Meteorite on a collision course with earth-- how to NOT see it coming?

February 6, 2013 4:08 PM

What are some reasonably-realistic scientific scenarios in which, assuming all the current technology we have at the moment was in working order, scientists might miss seeing a potentially catastrophic meteorite on a collision course with Earth until it was actually entering the atmosphere? I assume that astrophysicists see potentially threatening asteroids/meteors coming months, if not years, in advance. But could there be a circumstance in which that might not be the case? Thanks!
posted by np312 to science & nature (18 answers total)

There are a lot of objects flying around in space, and we've got the tiny-est understanding of only a tiny tiny fraction of them. I don't think it's too implausible that there are objects on a collision course which we just haven't seen yet.

However, I do think that a large object would probably be seen sooner than as it entered the atmosphere. A really sneaky asteroid or meteor might show up a few weeks before impact, but my sense is that at some point it becomes bright enough to be noticed, at which point it's immediately tracked and it's trajectory calculated.

I don't know if this is plausible, but an object with a very low albedo, e.g. black, would be much harder to notice than your standard white ice-meteor. Or if it was noticed, its size would be improperly calculated.

There's also the idea that it might come in directly from the sun, or behind a moon, but even those are pretty dodgy, since there are satellites that can see around the moon, and it, you know, moves out of the way constantly. Also, telescopes are pointed at the suns and notice little sun flares and spots, so a meteor might be picked up there too.
posted by brenton at 4:17 PM on February 6 [3 favorites]

Small Asteroid 2009 VA passed incredibly close to the Earth (within 14,000 of the Earth's surface) in November 2009, and was discovered only about 15 hours before the close approach. It was quite small, though. 2008 TC3 was discovered under similar circumstances on a trajectory to hit Earth with less than 24 hours notice. It entered the Earth's atmosphere on October 7, 2008 and exploded 23 miles above the desert in Sudan.
posted by The World Famous at 4:18 PM on February 6 [1 favorite]

We observe things because either they reflect radiation, or generate it -- We've just started detecting 'dark' asteroids', but if you have a "smallish" object that's not reflecting much and not radiating much heat, it's going to go a long time without being detected by our current technology. I don't know about entering the atmosphere, though: we track very small items within a few thousand miles of the earth's surface, and well outside of the atmosphere; geosynchronous orbit is 22,000 miles, 10% of the way to the moon, so everybody's pretty well aware of what's nearby. Past the moon, though, that's where it gets pretty fuzzy. But, even within the moon's orbit, it could only be less than a day, within hours, before it hits. Take in the delay of people debating what exactly they've seen: it's not like, at 50,000 miles away, trying to track something moving so fast, that people will instantly know what they just saw. It may take a while for scientists at different spots on the earth to compare data and evaluate speed and trajectory. A sufficiently fast object, coming straight in (not across, giving us hundreds of thousands of miles of time to collect data) might be too fast to predict enough for early warning.
posted by AzraelBrown at 4:20 PM on February 6

Very easy. Just have it be interior to the Earth's orbit, so that it's only up in the daytime.

If we're going to get the kiss of death from a spacerock, that's the scenario my money's on.
posted by BrashTech at 4:21 PM on February 6 [1 favorite]

These aren't great, but hey:
1) A comet that has an elliptic orbit which finally takes it close to the sun. Being partially made of ice, it splinters and the ejection alters the path enough to send it to Earth.
2) A mini-planetoid (think Pluto) we are normally not worried about, simply because it's known not to cause problems for Earth. But then it collides with an actual planet, spraying real debris.
3) Any aliens sophisticated enough to send out a probe to other solar systems would be able to make it dark enough to hide from us.
posted by tintexas at 4:22 PM on February 6

This very topic comes up in Bill Bryson's A Short History of Nearly Everything. The conclusion he came to with the aid of a scientist was that it was extremely likely that we wouldn't see it unless someone happened to have a telescope pointed in the right direction at the right moment and even if they did manage that up to a year in advance, we wouldn't have the technology to do anything to stop said meteor.

The theme of the book is pretty much we could all die at any moment via asteroid, geothermal vent, killer bacteria, etc and we should all be very grateful to be alive. Also fun science gossip!
posted by chatongriffes at 4:26 PM on February 6 [2 favorites]

It is not totally implausible that an iron meteorite might be covered with graphite powder, and in that case it would be mighty difficult to spot with a telescope.
posted by Chocolate Pickle at 4:33 PM on February 6

In 2008, NASA was under a congressional mandate to discover at least 90% of Near Earth Objects (NEOs) with diameters larger than 1 km within ten years. And there was talk at some time (not sure what happened) about extending this to smaller sizes, too.

