Tunguska Question

[Schneibster]

That's indeed the common nuclear-physics usage---with the odd effect that, say, U235 might be said to release both a 31 keV gamma ray (from excited states of the 231Th daughter nucleus) and a 90 keV x-ray (from inner-shell excitation of the daughter atom). Ugh. 235mU has a 70 eV "gamma ray"---that's what I call FUV, but it comes from the nucleus.

But when I'm wearing my astronomer hat, almost everything comes from atomic/electronic processes, so I tend to think: if you need a semiconductor fab to build the detector, it's an x-ray. If you need a machine shop and a screwdriver, it's a gamma ray.

Technically speaking, yeah, gammas come from strong force interactions, and X-rays from EM interactions; thus, gamma from the nucleus and X-rays from the electrons. But there's nothing to distinguish a particular photon once it's left the source, and by that definition there are gamma rays that are indistinguishable from deep UV, a ridiculous situation since there is no measurable difference between them. I therefore go with the "astronomer hat" all the time, and distinguish by the energy of the photon, not by its origin, just as you do, ben.

 

[Schneibster]

Let me just check.
You're saying anyone close enough to be irradiated would be flash boiled as well, so the putative gamma ray peak becomes sorta...marginal?

More than that. The gamma ray peak from the initial flash created at hundreds of millions of degrees is only visible inside the fireball; it's mostly absorbed by the bolide and compressed atmosphere, heating them to hundreds of thousands of degrees. You get some initial gamma flux outside the fireball, but anyone close enough to be seriously harmed by it is quickly dead from, as you say, being flash boiled, if not vaporized.

 

[Dr. Trintignant]

I have never touched the physics of how such a high temperature could come to be; I'm pretty certain that to get a result that's within even multiple orders of magnitude of the correct one, you're going to need numerical simulations, which ain't happenin' on this here cute little laptop anytime this year.

I found some simulations here:
http://www.psi.edu/~betty/chicx3d.html

The temperature plot indicates a maximum temperature of around 10,000 K. This seems plausible to me; I'd expect there to be a limit to the heating as vaporized material will only stay in the asteroid path for so long.

Suppose we accept Ben's gamma-ray cutoff of 50 keV; that's a frequency of about 1.2e19 Hz. Online calculators say the BB peak at 10,000 K is about 1.04e15 Hz. Running these numbers through Planck's equation (and skipping the intermediate math), it looks like the proportionality constant is something huge; more than 10^25000 (the exponent in the denominator of Planck's equation really makes a difference). So the amount of gamma radiation at that temperature is effectively zero.

(as usual, I welcome any criticisms of my calculations...)

- Dr. Trintignant

 

[Schneibster]

They're modeling something completely different from Tunguska; a 10km object hitting the ground, not a 30m object disintegrating 30km up and turning into a fireball. Not only that, but the model doesn't appear to cover the atmospherics prior to impact. I suggest that the model is insufficient evidence to prove your case; Tunguska is much better parameterized, and has required considerably more in the way of modeling, resulting in a very recent new model of what happened.

ETA: re, "stay in the path only so long." Does milliseconds or less work for you? I can't see air escaping from the path of a hypersonic object at the speed of sound very easily, and that's essentially all the faster it can move.

 

[Dr. Trintignant]

They're modeling something completely different from Tunguska; a 10km object hitting the ground, not a 30m object disintegrating 30km up and turning into a fireball.

My apologies; I was thinking we were still talking about the Chicxulub impact.

Tunguska is much better parameterized, and has required considerably more in the way of modeling, resulting in a very recent new model of what happened.

I'm not particularly knowledgeable here, but wasn't the modeling largely based on how the explosion interacted with the ground (the tree patterns and such didn't have easy explanations)? However, I'm not sure how that would change the peak temperatures you would see in the blast; only the total energy content. After all, by the time the blast expands a small amount, any very high temperatures would have already dissipated.

ETA: re, "stay in the path only so long." Does milliseconds or less work for you? I can't see air escaping from the path of a hypersonic object at the speed of sound very easily, and that's essentially all the faster it can move.

Wouldn't the shockwave carry material away at hypersonic speeds? And there is radiative transfer to consider. I suspect that I could put together a conservative upper-bound on the temperature based on blackbody radiation; I'll see if I can do the calculation later tonight. It's just that millions of degrees is a very large number; high enough that the radiation alone may be far higher than the power absorbed through atmospheric friction. But I don't yet have numbers to back me up on that one.

