Slashdot Mirror
Schoolboy Corrects NASA's Math On Killer Asteroid
spiracle writes "A German schoolboy, Nico Marquardt, has revised NASA's figures for the chances that the Apophis asteroid will hit earth. Apparently if the asteroid hits a satellite in 2029, its path could be diverted enough to cause it to collide with Earth on the next orbit, in 2036. NASA had calculated the chances as 1 in 45,000 but the 13-year-old, in his science project, made it 1 in 450. NASA agreed." Update: 04/16 16:47 GMT by Z : This is not entirely accurate, it turns out — more details.
Not peer reviewed.
NASA previously estimated the chance "Apophis" the asteroid would strike earth in 2027 was 1 in 45,000. But a german schoolboy, Nico Marquardt, pointed out that NASA overlooked the probability the asteroid would strike one of the 40,000 sattelites orbiting Earth and enter a new solar orbit intersecting Earth in 2036. A german newspaper reports that NASA now concurs the chance this will happen is about 1 in 450. If the 200 billion tonne ball of iridium and iron stikes the planet then it's literally light's out for earth: 800 foot tidal waves followed by an indefinite period of dust cloud covered darkness, not to mention metal vapor in the atmosphere. The original Slashdot discussion was in 2007 when the odds were better. At that time it was known that there was a small risk of a gravitational slingshot dropping it into the 2036 collisional orbit, however, to do so the asteroid had to pass through an improbable 400 meter wide strike zone to be properly deflected, as described in 2006 in Popular Science from 2006. Today's announcement of the new finding is here and here.
Some drink at the fountain of knowledge. Others just gargle.
one hundred, do you by chance work at NASA? :P
Sigs are too short to say anything truly profound so read the above post instead.
And the 2038k problem solves itself, thus vindicating Ken Thompson and pessimists everywhere.
I have a theory that the truth is never told during the nine-to-five hours. - Hunter S. Thompson
I hope that kid won the science competition he was in!
"... and for my science project, I proved NASA wrong and made a discovery of potentially epic proportions..."
Kindof tough to follow that one.
... it will create a ball of iron and iridium 320 metres (1049 feet) wide and weighing 200 billion tonnesNext week: 13 year old boy discovers new chemical reaction in which a combination of scientifically illiterate PR bunnies and sub-editors produces large quantities of bullshit.
And this observation is applicable for exactly what reason? Are you claiming that NASA management is screwing up the calculations? Or are you just talking out of your ass and trying to insinuate that NASA is always incompetent in whatever it does?
Btw, in case you are not aware, the NEO office is at JPL--not JSC. And JPL is run by Caltech for NASA--not directly by NASA.
Now that we have that cleared up you should feel free to continue your bullshitting and insinuating via hearsay.
Thoughts of another schoolboy:
"But if we make it strike the Earth and not one of those sattelites in 2029, the probability of it striking the Earth in 2036 is NIL. NASA agreed."
Here is the original article, in German, from the German newspaper. It looks like a professor helped him (Professor Spahn from Potsdam University). Bild is semi-infamous in Europe for sensationalizing stories but at least we know that the boy is real if nothing else...
Let's see. We begin with the original source of data, "telescopic observations." Good, but perhaps a bit, shall we say, lacking in nine-digit precision. Then we add the element of a bright schoolboy (always a favorite in the papers) doing something big and being validated (instantly!) by "NASA" (not a person, but apparently the entire agency). Oh, and "NASA" told "ESA", but we still don't have the identity of anyone other than the putative schoolboy.
So far, doing well.
Then we hit the big problems. First, we have the scare factor of "40,000" satellites surrounding Earth. Most of which, actually, are in LEO, with a few more in geosynchronous orbit. Which makes the space around the Earth only about 99.999% empty space, rather than a few more nines. As it turns out, space is big.
But it sounds good to imply that somehow there's this asteroid belt around the earth, and that the "killer" asteroid might hit a satellite.
Well, WHICH ONE? They have a lot of different masses, they are going in different directions, and we pretty much have to get a specific momentum change in the right direction in order to get just the right perturbation. Hitting a small piece of space junk is one thing, but the variation in weight of those "40,000" satellites is orders of magnitude. And that makes a big difference in orbital perturbation, even if the difference in orbital velocity is small compared to the velocity of the asteroid. We're talking about a subtle effect here.
And let's not figure in things like elastic collisions, off-center collisions, pieces flying off, or anything else. Nope, it's gonna happen perfectly, just like that seven-ball four-cushion bank shot we all can hit again and again.
Heck, they even called the pocket. Right into the Atlantic, after an orbit measuring in the decades. Now I will grant that the orbit is pretty well known, but again, that little "satellite assist" must be just precise as heck.
A nice touch gives us the "destroy both coasts and darken the world indefinitely." While it's good to be so certain, couldn't they be more specific about the method of destruction? Seeing as how they apparently know everything else, and all.
And finally, we have the 450:1 odds. Not 500:1, and certainly not 1000:1, but exactly 450. Cool. About as believable as my old homework excuses, but infinitely cooler. Can you say "significant figures"? I knew you could.
I think it's what you get when you let AFP (my source of news of the world for sure) loose in spring.
Now lets burn down the observatory so this can never happen again!
> 26,000 Hirshimas
So a little less than 1 Mt St Helens then.
And now Hollywood can turn the German boy into an American boy, chuck the complex math for a backyard telescope, name the asteriod after the boy, throw in a baby to add drama and get Morgan Freeman to play the President... Oh wait... ...never mind.
