19 Comments

Well put and entertaining, as always.

What would you write if Musk's new Department of Government Efficiency (DOGE) asked you to explain how to fix US regulations - in plain language so the public understands why it is in their best interest? What's the goal? What are the societal benefits of the changes? Nuts and bolts, how do you make it happen?

Expand full comment
author

V,

The Gordian Knot Group proposes replacing the current nuclear power regulatory regimee with Underwriter Certification (UCert). A brief description of UCert is at https://jackdevanney.substack.com/p/underwriter-certification-of-nuclear-4fe

A 100 page monograph for congressional staffers who will have to draft the legislation is at https://gordianknotbook.com/download/underwriter-certification-of-nuclear-power

Half-measures will not work. Any "reforms" which end up with a bureaucrat whose over-riding priority is safety (aka preventing a release) having the last say will leave us in the current mess.

Support for more nuclear is now over 70% (Bisconti, 2024) Normally such legislation could be sold with a hard enough push. There are at least two problems.

a) Such legislation will be strongly resisted in districts that are the beneficiaries of the current taxpayer rip offs. The majority of these districts are red. Trump would almost certainly need Democrat votes in the House to get the legislation enacted. Dems are in no mood to give Trump anything.

b) I see no evidence that Trump regards nuclear power as an important issue let alone a priority.

Sadly, I don't expect any substantial changes,

Expand full comment

My current standard comment in social media mentions of nuclear regulation is that Congress should solicit third party bids for alternative regulation systems to handle advanced reactors, and potentially all reactors if that succeeds

The Gordian Knot system would be my choice but it also has the virtue that people don't have to directly prefer it to the NRC, but rather approve of the application process where it could be the winner

Expand full comment
author

Smope,

It's an intriguing idea, but I don't like the advanced/non-advanced (whatever that means) distinction. Conceding non-advanced nuclear (however defined) is not good enough undermines the whole argument.

I think we have to take it in steps.

1) The first step is replacing LNT with a harm model that recognizes repair, such as SNT.

2) If and only if we can do that, then a radiation exposure compensation program based only on the individual's maximal dose-rate profile become insurable.

3) Once that compensation program is enacted and we enforce adequate buffer zones around each plant, then it should be possible to argue that the NRC is superfluous.

But nothing happens as long as LNT is the established dogma. If we go after LNT where it's weakest, the L, then it is just possible we can make an end run around the careerists. The argument is so strong an independent minded person like Musk or Ramaswamy could be persuaded to go against the 'experts' at EPA and NRC for whom they have little regard. And he just might be able to convince Trump to push Congress to legislatively replace LNT with SNT. The change might be able to fly under the radar as a technical modification to the No Threshold model.

Expand full comment

You could have a sigmoid with a threshold as well. I think the name SNT is beneficial because it shoves their poor curve fitting skills right back in their faces.

Expand full comment

I got confused? Lewis or Lucy? Which names stays? The immaculate influencer turned nuclear energy activist? :-D

I tease.

Expand full comment

There may not be a threshold in the sense you intend but could it not be argued there is something that term captures, perhaps poorly: People would be actually less healthy if all radiation was blocked. (Well, you can't block internal sources, so I mean just all external sources.)

Expand full comment
author

Max,

Pls see the pieces on hormesis. You can get to them thru the site directory in the navigation bar. Hormesis is a swamp we do not want to enter. Stick to blasting LNT. Embrace no threshold. You guys were right all along. Just need to tweak the No Threshold model a bit. That's the way forward.

I won't respond to any more handwaving about hormesis. If you want to argue for hormesis,

come with a completely defined model, code that will convert an arbitrary dose-rate profile into a quantitative increase/decrease in cancer. Otherwise don't come at all.

Expand full comment

Nothing makes sense in the field of radiation risks. The field was deliberately undermined by maintaining confusion between acute high-dose-rate exposure and chronic low-dose-rate exposure. I increasingly believe that everything was done, including fraud, to minimize the number of cancers caused by the nuclear explosions in Japan.

https://danielcorcos.substack.com/p/radiation-the-other-conspiracy-of

Expand full comment

I wonder how LNT relates to Radon. Is all that testing necessary? Are the guidelines too stringent?

Expand full comment
author

World,

The EPA radon recommendations are based on LNT which claims we cant repair DNA damage. The radon decay chains are mostly alpha emitters so the damage is similar to teh radium dial painters and the repair processes are the same. If SNT were applied to radon dose rate profiles, the recommendations would be entirely different.

Having said this, radon is not a a fissionproduct and is not a player in NPP releases, so I have never done the SNT radon numbers. Perhaps you can con a member of the choir to do them for you.

Expand full comment

I will accept that challenge. I will play devil's advocate in another forum. This is likely to be a lengthy debate, and as Jack says, radon is not a player in NPP releases, however, as Jack also says, radon is the chosen field of battle for anti-nukers wanting to defend LNT, and belief in LNT is the foundation of their arguments on safety.

