Japanese N-Plant Explosion

[keith1]

It may be time to expand this discussion to time of jet stream travel and location of fall-out to the americas:

courtesy:http://virga.sfsu.edu and http://www.goes.noaa.gov

 

[Bowser]

The media here at home is assuring us that there would be no fallout to be concerned about; nonetheless, people are buying iodine pills. I'm not worried (Oregon).

 

[Trippy]

After I had posted...

Cross post then.

 

[iceaura]

Where do you draw the line iceaura?
- - -
This gives you, when you weigh it against the probable outcomes of that event occuring, mitigation measures can be put in place to reduce the risks to acceptable levels.

So there's hope that you will deal with the posted issues and questions after all. Cool:

What are the acceptable odds for semi-permanent loss of the upper Mississippi River? I think 100,000 to one is far too likely, for that event. I want hundreds of millions to one, and I want them calculated competently - not the crapola we've been getting from industry-friendly and otherwise co-opted agency agents who even think (much less say in public) that a series of near misses is evidence of safety.

How probable does an outcome have to be - and how is that calculated - before it has to be handled in the "mitigation to acceptable levels" efforts?

A Japanese nuclear power plant that cannot withstand even the size earthquakes and tsunamis that have actually hit its geologically local environment in the past fifty years - that is put at risk of meltdown by, say, the 9.2 that hit offshore of Anchorage in 1964 - is not safe by any reasonable standard. And all the assurances of nuclear power plant safety that preceded this week's events are revealed to be garbage

once again.

 

[DwayneD.L.Rabon]

Well, The Status still remains the same at this point, destroy the Nuclear Facility and remove it in pieces. Which allows continous volitile radioactive materials to be removed, at various volumes (pieces sizes) reducing hazards to personell.
If the entire Facility is frozen, when it is bombarded by shells from a U.S. Navy Battleship Cannon Shell or Japan Defense Force tank or ship it will brake in to pieces, the cannon shell will penitrate right to the core, (even through ground dirt if it is frozen) smashing the core into pieces. After the reactor cores are bombarded with high velocity round, demolish the rest of the Nuclear Facility the same way with a larger shell round from the battleshipor tank.
When the base of the facility is open and visable, start clean up by tractor Dividing material and piling it in sections. The emitted radation will be distributed to the Square Foot or Meter deminishing radation exsposer. Cleaning also then becomes more managable and quicker.
Before the actual undertaking of the bombardment and during the bombardment of the Nuclear Facility, spray a Chemical that turns to jelly when contacted with the atmosphere, use several local insectcide mist spraying airplanes, if there is not a special jell that is avaiable to the region or in insuffcient quanity, just use normal food gelatin, should be tons of it in powder form, prehaps also some kind of color dye that can provide clean up crews with a visable maker as to the area that needs direct attention, it would also show any vapor drift patterns. (Latex paint make work in a seperate plain. Given the current weather temptiures and the Frozen building and its Frozen ground works the jello would gel trapping the Dust materials,gases and vapors escaping from bombardment, and their location of fall would be marked by the color dye.
The So then the difficulty would now be to locate a cooling agent such as nitrogen in suffcient concentration to cool the entire ground works. the collest tempitures will most likly occur about 1:30 AM.
Considering the reactors have been saturated with sea water it will make a nice ice cube.

Dwayne D.L.Rabon

 

[Dywyddyr]

Somebody please delete the inane drivel that is post #165.

 

[ElectricFetus]

The radiation around the reactor is now at 8,217 uSv per hour[1], but that dosage unit is probably meaningless to most, ~3 mSv or ~3000 uSv is roughly the average amount of background radiation one is exposed to in year... still sounds bad, ok how about this, a chest X-ray is 100-200 uSV[2] so that max dosage outside the reactor is equal to 82-41 chest x-rays an hour. A CT scan is anywhere between 1.5-30 mSv (1500-30,000 uSv) so that means the max dosage outside the reactor is equal to 5 to just 1/4 a CT scan... yes that right we exposed hospital patients to more radiation then whats outside these reactors! Now I can't imagine how much lower the radiation levels will get after being taken by the jet stream to America!

