If the meltdown at Japan's stricken nuclear plant goes total, experts don't expect to see a "China syndrome" scenario or a Chernobyl-style conflagration. But the situation would be worse than it is now — which is why Japanese authorities have been working so hard to stabilize the situation at the Fukushima Dai-ichi complex. Emergency operations were briefly suspended on Wednesday, Japan time, due to a spike in radiation levels.

Just how much worse could things get? That's a matter of debate.

Princeton nuclear physicist Frank von Hippel suggests that we're already seeing the major effects of the meltdown, in the form of periodic releases of radioactivity from the reactors as well as from a fire-damaged storage facility for nuclear fuel rods.

"In a sense, the worst has happened already, with the fuel releasing much of its volatile radionuclides," von Hippel said today during an msnbc.com chat about the nuclear crisis. "The major danger from a meltdown would be a low-probability steam explosion if the molten core fell into a pool of water."

On the other end of the spectrum, a big explosion is exactly what Masashi Goto, a former nuclear design engineer at Toshiba, is worried about. During a Tokyo news briefing presented by the Citizen's Nuclear Information Center, which is generally critical of nuclear power, Goto said the molten core could spark a steam explosion ... or another hydrogen gas explosion like the ones that have rocked the reactor complex over the past few days. He said a full core meltdown could also set off a fresh nuclear chain reaction, much like the one that occurred in 1999 at Japan's Tokai processing plant.

"It could trigger the resumption of criticality," Goto said.

Experts on nuclear power say that the seriousness of the Fukushima Dai-ichi currently rates somewhere between Pennsylvania's 1979 Three Mile Island incident, in which the reactor's core melted down halfway but was kept contained within the facility; and the 1986 Chernobyl incident in Ukraine, in which a raging, uncontained fire spread radioactive contamination throughout Europe.

Japanese authorities estimate that 70 percent of the fuel rods are damaged at the complex's reactor No. 1, with 33 percent damage to the rods in reactor No. 2. The storage pool for fuel rods at reactor No. 4 has weathered two fires and an explosion, which apparently led to the release of radioactive plumes through gaps in the reactor building's walls. Meanwhile, water levels are reportedly dropping in other storage pools.

Water is the key
To head off further damage, Tokyo Electric Power Co. is talking about adding water and boric acid into the reactors and the storage pools, from helicopters as well as fire trucks. The water replaces the liquid that's boiling off because of the fuel rods' residual radioactive decay heat, while the boric acid helps slow down the nuclear fission rate.

Keeping the fuel rods covered with water is the key to heading off the worst-case scenario, said Elmer Lewis, professor emeritus of mechanical engineering at Northwestern University. "If you can keep any liquid at all in contact with them, they would probably not melt," he told me. Of course, you still have the problem of dealing with the radioactive steam released in the reaction. That's a problem the Japanese have become all too familiar with over the past few days.

"You don't want to be downwind, let's put it that way," said Norman McCormick, a professor emeritus of mechanical engineering at the University of Washington who is co-author of the soon-to-be-published book "Risk and Safety Analysis of Nuclear Systems." McCormick added, however, that the radiation risk would drop off sharply with distance.

"Dilution is the solution," he told me. "It's not a desirable thing, but it's the least undesirable thing."

Even under the best-case scenario, the messy cooldown process would have to continue for weeks. "It's going to be a long endgame, because they are going to have to keep cooling in there, and they will have to rig up something for the longer term," Lewis told Reuters.

The worst-case scenario
So what happens under the worst-case scenario? Let's assume for just a moment that fuel rods become exposed and go into full meltdown.

The fuel rods are actually long, hollow poles of zirconium metal, filled with pellets of uranium oxide fuel. That zirconium cladding is the first thing to go, melting at a temperature of about 2,000 degrees Fahrenheit (1,200 degrees Celsius). The uranium pellets' melting temperature is significantly higher — about 5,000 degrees F (2,800 degrees C). If the fuel pellets melt down, that creates a molten lava-like mess known as "corium."

isotype.com / Reuters / Source: Deutsches Atomforum

Experts say it's possible for the molten core material to trap water in such a way to create a steam explosion, causing radioactive material to blast outward. That would be a bad thing, but as von Hippel noted, that's seen as a low-probability event. There's an even lower probability for a "China Syndrome" scenario, in which molten core material finds its way out of a reactor's container vessel and is released into the environment (supposedly falling all the way down to China, or in this case, New York). The reactors' containment shells are designed to catch the molten core material in a basin of steel, concrete and graphite.

