Deflecting or destroying dangerous asteroids

[BeamRider]

I have tried to search for other posts on this, but have been unable to figure out how to search within this forum, where it clearly belngs, so here goes.

Protection against Earth-approaching asteroids is a topic that comes to mind every time we see headlines about a "surprise" close approach of a wandering hunk of rock, but then fades for most of us. But what about the surprise that turns out to have our name on it - and won't go away? What are the impediments to deploying a serious protection system against those?

For one, there is the "giggle factor", but I think it is diminishing with each new "surprise" visit. Probably the remaining big factor is the assumption that any defense system would be staggeringly expensive, requiring a diversion of investment that could be used in more "productive" ways. The assumption here is that such a system would have no other use than to protect us from the "nasty surprise" that may never come, and would otherwise sit idle.
Maybe there is a way around this objection - if not to significantly reduce the cost, to at least derive some secondary benefit from it. Recent attention to asteroid capture missions got me to thinking about the possibility of "fighting fire with fire"... or in this case fighting asteroids with asteroids!
My first thoughts were occasioned by the concept of capturing an asteroid in lunar orbit... could it be stored there as a mass to be directed into the path of a dangerous visitor? I soon realized that to direct even a smallish asteroid out of lunar orbit (rather suddenly, on demand) would require prodigious amounts of force that would probably require development of thrusting engines orders of magnitude greater than we have now.
Then I considered whether an asteroid could be maintained in a co-orbital trajectory (figure-8?) around Earth AND Moon. Would this allow an orbital velocity to be maintained that was much higher than that required to orbit either body? The mental picture I had was of a mass using the deep gravity well of the Earth to continuously return to the Moon's shallower one, to be returned toward Earth, etc. Of course the mass could approach the Moon quite closely without an atmosphere to interfere. Might it be possible to use relatively small trajectory changes to cause the mass to be ejected from this co-orbit, to smack an unwelcome visitor?
Now even if the basic mechanics would work (and I am becoming more skeptical that they would), this situation would require constant fine-tuning to avoid either losing the mass into space, or having it impact one or the other of its "parents", potentially causing the kind of destruction it was meant to prevent! The investment in keeping this monster pinball game under control might not be sustainable, either.
But what about a sling, perhaps at a Lagrange point? Would it be possible to connect two asteroid masses and spin them around each other?! The required potential energy could be raised gradually with more or less known technology, but released suddenly as kinetic energy with an explosion to sever the connection between the two masses. Yes, the connection would have to be formed from materials suitable for a space elevator, at least! And of course, the "recoil" would consist of the "other" mass flying away just as rapidly, hopefully into deep space.
And what about the "real-world" secondary benefits I suggested would be necessary? Well, a space probe manufacturing / launch facility on one or both of the masses would enjoy artificial gravity, and to launch... just kick the probe over the edge! Getting ON this giant merry-go-round might be trickier than getting OFF however!

Okay, this is my bizarre idea.  Have at it!

[whitelancer64]

For the record, figure 8 orbits are possible, but they are not stable, since any perturbation would deflect it from the orbit. Given that the Moon is moving around the Earth and the Sun also exerts gravitational force on the asteroid, it would probably not remain in a figure 8 orbit for more than one orbital period before being deflected off-course.

A better (and far simpler) means of deflection is to direct a nuclear weapon at the incoming asteroid. The goal would not be to destroy the asteroid, but to detonate the weapon sufficiently close to the surface to pick up mass from the asteroid and blast it away (causing thrust in the opposing direction), thereby deflecting the asteroid from its original course.

The primary problem with this is it must be done a long time in advance, so prior warning is required, on the order of years of lead time.

There are multiple other proposals for deflecting away an asteroid, but they would require even longer lead times to prepare and execute, such as albedo change, and gravity tractor.

[BeamRider]

Thanks for your comment, and I am not surprised an Earth-Moon orbit is not stable. Do you have any insight into whether the "pumping" of energy into an object in such an orbit could be effected the way interplanetary probes use planet flybys, assuming the stability of the orbit is maintained through other active measures?

Interesting that you mention the problem of known methods of deflection requiring long lead times, because they exert gentle continuous forces. This is the specific problem I was addressing with the double-asteroid sling, although I did not articulate it explicitly. This approach seems to me to potentially allow for the storage of large amounts of energy, accumulated gradually with not-too-advanced technology, but released instantaneously. It seems like it would be able to deliver a HARD hit relatively quickly, and at a distance, either deflecting the target or perhaps even shattering it. Thoughts?

