What is going on here?
In the aftermath of the recent earthquake and tsunami in Japan, two nuclear power stations on the east coast of Japan have been experiencing problems. They are the Fukushima Daiichi ("daiichi" means "number one") and Fukushima Daini ("number two") sites, operated by the Tokyo Electric Power Company (or TEPCO). Site one has six reactors, and site two has four. The problematic reactors are #1, #2, and #3 at site one, which are the oldest of the ten and were due to be decommissioned this year.
In short, the earthquake combined with the tsunami have impaired the cooling systems at these reactors, which has made it difficult for TEPCO to shut them down completely. Reactor #1 is now considered safe after crew flooded the reactor with sea water. Reactor #3 was starting this process as this was originally written (6:00PM CST/11:00PM GST on March 13th). Site crew began preparing to add sea water to reactor #2 around 7:30AM GMT on March 14th, if a cooling procedure does not work.
The four reactors at site two did not have their systems impaired and have shut down normally.
Can this cause a nuclear explosion?
No. It is physically impossible for a nuclear power station to explode like a nuclear weapon.
Nuclear bombs work by causing a supercritical fission reaction in a very small space in an unbelievably small amount of time. They do this by using precisely-designed explosive charges to combine two subcritical masses of nuclear material so quickly that they bypass the critical stage and go directly to supercritical, and with enough force that the resulting supercritical mass cannot melt or blow itself apart before all of the material is fissioned.
Current nuclear power plants are designed around subcritical masses of radioactive material, which are manipulated into achieving sustained fission through the use of neutron moderators. The heat from this fission is used to convert water to steam, which drives electric generator turbines. (This is a drastic simplification.) They are not capable of achieving supercritical levels; the nuclear fuel would melt before this could occur, and a supercritical reaction is required for an explosion to occur.
Making a nuclear bomb is very difficult, and it is completely impossible for a nuclear reactor to accidentally become a bomb. Secondary systems, like cooling or turbines, can explode due to pressure and stress problems, but these are not nuclear explosions.
Is this a meltdown?
Technically, yes, but not in the way that most people think.
The term "meltdown" is not used within the nuclear industry, because it is insufficiently specific. The popular image of a meltdown is when a nuclear reactor's fuel core goes out of control and melts its way out of the containment facility. This has not happened and is unlikely to happen.
What has happened in reactor #1 and #3 is a "partial fuel melt". This means that the fuel core has suffered damage from heat but is still largely intact. No fuel has escaped containment. Core #2 may have experienced heat damage as well, but the details are not known yet. It is confirmed that reactor #2's containment has not been breached.
How did this happen? Aren't there safety systems?
When the earthquakes in Japan occurred on March 11th, all ten reactor cores "scrammed", which means that their control rods were inserted automatically. This shut down the active fission process, and the cores have remained shut down since then.
The problem is that even a scrammed reactor core generates "decay heat", which requires cooling. When the tsunami arrived shortly after the earthquake, it damaged the external power generators that the sites used to power their cooling systems. This meant that while the cores were shut down, they were still boiling off the water used as coolant.
This caused two further problems. First, the steam caused pressure to build up within the containment vessel. Second, once the water level subsided, parts of the fuel rods were exposed to air, causing the heat to build up more quickly, leading to core damage from the heat.
What are they doing about it?
From the very beginning, TEPCO has had the option to flood the reactor chambers with sea water, which would end the problems immediately. Unfortunately, this also destroys the reactors permanently. Doing so would not only cost TEPCO (and Japanese taxpayers) billions of dollars, but it would make that reactor unavailable for generating electricity during a nationwide disaster. The sea water method is a "last resort" in this sense, but it has always been an option.
To avoid this, TEPCO first took steps to bring the cooling systems back online and to reduce the pressure on the inside of the containment vessel. This involved bringing in external portable generators, repairing damaged systems, and venting steam and gases from inside the containment vessel. These methods worked for reactor #2 at site one, prior to complications; reactors four through six were shut down before for inspection before the earthquake hit.
In the end, TEPCO decided to avoid further risk and flooded reactor #1 with sea water. It is now considered safely under control. Reactor #3 is currently undergoing this process, and reactor #2 may undergo it if a venting procedure fails.
The four reactors at site two did not have their external power damaged by the tsunami, and are therefore operating normally, albeit in a post-scram shutdown state. They have not required any venting, and reactor #3 is already in full cold shutdown.
Is a "China Syndrome" meltdown possible?
No, any fuel melt situation at Fukushima will be limited, because the fuel is physically incapable of having a runaway fission reaction. This is due to their light water reactor design.
In a light water reactor, water is used as both a coolant for the fuel core and as a "neutron moderator". What a neutron moderator does is very technical (you can watch a lecture which includes this information here), but in short, when the neutron moderator is removed, the fission reaction will stop.
An LWR design limits the damage caused by a meltdown, because if all of the coolant is boiled away, the fission reaction will not keep going, because the coolant is also the moderator. The core will then only generate decay heat, which while dangerous and strong enough to melt the core, is not nearly as dangerous as an active fission reaction.