But what if the containment shells have somehow become cracked or otherwise compromised, due to Japan's earthquake or explosions? "There's always been a debate within the technical community over whether [nuclear core material] can melt and mix with the soil," Lewis told me. "It's hard to model something that's never happened."

Physicist David Albright, a former U.N. weapons inspector who is now president of the Institute for Science and International Security, noted today on MSNBC that there's been a "slow bleeding" of radioactive pollution into the environment. The worst-case scenario, he said, would result in a release of radioactivity on a level that's "probably not as much as Chernobyl, but nevertheless a very significant release."

In the most extreme case, the Japanese might have to consider following the Chernobyl example and "dump sand or concrete on the open wound," Nathan Hultman, an energy policy expert at the University of Maryland, told me.

"We don't have a lot of experience with this," Hultman said.

The best-case and the worst-case scenario both end with years of cleanup and reconstruction, not only due to the nuclear crisis, but also due to the wider challenge of dealing with the other effects of the earthquake and tsunami. How will Japan cope? How will the wider world respond? Feel free to weigh in with your comments below.

More on Japan's crisis:

• Q&A: Clearing up nuclear questions
• Nuclear worries focus on spent fuel pool
• Robots to the rescue in Japan? Not yet
• Cosmic Log archive on the Japan crisis
• Special report on the disaster in Japan

Join the Cosmic Log community by clicking the "like" button on our Facebook page or by following msnbc.com science editor Alan Boyle as b0yle on Twitter. To learn more about Alan Boyle's book on Pluto and the search for planets, check out the website for "The Case for Pluto." 

As the search for survivors and grim recovery of bodies continues following the devastating one-two punch of the 9.0 magnitude earthquake and subsequent tsunami in Japan, researchers are weighing what types of robots could be most helpful.

There are ground-based robots, for example, designed to climb up and down piles of rubble and slither into otherwise inaccessible cracks to look for survivors. Other robots are designed to work underwater, looking for survivors in cars that fell off bridges and to check the integrity of infrastructure.

Robin Murphy, director of the Center for Robot-Assisted Search and Rescue (CRASAR) at Texas A&M University, in College Station, and one of the world's top experts in rescue robotics, told me today that "ground robots are not going to be much use in this disaster."

That's mostly because ground robots are designed to go places where dogs and people can't — such as into piles of rubble more than 20 feet high. Those types of piles are created when parking garages and skyscrapers collapse.

"A lot of those types of structures didn't collapse," Murphy said. "And if they were down by the tsunami, they got completely washed away."

Instead, most of the rubble piles are shallow enough for people and dogs to search, who are more manageable and work faster than current rescue robots such as the Active Scope Camera, a snakelike robot with a camera for a head that can slither into tight spaces and check for survivors.

Earlier reports suggested that the robotic snake would be released in hard-hit Sendai by a team from Tohuku University led by Satoshi Tadokoro, but that idea "got ruled out very quickly," Murphy said.

The same goes for a second ground robot called Quince, which has tanklike tracks on its body and wheels that can roll over rubble and climb stairs.

Murphy's organization, however, is seeing a need for robots that can maneuver underwater and use sonar and acoustic cameras to search for survivors and check the integrity of vital infrastructure.

"And that makes sense when you think of all the water that's there and you’ve got ports, bridges, pipelines and other critical infrastructure underwater," she said. These underwater robots are small and attached to a tether so they can't be swept away in currents.

The Center for Robot-Assisted Search and Rescue has offered its assistance to Japan is on standby awaiting an official invitation to join the effort. The U.S. Army Corps of Engineers may already be using one robot, Seabotix, which they successfully used to investigate bridge a seawall damage following the 2010 Haiti earthquake.

"What we would be doing is bringing in a different kind, with a different type of expertise to complement what they are doing," she said.

More on search and rescue robots:

• A robot cut out for mine rescue work
• Snake robots could disable explosives
• Shape-shifting plane to aid maritime rescues
• No obstacle too high for climbing snakebots

Last updated 11:25 p.m. ET:

Three days after a catastrophic earthquake and tsunami hit Japan, the situation at the Fukushima Dai-ichi nuclear complex has turned into the biggest uncertainty of the crisis. Recovering from the seismic event will take tens of billions of dollars and years of work — but if the nuclear situation goes the wrong way, that would add dramatically to the disaster's cost.

