What if cooling in one or more of the reactors at the Fukushima nuclear plant were lost?
Richard Lester, chair of the department of nuclear science and engineering at the Massachusetts Institute of Technology (MIT) in Cambridge, emphasizes the "very, very" unlikely possibility of that scenario. But if it were to occur, the inherent heat of the radionuclides would cause the fuel in the reactors to melt. Here's what would happen next.
In the event of a meltdown, the fuel could melt through and flow out of the primary pressure vessel, falling into the so-called core capture chamber which sits below the reactor for this very purpose.
That vessel has water that would hopefully cool the molten fuel down, eventually ending the crisis. If this didn't happen, however, a steam explosion could blow out the primary containment vessel, spewing massive quantities of radioactive aerosols as well as particulates. With towns evacuated at a perimeter of 30 kilometers, the lethality of that release "would depend on the winds," says Lester.
How would this compare to the disaster at Chernobyl? As noted in the New Scientist: 
At Chernobyl the pressure vessel was breached and the reactor had no containment. There, the core itself burned fiercely, largely because it was made of graphite - which was used as the moderator… once the reactor exploded the graphite made the situation worse, because it burned so readily. The fires carried radioactive material from the reactor core high into the atmosphere, where it spread far and wide. This could not happen at Fukushima Daiichi, as it does not use graphite as the moderator.
Chernobyl fires, fed by the graphite, burned for 2 weeks, driving radioactive material into the stratosphere. By contrast, the worst-case example at Fukushima, a one-time steam explosion, would drive some material into the atmosphere. But it would not have the power or duration of graphite fires. "This is not a Chernobyl-like situation," says Lester, both in terms of the graphite difference and the fact that in Chernobyl the reactor went supercritical.
Still, the situation could be dire if the storage pools that contain spent fuel rods run dry, which may have happened yesterday and could presumably occur if the plant was abandoned. (Experts say that because the plant's storage pools are not surrounded by containment vessels, they could spread radiation far more easily.)
Once the storage pool has run dry, then refilling it could cause additional short-term problems. (That's assuming workers can reach the pool.) Dumping water on the hot fuel rods would create a burst of steam that could take radioactive material with it into the atmosphere. "You might even have a steam explosion," Lester says. "That would be very bad." Just how much radiation went up would depend on what had ruptured or melted. The rods' zirconium-alloy cladding would melt at roughly 2000°C, whereas the uranium oxide pelts inside the cladding would melt at about 2800°C.
Adding water could also cause the fuel rods to rupture. The zirconium alloy interacts with steam at roughly 1200°C, as water molecules split and the liberated oxygen atoms bind with the zirconium. Such oxidation, which gives off its own heat, would likely cause the cladding to crumble and could even set the fuel aflame. That would drive exposed radioactive materials into the atmosphere.
If the fuel has melted, then adding water could even cause a brief nuclear chain reaction, says Charles Ferguson, a nuclear engineer and president of the Federation of American Scientists in Washington, D.C. That's because the melted fuel could gather in the bottom of the pool. The water, if it remained, could then act as a "moderator" to slow down the neutrons the fuel was emitting to the point where they could drive a brief chain reaction. That would create an intense pulse of gamma rays and neutrons that could be lethal to workers applying the water who were not sufficiently shielded.
However, not everyone believes that adding water back into the pool will lead to such dire short-term consequences. Mujid Kazimi, a nuclear engineer at MIT, points out that many of the radioactive compounds in the fuel rods would have already escaped once the rods ruptured. Kazimi says that the collapse of the fuel into a more compact mass will make a chain reaction less, not more, likely, as it means more neutrons will just fly out of the fuel. "I think there is a chance [of chain reaction], but I don't think that the chance is great."
All agree that the goal should be refilling the pool, as the long-term benefit of cooling the fuel will outweigh the short-term consequences of pouring in water. The engine that ultimately drives the release of radioactivity is the heat generated by nuclear decay in the fuel, Lester says, so as long as that heat is there, the radiation will eventually come out. So cooling the fuel is the only way to reduce the total radiation release, even if it means throwing up more radiation in the short run.
The situation is fiendishly complex, since a radioactivity release from either the reactor or the pools that was too high would mean that crews could not operate on site. Their absence makes further meltdowns possible, increasing the chance of yet more radioactivity.
But Lester is cautiously optimistic about the situation stabilizing, as is nuclear energy expert Tom Cochran of the Natural Resources Defense Council in Washington, D.C. "For a while I feared total abandonment—but the risk of that seems to have dissipated," says Cochran.
But that doesn't mean we're out of the woods, says Lester: "One can't at this point rule out any possibilities except that things can never be like they were before" at Fukushima.