Fukushima meltdown hastens decline of nuclear power

[biffvernon]

As we agree the information we want is lost in statistical noise. That doesn't mean it's not there. We do know plausible mechanisms for the interaction of ionising radiation and biological material and we do have the precautionary principle.

[Little John]

As we agree the information we want is lost in statistical noise. That doesn't mean it's not there. We do know plausible mechanisms for the interaction of ionising radiation and biological material and we do have the precautionary principle.

Itt doesn't mean its not there, but it does mean that it is statistically insignificantly small in terms of its effect. If it was significant then it wouldn't be merely noise, by definition. That noise exists and existed independent of and prior to the incident at Chernobyl. It hasn't suddenly appeared since Chernobyl. If it had then again, by definition, it wouldn't be noise. Therefore the very most we can say is that some of that noise may or may not be attributable to Chernobyl. However, since the level of noise is insignificantly small in the first place and since it already existed previously, we can say with a very high degree of confidence that whatever component of it may or may not be attributable to Chernobyl, that component has to be vanishingly small.

This is basic statistics B and I can't believe you don't understand this.

To be clear here. I am making no claim about the potential for a nuclear accident, in principle, to cause widespread lethal cancers. We both know how ionising radiation works. What I am saying, though, based on the documented history of every major nuclear accident over the last several decades, is that such an accident causing such a level of cancers simply has not happened as a matter of recorded and measured fact.

The above is unarguable and on the record. It's not my opinion.

[biffvernon]

Oh dear....

Something can be statistically insignificant, lost in the noise, but yet be very significant, causing people to die.

For example, say x number of people die each year because of a cancer that was induced by a high energy cosmic ray and y number of people die because of a cancer induced by the nuclear industry. You are not justified in saying the nuclear industry has not caused deaths just because you can't tell whether the deaths are part of x or part of y or caused by any one of the other numerous carcinogenic substances that we frequently encounter, whose individual frequencies vary in often unknown ways so cannot be disentangled from single events such as Chernobyl.

And I'm not saying that I know that Chernobyl killed more than 37. I'm just saying that anyone who claims that Chernobyl killed only 37 and could not have killed thousands is not justified in the claim.

[Little John]

Oh dear....

Something can be statistically insignificant, lost in the noise, but yet be very significant, causing people to die.

For example, say x number of people die each year because of a cancer that was induced by a high energy cosmic ray and y number of people die because of a cancer induced by the nuclear industry. You are not justified in saying the nuclear industry has not caused deaths just because you can't tell whether the deaths are part of x or part of y or caused by any one of the other numerous carcinogenic substances that we frequently encounter, whose individual frequencies vary in often unknown ways so cannot be disentangled from single events such as Chernobyl.

And I'm not saying that I know that Chernobyl killed more than 37. I'm just saying that anyone who claims that Chernobyl killed only 37 and could not have killed thousands is not justified in the claim.

If the number of people who are known to contract cancers in the absence of exposure to the ionizing radiation following a nuclear accident is x and, following a nuclear accident, the number of people who contract cancer is still roughly x then we can, with reasonable confidence, say that the nuclear accident has not contributed in any significant way to a rise in the number of cancers. The above happens to be, by the way, an accurate description of the highly recorded and deeply analyzed results of many years of rigorous scientific study surrounding the cancer rates following the Chernobyl accident.

I should note, none of the above is particularly contentious. It's just the recorded facts. Frankly, your persistence in denying these incontrovertible facts is becoming somewhat surreal. I can only assume that this topic is something of a blind spot for you.

[biffvernon]

'Expert' opinion on Chernobyl deaths shows a very big range, from 37 to many thousands, so please let's not get personal about this.

It doesn't matter how rigourously you analyze highly recorded data, if the information isn't there you won't get an answer. A lot of people die from cancer, almost all of them nothing to do with nuclear power generally or Chernobyl in particular. Cancer rates have varied with time as we have adopted different lifestyles, become exposed to new substances and learnt not to die of other causes. Death may occur decades after the triggering cause. It is just impossible to know. Epidimiological methods don't work when the signal to noise ratio is too small. The general is made up of many particulars. Here are two. My sister died of cancer but the cause of her cancer will ever be known. We don't enven know whether she was out walking in the rain a few days after Chernobyl and suffered a fatal nuclear disintegration. (Infinitesimally unlikely but not a zero probability.) My son was not out playing in the rain that day because I had, acting on the precautionary principle, kept him indoors at the time. He's still alive but there is no possible way to tell whether this is a result of my action. (Infinitesimally unlikely but not a zero probability.)