How did all this happen, and how could it end? Different folks have different answers, depending on how they feel about nuclear power. Here's a roundup of the best answers I've been able to put together — accompanied by an invitation to add your own sources and perspectives as comments below:

Has there been a nuclear meltdown?

Authorities say partial meltdowns have probably occurred at three of the Fukushima Dai-ichi plants.

To understand what a "partial meltdown" means, we need to discuss how the reactors are constructed. Under normal conditions, the plants produce power by sustaining a controlled nuclear reaction inside a pressure vessel. Chain reactions in the nuclear core's uranium-filled fuel rods heat up water, generating steam that turns turbines to generate electricity. That steam is circulated through a cooling system and returned to the pressure vessel as water to keep the cycle going. The uranium oxide fuel is contained inside sheaths of zirconium metal that can withstand temperatures of 2,200 degrees Fahrenheit (1,200 degrees Celsius).

Control rods can be inserted between the fuel rods to shut down the main chain reaction in the uranium. But the water-circulating cooling system is needed as well to bring the temperature down while the radioactive decay subsides.

isotype.com / Reuters / Source: Deutsches Atomforum

The problem is that the power for the cooling system was cut off when the earthquake hit. Then the backup diesel generators were knocked out of commission by the tsunami. Backup batteries could keep the cooling system going for only about eight hours more. The plant's operator tried to bring in mobile generators to restore power, but the connections reportedly didn't match up.

Meanwhile, residual heat from radioactive decay continued to build up, and water continued to turn to steam. Eventually, the fuel rods became exposed. The temperatures apparently reached the melting point for the fuel rods' zirconium sheaths. That can result in uranium oxide fuel falling to the bottom of the pressure vessel — which is what some experts mean when they talk about a partial meltdown. Other experts, however, would reserve that term for a situation in which the nuclear fuel makes its way out of the pressure vessel but stays within a steel-and-concrete containment shell that surrounds the reactor. 

Is that why radioactive material escaped?

At the three Fukushima Dai-ichi reactors that are apparently experiencing partial meltdowns, the nuclear fuel is still contained within the pressure vessel. The radioactive material is not coming from the core itself. At reactors No. 1 and No. 3, the material is contained in steam that has been released from the vessels. Plant operators opened the steam valves to reduce the risk of a high-pressure explosion inside the vessels — in effect, letting off steam to keep the lid from blowing off a pressure cooker. The steam contains radioactive cesium-137 and iodine-131, which are byproducts of the uranium reaction. The authorities said the radioactivity in that steam is still below regulatory limits and should not pose any health risk.

Despite those reassurances, authorities ordered an evacuation of the area within a 12-mile (20-kilometer) radius of the Fukushima Dai-ichi plant, and have distributed stable iodine to evacuation centers as a precaution. If people are exposed to significant amounts of radioactive debris, taking doses of iodine can prevent the uptake of radioactive iodine and reduce the risk of thyroid cancer. 

Right now the radioactive plume is blowing out to sea, which means it's not wafting over Japanese population centers. It is wafting over the Pacific, however, and the U.S. Navy found that air crew members from the aircraft carrier USS Ronald Reagan were exposed to low-level contamination. The Navy says the crew members were decontaminated with soap and water, and all U.S. ships have been moved out of the downwind direction. Apparently, no harm was done.

Nevertheless, the contamination incident was worrisome to nuclear physicist Frank von Hippel, a former Clinton administration official who is now co-director of Princeton University's Program on Science and Global Security.

"I was surprised how high the radiation levels are," von Hippel told me. 

So what's being done?

Plant operators have been pumping cool seawater into the pressure vessels to replace the water that's being lost as steam, in an effort to keep the fuel rods from heating up further. They've added boric acid to the seawater, because boron suppresses the nuclear reaction and could accelerate the cooldown. Authorities were reluctant to turn to  this strategy because the seawater is so corrosive that it ruins the reactors for future power generation. But that's better than having the meltdown progress to an even worse stage.

What about these hydrogen explosions?

When the seawater hits the hot zirconium rods and uranium fuel, some of it is broken down into hydrogen and oxygen gas. Venting the steam allowed that hydrogen and oxygen to escape and build up between the pressure vessel and an outer structure that protects the reactor from the elements. At reactors No. 1 and No. 3, the hydrogen ignited, blowing the roof off the outer structure in each case. However, the pressure vessel and the steel containment shell remained intact. It's important to note that the hydrogen blast was not the result of any sort of atomic or "H-bomb" explosion, but was a purely chemical reaction.

