In the summer of 1967 I worked on dispersal of "air pollution" plumes and got curious about the concentrations measured in "dosages", so I looked at the original papers -- they were all Declassified but it was pretty clear who wanted to know how well "pollution" plumes could be predicted, and why. The good news is, we found that the answer to the real question was, "BAD IDEA!" -- you can't predict worth a damn.
I don't know what models you were looking at in 1967. Were Lagrangian models even around then? The models have come along way in the last 60 years. HYSPLIT has spent a lot of effort on the interaction between clouds and aerosols, and there has been a great deal of validation work.
It's not an area I know a great deal about; but I have talked to a world famous meteorologist and he feels the capability is needed and useful. Of course, we are dealing with a chaotic system, so the accuracy of the predictions will fall off pretty quickly. But in this case all we need is a day or two to be extremely useful.
I would also point out, that such a model is needed to do pre-attack risk evaluations, running the model against hypothetical weather patterns to get insight on the risks involved, and develop intelligent response policies. Right now we are flying blind.
Dear Jack: Based on the analysis of M. Herschel Specter, the likelihood of any plume event from an attack on al-Barakah is extremely low. Central to his analysis are the extreme safety margins that have been built into modern nuclear power reactors with large dry containments. We are *not* talking about Chernobyl, which lacked a containment and generated a radioactive plume.
Please see Herschel's 2019 analysis "Become a Nuclear Safety Expert, Rev. 2" at https://www.nuceng.ca/refer/Specter-2019.pdf. The physics of modern light water reactors argue against plume dispersal because the radioactive material of concern tends to "plate out" while inside the containment, based on Sandia Labs analysis.
This plating out behavior of radioisotopes was observed at both Three Mile Island in 1979 and Fukushima dai-ichi in 2011. Nuclear fuel is a dense ceramic-like solid. Its density is twice that of lead.
Obviously, it depends on the weapon. The USA GBU-57 can penetrate 18 meters of concrete. It will bust a big hole in just about anything. And a sophisticated attacker can use sequences of these things as we did in the earlier attack. Such an attack will result in a Chernobyl sized plume, but probably more spread out which is a good thing. We need the capability I asked for, if only to protect ourselves against an adversary which has essentially the same capability that we have.
I don't know what weapons the Iranians have. But nuclear plants have other vulnerabilities. For light water reactors these include the spent fuel pool and a station black out producigna Fukushimalike situation
The SOARCA study that Specter cited was contradicted before it got off the press.
Yes most of the cesium willplate out ina TMI like casualty, but there are other ways of releasing iodine and cesium as Fukushima proved. The best defense in a SBO is early and frequent venting. Kind a nice to have an idea of where the plume will go and how much damage it will do. Likke I said, right now we are flying b,ind.
To Ken's comment, without any guidance at all, people will assume the worse and panic. A good plume model combined with SNT will make the point that even a Chernobyl is can be handled.
Jack: For an Iranian weapon to be able to harm al-Barakah, there must be a delivery means and a warhead. Iran's air force has been essentially eliminated. The warheads on Iranian missiles can damage apartment buildings and flammable fossil fuel infrastructure. Their effectiveness against a hardened target like al-Barakah is questionable.
We are veering off topic. Iran may or may not be able to produce a big release at Barakah. The spent fuel pools are much more exposed than the reactor itself and I'm pretty sure they use dense packing. Pls see
But the Russians almost certainly have that capability in Ukraine.
But the issue on the table is: do we need a good plume model and a good radiation harm model so that we both prepare for such a release (or any release) and intelligently react if one happens?
The problem is NRC's plume program MACC2 is a simple Gaussian
plume model. It can't model real weather pattern and can only produce completely misleading dose rates. Worse, the NRC uses LNT to convert the resulting does rate profiles to cancer incidence. The combined result is panicked evacuations which nearly always will cause far more harm than the radiation.
How thick is the concrete containment? Could a missile penetrate it?