Righhht.... sure, Lord Apathy, we believe ya....
That's the depressing part.
To answer your question: Probably a few months after the 2027 encounter (and hypothetical collision with the satellite), but at that point, it'll be impossible to do anything about it in the 9 years between 2027 and 2036.
The right strategy is to use the 20 years between now and 2027 to build an orbiter/lander (with a big-ass nuke, nuclear reactor powering a big-ass laser, or big-ass solar sail of reflective/absorptive paint -- and as much as I like nukes, the big can of paint's probably the best way to go -- attached).
We use the 20 years to build the orbiter/lander. We send it up to rendezvous or orbit in 2027. If Apophis smacks into a satellite (or we're just unlucky), we'll have an orbiter and countermeasures in orbit around the asteroid on that pass, and those countermeasures will have nine years in which to do their work. A nuke's pretty cool, but it can't compete with nine years of momentum transfer from the sun shining on a rock painted white on one side and black on the other side.
Suppose we cut it short and by 2027 we still don't have any good countermeasures - just a crappy-ass nuke as a last-ditch measure. Even if we go this route, we've still got 9 years for this orbiter to give us an exact gravity map of this object, and we'll have a couple of years after that to figure out where to land the nuke for maximum trajectory deflection away from the earth. (Hell, if we get the orbiter up there early enough in 2027, we can blow the nuke at/near closest approach to Earth and guarantee a miss in 2036!)
But we're short-sighted. So we'll do nothing between now and 2027. And odds are it'll sail on by in 2027 and we'll conclude that the odds of an impact in 2036 are only one in a few tens of thousands. But what an irony -- if we're wrong, then it'll be too late in 2028 for us to send anything to catch up to the rock and do anything about it. For the sake of a month's pork-barrel spending in Iraq, we'll condemn a few billion of our fellow humans to certain death in 2036.
If it's not Apophis, it'll be some other rock in the next few centuries. Just like the dinosaurs, we'll go extinct because we don't have a space programme. Unlike the dinosaurs, this time around, we'll deserve it.
Well, think on the bright side.
At least we don't have to worry about fixing the 2038 UNIX 32-bit date bug any more.
We just need to make sure it is in a place like Germany, France or Russia where they still teach terrorist material like mathematics, physics and chemistry in school.
Baker's Law: Misery no longer loves company. Nowadays it insists on it
http://www.sigsegv.cx/
I'd love to see the computer-cluster this kid calculated the 100k+ iterations of the many-bodied system of time-retarded lagrangians required to solve something like this. His parents power bill has to be insane. Considering uncertainties involved in orbital trajectories and timing for asteroids like this, 100k might even be a low number of runs for something like this. The number of satellites in orbit, their varying masses, uncertainty in the current un-colided trajectory, etc. can't possibly create a situation where you have improved odds of impact anyways. There is actually a greater solid angle of impact for collisions that would decrease the likelihood of eventual earth impact than increase it. Maybe these odds are after the most favorable possible satellite impact plus the help of magical space faries?
It appears we have a 200 billion tonne asteroid in a possible collision against a satellite weighing between 200KG and several tonnes
I'd say it's more like the haystack hitting the needle.
Even so, it would be nice to see the math. The only place I have seen such equations solved is in Feynman's Physics volumes, which unfortunately I lost to Katrina.
What is the error estimate on the precise trajectory of the asteroid and its velocity? How can they arrive at a 400 m window, when they don't even have a good tracking of all the space junk in orbit? How many satelites were taken into consideration in reaching the 1:450 number? Can these really be ignored if the trajectory is to be computed this precisely? Have all the calculations taken into account numerical precision associated with floating point representation? Have the gravitational effects of the other planets been adequately accounted for? With what precision?
Just questions it would be interesting to look at to assess how these figures are arrived at.
It wouold be instructive to see what figures NASA or the German schoolboy used in their equations.
26,000 Hiroshimas?! Why, that's almost an Africa!
I hate printers.
No need to translate it to make it hilarious:
a) The source "newspaper" is Germany's biggest tabloid - with as much knowledge on astrophysics as a kindergarten kid
b) No 13 year old German kid says "stuerben"
Live today, because you never know what tomorrow brings
As it turns out, space is big
You may think it's a long way down the road to the chemist's, but that's just peanuts to space.
I am one of many. My idea is not unique, nor do I expect my voice alone to sway you. I speak in a chorus of opinion.
This just in: Kindergarten kid corrects 13 year old student's earlier correction of NASA calculation.
Chance of impact now 1 in 4.
Toddler's have be banned from using calculators for fear they will doom us all.
Doom Us All, I tells ya!
Genesis 1:32 And God typed
Not so, Sir. They EVOLVED into Italians.
How amazed would you be to suddenly find that you just forgot what I wrote and you needed to reread my post.... again.
Sorry to burst the balloon, but apart from the one German article that was picked up by AFP, there's no source for this story. And NASA and the ESA deny ever saying that the schoolboy was right. It seems that the schoolboy's sums were wrong, and NASA's original workings are right. More info: http://www.theregister.co.uk/2008/04/16/esa_german_schoolboy_apophis_denial/ I'd file this under 'web hoax' or 'lazy journalists pick up on anything sensational'
Just in case anyone still believes we'll all be killed by an asteroid in 2029 or 2036, here's an article from El Reg, claiming that the boy got it all wrong.
I guess we'll have to live with the miniscule 1 in 45,000 chance.
link to article:
http://www.theregister.co.uk/2008/04/16/esa_german_schoolboy_apophis_denial/