As for the reactor, here a diagram of the reactor (a GE/Hitachi BWR Mark 1):

The most recent events is with reactor #2: they could not pump more seawater into it because its release valve got stuck close, and it pressurized until the valve, or something burst, now what burst I don't know but I'm guessing not any part with a meter of concrete on top if it, I would guess piping. Now in the event of a full meltdown (which has not happened yet to my knowledge) the reactor in the center would turn into a giant mass of molten metal and fall down into the drywell and wetwell below that, this separates the molten mass and cools it, once spread out thinly by this architecture it cools naturally. In short these power plants can contain a full meltdown if it comes to that. The amount of radioactive material that would leak out would merely be what could be carried away in gas or vapor with the steam.

Compare that to Chernobyl which had no containment system to speak of and a good percentage of its nuclear fuel literally burned up (as in on fire!) on its parking lot! Chernobyl was the worse nuclear disaster imaginable: a reactor with no containment domes explodes and spews radioactive material over a city, with the wind taking it over a continent! Yet the total death toll from radiation sickness and the confirmed cancer rate increase and deaths of (thyroid cancer from radioactive iodine was the only cancer increased verifiable detected. Leukemia, lung cancer, etc, all the others had no verified statistical increase)[3] Was less than 100 people... that right the worse nuclear disaster ever, one that release thousands of time more radioactive material then this present nuclear accident ever could, has a confirmed death toll of less than 100 people, now how many people died in this recent Earthquake and Tsunami? Oh that right death toll is at over 2400 confirmed dead. So in short japan's present nuclear "emergency" is not likely to kill anyone and yet we already have thousands dead of things that really kill people like earthquakes and tsunamis!

[1]http://en.wikipedia.org/wiki/Timeline_of_the_Fukushima_nuclear_accidents#Tuesday_15_March
[2]http://www.radiologyinfo.org/en/safety/index.cfm?pg=sfty_xray
[3]http://www.iaea.org/Publications/Booklets/Chernobyl/chernobyl.pdf

 

[Trippy]

So there's hope that you will deal with the posted issues and questions after all. Cool:

Cut the crap.

What are the acceptable odds for semi-permanent loss of the upper Mississippi River? I think 100,000 to one is far too likely, for that event. I want hundreds of millions to one, and I want them calculated competently - not the crapola we've been getting from industry-friendly and otherwise co-opted agency agents who even think (much less say in public) that a series of near misses is evidence of safety.

How probable does an outcome have to be - and how is that calculated - before it has to be handled in the "mitigation to acceptable levels" efforts?

As I understand risk management (in broad, international terms) that depends entirely on things like government legislation which make statements like "There is to be no more than an N% risk of exposure in excess of Xmrem as a result of plant failures" or it might be written in terms of total expected exposure over the lifetime of the plant, which is (as I understand it), would be for every foreseeable contingency the amount of exposure generated multiplied by the probability of that event occuring.

"A Japanese nuclear power plant that cannot withstand even the size earthquakes and tsunamis that have actually hit its geologically local environment in the past fifty years - that is put at risk of meltdown by, say, the 9.2 that hit offshore of Anchorage in 1964 - is not safe by any reasonable standard."[sup][citation needed][/sup]

Seriously though. You don't know what you're talking about.
This is the First M9 earthquake to hit Hit Japan in (at least) 1500 years.

 

[iceaura]

As I understand risk management (in broad, international terms) that depends entirely on things like government legislation

Well, what do we want that legislation to say?

Clearly it is too lax, at the moment - we have nuclear power plants that can't handle the size earthquakes and tsunamis that have occurred right next door across the ocean, in the same geological setup. We have some built over faults in the headwaters of the Mississippi River that can't handle the size earthquakes known to have occurred in recorded history a few hundred miles away.