- Dr. Trintignant

 

[Schneibster]

My apologies; I was thinking we were still talking about the Chicxulub impact.

No problem; the context and flow of the conversation left that open as a valid interpretation. I should have been more specific.

The big difference is that the Chicxulub object didn't disintegrate before impact.

I'm not particularly knowledgeable here, but wasn't the modeling largely based on how the explosion interacted with the ground (the tree patterns and such didn't have easy explanations)?

Well, the problems with previous simulations were due to that. And don't forget that there's a lot more computer power out there, both because of newer hardware, and because we're discovering new mathematical algorithms, as well as new ways to implement old ones. Parts of simulations that before had to be hard-coded or entered by hand, can now be calculated by sub-simulations of their own and fed into the main simulation during computation. Small, finicky details that could be hand-waved away by referring to the difference between the hard-coded values and the real world are now much closer to the real world, and so the results can be checked against far more detailed representations of the evidence. And if they don't agree then we can go looking for why, and discover some new science doing it.

However, I'm not sure how that would change the peak temperatures you would see in the blast; only the total energy content. After all, by the time the blast expands a small amount, any very high temperatures would have already dissipated.

We're talking about a region of space that's 30m wide; that's pretty small on the scale of Chicxulub. It's probably well below the resolution of the data used in that simulation.

You're going to see temperatures like that in front of a dinosaur killer, too, and probably some high-frequency radiation from much the same cause, but nothing that is line-of-site with it is going to be solid matter after the impact. The crater is going to be 150km wide; and stone will not stand on stone within 2000km. That's two-thirds the width of the US, just to give you an idea. It's a catastrophe of barely imaginable proportions; we're talking about an area it would take you months to walk across that will contain no building, no tree, nothing that sticks up above the ground in any way; and in its center a hole so big you can't see the far edge. The material that used to be in that hole is mostly floating in the stratosphere; some of it might even be more-or-less permanently in orbit or deposited on the face of the Moon.

There will be no evidence of anything but heat, heat, pressure, heat, more heat, more pressure, and more heat. Coming within a thousand km of the crater within a month of its formation would be as much as your life is worth, assuming you were underground or underwater on the far side of the world when it hit; if you weren't, you're not alive. Any evidence of what happened is rock vapor.

I'd trust a numerical simulation on the scale of that event to tell me anything about relatively minor thermal effects on the front face of that descending armageddon about as much as I'd trust said numerical simulation to tell me the underwear size of Xythron the Insistent of Aldebaran V. We're talking many, many orders of magnitude larger than Tunguska. Sure, stuff like that happened; but the material objects it happened to simply no longer exist. The nuclei probably don't even have the same electrons any more. In the face of which, who cares? Just figure how many joules over how much area, average it, and let it go at that. Whatever temperature that gives you, it's not going to have much to do with what really happened, but as long as it creates plasma, who's to know the difference?

Wouldn't the shockwave carry material away at hypersonic speeds?

Not faster than the bolide is coming down. A 10km object isn't going to slow down in the atmosphere; as far as it's concerned, the atmosphere might as well be an extra six inches of rock. A 30m object will be much more affected, but that's what they're saying happened; by the time it was 120km deep in the atmosphere, it didn't exist as an integrated solid object any more, it was a fireball of plasma. Completely different situation.

And there is radiative transfer to consider.

Sure; most of the radiation had to be absorbed right there, in order to create that plasma. But that doesn't mean the radiation didn't exist.

I suspect that I could put together a conservative upper-bound on the temperature based on blackbody radiation; I'll see if I can do the calculation later tonight.

You're not going to do that without considering the numerical simulation; you're going to be averaging behavior that you need to specify in far greater detail to see what really happened in order to even approach it without a computer you could heat your house with. Seriously, you need to think about what we're talking about here; temperatures, pressures, and physical situations that we create microseconds at a time, and separately, one at a time, in labs, not the whole shootin' match.

It's just that millions of degrees is a very large number; high enough that the radiation alone may be far higher than the power absorbed through atmospheric friction. But I don't yet have numbers to back me up on that one.

- Dr. Trintignant

See, temperature is a derived quantity; it's a function of the average kinetic energy of a collection of atoms, ions, or molecules. How small a sample would you like to average over? Shall we stick with plasma that's within a centimeter of the front edge of the bolide? If so, not only will you need the home-warmer computer, you'll need a numerical simulation that neither you nor I know how to even begin to write. If you're going to average over an area that includes ANY PORTION of the solid matter of the bolide, or for that matter any air not in the shockwave directly in front of it, you're going to get answers that aren't in the millions of degrees but in the tens or hundreds of thousands.

