Radar observations made during this week's close encounter with the asteroid Apophis have ruled out the risk of a catastrophic cosmic collision in 2036, NASA says. Experts say it'll be much farther away at that time than it is right now.

The crucial readings came on Wednesday when the space rock, which is thought to measure at least 885 feet (270 meters wide), approached within 9 million miles (14.5 million kilometers) of Earth. NASA is monitoring Apophis with its 230-foot (70-meter) Goldstone radio dish in California. Optical readings also have come in from the Magdalena Ridge Observatory in New Mexico and the Pan-STARRS observatory in Hawaii.

The bottom line? "We have effectively ruled out the possibility of an Earth impact by Apophis in 2036," Don Yeomans, manager of NASA's Near-Earth Object Program Office at the Jet Propulsion Laboratory, said today in the all-clear news release. "The impact odds as they stand now are less than one in a million, which makes us comfortable saying we can effectively rule out an Earth impact in 2036. Our interest in asteroid Apophis will essentially be for its scientific interest for the foreseeable future."

Jon Giorgini, who developed JPL's online Horizons database to keep track of solar system objects, would go even further. He says that according to calculations based on the Goldstone data, Apophis will probably pass by Earth at a distance of 36 million miles (58 million kilometers, or 0.39 AU), and absolutely no closer than 14 million miles (22 million kilometers, or 0.15 AU). "That is a very extreme minimum," he told NBC News. "Nothing else plausible can get you closer."

Apophis, a.k.a. 2004 MN4, created a huge splash when it was discovered in 2004 because the initial assessment of its orbit gave a 1-in-40 chance of Earth impact in 2029. That would be catastrophic: The space rock is big enough to wipe out a city if it struck land, or create killer tsunami waves if it splashed into the ocean.

Additional orbital data quickly eliminated the risk for 2029, but showed that it would pass within 20,000 miles (32,000 kilometers) of our planet at that time. That's so close that Earth's gravitational field will perturb Apophis' orbit. The experts worried that if the asteroid passed through a particular half-mile-wide zone in space, known as a "keyhole," its orbit would be perturbed just enough to set up a smash-up during the 2036 encounter. Fortunately, the latest observations indicate that Apophis will miss the keyhole by a long shot.

Did I just hear a cosmic sigh of relief?

UH / IA

The asteroid Apophis, highlighted here by a white circle, was discovered in June 2004.

There are still a few uncertainties surrounding Apophis: Astronomers don't yet have enough data to determine how the asteroid is spinning or how solar radiation is affecting its orbital path — a phenomenon known as the Yarkovsky effect. Giorgini said that even under the worst-case scenario, the effect won't push Apophis into a collision in 2036. But there could conceivably be other risky encounters in the decades or centuries ahead.

"There's a non-linear amplification that can really move it around more," Giorgini said.

Also, there are questions about Apophis' exact size. Just this week, readings from the European Space Agency's Herschel space telescope suggested that the asteroid may be nearly 20 percent bigger than previously thought. But that larger size estimate is based on the assumption that Apophis is a spheroid, and astronomers already know that it's elongated.

"We're not seeing that larger size in the radar data," Giorgini said.

By the end of next month, continued radar observations from Goldstone as well as the Arecibo Observatory in Puerto Rico should give astronomers a much better fix on Apophis' spin and its size. When those factors are fully accounted for, the Jet Propulsion Observatory will update its official risk assessment for Apophis — and could take this bad boy off the hit list for good.

Update for 6:30 p.m. ET: Clark Chapman, senior scientist at the Southwest Research Institute, weighed in on the current state of the asteroid hunt in an email:

"One thing you should be aware of, and might mention, is that the next Planetary Defense Conference, an every-two-year international meeting, will be held April 15-19 in Flagstaff, Arizona. ... Some presentations are already listed in the program, which should be finalized a week from now, which is the due date for abstracts.

"An interesting tie-in with the new observations of Apophis is that a similar thing happened with 2011 AG5 a few weeks ago, when observations with the huge Gemini telescope in Hawaii showed that it would, in 2023, miss the roughly 350-km-wide 'keyhole' and, therefore, not strike the Earth in 2040.  Prior to these critical observations, the chance of a 2040 impact was unusually high (though still low in everyday terms) at 1 in 500.

"A point to be realized is that while the chances of impact in these cases are very low by ordinary standards, they aren't zero, and the consequences of an impact could be very terrible, so it is important to plan and prepare for the possibility of impact until it is ruled out.