Is the situation getting worse?

Yes. Authorities say another blast has been heard at reactor No. 2 at the Fukushima Dai-ichi plant. Details are sketchy, but the plant's owner, Tokyo Electric Power Co., said the explosion occurred near the reactor's suppression pool, a water reservoir that's part of the cooling system. A government spokesman said the pool was damaged, and there was concern that the No. 2 containment shell may have been breached. "A leak of nuclear material is feared," The Associated Press quoted Shinji Kinjo, a spokesman for Japan's nuclear safety agency, as saying.

Kinjo said radiation levels rose from 73 microsieverts before the blast to 11,900 microsieverts (11.9 millisieverts) three hours afterward. To put that figure in perspective, the U.S. Nuclear Regulatory Commission says that occupational exposure for adults working with radioactive material must be limited to 50 millisieverts per year.

In a nationally televised statement, Japanese Prime Minister Naoto Kan said "the level seems very high, and there is still a very high risk of more radiation coming out." Kan told people living within 19 miles (30 kilometers) of the Fukushima Dai-ichi complex to stay indoors to avoid radiation sickness.

AP also quoted Chief Cabinet Secretary Yukio Edano as saying that a fourth reactor at the complex was on fire and that more radiation had been released. "Now we are talking about levels that can damage human health. These are readings taken near the area where we believe the releases are happening. Far away, the levels should be lower," Edano said. Later, AP reported that the fire at reactor No. 4 was extinguished.

What about the nuclear fuel stored at the site?

The spent fuel rods at the Fukushima facility are stored in pools of water above the reactor. Plant operators have signaled that water levels were falling at reactor No. 1's storage pool, suggesting that the cooling system is failing. "It's on a slower fuse," von Hippel said, "but on the order of a week or so, it could boil down to the level of fuel."

What's the best-case scenario?

The seawater gambit keeps temperatures inside the pressure vessels under control for the next few days. During that time, the residual heat of radioactive decay dissipates, and operators no longer need to release steam from the vessels. Eventually, electrical power is restored to the cooling system, and each vessel's core can be removed. 

What's the worst-case scenario?

Authorities can't cool down the cores, and temperatures rise to the point that the uranium fuel melts into a mess on the bottom of the pressure vessel. The concrete-and-steel containment floor beneath the vessel has been built to contain a full core meltdown — but experts can't completely rule out the possibility of a breach that causes the highly radioactive material to escape into the environment.

Right now, the situation at Fukushima Dai-ichi is analogous to the Three Mile Island incident of 1979, which involved a partial core meltdown and a release of radioactive gases — but no breach in the reactor vessel. "It's at least as bad as Three Mile Island," von Hippel said. But if the nuclear fuel breaks out of the vessel, the situation could turn into something more like the 1986 Chernobyl nuclear accident in Ukraine, which sparked fatal cases of radiation sickness and spread contamination across a wide swath of Europe.

How long will this go on?

Even under the best-case scenario, it will take years to clean up the mess. "When you're dealing with spent fuel, you don't put it in cool, dry casks until three years after the reaction has stopped," von Hippel said.

More explanations of the nuclear situation:

• Radiation health risk remains low, experts say
• BoingBoing: Inside the 'black box' of nuclear power 
• Brave New Climate: Fukushima accident explained
• Scientific American: Beware the fear of nuclear ... FEAR!
• The Great Beyond at Nature: Anatomy of a meltdown
• Cosmic Log archive on the earthquake and tsunami

Princeton physicist Frank von Hippel answers your questions about the nuclear situation in Japan during an online chat at 4:30 p.m. ET Tuesday, March 15. Click here to submit your questions in advance and participate in the chat.

Have you found some particularly good explanations for what's going on, or are there some burning questions yet to be addressed? I've purposely stayed away from discussing the potential health risks in depth, because my colleague JoNel Aleccia is handling that important angle. But if you'd like to shed more light (rather than heat) on the situation at Fukushima or Japan's other stricken reactors, please feel free to add your comments below.

Join the Cosmic Log community by clicking the "like" button on our Facebook page or by following msnbc.com science editor Alan Boyle as b0yle on Twitter. To learn more about Alan Boyle's book on Pluto and the search for planets, check out the website for "The Case for Pluto." 

