Wednesday, 26 January 2011

Mr Huhne Goes To (Radioactive) Hinkley

Readers will know of the high esteem in which we hold Chris 'Crapper' Huhne. Join us, then, in our devout hope that his visit this week to Hinkley Point hasn't injured his health.

For this is the site of one of EDF's recently-acquired nukes (Hinkley 'B'), and also where they intend to build Hinkley 'C' (this year, next year, some time, never). 'A' was
a grotty old Magnox reactor, decommissioned in 2000.

As part of its preparation for the new plant, EDF commissioned engineering firm AMEC to carry out a survey of radioactiv
ity at the site surrounding the existing plant. The survey & data are in the public domain, and have been independently assessed by one Prof Chris Busby, whose conclusions are potentially alarming for the citizens of Somerset (including some of this blog's nearest, dearest, and good friends).

He reckons that the ra
diation there exceeds background levels for that part of the world; and that the readings indicate the presence of rather a lot of enriched uranium - which doesn't occur in nature. The only obvious source would be Hinkley 'B' (which uses enriched uranium, whereas 'A' didn't). But how would it have got into the ground outside the power plant ?

Well I certainly don't know. 'A' and 'B' both had their share of leaks and accidents over the years - past standards were pretty lax - but large amounts of enriched uranium would seem unlikely. We should also note that Prof Busby is by no means uncontroversial, & that's putting it mildly.

The immediate responses from EDF and the Environment Agency have been just bluster, but one imagines they will be scurrying around behind the scenes to come up with a proper answer to the prima facie case made by Busby.

In the meantime, let's all hope Huhne hasn't been harmed, eh ?
(& not forgetting the population of West Somerset). Ironically, Huhne has just signalled a significant increase in the cap on cleanup liabilities for nuclear generators: his EDF hosts will be hoping they aren't the first to wear this new £1 billion cap.

ND

32 comments:

Blue Eyes said...

Surely Huhne's job now is to delay the start of any nuclear/clean coal/solar/wind projects until the price of gas collapses.

Nick Drew said...

... is the correct answer ! (and not to give the subsidy-wallahs any free £££ in the mean time)

a £100 loft insulation voucher for that man

Blue Eyes said...

I have something much better than an insulated loft: people who live below me and people who live above me!

Anonymous said...

Maybe the powers that be will design more efficent nuclear power stations, a lot of the old ones were designed to create plutonium so the west could threaten the east, that if they launched a nuke attack on the west they would be nuked likewise and of course visa-versa, electricity was an easily sellable by product the story going back years was that nuke power would give unlimited cheap power, and they were believed. Once they design a plant that can use the lethal by product as a source of power maybe, just maybe "unlimited" power from nuclear power may be acceptable.
"The most important isotope of plutonium is plutonium-239, with a half-life of 24,100 years" (wikipedia) with others more or less

Anonymous said...

look, please stop endearing 'prof' busby with more scientific 'clout' than he deserves. Judging by his cv he's prof only by virtue of a 'visiting' prof-ship - basically access to a library card - and he hasn't done any hands-on research beyond a single years post-doc way back when. Somebody here mentioned he had two phd's the other day - a look at his cv says he dropped out of the first and has, since his post-dac days, spent all his time talking, not doing, and committees rather than actively doing. The lack of hands-on data handling shows in this report.

if you submitted graphs like that shown on page 6 or 10 of the report for a fist year undergrad physics project you'd struggle to pass. Sorry, its just not good enough - and lots not talk about detection limits error analysis analytical protocols or even references. This kind of scaremongering report is really that, scaremongering. Besides which, if this were a stock report, written by a vested interest, you would treat it with a whole heap more scepticism.

finally, I'd like to point the prof in the direction of the Gabon natural reactor where elevated 235/238 isotope ratios did occur - naturally.

Anonymous said...

excuse spelling - am in a rush. :(

Anonymous said...

'prof' busby who on earth is he, never heard of him the great british public has bombarded for years will the talk of un limited cheap nuclear fuel and it has yet to make an appearance, it is nothing to do with scare mongering, so far science has NOT delivered, they have delivered on massive cleanup bills, Maggie thought she could get someone interested in BNF no one was interested too great a liability for clean up and on going storage. If nuclear power stations are perfetly safe what would you say to one within 5 miles of your house in the center of London.
Sorry Nick, couldn't resist it

Anonymous said...