Suppose you take one atom, the fastest one you can find. What is its temperature? Does temperature even have any meaning for that atom? What if it is an ion, and it decelerates when it hits another ion, or an atom? What will happen then, when we decelerate the charged particle? Does that have anything to do with blackbody radiation? Sure it does- blackbody radiation will be composed of billions upon billions of such collisions. But it's meaningless to talk about the temperature of the atom.

See what you come up with, and have fun, but don't confuse it with the real world.

 

[INRM]

What's a blackbody peak?

What's the Z-machine (it recorded 2 billion degrees, highest temperature)

And what's the Sandia Supercomputer


INRM

 

[Dr. Trintignant]

Well, the problems with previous simulations were due to that. And don't forget that there's a lot more computer power out there, both because of newer hardware, and because we're discovering new mathematical algorithms, as well as new ways to implement old ones.

Indeed; in fact, one of the recent Tunguska articles mentions that previous simulations didn't even use momentum in the calculations.

You're going to see temperatures like that in front of a dinosaur killer, too

I'm not entirely sure of that either, although it is more likely in that case.

It's a catastrophe of barely imaginable proportions; we're talking about an area it would take you months to walk across that will contain no building, no tree, nothing that sticks up above the ground in any way;

No doubt about that--the near-UV radiation would be enough to fry anything in hundreds or thousands of kilometers. I'm just trying to get an idea about the gamma pulse in particular.

You're not going to do that without considering the numerical simulation; you're going to be averaging behavior that you need to specify in far greater detail to see what really happened in order to even approach it without a computer you could heat your house with.

But I can make some very conservative assumptions to see if things are even plausible.

Shall we stick with plasma that's within a centimeter of the front edge of the bolide?

I don't think that matters. Radiative transfer is a function of surface area. The depth distribution is just a matter of detail.

Suppose you take one atom, the fastest one you can find. What is its temperature? Does temperature even have any meaning for that atom?

While I agree that temperature is a statistical quantity, and we can't necessarily claim that there is no atom with a specific energy, we can just the same use equations like Planck's Law to make estimations.

So let's give it a shot. As I mentioned, I'm going to make what I think are very conservative assumptions here.

I'm going to compute how long it would take the asteroid to dissipate all of its kinetic energy via radiation, assuming a surface temperature of 1,000,000 K. I think 1e6 K is probably too low to see much gamma radiation, but let's use it since it's a round number. And of course I'm completely ignoring other means of heat loss here, again just to be conservative.

Anyway, if the time we come up with is very low compared to the time the asteroid spent in the atmosphere, we can probably conclude that our temperature assumption is implausible. An object can't heat up if it's losing energy faster than it's picking it up.

The frontal surface area of the asteroid is about 700 m^2, assuming a 30 m diameter sphere. Obviously the temperature distribution across this is going to vary and need a nasty integral to compute exactly; so instead, consider a small 100 m^2 patch on the front. Assume this is fairly flat and of even temperature. This isn't quite true, but the surface area that we're neglecting should more than "make up" for this fact.

The Stefan-Boltzmann equation goes like this:
P = 5.6696e-8 W*m^-2*K-4

Filling in the numbers:
P = 5.6696e-8 W*m^-2*K-4 * 100 m^2 * (1e6 K)^4 = 5.67e18 W

The newest kinetic energy numbers indicate 3-5 megatons, but let's go with the older ones, which were more like 15 Mt. That's 6.3e16 J.

So dividing things out, we see that a 100 m^2, 1e6 K patch on the asteroid would dissipate all of the kinetic energy of the asteroid within about 0.011 seconds. Asteroids are fast, but not that fast--it lasted a few seconds even deep in the atmosphere.

Hence I'm skeptical that a surface temperature of 1M K could be maintained, let alone 10M K or more. That T^4 part of the Stefan-Boltzmann equation makes for a really amazing heat transfer. And even though much of that energy would be immediately reabsorbed by the surrounding gas/plasma, that doesn't change the ultimate result because we're just trying to see if any layer can maintain that temperature.

In any case, I'd like to see any objections you might have. I tried to make the most conservative possible assumptions but it's certainly possible that I missed something.

- Dr. Trintignant

 

[Schneibster]

Here's the problem with doing it that way: the density of the atmosphere is not constant, so the period when extreme heating occurs is all at the end.