The containment vessel at Fukushima should be strong enough to resist breaching even during a decay heat meltdown. The amount of energy that could be produced by decay heat is easily calculated, and it is possible to design a container that will resist it. If it is not, and the core melts its way through the bottom of the vessel, it will end up in a large concrete barrier below the reactor. It is nearly impossible that a fuel melt caused by decay heat would penetrate this barrier. A containment vessel failure like this would result in a massive cleanup job but no leakage of nuclear material into the outside environment.
This is all moot, however, as flooding the reactor with sea water will prevent a fuel melt from progressing. TEPCO has already done this to reactor #1, and is in the process of doing it to #3. If any of the other reactors begin misbehaving, the sea water option will be available for those as well.
What was this about an explosion?
One of the byproducts of reactors like the ones at Fukushima is hydrogen. Normally this gas is vented and burned slowly. Due to the nature of the accident, the vented hydrogen gas was not properly burned as it was released. This led to a build up of hydrogen gas inside the reactor #1 building, but outside the containment vessel.
This gas ignited, causing the top of the largely cosmetic external shell to be blown off. This shell was made of sheet metal on a steel frame and did not require a great deal of force to be destroyed. The reactor itself was not damaged in this explosion, and there were only four minor injuries. This was a conventional chemical reaction and not a nuclear explosion.
You see what happened in the photo of the reactor housing. Note that other than losing the sheet metal covering on the top, the reactor building is intact. No containment breach has occurred.
At about 2:30AM GMT on March 14th, a similar explosion occurred at the reactor #3 building. This explosion was not unexpected, as TEPCO had warned that one might occur. The damage is still being assessed but it has been announced that the containment vessel was not breached and that the sea water process is continuing.
Around 7:30AM GMT on March 14th, it was announced that the explosion at reactor #2 has damaged the already limping cooling systems of reactor #2. It may also receive the sea water treatment if they are unable to use a venting procedure to restart the cooling systems.
Is there radiation leakage?
The radiation levels outside the plant are higher than usual due to the release of radioactive steam. These levels will go down and return to their normal levels, as no fuel has escaped containment.
For perspective, note that charts detailing detrimental radiation exposure start at 1 Gy, equivalent to 1 Sv; the radiation outside the problematic Fukushima reactors is being measured in micro-Svs per hour. The highest reported levels outside the Fukushima reactors has been around 1000 to 1500 micro-Svs per hour. This means that one would have to stay in this area for four to six weeks, 24 hours a day, without protection in order to experience the lowest level of radiation poisoning, which while unpleasant is not normally fatal. And this level will not stay where it is.
Also note the chart of normal radiation exposure levels from things like medical x-rays and airline flights.
There have also been very minor releases of radioactive reactor byproducts like iodine and cesium along with the steam. This material is less radioactive than the typical output of coal power plants. It is significant mainly as an indicator of the state of the reactor core.
I read that there's a plume of radioactive material heading across the Pacific.
In its current state, the steam blowing east from Japan across the pacific is less dangerous than living in Denver for a year. If it makes it across the ocean, it will be almost undetectable by the time it arrives, and completely harmless as the dangerous elements in the steam will have decayed by then.
What's this about fuel rods being exposed to the air?
When the coolant levels inside the reactor get low enough, the tops of the fuel rods will be exposed to the air inside the containment vessel. They have not been exposed to the external atmosphere and the containment vessels are all intact.
Can this end up like Chernobyl?
No, it cannot. for several reasons.
Chernobyl used graphite as a neutron moderator and water as a coolant. For complicated reasons, this meant that as the coolant heated up and converted to steam, the fission reaction intensified, converting even more water to steam, leading to a feedback effect. The Fukushima reactors use water as both the coolant and the neutron moderator, which means that as the water heats up and converts to steam, the reaction slows down instead. (The effect of the conversion of water coolant to steam on the performance of a nuclear reactor is known as the "void coefficient", and can be either positive or negative.)
Chernobyl was designed so that as the nuclear fuel heated up, the fission reaction intensified, heating the core even further, causing another feedback effect. In the Fukushima reactors, the fission reaction slows down as the fuel heats up. (The effect of heating of the nuclear fuel on the performance of a nuclear reactor is known as the "temperature coefficient", and can also be positive or negative.)
Chernobyl's graphite moderator was flammable, and when the reactor exploded, the radioactive graphite burned and ended up in the atmosphere. The Fukushima reactors use water as a neutron moderator, which is obviously not flammable.
Note that while Chernobyl used light water as a coolant (as distinct from heavy water), it was not a "light water reactor". The term LWR refers strictly to reactors that use light water for both cooling and neutron moderation.
The news said this was the worst nuclear power accident since Chernobyl, though.
It's the only nuclear power plant accident of its type since Chernobyl. It's easy to be the worst in a sample size of one.
Is this like Three Mile Island?
There are similarities. The final effect on the world is likely to be similar: no deaths, minimal external contamination, and a tremendous PR disaster for the nuclear industry due to bad reporting by the media.
How can I keep up with developments?
The western media has been very bad about reporting this event, due to a combination of sensationalist reporting, ignorance, and the use of inexact or unexplained terminology.
One of the safe sources of information is the TEPCO site, which has been posting press releases on a regular basis. Unfortunately, this site is often unresponsive due to the immense traffic it is receiving.
The important thing to remember is that most of the "experts" appearing on the news are engaging in speculation. Very few of them are restricting themselves to what they can be sure about, and those that are have often been misrepresented.