iron, volatiles (rocket fuel!) and other materials.

[movieman]

Originally Posted by Astronomer

Ah, I get it now. That's actually something worth exploring. I still think Bruce Willis should be involved, though.

MIT studied this in the 60s or 70s; they intended to detonate 100MT bombs within 50 feet of the asteroid surface using Apollo-era technology. It was all technologically possible, but would have cost tens of billions of dollars.

The big problem is actually tracking the asteroids well enough to know whether it's going to hit the Earth. If you can predict an impact a few years ahead then there are a lot of options for avoidance, such as the gravity tractor previously mentioned; but predicting that accurately is very hard and if you get it wrong you could turn a near miss into an impact by trying to avoid it. If you have to stop an asteroid when it's only a few weeks away then nukes are probably your only option at this time.

[Skyler]

Isn't Apophis' asteroid made out of naquadah, and hence tremendously dangerous to nuke?

Or is this a different asteroid we're talking about?

[efkelley]

Talking about the actual asteroid: http://en.wikipedia.org/wiki/99942_Apophis

If we were in the Stargate universe, then nuking a naquadah asteroid might be cause for concern. But I'd do it anyway. Just to see what happened.

[Skyler]

I would do it as it was crossing the atmosphere boundary of Earth and win the "Destroy the Earth" prize.