[Little John]

'Expert' opinion on Chernobyl deaths shows a very big range, from 37 to many thousands, so please let's not get personal about this.

It doesn't matter how rigourously you analyze highly recorded data, if the information isn't there you won't get an answer. A lot of people die from cancer, almost all of them nothing to do with nuclear power generally or Chernobyl in particular. Cancer rates have varied with time as we have adopted different lifestyles, become exposed to new substances and learnt not to die of other causes. Death may occur decades after the triggering cause. It is just impossible to know. Epidimiological methods don't work when the signal to noise ratio is too small. The general is made up of many particulars. Here are two. My sister died of cancer but the cause of her cancer will ever be known. We don't enven know whether she was out walking in the rain a few days after Chernobyl and suffered a fatal nuclear disintegration. (Infinitesimally unlikely but not a zero probability.) My son was not out playing in the rain that day because I had, acting on the precautionary principle, kept him indoors at the time. He's still alive but there is no possible way to tell whether this is a result of my action. (Infinitesimally unlikely but not a zero probability.)

Show me a single publicly moderated study where there has been shown to be a significant rise in any cancers following the Chernobyl accident on the whole of the European continent.

If you can't then every thing you have said on this up to now (in terms of Chernobyl being responsible for such a rise) is quite simply incorrect no matter how you may spin it.

If, on the other hand, you merely wish to make the the less specific case that nuclear is bad and it's probably better to play it safe and not have it in the first place, then fair enough. But, please, stop trying to suggest that previous accidents provide evidence in support of this precautionary principle because they quite simply don't. In fact, they provide evidence to the contrary when compared to deaths arising from the use of other energy sources.

[biffvernon]

No I can't show you a single publicly moderated study where there has been shown to be a significant rise in any cancers following the Chernobyl accident on the whole of the European continent. Such a thing does not exist because it cannot exist, for the reasons I gave above.

I haven't said that Chernobyl is responsible for a rise. I've just said that we cannot say that it is not responsible for a rise.

We have a hypothesis - Chernobyl caused a rise in cancers. That hypothsesis has not been falsified. What we do have is a mechanism in biology and a precautionary principle. That is one of the several reasons why I'm against nuclear power.

Now, less of the personalised strawmen, please.

[RenewableCandy]

As a small divertissement from the above statistical gymnastics, there are people I know, quite sensible people with an allotment, who say that broccoli(sp?) protects one against the effects of a mild radiation dose. I presume it's either the "mopping-up free radicals" job or else increases the ability of cells to repair themselves (or snuff it gracefully and be replaced). Can anyone on the Panel shed any light on this?

[Little John]

As a small divertissement from the above statistical gymnastics, there are people I know, quite sensible people with an allotment, who say that broccoli(sp?) protects one against the effects of a mild radiation dose. I presume it's either the "mopping-up free radicals" job or else increases the ability of cells to repair themselves (or snuff it gracefully and be replaced). Can anyone on the Panel shed any light on this?

The most recent research suggests that very mild doses of radiation are actually good for you. I shit you not.

It turns out that there is not the simple linear relationship between exposure and cell damage that was originally thought. Rather, the slope of the damage line has a hockey stick shape to it near the bottom. A small dose of ionizing radiation seems to trigger cell repair mechanisms in the body to go into overdrive and, in doing so, more than compensate for any damage done by the radiation. Thus, people who are exposed to low levels of radiation are slightly less likely to get cancer than those who are not exposed at all.

[Little John]

No I can't show you a single publicly moderated study where there has been shown to be a significant rise in any cancers following the Chernobyl accident on the whole of the European continent. Such a thing does not exist because it cannot exist, for the reasons I gave above.

I haven't said that Chernobyl is responsible for a rise. I've just said that we cannot say that it is not responsible for a rise.

We have a hypothesis - Chernobyl caused a rise in cancers. That hypothsesis has not been falsified. What we do have is a mechanism in biology and a precautionary principle. That is one of the several reasons why I'm against nuclear power.

Now, less of the personalised strawmen, please.

This is a complete mangling of the concepts of science. It is akin to asserting that because no-one can falsify your claim that a goblin created the universe and lives in the sole of your shoe, then on the basis of the precautionary principle, we cannot discount your claim. You claim cannot be falsified because it is unfalsifiable. In scientific terms, this means it is basically irrelevant in terms of it being any kind of evidential basis for asserting that it has caused lots of cancers.