In the past year, massive earthquakes have rattled three corners of the Pacific plate — Chile, New Zealand, and Japan. Is the fourth corner of the plate — the west coast of North America — the next to go?

The possibility is getting a serious look — and air time. Journalist Simon Winchester, who often covers geological history, wrote a piece speculating about a major quake along the west coast in the March 13 issue of Newsweek and spoke about it this morning on the Today show.

According to Winchester, there is little doubt that earthquakes happen in clusters — a major event on one side of a tectonic plate is often followed weeks or months later by another major event on the far side of the plate.

"It is as though the earth becomes like a great brass bell, which when struck by an enormous hammer blow on one side sets to vibrating and ringing from all over. Now there have been catastrophic events at three corners of the Pacific Plate — one in the northwest, on Friday; one in the southwest, last month; one in the southeast, last year," he writes in Newsweek.

"That leaves just one corner unaffected — the northeast. And the fault line in the northeast of the Pacific Plate is the San Andreas Fault, underpinning the city of San Francisco."

Speaking with Meredith Vieira on the Today show, Winchester also emphasized the potential of a massive rupture on the Cascadia fault, which runs off the coast from Northern California to British Columbia. If that fault were to rupture, "it would cause not just terrifying problems on land, it would also generate a tsunami and that's a big, big problem," he said.

Earthquake clusters
John Rundle, an expert on earthquake dynamics at the University of California at Davis, said evidence is mounting that major earthquakes do cluster in space and time, but that the New Zealand quakes were too small to count (magnitude 6.3 for Christchurch) in this current spate of events.

Even taking out the New Zealand events, there are still four major earthquakes at least as big as magnitude 8.6 in the last seven years – the two events in Sumatra in 2004 and 2005, the Chilean earthquake in 2010, and Japan. This is unusual, Rundle said.

"The question is, is it so unusual that these things are causal? We don't know that, but we used to believe that these things were independent and you would see random clustering, but this has become so pronounced that you have to think there might be some correlation to it," he told me today. "In fact most of our models that we use do show correlations of earthquake events."

Rundle added that he believes that earthquakes are correlated and these big events are correlated. The question is, are they correlated because there is a surge in plate motion or do these things just align in spurts when looked at over spans of thousands of years?

Regardless, he noted, that statistically speaking, faults along the west coast of North America are ripe to rupture. For example, the Cascadia fault ruptures about every 500 years. It last ruptured in 1700, an earthquake that triggered a tsunami that was felt clear across the ocean in Japan.

The last big event on the San Andreas in southern California was in 1857 and historical analyses show they occur about every 160 years, "which would lead one to think we are pretty much due for one of these magnitude 7 plus events on Carrizo Plain, which could obviously grow into a magnitude 8 if we are unlucky," Rundle said.

In Northern California, the biggest threat is on the Hayward fault, which last ruptured in 1868 and the recurrence interval is about 170 years. "So, we are definitely in the window where we could see a Hayward event," he noted.

Examined in this light, the west coast is due for a major quake whether or not a major event on one side of a tectonic plate can trigger a major event on the other side. Rundle noted that according to his modeling, there is a correlation between the clustering activity and the periodicity of when faults rupture. "There is a definite association," he said.

Aftershock concern
The concept of clustering, however, remains controversial. Andrew Michael, a geophysicist with the U.S. Geological Survey in Menlo Park, California, addressed the question of whether one major earthquake triggers the next in a summary about the Chile quake. There, he wrote, "probably not … the global rate of earthquake energy release shows little, if any, difference between times when very large earthquakes have been common and when they have not."

In an e-mail to me today, he said "[I] have revisited the calculations and the conclusions remain the same. There is no evidence of global large-earthquake clustering beyond localized aftershock sequences."

These aftershock sequences are what have Rundle the most concerned. The Japan earthquake was a magnitude 9. According to aftershock analyses, there should be one magnitude 8 aftershock and for each magnitude 8 there should be ten magnitude 7 earthquakes. So far, the largest aftershock was 6.8.

"Also, the aftershocks, appear to be migrating from the Sendai region down towards Tokyo. So I am extremely concerned that there may be a magnitude 8 earthquake in the relatively near future near Tokyo … and in that case you are talking about massive suffering," he said.