Seriously though. You don't know what you're talking about.
This is the First M9 earthquake to hit Hit Japan in (at least) 1500 years.

So? Once every 1500 years would be far too likely an event to be left out of the "mitigation", with meltdown at stake. With a thirty year life span, that's a one in fifty shot.

And the 9.2 right across the narrow part of the Pacific, on the same plate, was less than 50 years ago.

 

[keith1]

The media here at home is assuring us that there would be no fallout to be concerned about; nonetheless, people are buying iodine pills. I'm not worried (Oregon).

Yes, the Pacific Ocean is a large filter and a long gauntlet for a fallout cloud to negotiate without extensive washout or thin-out...especially during this wet season. Low probability of a successful crossing of an accumulated toxicity...

Having said that, I eat a lot of Sushi seaweed (coincidentally, not unlike the large amounts consumed by the Japanese) and yogurt, both high in Potassium Iodide.

 

[Trippy]

Well, what do we want that legislation to say?

I'm quite happy with my Legislation.

Clearly it is too lax, at the moment - we have nuclear power plants that can't handle the size earthquakes and tsunamis that have occurred right next door across the ocean, in the same geological setup. We have some built over faults in the headwaters of the Mississippi River that can't handle the size earthquakes known to have occurred in recorded history a few hundred miles away.

Much of this is too ridiculous for words.
For example this: "...that can't handle the size earthquakes known to have occurred in recorded history a few hundred miles away..."

So? Once every 1500 years would be far too likely an event to be left out of the "mitigation", with meltdown at stake. With a thirty year life span, that's a one in fifty shot.

And I suppose you think that a one in 1500 year event has a 100% chance of occuring after 1500 years?

And note, I'm fairly sure I said it's been at least 1500 years, that's just the records that I've been able to find.

And the 9.2 right across the narrow part of the Pacific, on the same plate, was less than 50 years ago.

LOL
You have no idea what you're talking about. There's 5300km between Fukushima and Anchorage - at least according to Travel Math(assuming you're referring to the 1964 EQ in Anchroage, you're being kinda vague about that).

Why is that significant
To give you an example - If I move 400km to the north, the seismic risk at least doubles.

The point being that the magnitude and frequency of earthquakes varies hugely over short distances - oh, and the Tsunami generated by that earthquake was less than what was experienced in Japan, so a system designed to those standards would have failed.

Incidentally, the residents of Miyako built themselves a Seawall to protect them from Tsunamis after the (then) village of Taro lost 48% of its population and 98% of its buildings in a 1933 Tsunami. The wall was completed in 1958 (just in time for the 1960 Tsunami). The Tsunami wall protected them from multiple Tsunamis, but you've probably seen the footage of what happened to it this time around:

I'm not sure how tall that is, I've seen one source that says 10m, and another that says 30m, I'm inclined towards 10m for a number of reasons, however the statement is obvious:
This structure was designed to protect from Tsunamis expected in this area, and up until last week, it has done so, successfully.

 

[iceaura]

ou have no idea what you're talking about. There's 5300km between Fukushima and Anchorage - at least according to Travel Math(assuming you're referring to the 1964 EQ in Anchroage, you're being kinda vague about that).

Why is that significant
To give you an example - If I move 400km to the north, the seismic risk at least doubles.

The geological setup can change easily in less than that - or not, as with the Alaska coastal earthquake risk compared with Japan's. They are usefully similar, for estimating risk - ring of fire, same plate, subduction, potential violence, etc.

oh, and the Tsunami generated by that earthquake was less than what was experienced in Japan, so a system designed to those standards would have failed.

No, it wasn't. http://wcatwc.arh.noaa.gov/64quake.htm In this chart the greatest height of the '64 tsunami at shore strike is given as about 20 meters, but evidence of greater heights has been found in more remote places. General height was two or three meters for hundreds of kilometers of coast.