Ps. I am not convinced of the validity of the arguments that are made against disruption of the asteroid - that it would make the problem worse by multiplying the number of objects.

[gospacex]

Nukes would work. Relatively modest development needed.

[whitelancer64]

Thanks for your comment, and I am not surprised an Earth-Moon orbit is not stable. Do you have any insight into whether the "pumping" of energy into an object in such an orbit could be effected the way interplanetary probes use planet flybys, assuming the stability of the orbit is maintained through other active measures?

Interesting that you mention the problem of known methods of deflection requiring long lead times, because they exert gentle continuous forces. This is the specific problem I was addressing with the double-asteroid sling, although I did not articulate it explicitly. This approach seems to me to potentially allow for the storage of large amounts of energy, accumulated gradually with not-too-advanced technology, but released instantaneously. It seems like it would be able to deliver a HARD hit relatively quickly, and at a distance, either deflecting the target or perhaps even shattering it. Thoughts?

Ps. I am not convinced of the validity of the arguments that are made against disruption of the asteroid - that it would make the problem worse by multiplying the number of objects.

A figure 8 orbit wouldn't "charge up" an asteroid, as probes do with a gravitational assist, if it gained speed it would be ejected out of the figure 8 orbit. Spinning two asteroids tethered together would, but that would be very difficult to aim thanks to gyroscopic effects. You'd also have to very, very, very precisely time the breaking of the tether to have any chance of hitting a target, and that's not likely, even minute errors would cause it to miss, in other words, there's no good way to aim such a system. Even our most precise rocket launches have scheduled course corrections to get our probes to their targets accurately.

Breaking apart an asteroid is undesirable for a few reasons. If done at close range you, to quote Independence Day, "risk turning one dangerous falling object into many."

However, when done a great distance away, most of the bits might miss Earth entirely, but then they a. would be much harder to track and b. would form a debris cloud heading towards Earth. The danger to our satellites would be pretty high even if all the bits were small enough to burn up in Earth's atmosphere.

[Ludus]

There will be a lot of money invested in learning to redirect asteroids as part of asteroid mining. The main issue is deltaV for very massive objects. It doesn't really solve anything to imagine hitting one very massive object with another. If you can produce so much delta V for one object as to precisely redirect it to do that, you could much more easily deflect the problem asteroid using the same techniques.

Asteroids don't have to be nudged very much to make them NOT intersect with the earth. To shove around one asteroid enough to make it precisely intersect with another's path is WAY harder.

One possible technique that could do quite a lot is to land a combination of an automated drill (maybe plasma) a big solar array and an EM catapult. This would be hundreds of tons but may be doable by SpaceX ITS Cargo Version. The setup would use Solar power to fling pellets at high speed away. It would very slowly consume the sateroids mass as propellant using Solar power. This kind of setup operating continuously to do some pretty impressive Asteroid herding.

[sanman]

Could the so-called "3rd/4th generation" nuclear weapons, which are in theory supposed to be able to asymmetrically direct their explosive power like a shaped charge, be more useful for this situation?

How could a nuclear blast be shaped to impart maximum effect on a target asteroid?

Would existing nuclear warheads be easily repurposed to this task, or would there have to be any special customizations / modifications?

[tyrred]

"Casaba Howitzer" would not be prohibitively difficult to field at close range to Earth, kludged onto an existing ICBM.  Creates a collimated beam of nuclear plasma, a directed energy weapon.  Not exactly "off the shelf" asset (as far as is public), but the studies for project Orion showed that at least theoretically it was possible... And that was decades ago.  Range and targeting could be a challenge, as the target trajectory needs to be deflected enough to miss Earth and it's constellation of satellites without reducing the target to a rubble pile of buckshot.  Most likely this approach would still be too little, too late.  We must first have the eyes to see where to direct the defense.

[M.E.T.]

I don't understand the standard response that blowing up the asteroid with a nuclear weapon is not feasible, because the cloud of resultant debris is supposedly just as dangerous as the single massive rock.

Surely it is the size of the single rock that makes it able to penetrate the protective blanket of the atmosphere, which means that a million smaller pieces will burn up in the atmosphere instead of impacting the ground and creating mega tsunamis or gigaton sized explosions.

Smaller is therefore better, in my view. So blowing it up would be a valid option.

[jgoldader]

The incoming has KE and gravitational PE, and that's all going to be dissipated.  It could happen all in one place, or in several.  A 20 meter piece can make a very powerful airburst with heat and shock wave (ref: Chelyabinsk).  Imagine a hundred of those, all at once, spread over a ~1500 km diameter area.  Figure one or two for every major European or US city, say.  Are you *sure* that's better than a single impact?

