SNT: Frequently asked questions.
1) Divide individual's dose rate profile (blue curve) into daily doses.
2) Compute the cancer mortality from each day's dose separately using an S-shaped acute dose response curve based on the atom bomb survivor cancer data. In upper frame insert, we see only the very bottom of the S.
3) Add up the daily mortalities adjusted for the chance of already contracting mortal cancer to obtain red curve.
4) Compensate each exposed person on the basis of his or her total expected mortality.
An SNT FAQ
Occasionally, the choir can be quite perceptive. Periodically, I receive questions, often quite good questions, about SNT. So I've decided to set up an SNT FAQ. Unlike most substacks, the FAQ will be updated as a matter of course. I will use Notes to notify the choir of any material changes. You can use Comments to suggest additional questions. Please be precise. Do not use dose when you mean dose rate nor dose rate when you mean dose rate profile. People experience neither a dose nor a dose rate. They experience a dose rate profile.
Where did SNT come from?
AFAIK, your preacher. I started out fitting an S-curve (logistic) to the Radiation Effects Research Foundation(RERF) atom bomb survivor cancer data, since this acute dose response was clearly S-shaped. Fitting a logistic curve to dose response data was not original at all. Logistic dose responses are near standard in many dose response studies; but to my knowledge the RERF had not tried such a fit to their data. I found I needed the 5-parameter logistic to get a decent fit, since the lower hook of the S was clearly much smaller than the upper.
But then I realized this did not solve the basic acute vs chronic issue. Out of ideas, I tried dividing the dose rate profile into repair periods and applying the S-shaped acute dose response to each repair period independently, simply because it was the simplest way to work around the problem. This worked far better than I expected.
The obvious repair period to start with was a day, which is an upper bound on the actual cell repair process times and a convenient measuring interval; but it took me a while to get there.
Where does SNT fail to predict the response seen by actual populations exposed to low dose rate profiles?
Simple SNT based on fixed repair period of a day is clearly too conservative. This shows up most obviously with the dial painters. So far, I have not run across a case where simple SNT under-predicts cancer incidence. But whenever the total dose is spread over a protracted period SNT is closer to reality than LNT by many orders of magnitude.
It is possible to make SNT more realistic by assuming a dose rate dependent repair period and a lower infinite acute dose incidence. But so far I've elected to stay with the simplest version, and accept the uncomfortably high conservatism. For policy purposes, it is good enough. For example, the EPA MIR (Maximum Individual Risk) dose rate limits go up by a factor of 300 to reasonable numbers.
One can make a weak argument that SNT's conservatism is required to cover all the possible uncertainties in dose rate profile, actual cancer mechanisms, and individual radiosensitivities.
Does SNT account for DNA repair?
Only indirectly. SNT claims that spikey dose rate profiles are far more harmful than flat profiles with the same total dose, which can only be the case if we have dose rate dependent DNA repair processes. SNT does not model the repair process directly, but the key input parameter is the assumed repair period.
Does SNT have a following in the scientific, toxicological, biological communities?
No. Most of these people have a vested interest in LNT. My attempt to present an SNT paper at the ANS Special Conference on LNT was rejected.
The few retirees who have had anti-LNT epiphanies argue for hormesis or at least a threshold. But this is qualitative handwaving, when we must have a fully defined, quantitative replacement for LNT, meaning a model which can convert any dose rate profile into a prediction of cancer incidence as LNT can. Almost all these people are scientists, who tend to be purists.
For them, SNT is not descriptive enough. It's an empirical kluge like the way we engineers handle cyclic metal fatigue and high temperature creep. We don't fully understand either the fatigue or the creep mechanisms. The lab data exhibits order of magnitude scatter, in part due to the dependence on minute flaws. If you sent steel from the same ladle to two different labs and they came back with the same creep or fatigue numbers, you'd assume the labs had colluded.
But with the help of a little statistics and some conservatism, we can construct quantitative models of fatigue and creep which allow us to build massive machinery subject to cyclic, high temperature loads, with tolerable results. It's not science; it's engineering. They are right; SNT is such a kluge.
Some of them believe that we should not even try to replace LNT with a fully defined model. The biology is too complex and is too different for each individual. They do not understand that without a fully defined radiation harm model regulation will be haphazard and subject to political whim. They do not understand that without a fully defined model, we will not be able to develop a radiation exposure compensation scheme, which depends only on each individual's dose rate profile. Radiation releases will continue to be subject to the American tort system. When we have a Fukushima sized release in the USA, we can expect close to a trillion dollars in claims. Nuclear plant liability will remain uninsurable. They do not understand that, without a quantitative replacement for LNT, LNT will not be replaced.
Last Energy is filling up their order book. They already have about 80 orders for their 20MW plants under power purchase agreements. That number of orders allows them to keep a supply chain operating. It is also the equivalent of an AP1000 ++. I am super interested to see if they can successfully build the units. They state their goal is to build thousands a year. None of these will be in the USA as long as LNT is the reigning paradigm. Chris Wright would be very open to your presentation that I suggested below your send to EPA. The DOE can sponsor research into your presentation.
Jack, I think the most important thing you said in this article is that SNT is an engineering solution - not a biological solution. It allows us to build safe and reliable Nuclear Power Plants without constantly revising the biology. This is the advantage of the over conservative results. The engineers can get about their business and the biologists can continue to explore the exact nature of the repair mechanisms. The biological research will NOT constantly revise the engineering. As you point out LNT - by its nature - does not allow AN engineering solution. Each engineering solution is flawed by definition under LNT. SNT allows engineering solutions that can be confirmed as "SAFE." LNT cannot use the term "SAFE" no matter the solution. As long as the regulator, or insurance adjuster cannot stamp "SAFE" on a NPP we cannot build them in the USA. This is why, even with large subsidies, we don't have order books filled with new NPP builds. It is the future liability risk, which under LNT is unlimited, that makes a utility, of any size, hesitant. If the risk of exposure were known and limited, it can be accounted for, costs calculated and cost per KWH or MWH known. I think you should approach the new leaders at the EPA directly with your proposal. The approach needs to be in a Power Point or graphic format with the benefits and weaknesses of LNT, and the benefits and weakness of SNT clearly laid out. This specific post is a fantastic outline to start with. The presentation needs to be at a lay level, about 9th grade HS. Your fantastic back notes and supporting material will supply the technical depth. The simple presentation will either get them searching your notes or will go in the trash. The presentation should emphasize the difference between an engineering solution and a biological solution, and should clearly justify changing the dosage exposure by an INCREASE of 3000 times. The FAA finally dropped the regulation of cell phones in airplanes because the science just did not justify it. Individual airlines could do what they wished but they could no longer use FAA as a cover for their desire to keep people from using their phones in flight.