What a crying shame that on such a critical issue as energy security, the debate as been comprehensively hijacked by charlatans and those, most notably Al Gore, talking their book.

As our seniors freeze to death in fear of the next gas bill, self-serving 'greens' lie, lie and lie again about the climate change, environmental risks and costs.

The French and Russians are laughing all the way to the bank as they prepare to put the UK over the barrel and rob us blind for each kW.

Lantzelot said...

The Busby study is flawed due to the simple fact that he has ignored the statistical uncertainties given by the EDF data (this is not the first time where he fabricates alarming results by adjusting the data in some way or another). If the uncertainties are included in each plot, they will cover the whole pictures, clearly showing that any kind of line can be drawn. The associated R-squared values that are shown in the report is an indication of how improbable the conclusions are. R-squared close to 1.0 shows very likely correlations. In Busby's study the R-squared values are 0.2 or less.

Furthermore, the measured activities are close to the lower limit of what is possible to measure. 60 Bequerel per kilo mud is a typical value from the EDF-study, Busby tries to scare people by saying that this is alarmingly high and that the expected background value should be about 20 Bequerel per kilo. Even if he would be right on this, the level of 60 Bequerel can be compared with 20 Bequerel in an ordinary banana of 150 grams. Hardly any reason for health concerns, especially as most people never eat a kilo of mud, but more often a banana.

hatfield girl said...

Perhaps it's our very own Semipalatinsk.

rwendland said...

"the measured activities are close to the lower limit of what is possible to measure. 60 Bequerel per kilo mud is a typical value ..."

Lantzelot, if that is the case why does the AMEC lab report for EDF claim it can measure, for example, U-235 to +- 0.00045 Bq/g (or 0.45 Bq/kg)? [on page 36 of the pdf (page 4 of the lab report)]

Are you saying the errors accumulate further down the analysis?

Nick Drew said...

thanks for the detailed feedback, Lantzelot & anons

seniors freeze to death in fear of the next gas bill

more likely to be electricity bills (there is plenty of gas), but I have a feeling that this is what will eventually call a halt to the current 'decarbonisation' thrust & the sooner the better

GDP trumps GHG, as I have been known to say

Anonymous said...

rwendland


To be fair, its not a brilliant report either, but it seems the key point is on page 20 - "Most data are presented as “less than values”, but where a radionuclide has been detected above the LOD, the uncertainties have been presented". If values are beneath the detection limit, then quoting actual numbers is meaningless.

Lantzelot said...

rwendland 10:25 PM:

Lantzelot, if that is the case why does the AMEC lab report for EDF claim it can measure, for example, U-235 to +- 0.00045 Bq/g (or 0.45 Bq/kg)? [on page 36 of the pdf (page 4 of the lab report)]

The 0.00045 Bq/g is the statistical uncertainty of the measured activity of 0.00211 Bq/g. That means a 21% statistical uncertainty on the measured activity of U-235. In some of the other samples the uncertainty is more than 50%. To me this indicates that the measurement is difficult, close to the limit of what is technically possible in the present situation.

If you have a 1 g sample then you have to measure for almost 8 minutes to get a single count of U-235. In order to get the 21% accuracy (and disregarding other correction factors that probably have increased the uncertainty) you need 20-25 counts. For a 1 g sample this would take up to 3 hours of measurement. I assume that the sample was larger though.


Are you saying the errors accumulate further down the analysis?

Yes, they should. For the present example you have the following activities:
U-235: 0.00211 +- 0.00045 (21% error)
Th-234: 0.040 +- 0.016 (40% error)

Then, following Busby's recipe, we get the ratio Th-234/U-235 = 18.96 to be compared with 21.3. To Busby this is an indication of a "Borderline" case that may have an enrichment slightly higher than normal. But Busby has not included the statistical errors, which in the present case would give the result
Th-234/U-235 = 18.96 +- 8.59, assuming standard error propagation from Gaussian distributions. As seen from this example the error bars certainly overlap with the 21.3 value for natural uranium. This is the case for most of the samples in the AMEC report.

I'd say that it is very dishonest of Busby to display all the plots and making all the statements without including any statistical uncertainties. But of course, if he did then anybody could easily see that he has nothing to show. Those who argue against new nuclear power plants at Hinkley Point should think twice before associating with Busby, this lowers their credibility.

rwendland said...

