Earlier this week, some workers at the Fukushima plant suffered from radiation burns when highly contaminated water got inside their protective clothing, as they waded towards a repair site. This prompted the ever cautious Japanese officials to warn that this might indicate a partial loss of containment.
My initial feeling about this was that it was very unlikely. Possible, yes, but not terribly likely, given that a breach of the pressure vessel would have drastically impacted efforts to circulate cooling water (pressure would drop rapidly and significantly, and probably stay down, regardless of how much water was pumped, for instance, which would be immediately detectable). It appears that my gut feeling was correct. While they haven't pinned down the source of the water yet, I'd put my money on damage to the reactor cooling system, which has likely developed a number of leaky valves, gaskets or pipes. The cooling system plumbing has been exposed to multiple hydrogen explosions, earthquakes, debris, and the corrosive effects of seawater. Enough water has been moving through it for a long enough period that even a relatively small leak of water containing fission products from the reactor will have had time to accumulate to dangerous levels inside the plant.
This radioactive water poses a problem to workers, first and foremost, and will significantly complicate repair efforts. Unfortunately, it may also pose a fairly significant environmental risk, depending on its source. If the leaks are occurring in places where the leaked water can come into contact with hot components, or if the leaks are a mixture of steam and water directly downstream from the pressure vessel, then a small, but significant amount of fission products (the main concerns are Cesium and Iodine, but in the long term, primarily Cesium 137, which has a half life of 30 years and has a decay chain that includes gamma decay) may be being dispersed from the plant. Even small amounts can accumulate over time in the near vicinity of the plant, and make uses like agriculture problematic for the near term future. It should be stressed that only water soluble products are likely to end up in the cooling water, so that means no plutonium or uranium are likely to be dispersed by the leaks, which is a very good thing.
This contaminated water also poses a disposal problem, something that has been bothering me for some time.
Removing the dissolved contaminants is probably technically feasible, but will be complicated and expensive, and until all of the radioactive water is removed, it presents a serious problem if it seeps out the containment buildings, and could pose as risk to the ocean environment. Ocean currents and biological mechanisms would likely remove the contaminants from the water column well before the decay. Eventually they would end up in the sediment on the sea floor, but not before first passing through a lot of plankton, squid, fish, whales, etc. All of which would have to be checked before being eaten, and shellfish from these areas might be a problem for as long as decades (note that Japan has had similar problems with this in the past, but with mercury). They are already discharging water from the plant into the sea, and so far dilution seems to be doing its thing.
In spite of the radioactive water problem, significant progress has been made in repairing the plant, and they've moved to using fresh water again for cooling for some of the reactors.
The other major story has been the presence of fission products in drinking water in areas somewhat distant from the plant. I'm glad to see that the Japanese authorities weren't taken off-guard by an entirely predictable phenomenon (probably down to a period of days when it was likely to occur). Rainfall containing fission products from releases at the plant will have been gone through it's normal cycling process, before ending up in reservoirs and streams used from drinking water. Japanese hydrologists should have a very good handle on how long that will take based on dates of rainfall and atmospheric radiation levels, and should be able to produce a decent prediction of when to be concerned (and a system for checking the water is already in place, obviously). Given that there haven't been any major releases at Fukushima in some time, this is probably the only pulse of concern that will occur.
All of this new information about contamination and damage at the plant puts this whole mess at #3 in terms of worst nuclear disasters, after Chernobyl and Kyshtym (which most folks haven't heard of, including me, until recently), and in front of Three Mile Island. It's not hard to beat out TMI, of course, since the impacts were pretty close to zero. Fukushima is likely to go down in history as a INES Level Four accident (note that scale is applied very inconsistently), though the extent of the local impacts will take some time to know in a substantial way. Probably some local land will be made off limits until the radiation levels are accurately mapped, and then a smaller area will be removed from agricultural or residential use, set aside and monitored.
All told, not great, but not too bad considering the age of the reactors, TEPCO's somewhat sketchy record, and the scale of the events that caused the problems in the first place. Much of this just highlights how much better things would be going with a modern reactor design. I'll likely post again later this weekend about Next Generation Nuclear Power.