 

[Dr. Trintignant]

Here's the problem with doing it that way: the density of the atmosphere is not constant, so the period when extreme heating occurs is all at the end.

Agreed, but all in 11 milliseconds? Even at 17 km/s, that is only 187 meters. Surely we can assume that the last few kilometers were at fairly constant atmospheric pressure--within a factor of two, at any rate.

- Dr. Trintignant

 

[Jimcalagon]

[...] Forget silly Armageddon movies; we're talking about a rock six miles across, basically a few mountains mashed together. Nothing we have is going to even make a dent in that. If we figure it out with enough lead time, we might, if we're very, very lucky, have enough thermonuclear weapons in the entire Earth's arsenal to move it far enough not to hit us. Now, how you gonna deliver said thousands of nuclear weapons across tens of millions of miles of empty space? Not to mention get them all to go off at the same time so you can generate enough impulse to move said rock.

Project Orion, that's how.

That could have delivered thousands of nuclear weapons across the solar system and, better than setting them all off at once (and changing a huge piece of rock on a collision course with the Earth into a million tuguska events!), it could then use their energy in a controlled manner to change the orbit of the rock to render it harmless for the forseeable future.

And all using 1960s technology.

 

[Hellbound]

[derail]
An idle thought I had while sitting here. I figured I'd toss it out to see if anyone thought it was worth a thread on investigating further.

What would be the feasibility of using some type of a charged particle weapon to change the electrical properties of an incoming object, causing it to be deflected by the Earth's own magnetic field? Assume that the technology can be created, I'm more interested in the theory. Specifically:
1. How much energy/mass would it take in, say electrons or protons fired at an asteroid to give it a significant net charge?
2. How much charge would it need?
3. Would the Earth's field even be powerful enough to cause a deflection?
4. Would this limit the effects of an impact, or would it be just as bad only different?

If anyone thinks this is interesting, I'll start it as another thread. If it's the most ridiculous thing you've heard this week, I won't bother .
[/derail]

 

[Jimcalagon]

[derail]
An idle thought I had while sitting here. I figured I'd toss it out to see if anyone thought it was worth a thread on investigating further.

What would be the feasibility of using some type of a charged particle weapon to change the electrical properties of an incoming object, causing it to be deflected by the Earth's own magnetic field? Assume that the technology can be created, I'm more interested in the theory. Specifically:
1. How much energy/mass would it take in, say electrons or protons fired at an asteroid to give it a significant net charge?
2. How much charge would it need?
3. Would the Earth's field even be powerful enough to cause a deflection?
4. Would this limit the effects of an impact, or would it be just as bad only different?

If anyone thinks this is interesting, I'll start it as another thread. If it's the most ridiculous thing you've heard this week, I won't bother .
[/derail]

Just off the top of my head (I'm definitely NOT an expert here) - 3 - nowhere near strong enough.

Possibly though, if you have a long enough time to work on the object, you might be able to get some kind of deflection from the solar wind or the sun's magnetosphere. Over a period of years this might be enough to significantly change the orbit. (Thunderbolts of the Gods-type stuff possibly )

However I don't believe that it would be possible to project a powerful enough beam from the earth (I presume you are alluding to) through the atmosphere at something halfway across the solar system and expect the beam to stay collimated enough for any kind of effect to be produced.

What might be better (if the electric charge idea has merit) might be to send a device to the object and let it create the charge locally (by projecting a beam of electrons FROM the object)...?

There is also the gravitational 'tug' idea, which might be effective.

In any case, I don't think that simply blowing the thing up would be any use - as mentioned above, this simply creates a large number of objects which are still on roughly the same orbit. Not a good idea, I would have thought.

 

[Hellbound]

Ah well, it was a thought I figured I was building a bridge from tissue-paper, but didn't know the numbers off-hand.

From the sound of it, a mass-driver type propulsion system sent to and placed on the rock might be a better choice...but we'd need better technology to do something like this (and enough warning time to know the thing was coming).

 

[Schneibster]

Agreed, but all in 11 milliseconds? Even at 17 km/s, that is only 187 meters. Surely we can assume that the last few kilometers were at fairly constant atmospheric pressure--within a factor of two, at any rate.

- Dr. Trintignant

I think you're still ignoring the degree to which the atmosphere thickens. I would argue that, yes, it is not unreasonable to expect that the vast majority of the heat would be dissipated in 11 milliseconds. And in fact, this is just about what happens with most bolides. And I can confirm it anecdotally.