"It was important to get these observations of AG5 in the autumn of 2012, because if it had turned out that AG5 was actually on an impact trajectory, it would have given us an additional year to mount a deflection mission and succeed in deflecting it from the 2023 keyhole. Without making a major observational effort with a very large telescope this autumn, the next routine observational opportunity wasn't until this coming autumn."

Update for 8:30 p.m. ET: One of NASA's experts on the asteroid threat and two former NASA astronauts have weighed in on the report about Apophis. David Morrison of NASA's Ames Research Center sent these comments via email:

"One possible angle is the recent proposal from [NASA Administrator] Charlie Bolden, based on a Keck study, that we retrieve a 7-meter carbonaceous near-Earth asteroid and bring it into lunar orbit. There are many questions about this idea, but the one I have in mind is our assumed ability, without Sentinel, to find 7-meter C-type asteroids in Earthlike orbits. If you can't find them, you can’t protect against them, or do anything with them as potential resources." 

Now here's an email from Ed Lu, a veteran of two space shuttle missions and an extended stay on the International Space Station. Lu now serves as chairman and CEO of the B612 Foundation, which is planning to launch the Sentinel space telescope to track half a million near-Earth asteroids:

"While it is great that Apophis is much better understood, and we know it won't hit us in 2036, the greatest danger from an asteroid strike is from the ones we haven't yet found.  Of asteroids larger than the one that struck Tunguska in 1908, we know of less than 1 percent of them.  And as David Morrison points out, we can't protect ourselves from the unknown asteroids (or make use of them either). The B612 Foundation Sentinel Space Telescope is going to work on finding and tracking these asteroids."

And here are some comments from Apollo 9 astronaut Rusty Schweickart, who has played a key role in raising awareness about the threats and opportunities presented by near-Earth objects. It was Schweickart who warned in the wake of Hurricane Katrina that asteroids like Apophis could spark a much more devastating "cosmic Katrina":    

"I'm hoping that you don’t follow the bad (surprisingly wide) precedent of stating that [the risk from] Apophis has been eliminated.  Please look on the JPL risk page  and especially the more detailed info and note that 1) The 2036 impact possibility is, while significantly reduced, still possible, and 2) that the 2068 impact possibility is now elevated ... to a level that exceeds what the 2036 impact was prior to this apparition.

"There’s certainly good news re the 2036 impact decreasing in probability ... but frankly it was 1 in 234,000 prior to the new observations ... not exactly an impact probability to worry one. (There are many NEOs with higher impact probability ... but no one pays attention to them ... they aren't the 'poster child' that Apophis is.) My personal reaction was one of surprise that the new 2036 impact was not zero!

"But/And ... there are more radar observations to integrate in ... as well as optical tracking both now and for the next several years.  Apophis isn't going away ... the impact possibilities are simply shifting around a bit with refinement of the tracking data. 2036 is now less probable; 2068 is now more probable (but still very low).

"Until JPL and the other guys get more data (enough to really define the Yarkovsky effect), we really won’t be able to get definitive data for longer time scales that we can rely on."

JPL's Giorgini said the risk assessment that Schweickart mentioned won't be full updated until after Goldstone and Arecibo finish their observational campaign in mid-February — so there may still be a non-zero risk listed until then. But Giorgini is confident that the 2036 risk will disappear when all the observations are factored in. (As of this writing, the estimated risk of collision is listed at 1 chance out of 10,989,000.) But you're right, Rusty: In order to eliminate the risk completely, astronomers will have to get more data about Apophis' physical characteristics. And then there are all those other unknown killer asteroids that might be out to get us...

More doomsday worries addressed:

• Asteroid 2012 DA14 won't hit Earth next month
• Asteroid 2011 AG5 won't hit Earth in 2040
• This year's big comets won't pose a threat
• Asteroid-hunting telescope to be launched in 2017

Alan Boyle is NBCNews.com's science editor. Connect with the Cosmic Log community by "liking" the log's Facebook page, following @b0yle on Twitter and adding the Cosmic Log page to your Google+ presence. To keep up with Cosmic Log as well as NBCNews.com's other stories about science and space, sign up for the Tech & Science newsletter, delivered to your email in-box every weekday. You can also check out "The Case for Pluto," my book about the controversial dwarf planet and the search for new worlds.

Planetary Society

Researchers are looking into the possibility of sending a swarm of "laser bee" satellites to deflect a potentially hazardous asteroid, as shown in this artist's conception.