Until we better understand the physical properties of NEOs, and there are likely to be many different classes (from metal chunks to loosely-bound aggregates), it's not easy to figure out what to do.  And we might not have the luxury of time to investigate a possible impactor far enough in advance to really tailor a response.

[RonM]

Big question is how far in advance of impact do we detect the object and when can it be intercepted. If there's very little time, as in the OP scenario, blowing up the asteroid won't work since all of the debris cloud will still hit the Earth. If we can hit the asteroid a year in advance then we don't need to blow it up, just a little thrust will cause it to miss. If it's too late to move the asteroid, but there's enough time for most of a debris cloud to dissipate and miss the Earth, then blowing it up will work. There's no one size fits all solution.

[ppnl]

One possible technique that could do quite a lot is to land a combination of an automated drill (maybe plasma) a big solar array and an EM catapult. This would be hundreds of tons but may be doable by SpaceX ITS Cargo Version. The setup would use Solar power to fling pellets at high speed away. It would very slowly consume the sateroids mass as propellant using Solar power. This kind of setup operating continuously to do some pretty impressive Asteroid herding.

High speed projectiles would waste energy which may be your most limited resource. Instead use massive projectiles at low velocity. Maybe some reasonably small multiple of its escape velocity. Also less wear on your launcher.

[gospacex]

The incoming has KE and gravitational PE, and that's all going to be dissipated.  It could happen all in one place, or in several.  A 20 meter piece can make a very powerful airburst with heat and shock wave (ref: Chelyabinsk).  Imagine a hundred of those, all at once, spread over a ~1500 km diameter area.  Figure one or two for every major European or US city, say.  Are you *sure* that's better than a single impact?

Yes, I'm sure.

A thousand 20 meter meteors airbursting at ~20 km up over an entire hemisphere can break a lot of glass and kill maybe 2000 people, in total, by shards etc.

An equivalent single 200 meter bolide won't be significantly slowed by atmosphere and would make a 5-10 km diameter crater. It's easily a city killer, or can create ocean-wide tsunamis 50 m in height.

[jgoldader]

Obviously, either would be "not good."  I wonder if a single impactor yields both the worst case scenario in terms of fatalities (probably a water impact w/tsunami) and the fewest, via a land impact in some very remote area. I tried to conjure the worst "multiple small impact" case I could imagine; in a "real" case, most of the little pieces wouldn't hit major cities.  But don't underestimate a Chelyabinsk airburst over a big city; it could well shatter pretty much all the glass in a major city, bring down building facades... I wouldn't want to be downtown.  And if that happened in many cities, well, not good.

But I do agree that a single impactor should give the worst-case scenario.  If I had to choose between one big impact in the ocean or the "shotgun" effect of a nuked parent body, I'd go for the latter.  It would be, as you suggested, at worst tens of thousands of deaths vs. potentially tens of millions.

Either way, not a good outcome.  And a circumstance I hope we don't face any time soon.

[whitelancer64]

I don't understand the standard response that blowing up the asteroid with a nuclear weapon is not feasible, because the cloud of resultant debris is supposedly just as dangerous as the single massive rock.

Surely it is the size of the single rock that makes it able to penetrate the protective blanket of the atmosphere, which means that a million smaller pieces will burn up in the atmosphere instead of impacting the ground and creating mega tsunamis or gigaton sized explosions.

Smaller is therefore better, in my view. So blowing it up would be a valid option.

then they a. would be much harder to track and b. would form a debris cloud heading towards Earth. The danger to our satellites would be pretty high even if all the bits were small enough to burn up in Earth's atmosphere.

[hop]

There is a large body of literature on planetary defense, based on actual calculations and experiments.

Presentations from the recent planetary defense conference in Japan may also be of interest.
https://www.youtube.com/channel/UCpq_pj8aLeFu-0i4Je8_JsQ

But all this is irrelevant. Anything small enough for us to redirect is too small for us to see in time while anything we can see in time is too large to redirect. Unless we start putting +20m telescopes in space there's no point in developing redirect capabilities. Better just to make peace with and live your days.

This is completely wrong. We routinely detect asteroids too small to be a concern, and objects capable of creating larger than city-scale devastation are readily detectable without 20m space telescopes. Very large objects can be deflected with enough advance warning.

Long period comets are much harder to deal with, but they are only a moderate fraction of the overall risk.