Lantzelot, trouble is that Busby is only using the U-235 and Th-234 (a U-238 indicator) results in his Uranium analysis. And all of those readings in the AMEC lab report seem to be above the LOD (Limit of detection), and in stated +- uncertainty ranges. So the LOD issue does not seem to affect his analysis.

I suspect there is +- error bar issue in his analysis, but it does not look to me like it is the LOD issue.

Bill Quango MP said...

I'm still confused. is the prof wrong? Is the data incorrect? Or am I am about to go all Silkwood?

Lantzelot said...

rwendland 10:25 PM:

Lantzelot, if that is the case why does the AMEC lab report for EDF claim it can measure, for example, U-235 to +- 0.00045 Bq/g (or 0.45 Bq/kg)? [on page 36 of the pdf (page 4 of the lab report)]

The 0.00045 Bq/g is the statistical uncertainty of the measured activity of 0.00211 Bq/g. That means a 21% statistical uncertainty on the measured activity of U-235. In some of the other samples the uncertainty is more than 50%. To me this indicates that the measurement is difficult, close to the limit of what is technically possible in the present situation.

If you have a 1 g sample then you have to measure for almost 8 minutes to get a single count of U-235. In order to get the 21% accuracy (and disregarding other correction factors that probably have increased the uncertainty) you need 20-25 counts. For a 1 g sample this would take up to 3 hours of measurement. I assume that the sample was larger though. So, an LOD issue or not, it is a measurement with low statistics.

Blue Eyes said...

My favourite nuclear "story" is still that the only place in Europe over which the path of the wind from Chernobyl passed twice was at Windscale, Cumbria.

Lantzelot said...

rwendland 10:25 PM

Are you saying the errors accumulate further down the analysis?

Yes, they should. For the present example you have the following activities:
U-235: 0.00211 +- 0.00045 (21% error)
Th-234: 0.040 +- 0.016 (40% error)

Then, following Busby's recipe, we get the ratio Th-234/U-235 = 18.96 to be compared with 21.3. To Busby this is an indication of a "Borderline" case that may have an enrichment slightly higher than normal. But Busby has not included the statistical errors, which in the present case would give the result
Th-234/U-235 = 18.96 +- 8.59, assuming standard error propagation from Gaussian distributions. As seen from this example the error bars certainly overlap with the 21.3 value for natural uranium. This is the case for most of the samples in the AMEC report.

I'd say that it is very dishonest of Busby to display all the plots and making all the statements without including any statistical uncertainties. But of course, if he did then anybody could easily see that he has nothing to show. Those who argue against new nuclear power plants at Hinkley Point should think twice before associating with Busby, this lowers their credibility.

Tim Worstall said...

We can go further as well.

Assume, for a moment, that Busby is correct in everything he says.

That numbers are 40 Bq above where they should be for the area, to 60 Bq instead of 20.

As he himself points out, Dartmoor is 330 Bq, naturally. So this "massive pollution" has just made one part of Somerset 10% more like another part of Somerset.

Not all that much to write home about.

BTW, this is also almost certainly less than the additional radiation from having a coal fired plant in hte same place.

Anonymous said...

Nitpick for TW: Dartmoor, as any fule kno, is in Devon.

Bill Quango MP said...

I was in a village in South Somerset today. They have radon gas detectors. What's that all about?

rwendland said...

Lantzelot, yes I agree with you. It is quite possibly not allowing for the error brackets that have caused Busby to erroneously conclude elevated enrichment in the samples. I suppose with enough samples you could do statistical analysis to unwind some of the bracket on a probability basis, but I doubt 28 samples is enough for that to be done reliably.

I've looked at Busby's "Highest EU sample", and you do need to go to the extreme edge of the error bracket to get it back to Natural U, but you do get to there. Sample TRE02 S2 0.4-1.3, where Busby lists a High EU, at the edge of the error bracket gives:

(0.046 + 0.021) / (0.00377 - 0.00068) = 21.68

which is a central Natural U ratio.

He also seems to have a data transcription error, muddling up the sample numbers and data. With sample TRE03 S2 0.45-1.7, another "High EU" according to Busby, I read the AMEC report numbers differently than Busby has listed:

Busby: 0.046 / 0.00377 = 12.20
AMEC: 0.046 / 0.00241 = 19.09

Busby seems to have listed these numbers incorrectly against sample TRE04 S4 2.5-4, so his ratios are currect, but listed against wrong sample numbers sometimes. I've not chased it down, but this means his enrichment level to depth mapping could well be wrong.