I really need to be absolutely clear here, though. I am certainly not denying that ionizing radiation in sufficiently high and chronic doses can cause cancer. I am also not denying that a nuclear accident could occur that could cause massive damage to people's health. There may even be an argument to be made on the basis of the precautionary principle that states that because of what we know about radiation damage it is probably not a good idea to have nuclear power. Any of these are legitimate, though arguable, positions. However, what you cannot do with any legitimacy is try and make out that nuclear accidents such as Chernobyl (or, indeed, any other major nuclear accident of the last several decades) provide evidence of the kind of widespread damage to people's heath that you fear may happen in any future accident.

Indeed, what can be stated categorically is that the radiation emitted from the Chernobyl accident was neither high enough nor chronic enough to have had any significant effect on the background levels of deaths from cancer in the European continent.

We can state this as a fact.

[mobbsey]

there are people I know, quite sensible people with an allotment, who say that broccoli(sp?) protects one against the effects of a mild radiation dose.

The idea is that highly nutritious food full of micro-nutrients charges you're body's storage system so that when exposed to radionuclides from a meltdown/weapons accident -- that's the actual radioactive chemical elements, not the radiation these elements produced -- you absorb very little.

This is the same idea as taking iodine tablets in the event of a nuclear meltdown, but it covers a wider range of pollutants. For example, all leafy veg is high in natural (non-radioactive) strontium; eat lots of leafy veg and you keep your bones well stocked with strontium so that in the event of exposure to a meltdown you won't absorb radioactive strontium.

This is one of the reasons why it's good to have a diverse diet, high in raw and unprocessed food. It's not just that the nutrients strengthen your immune system and the antioxidants protect your DNA, keeping your body topped up with nutrients reduces (but can never eliminate) the uptake of pollutants which mimic nutrients -- not just radioactive pollutants, but chemicals in general.

The difficulty is that so called "internal emitters" are not routinely included in the standard calculations for radiation exposure -- and it's the counting of risk from internal emitters which leads to proportion of the difference in results between official IAEA risk estimates and those of other scientists.

[biffvernon]

However, what you cannot do with any legitimacy is try and make out that nuclear accidents such as Chernobyl (or, indeed, any other major nuclear accident of the last several decades) provide evidence of the kind of widespread damage to people's heath that you fear may happen in any future accident.

There's the strawman again. I've only said we can't be sure widespread damage has not occured.

Chernobyl may not have had a statisically significant effect above background levels but, considering the size and noisiness of that background level, we cannot say that Chernobyl did or did not kill thousands. I may have mentioned that before, so this is getting a bit repetetitive.

What we mustn't do is conflate purely statistical reasoning with views about the linear no-threshold model (LNT) and radiation hormesis, about which there is no consensus view amongst researchers. Your 'It turns out...' statement is premature.

Wikipedia has a fair account of the state of the science at http://en.wikipedia.org/wiki/Radiation_hormesis

[biffvernon]

Inevitably, other views are available, such as this: http://www.shtfplan.com/headline-news/u ... e_05252012

[Little John]

However, what you cannot do with any legitimacy is try and make out that nuclear accidents such as Chernobyl (or, indeed, any other major nuclear accident of the last several decades) provide evidence of the kind of widespread damage to people's heath that you fear may happen in any future accident.

There's the strawman again. I've only said we can't be sure widespread damage has not occured.

Chernobyl may not have had a statisically significant effect above background levels but, considering the size and noisiness of that background level, we cannot say that Chernobyl did or did not kill thousands. I may have mentioned that before, so this is getting a bit repetetitive.

What we mustn't do is conflate purely statistical reasoning with views about the linear no-threshold model (LNT) and radiation hormesis, about which there is no consensus view amongst researchers. Your 'It turns out...' statement is premature.

Wikipedia has a fair account of the state of the science at http://en.wikipedia.org/wiki/Radiation_hormesis

Well, yes, as a matter of fact, we can be sure of that. Show me the numbers to contradict the above. As for your view on" statistical reasoning"; what precisely would be your preferred method of objective analysis and derivation of conclusions from a large data set of evidence?

[biffvernon]

I can't show you the numbers! That's why we can't be sure.

It doesn't matter how big the data set is - if it's too noisy the only safe conclusion is that we don't know.