To do this, they have to monitor the (preferably entire) sky over and over and look for things that move. The motion gives them insight into the distance (the more epochs the better), and the brightness to estimate the size of the object.

You could miss an NEO if you don't monitor the *full* sky regularly. This is a lot of work! Plus, you can miss out on portions of the sky you mean to monitor if it is cloudy.

If the monitoring is done at optical wavelengths, the size estimates can be pretty bad, because the albedo is unconstrained and just assumed to be a constant. So you could just dismiss an NEO as not being catastrophic because you think it's smaller than it really is.

Monitoring at infrared wavelengths, say with the WISE and Spitzer missions, provides better information about the size. And it can discover dark asteroids that would be missed at optical wavelengths, as mentioned above. However, infrared observations are tough to do from the ground, so you have to go to space. This is expensive, and the missions are short-lived because of the need for coolant. Also, not all infrared missions are full-sky (e.g. Spitzer). In this example, both the Spitzer and WISE missions are over.

No matter how you do the initial discovery of an NEO, you can "lose" it if you don't continue monitoring. People write telescope proposals to continue montoring e.g. WISE-discovered NEOs, but time allocation committees ultimately have to decide if it is the most compelling science. If the wrong person writes a poor proposal, a catastrophe might result!
posted by pizzazz at 4:59 PM on February 6 [2 favorites]

The B612 Foundation is the place to go for this info. from their site "The B612 Foundation is a 501(c)3 non-profit charitable organization started by a group of notable scientists, including two former NASA astronauts. Our vision is to harness the power of science and technology to protect the future of our planet, while extending our reach into the solar system." "Our solar system is a busy place. Ninety-nine percent of Near-Earth Asteroids larger than 40 meters have not been mapped and therefore we do not know if a collision with Earth is imminent."
posted by anon4now at 5:49 PM on February 6 [1 favorite]

Another mission that doesn't exist yet is JPL's NEOCAM. It will also focus on the smaller but still dangerous NEOs. neocam.ipac.caltech.edu
posted by pizzazz at 7:08 PM on February 6 [1 favorite]

If something was going fast enough, say a notable fraction of the speed of light, (Hundreds of times faster than known bodies orbiting the sun) we wouldn't notice it until it was close to hitting us, timewise. An asteroid going 0.1c would take 13 seconds to travel from the orbit of the moon to the earth.

The chance would be very rare. It would be something from outside the solar system, created by a violent star event (like going nova) moving so fast that Earth's (or even the Sun's) gravity have a chance to trap it and draw it into a collision. Essentially some fragment of a planet traveling hundreds of light years, to bullseye Earth. As far as I know we've never spotted anything like it. Of course, that's kind of the point.
posted by Ookseer at 8:20 PM on February 6

Bill Bryson (and his scientist) are right: a destructive meteor could fall from the heavens at any moment, and the first thing anyone would notice would be the enormous crash of it hitting the ground. Maybe it's happening right now. Look out your window!

Incidentally, I personally find this kind of poetic. In spite of the amazing state of modern technology, communications, transportation, etc, we're still vulnerable to the most brutish of disasters: at any moment a heavy thing could fall from a high place and destroy a city.

The basic problems are the following: 1) the sky is very big 2) potentially destructive objects can be very small 3) potentially destructive objects don't spend much time close to the Earth and are hence likely to be very dim.

I often get questions from the public like "I was driving the other night and saw a strange flash in the sky. It was sort of in the south east, between 11 pm and midnight. What was it?" People seem to imagine that astronomers are continuously monitoring the whole sky and there's some sort of "mission control" room where every blip in the sky is recorded. Then "mission control" sends out daily announcements "here's what happened in the sky last night." This is... not the case.

Getting back to the basic problems above: asteroids don't emit light themselves, only reflect light from the sun, and they're small and far from both the sun and the Earth, so the total amount of light that reaches the Earth is very small. Therefore to look for them, you want to use a big telescope, with a big primary mirror to collect a lot of light, so that you can actually detect them with a reasonably short exposure.

The problem with using a big telescope is that you have to take all that light and focus it down to a small area so that it falls onto a detector. That means high magnification which means small field of view. If you could build as big a detector as you wanted, this wouldn't be such a problem. However, today, the combination of detector pixel size, overall detector size, and the optical design of a telescope to take light from a big mirror and make it fall on the detector place a limit on how large your field of view can be.