More stories on earthquakes:

• Big quake question: Are they getting worse?
• Could big quake happen here? Yes
• Big Chile quake sparks swarm of faraway quakes
• Massive Chile quake may have set California shaking

John Roach is a contributing writer for msnbc.com. Connect with the Cosmic Log community by hitting the "like" button on the Cosmic Log Facebook page or following msnbc.com's science editor, Alan Boyle, on Twitter (@b0yle).

NASA

Images from NASA's Terra satellite show the coastline of Japan's Honshu island in the area around Sendai before and after Friday's earthquake. The left image is from Feb. 26, and the right image is from today. The images are color-coded to reflect surface composition rather than what the eye would see. The "Flood" label helps you gauge the extent of the flooding caused by the tsunami that followed the quake.

This week's earthquake caused the main island of Japan to shift as much as 13 feet to the east, seismologists say. That may sound like a shocker, but it's just one of the natural changes that come along with an 8.9-magnitude temblor — like the 1.8-microsecond speed-up of Earth's daily rotation and the 6.5-inch shift in Earth's axis.

The eastward shift was documented by Japan's Geonet network of GPS monitoring stations, based in Tsukuba, said Ken Hudnut, a geophysicist at the U.S. Geological Survey's Earthquake Hazards Program in Pasadena, Calif. Similar shifts took place during last year's 8.8 earthquake off the Chilean coast, as well as the 9.1 earthquake near Sumatra that caused a disastrous tsunami in 2004.

"It's the same phenomenon in all three cases," Hudnut said. The movement is linked to the release of the strain that builds up when one tectonic plate grinds against another in a subduction zone. 

"What's going on is that the plate going down drags along with it the upper plate as strain is stored in between earthquakes," he explained. "When the earthquake occurs, the upper plate lurches eastward over the subducting plate. The oceanic plate that's going down is relatively rigid, but the upper plate is like a wedge of material that's more elastic. So picture that upper wedge as being almost like an accordion that's being compressed between the times of earthquakes. It's like a spring. You're loading up the spring between earthquakes — in other words, you're compressing the eastern edge of the spring toward the main island of Japan. The earthquake allows that material to spring out toward the east."

Japan's network of 1,200 GPS monitoring stations, operated by the Geographical Survey Institute, shows a maximum springing-out effect of 13 feet (4 meters), with an average displacement of about 8 feet (2.5 meters) along a stretch measuring more than 300 miles (500 kilometers).

Everything that links GPS readings to maps, ranging from driving directions to property records, will have to be changed as a result of the shift, Hudnut told me. "Their national network for property boundary definitions has been warped," he said in an e-mail. "For ships, the nautical charts will need revision due to changed water depths, too (of about 3 feet). Much of the coastline dropped by a few feet, too, we gather."

We're starting to get pictures from space that document how the coastline has changed due to the earthquake and the tsunami. The NASA photos at the top of this posting show the coastline around the city of Sendai, which was one of the hardest-hit areas. The left photo was taken on Feb. 26 by the Moderate Resolution Imaging Spectrometer, or MODIS, aboard NASA's Terra satellite. The right photo was taken today by the same instrument. You can see that wide stretches of the coast are still flooded, in part because barriers erected at the coastline's edge are now keeping water in rather than keeping it out.

Other satellite pictures, distributed by Google, provide a closer-in view of the devastation caused by the tsunami. In each of the before-and-after sets below, the left picture was taken before the earthquake and the right picture was taken afterward. We've put together an eye-opening slideshow of before-and-after imagery that gives you control of the slider. And you can check out Google's blog posting and this Picasa Web album for still more.

Google / GeoEye / DigitalGlobe

Satellite photos provide before-and-after views of Kamaishi, a coastal city north of Sendai in Japan.

Google

These photos show before-and-after views of Japanese communities, with the left views taken before the earthquake and the right view taken today. The upper set shows Yuriagi and the lower set shows Yagawahama, both in Japan's hard-hit Miyagi prefecture.

Update for 7:25 p.m. ET March 13: I've updated the figures for the change in the day's length and the position of Earth's axis to reflect fresh figures from Richard Gross, a geophysicist at NASA's Jet Propulsion Laboratory.