I'm not sure how tall that is, I've seen one source that says 10m, and another that says 30m, I'm inclined towards 10m for a number of reasons,

I'd be inclined closer to 5 meters, unless the street signage in Japan is fifty feet in the air.

And I suppose you think that a one in 1500 year event has a 100% chance of occuring after 1500 years?

You can't help yourself, can you. It's a sort of keyboard Tourette's.

My only dubious assumption there was a uniform probability distribution of quake risk. We know better - they come in flurries and lulls, and the risk rises over time after the lull. But that implies even worse, for the common sense of the people who are downplaying the risk of hosting nukes all over the place.

This structure was designed to protect from Tsunamis expected in this area, and up until last week, it has done so, successfully.

So is that kind of "expectation" your suggested standard for nuclear power plant construction?

Or should we expect a bit more, from people who construct nuclear reactors on islands in major river systems?

Since we are back to dealing crap, btw, we can notice that the Ojibwe words the French slurred into one word, which French munching was exapted into English as "Mississippi", are often translated as "Father of Waters". http://www.infoplease.com/spot/aihmnames1.html So even if Mississippi were an Ojibwe word, which it isn't, "Father of Waters River" might seem a reasonable name.

But really, the better course would be to ponder the meaning of "derived", as it applies to place names.

 

[Trippy]

The geological setup can change easily in less than that - or not, as with the Alaska coastal earthquake risk compared with Japan's. They are usefully similar, for estimating risk - ring of fire, same plate, subduction, potential violence, etc.

(Very) different geological setting. So no, not so much.

No, it wasn't. http://wcatwc.arh.noaa.gov/64quake.htm In this chart the greatest height of the '64 tsunami at shore strike is given as about 20 meters, but evidence of greater heights has been found in more remote places. General height was two or three meters for hundreds of kilometers of coast.

I'm sorry, for some reason I was under the impression that we were talking about the effects of the Tsunami in Japan, which were less than 10m.

I'd be inclined closer to 5 meters, unless the street signage in Japan is fifty feet in the air.

Oh, sure, if you want to assume that the street is at MSL.
But if the street is 5m above MSL, and wall is 5m above the street, then the Tsunami wall would be 10m tall.

You can't help yourself, can you. It's a sort of keyboard Tourette's.

My only dubious assumption there was a uniform probability distribution of quake risk. We know better - they come in flurries and lulls, and the risk rises over time after the lull. But that implies even worse, for the common sense of the people who are downplaying the risk of hosting nukes all over the place.

The only valid thing you've said here is this: "and the risk rises over time after the lull"

Your assertion over estimates the risk. My recollection is that risks are most accurately modeled as a biased coin toss.

If we treat a 1 in 1000 year event as having a 1 in 1000 chance of occuring, then the probability of it occuring (if I'm remembering my maths correctly) in a 50 year period is more like 4.98% and the chance of it occuring in a 1000 year period is actually more like 63%.

So is that kind of "expectation" your suggested standard for nuclear power plant construction?

Or should we expect a bit more, from people who construct nuclear reactors on islands in major river systems?

Again, we come back to these unreasonable assumptions of yours. By your logic, everything in North America should be built as if it's in California.

And do you understand the usage of 'Expectation' in statistics?

 

[Bells]

Anywho, dick measuring aside..

Japanese authorities also today informed the IAEA at 04:50 CET that the spent fuel storage pond at the Unit 4 reactor of the Fukushima Daiichi nuclear power plant is on fire and radioactivity is being released directly into the atmosphere.

Dose rates of up to 400 millisievert per hour have been reported at the site. The Japanese authorities are saying that there is a possibility that the fire was caused by a hydrogen explosion.

http://www.iaea.org/newscenter/news/tsunamiupdate01.html

To put that into perspective:

 

[Trippy]

As far as the Tsunami goes, to the best of my ability to discern...