How do you solve a problem like Apophis? For years, researchers have been tracking the asteroid to determine whether or not it could get into a catastrophic smash-up with Earth on Friday the 13th in April 2036. Last week, Russian reports suggested that Moscow is thinking about launching a probe to Apophis as early as 2015 to track the threat, but U.S. experts say it's the wrong idea for the wrong asteroid.

So what's the right idea? They have a modest proposal that just might save the planet someday.

First, here's a little more background on Apophis: The 885-foot-wide (270-meter-wide) space rock was discovered in 2004 and set alarm bells ringing when experts said it had a 1-in-40 chance of hitting Earth in 2036, due to the uncertainty about its orbit. Since then, astronomers have gotten a better fix on Apophis' parameters and set the chances of collision at 1 in 240,000 or so. Some uncertainty remains, because if the asteroid passes through a relatively small region of space known as a "keyhole" in 2029, Earth's gravitational pull would deflect its course just enough to guarantee the hit in 2036.

Last week, the Russian Academy of Sciences recommended sending a small, radioisotope-powered probe with a radio beacon on board, so that astronomers could get a high-precision fix on Apophis' trajectory. "From a technical point of view, the mission could be started for implementation from 2015," the academy was quoted as saying by the RIA Novosti news service.

I ran that idea past several experts on the asteroid threat, and the consensus was that you almost certainly wouldn't need that kind of mission to rule out a collision threat. Retired astronaut Rusty Schweickart, who has become deeply involved in studying potential threats from near-Earth objects and how to address them, noted that there's an excellent opportunity for high-precision, Earth-based observations of the asteroid coming up in early 2013. Another opportunity comes along in 2021. Chances are that the upcoming observations will turn Apophis into a complete non-threat.

"Why, given this circumstance, the Russians would send a probe to Apophis in order to pin down its impact probability is indeed puzzling," he told me in an email.

Scientific value
That's not to say a trip to Apophis would be totally useless. Far from it.

"I would support a mission to study Apophis, either a lander or an 'orbiter' rendezvous mission," David Morrison, director of the SETI Institute's Carl Sagan Center for Study of Life in the Universe, wrote in email. "I wish the Russians luck. This fascinating object is a favorable target mostly because it will come so close to Earth in 2029, so that ground-based observations can be made in synergy with in situ measurements."  

Clark Chapman, a planetary scientist at the Southwest Research Institute, agreed that there are plenty of good scientific reasons for studying Apophis up close. "It is expected that the tidal effects on the body could change its characteristics markedly ... and I know a number of colleagues who would like to see such a mission," Chapman wrote. But based on RIA Novosti's report, the kind of mission that the Russians are contemplating may not fill the bill. In fact, it could do more harm than good.

"I wouldn't totally dismiss the concerns about physically interacting with Apophis during its close approach," Chapman said. "If a spacecraft were actually to dock with it, there is the possibility of interfering with the natural responses that are of interest ... and, in the extremely low-probability case that Apophis were right at the threshold of passing through a keyhole, tampering with it could raise legal issues.  I'm not aware that anyone has deeply thought through these concerns."

The right target
All three of the experts are giving deep thought to another potentially threatening asteroid, a 460-foot-wide (140-meter-wide) rock that's been designated 2011 AG5. Right now, the rock has been given a 1-in-500 chance of hitting Earth on Feb. 5, 2040, due to the uncertainties about its orbit and its location relative to a cosmic keyhole in 2023.

Schweickart has been calling on NASA to start making preliminary plans for a mission to 2011 AG5, just in case we need to get a better fix on its orbit between now and 2023. If observations over the next couple of years eliminate the possibility of a threat, hallelujah! But if they don't, and if we're still worried about AG5, we'll have a head start on the asteroid deflection campaign.

"So it would seem that Russia is heading toward the wrong object!" Schweickart wrote. "There could hardly be a better example of the need, not currently in place, for international coordination when it comes to NEO [near-Earth object] impact analysis and deflection planning. So why aren't we?? Ask NASA. Please."

Chapman had a similar assessment: "I agree with Rusty that precisely determining the orbit of Apophis seems like a low priority at this time.  And I also agree that, as of now, 2011 AG5 is the near-Earth asteroid of interest."

Back in February, NASA's Jet Propulsion Laboratory issued a statement noting that there'll be good opportunities for observations of 2011 AG5 in late 2013 and late 2015. "I fully expect we will be able to significantly reduce or rule out entirely any impact probability for the foreseeable future," said Don Yeomans, manager of NASA's Near Earth Object Program Office.