To put this in perspective, typical fields of view are a few arcminutes (60 arcminutes = 1 degree) on a side. The Advanced Camera for Surveys on the Hubble Space Telescope had a field of view of 3 am x 3 am, which means you need 16 million exposures to tile the entire sky. The most prominent wide field camera is probably the MegaCam on the Canada France Hawaii Telescope, which has a field of view of 1 degree x 1 degree. That still means 40,000 exposures to tile the sky. Multiply those numbers by two or three since it's standard to take several exposures of the same patch of sky to remove the effect of cosmic rays that hit the camera during the exposure. With MegaCam, if you want to cover the whole sky once per year, you can only tolerate exposures of about a minute, and that's assuming the telescope does nothing else. It's just a lot of telescope time to monitor the whole sky, and both of those telescopes have many other scientific priorities.

There are three separate projects that are building telescopes that will be dedicated to monitoring the entire sky and which state that cataloging near-earth asteroids are one of their primary goals. Two have been in the works since the early 2000's: LSST, the (Large Synoptic Survey Telescope) and Panstarrs. When I first heard of them, they were both planning to be taking data more or less now. Now, they're planning to start taking data circa 2020. Both projects plan to cover the whole sky once per week with exposures of about fifteen minutes if I recall. They do this basically by designing the telescope with field-of-view as the only consideration, and hence achieve what is by the usual standards an enormous field of view. The other project, B612 mentioned above, is privately funded and involves Ball Aerospace. It looks like they're on track to beat the other two projects, but proponents of a project are always optimistic... we'll see.

Once one of those three projects is running for five or ten years, the threat of sudden asteroid disaster will recede and the state of things will actually be what people seem to imagine it is now: we'll have a more or less complete catalog of object that can do significant damage and we'll have lots of warning if one is going to hit the earth. But that's not the state of affairs today.

I've had some frustrating conversations with colleagues about this subject, where they say something like "Yeah, but they're just asteroids, they're not interesting. We should dedicate those resources to [some other project]." Monitoring the sky for near earth asteroids is probably the only thing that could be considered a real practical application of astrophysics. Everything else in astrophysics is fascinating and sometimes philosophically profound (understanding our origins by studying cosmology and planet formation come to mind), but it really cannot be considered to have a practical application.
posted by ngc4486 at 2:50 AM on February 7 [2 favorites]

ngc4486: Pan-STARRS has actually been in operation for a few years. It doesn't have a large enough collecting area to achieve the congressional mandate, though. Even the proposed 4-telescope version (PS4) wouldn't be sufficient, and it's currently not funded, so couldn't be counted on. LSST takes two 15-second exposures every 3 nights, covering the entire sky visible from the Southern hemisphere at that rhythm.

As luck will have it, I was on a telecon today where NASA's current plans for finding killer asteroids were discussed. Basically, they want to take both approaches - a space-based mission (very expensive - hundreds of millions of $, probably) plus a ground-based search (comparatively cheap), as they can find different sorts of objects with each.

LSST could meet the current Congressional mandate (finding 90% of all Earth-crossing asteroids >140m in diameter, basically those large enough to destroy a city) in 12 years of operation, but the current goal is funding for ten years (as that's what we need for dark energy science)... If LSST operates for 12 years, that would mean the Congressional goal that was originally set as an aim for 2020 would be met somewhere around 2033. NASA's apparently interested in finding objects as small as 50m across (as those are big enough that astronauts could land on them), but presumably we care less about finding a complete set of such objects.

The main reason it would take LSST so long to find these objects, and that we'd only get to ~90%, is that (as was mentioned upthread) from Earth, we can't spot asteroids that are hanging out in the direction of the Sun (imagine asteroids on an approximately counter-Earth orbit so they generally stay on the opposite side of the Sun from us). The good news is that it would take a fair amount of time for such an object to work its way around towards us.

Given that their aims are dead on the congressional mandate, I think B612 must be angling for NASA funding ultimately. This is the first I'd heard of the project, though - it's had zero impact in the astronomical community.

tl;dr: Basically, with an extended LSST survey or LSST + a space mission, we can stop being concerned about unknown killer asteroids coming to get us, and just worry about undiscovered comets showing up and ending civilization.
posted by janewman at 2:45 PM on February 7

On Cosmic Log today:

However, only about a third of the objects between a kilometer and 100 meters (330 feet) are being tracked. And NASA has detected only a small proportion of the estimated 1 million asteroids that are smaller than 100 meters but still capable of doing significant damage — asteroids like 2011 DA14.

"It's an effort that will take another decade or two," said Lindley Johnson, program executive for NASA's Near-Earth Object Observations Program.

posted by unliteral at 5:25 PM on February 7

2008 TC3 previously.
posted by fantabulous timewaster at 1:29 PM on February 8

We do have some items with mass hit the earth's atmosphere at near light speed. Thankfully nothing big enough to cause more than a flash in the sky... so far.
posted by Admira at 12:23 AM on February 11

Thanks everyone!!
posted by np312 at 2:04 PM on February 11

How was the news today. Meteorites hit earth.
posted by unliteral at 4:41 AM on February 15