Preventative Measures and Why We Have Them
Something that's been annoying me on Twitter lately is the cries of horror and outrage that 200,000 people have been evacuated from the area around the Fukushima Daiichi nuclear reactors and that iodine tablets have been distributed. This isn't something I can address effectively in 140-character splurts, so I'll talk about it here.
These complaints seem to assume that if people are being evacuated and given medical supplies, then something awful must have already happened and that the measures being taken are remediative. I have to conclude that none of these people have ever lived in an area prone to extreme geological/meteorological events. I've lived in several, which means that I have been through tornado warnings, flood warnings, hurricane warnings, and snow warnings. One common theme of all these events is that they all feature preventative measures: when it looks like there might be a flood, you start sandbagging the riverbank, when it looks like there might be a tornado you go down to the basement, when it looks like there might be heavy snow affecting travel safety you close schools. You take these measures so that if something bad does happen, you will be in the safest place possible.
In the case of a nuclear accident, preventative measures run many layers deep. Measures such as evacuation and distribution of iodine take place very early in that chain: once the chance of a containment breach N hours in the future is X%, you evacuate the surrounding area and hand out medical supplies so that anyone who needs them already has them on hand. The last fucking thing you want to be doing during a containment breach is worrying about people getting exposed, or having to put more people into a hot area in order to evacuate others, so you get everyone in a wide radius around the reactor out of harm's way very early on. This is particularly important in the case of a massive geographical disaster: you have to factor in travel delays due to damage to the transportation infrastructure, so you have to tell people to get the fuck out even earlier in order to give them plenty of time to get to safety.
Happily, it looks like all the reactor cores have been successfully cooled down. Some of the Fukushima Daiichi reactors had to be cooled with seawater due to boil-off; they had to vent some steam to reduce the pressure in the reactor, which lowered the coolant level, and they replaced it with seawater with lots of boron dissolved in it. (Boron acts like flypaper for neutrons, capturing them and preventing further fission reactions.) If they had to flood the pressure vessel, then the reactor is probably a write-off. This is why you've heard people talking about boron/seawater being the "last resort" option; it is guaranteed to kill off any ongoing primary or intermediate fission interactions, but it also ruins the reactor for future use. Think of it as like spraying a kitchen fire with a fire extinguisher: you've put the fire out, but you'll have to throw out the food you had to hose down.
The Fukushima Daiichi reactors are all roughly 40 years old. In that intervening time, boiling-water reactor designs have improved substantially. GE Hitachi's Advanced Boiling Water Reactors (ABWRs) and Economic Simplified Boiling Water Reactors (ESBWRs) don't suffer from the need for active pumping that caused much of the trouble with Daiichi 1 and 3 (which are also GE Hitachi designs, but from an earlier generation). Two ABWRs are slated to go online at Fukushima Daiichi in 2016; it would be great if the damaged reactors could be replaced with ABWRs or even ESBWRs.
Finally, I just have to give a shout-out to the engineers who built this plant in the first place. For the most part, that plant successfully survived an earthquake seven times stronger than what they designed it for. That's some good construction work. It looks like a substantial amount of the problems that happened after the coolant system failure, such as problems connecting the second set of backup batteries and generators that were trucked in, could have been prevented by standardization that's only developed in the last 40 years; when your plants are standardized, it's much easier to make sure you're getting the right parts to the right place. I hope that Japan takes advantage of the improvements in design and advances in engineering standards that have taken place since Fukushima Daiichi went online, and gets the site back up and running better than it was before.
These complaints seem to assume that if people are being evacuated and given medical supplies, then something awful must have already happened and that the measures being taken are remediative. I have to conclude that none of these people have ever lived in an area prone to extreme geological/meteorological events. I've lived in several, which means that I have been through tornado warnings, flood warnings, hurricane warnings, and snow warnings. One common theme of all these events is that they all feature preventative measures: when it looks like there might be a flood, you start sandbagging the riverbank, when it looks like there might be a tornado you go down to the basement, when it looks like there might be heavy snow affecting travel safety you close schools. You take these measures so that if something bad does happen, you will be in the safest place possible.
In the case of a nuclear accident, preventative measures run many layers deep. Measures such as evacuation and distribution of iodine take place very early in that chain: once the chance of a containment breach N hours in the future is X%, you evacuate the surrounding area and hand out medical supplies so that anyone who needs them already has them on hand. The last fucking thing you want to be doing during a containment breach is worrying about people getting exposed, or having to put more people into a hot area in order to evacuate others, so you get everyone in a wide radius around the reactor out of harm's way very early on. This is particularly important in the case of a massive geographical disaster: you have to factor in travel delays due to damage to the transportation infrastructure, so you have to tell people to get the fuck out even earlier in order to give them plenty of time to get to safety.
Happily, it looks like all the reactor cores have been successfully cooled down. Some of the Fukushima Daiichi reactors had to be cooled with seawater due to boil-off; they had to vent some steam to reduce the pressure in the reactor, which lowered the coolant level, and they replaced it with seawater with lots of boron dissolved in it. (Boron acts like flypaper for neutrons, capturing them and preventing further fission reactions.) If they had to flood the pressure vessel, then the reactor is probably a write-off. This is why you've heard people talking about boron/seawater being the "last resort" option; it is guaranteed to kill off any ongoing primary or intermediate fission interactions, but it also ruins the reactor for future use. Think of it as like spraying a kitchen fire with a fire extinguisher: you've put the fire out, but you'll have to throw out the food you had to hose down.
The Fukushima Daiichi reactors are all roughly 40 years old. In that intervening time, boiling-water reactor designs have improved substantially. GE Hitachi's Advanced Boiling Water Reactors (ABWRs) and Economic Simplified Boiling Water Reactors (ESBWRs) don't suffer from the need for active pumping that caused much of the trouble with Daiichi 1 and 3 (which are also GE Hitachi designs, but from an earlier generation). Two ABWRs are slated to go online at Fukushima Daiichi in 2016; it would be great if the damaged reactors could be replaced with ABWRs or even ESBWRs.
Finally, I just have to give a shout-out to the engineers who built this plant in the first place. For the most part, that plant successfully survived an earthquake seven times stronger than what they designed it for. That's some good construction work. It looks like a substantial amount of the problems that happened after the coolant system failure, such as problems connecting the second set of backup batteries and generators that were trucked in, could have been prevented by standardization that's only developed in the last 40 years; when your plants are standardized, it's much easier to make sure you're getting the right parts to the right place. I hope that Japan takes advantage of the improvements in design and advances in engineering standards that have taken place since Fukushima Daiichi went online, and gets the site back up and running better than it was before.