Zapping vs. whacking
Even if Apophis and 2011 AG5 turn out to be totally harmless, it's not such a bad idea to have the plans for a "reference mission" to a worrisome asteroid ready to put into effect. And if it turns out that a future asteroid really is on a collision course, what should we do about it?

That brings us to another angle in the debate over near-Earth impacts: Some have suggested sending in the nukes. Others have proposed launching a "gravity tractor" to shepherd a threatening asteroid into a non-threatening orbit. Researchers at the University of Strathclyde in the Scottish city of Glasgow have been working on a different approach: using a swarm of solar-powered, laser-equipped satellites to blast away at the rock and deflect it.

They say their system could be effective with objects that are smaller than Apophis — say, 150 feet (50 meters) across, like the object that is thought to have blasted a Siberian forest to bits in 1908.

"We could reduce the threat posed by the potential collision with small- to medium-size objects using a flotilla of small agile spacecraft, each equipped with a highly efficient laser which is much more feasible than a single large spacecraft carrying a multi-megawatt [laser]," engineering professor Massimiliano Vasile said in a news release. "Our system is scalable, a larger asteroid would require adding one or more spacecraft to the flotilla, and intrinsically redundant. If one spacecraft fails, the others can continue."

Schweickart said researchers have been talking about zapping asteroids  for a few years. "I've never been able to convince myself that they've ever done any cost-effectiveness investigation ... assuming that they can actually solve the multitude of other technical issues inherent in the concept," he wrote.

It may well be that crashing one of the swarming satellites into an asteroid would have more of an effect on its trajectory than having the whole flotilla aim lasers at it, Schweickart said. In his email, he laid out the more likely scenario:

"It is difficult to take this proposal seriously when compared with the already available kinetic impact [KI] solution.  About the only advantage it has over a simple kinetic impact is that it would presumably be able to execute a precise orbit change (vs. the KI approximate orbit change).  However, in the last few years it has become generally accepted in the NEO deflection 'community' (such as it is!) that a gravity tractor capable 'observer' spacecraft would be pre-positioned to observe, verify, and ultimately precisely adjust (if necessary) the KI-generated orbit change.  The cost, current availability and simplicity of the KI/GT deflection concept is still the 'standard' against which other systems will have to be compared... and especially re their cost effectiveness."

Morrison agreed: "What is needed for deflection is to change the momentum of the asteroid in a controlled way. Surely a simple impact by a fast-moving interceptor is the most efficient way to do that, since the spacecraft is not required to brake and rendezvous, and then fly in formation with, the target."

Former astronaut Ed Lu, president and CEO of the B612 Foundation, also cast a vote for hitting an asteroid with something, if that's what needed to be done: "I couldn't agree more with Rusty's comment about how the real-life feasibility of such schemes needs to be taken into account.  The same could be said of suggestions that asteroids could simply be 'painted.'"

So it sounds as if the right course of action is to draw up the plans for a quick trip to a potentially hazardous asteroid, just in case we need some up-close reconnaissance, and working out a strategy for giving it a good whack if necessary. NASA's OSIRIS-Rex mission, due for launch in 2016, might serve as a good first step. The target for that $800 million mission is the asteroid 1999 RQ36, which has an ever-so-slight chance of threatening Earth in 2182. By then, we should have our anti-asteroid strategy well in hand. And who knows? We might even have to put the strategy into practice long before then.

Are we on the right track to avoid the dinosaurs' fate? Where would you rank the asteroid threat on your list of worries? Feel free to register your opinion in the vote above, and/or the comment space below.

Update for 1:40 a.m. ET April 13: I've made a few tweaks to the story in consultation with Schweickart. In an email, he said a successful asteroid deflection campaign requires two missions: first, an observer spacecraft to monitor the asteroid's position, and then a kinetic impact spacecraft to do the whacking:

"The observer spacecraft is more properly designated a 'transponder/gravity tractor' spacecraft ... the gravity tractor being potentially needed to adjust the NEO orbit slightly after the KI impact. Not either/or, but both/and. There should never be a KI deflection without a transponder/GT spacecraft in place before, during and after the KI impact. It's both/and."

Got it? On it!

More about asteroids:

• Asteroid debate rises to next level
• To fight off asteroids, humans must cooperate
• Largest digital camera hunts killer asteroids
• Here's how to counter a killer asteroid
• Interactive: Close encounters of the asteroid kind

Connect with the Cosmic Log community by "liking" the log's Facebook page, following @b0yle on Twitter or following the Cosmic Log Google+ page. You can also check out "The Case for Pluto," my book about the controversial dwarf planet and the search for new worlds.