Forecast wave heights:

Observed wave heights:

 

[iceaura]

Uh oh: http://www.sify.com/news/explosion-...structure-news-international-ldplugchafa.html

Context: the explosion in F3 had probably damaged F2, and the venting valves for the containment structure of F2 seemed to be stuck afterwards. Hydrogen pressure and risk in F2 has been a worry since.

It's beginning to look as if we were very lucky that 4,5,and 6 were essentially shut down for maintenance.

Your assertion over estimates the risk

My assumption of a uniform probability distribution over the entire expected time interval between events is unrealistic, given the known mechanisms - and that assumption underestimates the current risk of a large earthquake in a vulnerable zone that hasn't had one in a while.

Again, we come back to these unreasonable assumptions of yours. By your logic, everything in North America should be built as if it's in California

Everything built on an island in the Mississippi River over an earthquake fault, whose failure is capable of taking out the upper river in one event, should be built at least to withstand a quake an order of magnitude greater than the New Madrid quake.

We know for sure that size quake can reasonably happen in our lifetimes, and a safety margin beyond it would be prudent.

If we treat a 1 in 1000 year event as having a 1 in 1000 chance of occuring, then the probability of it occuring (if I'm remembering my maths correctly) in a 50 year period is more like 4.98% and the chance of it occuring in a 1000 year period is actually more like 63%.

Uh, yes? And that differs from my 1 in 50 cuff estimate over thirty years how? Thing is, you take any 30 years out of 1500 with a uniform distro, the probability is 1/50 the probability of the whole 1500 seeing the event. Since 1500 is the time "expected", the interval for which we take even odds, we take 1/50 of even odds for any thirty year span - a 50 to 1 shot.

But if you don't like that kind of estimation, by all means use your numbers - they are worse, for those claiming safety. Or use a more realistic and mechanism based assumption of rising probabilities over time, as strain builds up on the fault - get worse odds yet, for a major quake region that hasn't seen a big quake in a while.

 

[Trippy]

Everything built on an island in the Mississippi River over an earthquake fault, whose failure is capable of taking out the upper river in one event, should be built at least to withstand a quake an order of magnitude greater than the New Madrid quake.

Based on what evidence?

We know for sure that size quake can reasonably happen in our lifetimes, and a safety margin beyond it would be prudent.

Based on what evidence?

Uh, yes? And that differs from my 1 in 50 cuff estimate over thirty years how?

I was fairly certain you would say that without seeing the intrinsic hypocrisy of the statement.

Thing is, you take any 30 years out of 1500 with a uniform distro, the probability is 1/50 the probability of the whole 1500 seeing the event.

Well, duh, I make the same assumptions you do, and apply the correct model.

Since 1500 is the time "expected", the interval for which we take even odds, we take 1/50 of even odds for any thirty year span - a 50 to 1 shot.

No.
If you apply a biased coin tossing model, it's 4.98%, not 5%. You would begrudge designers that 1 in 5000, so why should you be let away with it.

But if you don't like that kind of estimation, by all means use your numbers - they are worse, for those claiming safety. Or use a more realistic and mechanism based assumption of rising probabilities over time, as strain builds up on the fault - get worse odds yet, for a major quake region that hasn't seen a big quake in a while.

All of this is nonsense, because it's based on a false assumption - that seismic risk analysis is calculated using a biased coin tossing model.

This is your assumption, and not a consequence of anything I have stated or implied.

I used the biased coin tossing model because I can remember the math, and do it on a calculator, and becuase in general it's not a bad approximation.

Addendum:
Oh yeah - and you left out the fact that it grossly over estimates the probability of a second large quake immediately after the first (Immeadiately after of course being a relative term here).

 

[iceaura]

We know for sure that size quake can reasonably happen in our lifetimes, and a safety margin beyond it would be prudent.
”
Based on what evidence?

Based on the actual occurrence of a large earthquake in the vicinity fairly recently, which nevertheless did not release the faults in question. There's a threat there, and we would be foolish to ignore it with so much on the line.