I am an amateur astronomer, and have seen many bolides. They increase in brightness fairly steadily to a peak as they enter and pass through the mesosphere, and then suddenly jump in brightness as they hit the top of the stratosphere, assuming they are large enough to make it that far, and that their angle takes them deep enough into the atmosphere to hit the boundary; you can see when it happens. Quite often, they will explode or fragment upon impact with the stratosphere, and rarely do they last long after that. Only once have I seen a bolide that I believe actually struck the surface, and it was out to sea by at least twenty miles, and more likely a hundred or so. It briefly was perhaps half the brightness of the full Moon, and ruined an astrophotograph I was exposing at the time; unfortunately, I was shooting East, and it was in the West, so I didn't even get a trail out of the shot.

The stratosphere extends from between eight and seventeen kilometers, to between thirty and fifty kilometers; this is the area where most larger meteors burn up. The thermosphere and mesosphere, collectively formerly called the "ionosphere," aren't thick enough to ablate them significantly, though they do heat to white hot in the mesosphere. The mesosphere/thermosphere boundary is generally where the small flecks of dust that are most meteors burn.

Significantly, the Tunguska object's disintegration occurred very close to the stratosphere/mesosphere boundary, which tends toward the lower, 30km figure near the poles; remember, this is in Siberia. This suggests that the bolide would have encountered the top of the stratosphere as a good approximation of a solid barrier, upon which most of its kinetic energy was converted to heat immediately on impact. This heat didn't merely vaporize the bolide, it turned it into an extremely hot plasma, which proceeded at hypersonic speed downward into the atmosphere, striking the surface and manifesting as an exploding fireball.

At an absolute minimum, I would expect to see some fairly hard X-rays; you may be right, gamma may be out of the question for so small an object, but we're in that ballpark.

 

[Schneibster]

Project Orion, that's how.

That could have delivered thousands of nuclear weapons across the solar system and, better than setting them all off at once (and changing a huge piece of rock on a collision course with the Earth into a million tuguska events!), it could then use their energy in a controlled manner to change the orbit of the rock to render it harmless for the forseeable future.

And all using 1960s technology.

OT, but actually not a bad idea. Probably worth a thread (I'd search to see if anyone has done it before first if I were you). Orion is a lot less problematic if you use it in space.

 

[Soapy Sam]

Agreed, but all in 11 milliseconds? Even at 17 km/s, that is only 187 meters. Surely we can assume that the last few kilometers were at fairly constant atmospheric pressure--within a factor of two, at any rate.

- Dr. Trintignant

Dr.T, where does the 17km/s figure come from?
I understood an object in a retrograde orbit (worst case scenario) could approach us at over 70km/s closing relative velocity.(I've seen numbers of 100kn/s mentioned). Would it decelerate so much in the brief passage through the atmosphere?

 

[Schneibster]

17km/s is the average velocity of asteroids with respect to Earth. Comets have higher average velocities- 40 to 70km/s. But we're talking about asteroids.

 

[BeAChooser]

Perhaps another explanation to consider:

http://www.bibliotecapleyades.net/ciencia/esp_ciencia_tunguska17.htm "Tunguska: The Fire In The Sky"


http://www.bibliotecapleyades.net/ciencia/esp_ciencia_tunguska20.htm "New Tunguska Crater Found?"

 

[Dr. Trintignant]

I think you're still ignoring the degree to which the atmosphere thickens.

Could be. I'm aware that it's (roughly speaking) an exponential dropoff. Of course what this means is that at any given altitude, there will only be a fixed percentage difference in pressure between that altitude and a fixed distance above that. 187 meters is something like a 2.5% difference--seems small to me.

Only once have I seen a bolide that I believe actually struck the surface, and it was out to sea by at least twenty miles, and more likely a hundred or so.

Neat. Might I ask specifically what your definition of "bolide" is? I've found several of them on the 'net, including one that only includes impactors.

Significantly, the Tunguska object's disintegration occurred very close to the stratosphere/mesosphere boundary

I thought it was much lower than that--5-10 km, not 30 km or more?

At an absolute minimum, I would expect to see some fairly hard X-rays; you may be right, gamma may be out of the question for so small an object, but we're in that ballpark.

Well, I will happily grant that if the heating occurs within tens of milliseconds, then you indeed should expect some rather energetic EM radiation, including X-rays.

I am still not entirely convinced that this would happen for a 30 m object, but this opinion is largely based on my ignorance of hypervelocity objects in the atmosphere. Perhaps sometime I will write a simple simulation that shows the balance between radiation and aerodynamic heating.

- Dr. Trintignant