More post-earthquake imagery:

• Slideshow: Japan before and after the tsunami
• Satellite photos show scope of Sendai damage
• Japan's devastation documented by satellites
• Slideshow: Japan struggles to recover from quake 

Join the Cosmic Log community by clicking the "like" button on our Facebook page or by following msnbc.com science editor Alan Boyle as b0yle on Twitter. To learn more about Alan Boyle's book on Pluto and the search for planets, check out the website for "The Case for Pluto." 

Japan has spent well more than $1 billion on earthquake prediction systems, including a network of more than 1,000 GPS-based sensors scattered around the country — and the payoff came today when Tokyo's residents were given up to a minute's warning that a Big One was on the way. That may not sound like much, but it's enough time for people to switch off their gas lines and get beneath a table or a door frame.

"The system functioned well, because warnings were seen on television across the country," Hirohito Naito, a seismic expert at the Japan Meterological Agency, told AFP.

The agency is in charge of quake preparedness as well as weather forecasting, and researchers have invested decades of effort into Japan's early-warning system. It's considered a model for the rest of the world, and U.S. researchers are adapting it for a system known as the California Integrated Seismic Network.

The system capitalizes on the fact that a seismic event sends out two types of shock waves: primary or P-waves, which move up and down; and secondary or S-waves, which shake from side to side. The P-waves travel faster but are weaker, while the S-waves are slower but do more damage. When Japan's system picks up the P-waves, it calculates how far away the source of the shaking is and issues an alarm while the S-waves are still en route. A warning can be broadcast via TV, radio, cell phones and home alarms less than 10 seconds after the P-waves are detected.

The early warning system isn't that useful for those who are close to the epicenter, because the S-waves come quickly behind the P-waves. But because Tokyo is about 230 miles away, that city's residents could have taken action as much as 80 seconds before the serious shaking began. As noted in this Technology Review report, that amount of time can give people a chance to stop a train, lower a crane, pull a car over to the side of the road, stop performing surgery in a hospital or get off an elevator in an office building.

That's assuming that you get the alarm immediately, of course. Some reports from Japan suggested that the alarms provided somewhat less advance warning, in the range of 15 to 30 seconds. This webpage from the Japan Meteorological Agency explains the early-warning system in much more depth.

Tsunami warnings worked
It takes longer to issue a tsunami warning, because that's dependent on an analysis of wave propagation from an undersea seismic source. The Japanese government issued a local warning three minutes after the quake struck. Technology Review estimates that residents in the hardest-hit coastal areas had 15 minutes of warning, and that Tokyo would have had at least 40 minutes to prepare.

Meanwhile, experts at the National Oceanic and Atmospheric Administration's Pacific Tsunami Warning Center (in Hawaii) and West Coast/Alaska Tsunami Warning Center (in Alaska) issued their first alerts nine minutes after the earthquake. They used computer modeling as well as readings from ocean buoys to track the waves as they sped across the Pacific at jetliner speeds. The wave-monitoring system has been beefed up significantly since the 2004 Indian Ocean tsunami, which pointed up gaps in the network.

Tsunami forecasters and emergency officials called for an evacuation of coastal areas in Hawaii, which were hit by walls of water measuring as much as 7 feet high.

"We called this right," Gerard Fryer, a geophysicist for the Pacific Tsunami Warning Center, told The Associated Press. "This evacuation was necessary. There's absolutely no question, this was the right thing to do."

Longer-term predictions
Could today's quake have been predicted days in advance rather than seconds in advance? In retrospect, maybe so: A 7.4-magnitude quake that hit Japan on Wednesday is now thought to be a foreshock heralding the bigger quake to come.

Two years ago, researchers looked at the records from Japan's crustal movement sensors and determined that large quakes could be anticipated by analyzing the "pre-signals" in the seismic data.

Then again, it's always easier to predict an event in retrospect. Five years ago, The Washington Post's Joel Achenbach wrote that Japanese geologists were sure the next Big One would take place southwest of Tokyo. Today's quake certainly qualifies as that Big One ... but it happened to the northeast, not the southwest.

More about earthquake prediction:

• 8.7 ideas for earthquake prediction
• Quake technology buys time for Japan
• Cosmic Log coverage of the Japan quake
• Can't predict the Big One? Bring on the pets!
• More from msnbc.com's Asia-Pacific news section

Join the Cosmic Log community by clicking the "like" button on our Facebook page or by following msnbc.com science editor Alan Boyle as b0yle on Twitter. To learn more about Alan Boyle's book on Pluto and the search for planets, check out the website for "The Case for Pluto."