As the debate now stands in Citizendium's summary {citizendium.org/wiki/Fear_of_radiation/Debate_Guide#LNT_and_radon,_Controversy_over_Figure_4}

the LNTers are winning with harsh ad-hominem attacks on Cohen and with blind arguments from authority on LNT orthodoxy. I believe their arguments are BS (Bad Science), and there is Good Science in Cohen's paper, which clearly shows a hormetic effect from low levels of radon. He thoroughly rebuts the criticism that his data can be explained with confounding variables. However, his arguments get deep into the statistics, and our readers (mostly journalists learning about nuclear power) won't be able to follow.

Have a look at the CZ summary. Follow the links to the original debate. Let me know if Citizendium should just drop the debate on radon and the figure critical of EPA's guidelines, or dig deeper and rebut the Dunning Krugers who spout statistics terminology, but misunderstand the science.

Expand full comment
author

I say again: bickering about the effect of tiny dose rates will get you nowhere. For every study that shows a hormetic effect, the antis can come up with a study that claims harm. The way to after your LNTer is to ask him to explain the radium dial painter data. LNT cannot. Ergo, LNT is another hypothesis shot down by an ugly fact.

Expand full comment
Nov 13·edited Nov 13

Show me one good study, as thorough as Cohen's, that supports LNT for low-dose radon. Darby is a "meta-analysis" of other studies. Averaging bad studies with good is BS, especially if the bad studies are as bad as Ed Calabrese has shown us.

The anti-nukers said "show me the data" for Cohen's plot. I did. I asked for the same for their European Code plot. They couldn't. I even made a couple more scatterplots from the raw data, testing their allegation that the 2% range on smoking was "cherry picked".

Cohen's data shows a 30% reduction in lung cancer rate. The uncertainty in lung cancer rate depends on entire county populations (millions). Case control studies with fewer individuals are a lot more uncertainty.

I don't want to give up the radon debate, and lose Figure 4 from Robert Hargraves. https://citizendium.org/wiki/Fear_of_radiation

I think Mike Conley would agree. He has a good section on radon in his new book, LNT Report.

I think the Kerala study makes the same point as the Radium Dial Painters. Should we include both? We can also just add a link, and keep the total reading down for our lame-brain journalists.

Expand full comment
author

Nobody's talking about giving up on the radon debate. What I'm talking about is attacking LNT where it is weakest, where defenders can't obscure the argument with cherry picking, inappropriate control groups, and a zillion confounding factors.

But suit yourself. You can try to convert your politicized, closed minded journalists the hard way, or the easy way.

Expand full comment

It is wrong to say that the anti-LNTer has to prove a threshold. The Null Hypothesis is "No Effect" at say 100 mGy. The pro-LNTer has to prove an effect at 50 mGy or 20 mGy. NCRP Commentary 27 basically says that of course we can't prove an effect at 50 mGy, it is too small. That is of course the point that it is too small to measure.

NCRP-27 says that Grant (Hiro/Nag survivor study) has strong support for LNT. But Grant admits their data is consistent with a threshold for males up to 750 mGy, for females up to 80 mGy, and no effect for the sex averaged linear model up to 80 mGy. I don't have to prove there is a threshold, Grant proved "no effect" to at least 80 mGy for me.

Expand full comment
author

Ken.,

The problem is we need a quantitative model that converts any dose rate profile to predicted cancer incidence. If you propose a zero threshold model up to say 80 mGy over some period you are making a very strong statement. It's one thing to say I am rejecting your linear model because you have not disproven the null hypothesis. It is quite another to claim absolutely zero effect up to some magic number.

By the way, what is the dose period for which your threshold holds. And what happens if that threshold is exceeded. I say again to replace LNT we must have a completely defined replacement, not just a rejection of LNT. See the pieces on Linear With Threshold.

All models are approximations. A well defined threshold model would be a far better approxiamtion of actual harm than LNT. And in terms of practical implications, there would not be that much difference between this model and SNT . (The model could be SNT translated by a threshold.) But standard SNT avoids making a very strong statement that you cant prove and may wellbe false.

Expand full comment

I see your point about a model for predicting cancer incidence by the regulator. However, I am speaking to the public who is terrified that 10 mGy lifetime dose to one person is so catastrophic that even dilution by factors of a million is still deadly to people further downstream from the LLW site that is an inconsequential source. For the public, I keep quoting the regulator who generally speaking knows that my LLW site is inconsequential. I think we have to convince the public that they have been worried about nothing before the regulator will be forced to listen and consider changing.

Expand full comment
author

My advice is: don't focus on the non-detectable "difference" between zero and negligible for the ridiculously tiny dose rates associated with waste sites. Go after LNT by showing it fails miserably for far, far larger dose rates (eg dial painters and other Green Table examples). Once we get rid of LNT and replace it by SNT, redo all the calcs and reset the limits. The essential first step is getting rid of LNT and to do that you must focus on high dose rates, not low.

Expand full comment