Since 1500 is the time "expected", the interval for which we take even odds, we take 1/50 of even odds for any thirty year span - a 50 to 1 shot.
”
No.
If you apply a biased coin tossing model, it's 4.98%, not 5%. You would begrudge designers that 1 in 5000, so why should you be let away with it.

Your objection to my estimating remains mysterious, but as there is no significance to my arguments I can't say I care to worry the matter.

All of this is nonsense, because it's based on a false assumption - that seismic risk analysis is calculated using a biased coin tossing model.

Which I mentioned - the more realistic models hand you more risk, for the Japanese situation. Very vulnerable, high threat area that hasn't had a major quake in a while. Strain building up. Geological situation known to produce very large quakes - 9.2, across the plate, not fifty years ago.

Waves from that one came ashore in Japan - people noticed them. Hints.

 

[Trippy]

Based on the actual occurrence of a large earthquake in the vicinity fairly recently, which nevertheless did not release the faults in question. There's a threat there, and we would be foolish to ignore it with so much on the line.

That's not evidence, it's speculation.

What evidence do you have that (for example) the New Madrid seismic zone is capable of producing an M9+ earthquake?

Which I mentioned - the more realistic models hand you more risk, for the Japanese situation. Very vulnerable, high threat area that hasn't had a major quake in a while. Strain building up. Geological situation known to produce very large quakes - 9.2, across the plate, not fifty years ago.

Waves from that one came ashore in Japan - people noticed them. Hints.

More bullshit.
Not every active fault is capable of producing an M9 earthquake.
Not even every active fault on the pacific rim is capable of generating an M9 earthquake.

 

[iceaura]

Now we can go back and compare the "alarmist" vs the "sensible" posts in this thread, for their connection to physical possibility, in the light of subsequent events so far:

http://www.nytimes.com/2011/03/15/world/asia/15nuclear.html?pagewanted=2&_r=1&hp

Tokyo Electric Power said Tuesday that after the explosion at the No. 2 reactor, pressure had dropped in the “suppression pool” — a section at the bottom of the reactor that converts steam to water and is part of the critical function of keeping the nuclear fuel protected. After that occurred, radiation levels outside No. 2 were reported to have risen sharply.

“We are on the brink. We are now facing the worst-case scenario,” said Hiroaki Koide, a senior reactor engineering specialist at the Research Reactor Institute of Kyoto University. “We can assume that the containment vessel at Reactor No. 2 is already breached. If there is heavy melting inside the reactor, large amounts of radiation will most definitely be released.”

Another executive said the chain of events at Daiichi suggested that it would be difficult to maintain emergency seawater cooling operations for an extended period if the containment vessel at one reactor had been compromised because radiation levels could threaten the health of workers nearby.

If all workers do in fact leave the plant, the nuclear fuel in all three reactors is likely to melt down, which would lead to wholesale releases of radioactive material - - -

What evidence do you have that (for example) the New Madrid seismic zone is capable of producing an M9+ earthquake?

An order of magnitude safety margin past the already and recently experienced, at least, seems reasonable and prudent to me.

What evidence do you have that the New Madrid quake was the largest possible in the general area? Seems unlikely, offhand - to presume an outlier.

Which I mentioned - the more realistic models hand you more risk, for the Japanese situation. Very vulnerable, high threat area that hasn't had a major quake in a while. Strain building up. Geological situation known to produce very large quakes - 9.2, across the plate, not fifty years ago.
”
More bullshit.

You do use that word a lot. Lessee: Japan has experienced at least three quakes about 8.5 or so in the past three hundred years - about every hundred and fifty years, a big quake set hits. The last set or flurry with a really big one was about 150 years ago. The plate margin involved generates large quakes with some frequency, and has since WWII - in Alaska, 9.2 about fifty years ago. So one would expect the design of any nuclear reactor to be able to handle the already experienced locally - 8.6 or so - with a safety margin for the possible as demonstrated abroad.

So 9 or so would have been perfectly reasonable and moderate - maybe even a little light.