We may not be able to eliminate the threat, but with modest effort we can significantly reduce it. In fact, we already have, the survey of >= km scale NEOs is nearly complete.

[hop]

Sure, we detected the Chelyabinsk meteor - as it happened. Well done!

We have also detected lots asteroids smaller than Chelyabinsk that did not impact. Impact risks are a numbers game. You don't have to identify every single one to reduce your overall risk.

It was only a 400-500kt blast that fortunately detonated at +25km altitude due to a shallow trajectory.

In the grand scheme of things, Chelyabinsk wasn't a big deal. Even Tunguska size objects aren't unless they hit near a population center. Sure, we'd like to detect and prevent these kinds of things, but in the worst case it's a disaster like a major hurricane, earthquake or tsunami, not a civilization killer. (edit: To be clear, I'm referring to Tunguska size events, not impacts in general!)

Fortunately we know of every kilometer-sized object and apparently have operational capability to deflect "very large objects", according to Hop.

That's not even remotely what I said. We have discovered most of the > 1km NEOs. We know that in principle, current technology is sufficient to modify their trajectories on multi-decade timescales.

Unfortunately that leaves a lot of 1km-20km craters, airblasts equal to armageddon and the extinction of humanity.

This is not accurate. See http://impact.ese.ic.ac.uk/ImpactEffects/ for example. The reason Congress initially gave NASA a mandate to find ~1km and larger objects is because those were the ones judged to be civilization killers. They later extended the goal down to 140m, because those pose serious regional risks. While the 90% goal for >140m NEOs won't be met soon with current resources, LSST and existing surveys are expected to reach  ~70% https://arxiv.org/abs/1706.09398

Discovery efficiency follows a curve, so completeness for, say >500m objects will be significantly higher.

As previously stated,

Unless we start putting +20m telescopes in space there's no point in developing redirect capabilities.
Better just to make peace with and live your days.

This is still a completely incorrect statement. Again, this is a well developed field of research, read the literature! No one in the field has called for a 20m space telescope for asteroid detection, the most serious proposal is for a quite modest IR telescope.

[jg]

Please take a look at http://fallingstar.com/home.php about the ATLAS project.

The first two ATLAS telescopes are coming on line this year.  You can see the discoveries on the minor planets center (or follow @fallingstarIfA for discoveries).  The discovery rate is about to go up a factor of 5 (the original corrector plates were defective, and new ones are/will be installed shortly).

This will get us good stats on small threatening NEO asteroids for the first time.

John Tonry (the PI) is a friend, and I can ask questions of him if the fallingstar web siteis unclear.

Note this is not a panacea: 40% of the sky is too near the sun to be observed.

[catdlr]

NASA plans to deliberately crash DART spacecraft into an asteroid

Tech Insider
Published on Jul 16, 2017


Large asteroids colliding with Earth is extremely rare. However, if that happens, it could mean the end of civilization as we know it. NASA isn't taking any chances. It is currently developing DART, a small spacecraft that will crash into an asteroid to deflect it from Earth.

[stefan r]

A better (and far simpler) means of deflection is to direct a nuclear weapon at the incoming asteroid. The goal would not be to destroy the asteroid, but to detonate the weapon sufficiently close to the surface to pick up mass from the asteroid and blast it away (causing thrust in the opposing direction), thereby deflecting the asteroid from its original course.

using wikipedia for the radius of underground nuclear tests we could melt from 40 to 120 meters of rock per megaton yield.  It may not be realistic or desirable to bury the nuke.  A surface explosion would ablate and vaporize rock and would melt much less rock than the subsurface test.  The cracked zone is 800 to 1200 m/Mt^1/3. 

Shock waves do not propagate through porous material the same way they propagate through crystalline solids.  Water ice transforms into liquid when compressed.  An explosion under a glacier will behave differently than an explosion under rock.  Underground nuclear tests do not give a precise model even if we could land and place the bomb.

If we have a 500m rubble pile and detonate a 1Mt nuclear device then we have a 499m rubble pile.  It becomes several pieces if the chunks are not bound tightly. 

4.184 x 10¹⁵J is one megaton.  A 500 meter asteroid (6 x 10⁷ m³ could weigh in at 10¹¹kg.  If we got 100% conversion of energy to thrust we would get 200 m/s change in direction.  I am not sure how to calculate the thrust from a surface explosion.  But I would guess less than 1%.  Which leaves 1 or 2 m/s.  Earth's radius is 6.3 x 10⁶m so you need the explosion more than 2 months before the asteroid contacts earth.