So it does rather look like Busby's case for enriched uranium is weak or non-existent. Certainly does not look convincing, as is, under close scrutiny.

rwendland said...

BQ: South Somerset & radon gas detectors.

Could be chance, or being careful. I was once randomly selected by NRPB to be asked to set up radon gas detectors in my house as part of national survey. Just a coincidence that I was interested in that - I never got to see the exact results for my house, but I do know I'm in a low Radon area.

Tim Worstall said...

Radon detectors: If you're in an area with granite underneath you the natural uranium in that granite breaks down, Radon being one of the daughter products.

So if you're in such an area (and of course, randomly elsewhere, to check on whether there is granite down there) you'd like to have a detector to see whether the gas is accumulating in your basement and thus whether you'd like to ventilate the house a bit more.

rwendland said...

Tim Worstall, to be fair to Busby, he is not claiming the levels themselves are alarming to the general population.

What he is trying to claim is that there must have been previously unrecorded levels of emissions or spillage. In particular more radioactive dust, either directly emitted when venting off with the CO2 coolant or out to sea in liquids which subseqently came back onshore, which feeds into his low-level radiation concerns.

His recommendations are to delay site prep work for more detailed samples to be taken and analysed.

Nick Drew said...

Tim - did commenter #5 on your own post ever come back with search details ? I assume not ...

Lantzelot said...

rwenland 12:14 PM and 12:46 PM

Busby seems to have listed these numbers incorrectly against sample TRE04 S4 2.5-4, so his ratios are currect, but listed against wrong sample numbers sometimes. I've not chased it down, but this means his enrichment level to depth mapping could well be wrong.

Yes, there is a mismatch in Busby's table, but only on the labels, they are displaced one step down with respect to its proper data.

What he is trying to claim is that there must have been previously unrecorded levels of emissions or spillage. In particular more radioactive dust, either directly emitted when venting off with the CO2 coolant or out to sea in liquids which subseqently came back onshore, which feeds into his low-level radiation concerns.

If we investigate the AMEC data closer, we see that the other isotopes identified with statistical uncertainties (i.e. with measured values, and not with a "<" value) all correspond to isotopes from the natural Uranium and Thorium decay chains. There is not a single isotope in any of the charts that is associated with nuclear fission products. If there is a leak from the nuclear power plants there would be signs of elevated levels of, say, Cs-137.

So, elevated levels of uranium only, where would it come from? Has somebody chipped off grains from the nuclear fuel before it went into the reactor, and then milled it into dust that would be dispersed in the environment?

My guess is that there are local variations in the natural uranium level that are not displayed in the report referred to by Busby (Beresford et al 2007). I am not based in the UK, if anybody knows of any more detailed study of the local geology, please give us a hint.

Nick Drew said...

thanks again chaps, very illuminating

& I imagine there will be more of this to keep us entertained in the years to come !

Lantzelot said...

Hi Nick, a late comment:
We have put our investigation of Busby's claim on our blog and forum. It is really nothing new from what was discussed here, but with a few enlightening pictures:
Blog post here: http://nuclearpoweryesplease.org/blog/2011/02/07/chris-busby-and-the-tall-tale-of-ten-tons-uranium-gone-missing/
Forum post here:
http://nuclearpoweryesplease.org/forum/viewtopic.php?f=16&t=113

Nick Drew said...

Lantzelot - yet again, thanks for following through, it is good to see your full analysis

the reasons my original post was rather guarded were twofold, and not very scientific: (a) ad hominem, based on Busby's track record; and (b) 10 tonnes !?

I am glad that a bit of rigour & science bore this out !

PS did you ever get to the bottom of the cryptic comment in Stanley Davies' comment at Tim's, here ?

http://timworstall.com/2011/01/26/radioactivity-at-hinkley/#more-21377

Lantzelot said...

Nick,
Thanks for the link to Tim's blog, I did not get what you meant the first time that you mentioned it.
I am at a loss to understand what Stanley Davis is aiming at, no web search of mine reveal any clues. Anyhow, Tim Worstall mentions some important things, the UK is de facto dispersing plenty of uranium through the coal power plants. So far I have not seen Busby trying to make up any cancer clusters in the vicinity of any coal plant.

I am sure that this will not be the last we hear from Busby regarding Hinkley Point, there may be plenty of "fun" ahead. :-)