How thick is the steel on their reactor vessels? Would a missile break it open? How much radioactive material could they push out of the building? If it is just steam from a rapid decompression, is that a worry?
Lacking a good plume model, could they just assume worst case (Fukushima) and tell everyone to close their windows and let it blow over. Looks like nothing but dessert for 20 miles in any direction.
If the Iranians attack, they could go for economic damage, not radiation dispersal. The damage could take years to repair.
As always the fear of radiation is more important than the actual risk. I agree with some comments below that releases from a missile strike are unlikely. The problem is if someone actually tries to do a conservative analysis and the public hears of it.
BTW, I worked on air pollution models decades ago. We stored actual weather data so that we could run our Atmospheric Boundary Layer model with real weather data.
Based on my experience, modern containment buildings are the most robust structures in the world. 3-4 feet of reinforced concrete with a rebar system that was designed to withstand an internal steam explosion and contain about 50 psig of pressure with minimal leakage.
Obviously a bunker buster would likely penetrate it.
But what about damaging the reactor coolant system and pressure vessel? Another layer of protective steel.,
Worst case scenario, we have a LOCA with a breach of containment The containment spray CS system would probably still function unless there were multiple missiles.
I believe that the UAE was required to install post Fukushima FLEX equipment. Portable equipment designed to keep the core cool and minimize the spread of contamination.
That’s definitely a beyond Design basis accident.
I hope they’re operators are as well trained as ours
But, as discussed, the fear and widespread panic, resulting from such an event would be quite damaging to the nuclear industry.
Jack is correct, the actual radiological impact and health effects would be minimal.
Prayers aren't enough. The probability of the next big release is 1.00. It is only a question of when.
Parts of this thread have gotten dangerously close to the Negligible Probability Lie
May I remind everybody of the Two Lies that killed nuclear power.
The Intolerable Harm Lie: Any big release of radioactive material would be so catastrophic that it cannot be allowed to happen.
The Negligible Probability Lie: The probability of a big release of radioactive material from a nuclear power plant is so low that we can just assume it won't happen.
The Intolerable Harm Lie begat the Negligible Probability Lie. We can't achieve zero big releases, and attempting to do the impossible has driven nuclear's does-cost to ten or more times its can, did and should-cost, to the tragic detriment of humanity. We would be far better off with more releases and should-cost nuclear, than without both,
But that does not mean we can just assume the harm from a big release will be "minimal" or "not detectable". We must do an efficient job of preparing for the next big release and reacting to it. That requires a good plume model and a radiation harm model that is consistent with undisputed biology.
Is that too much to ask from a nuclear establishment that eats up scores of billions of dollars per year of taxpayer money with little or nothing to show for it?
On March 22, 2026, Erik Townshend's Substack published a detailed article, "Worst-Case Scenarios if Nuclear Power Plants are Targeted in the Iran Conflict - Sobering thoughts on what's possible if involved governments resort to the unthinkable... "
You are making my point. The only way to counter vague claims of "catastrophic consequences" is to have the capability the GKG is asking for. But if tha harm model used is LNT, the results will be catastrophic and totally misleading.
In the summer of 1967 I worked on dispersal of "air pollution" plumes and got curious about the concentrations measured in "dosages", so I looked at the original papers -- they were all Declassified but it was pretty clear who wanted to know how well "pollution" plumes could be predicted, and why. The good news is, we found that the answer to the real question was, "BAD IDEA!" -- you can't predict worth a damn.
Jess,
I don't know what models you were looking at in 1967. Were Lagrangian models even around then? The models have come along way in the last 60 years. HYSPLIT has spent a lot of effort on the interaction between clouds and aerosols, and there has been a great deal of validation work.
It's not an area I know a great deal about; but I have talked to a world famous meteorologist and he feels the capability is needed and useful. Of course, we are dealing with a chaotic system, so the accuracy of the predictions will fall off pretty quickly. But in this case all we need is a day or two to be extremely useful.
I would also point out, that such a model is needed to do pre-attack risk evaluations, running the model against hypothetical weather patterns to get insight on the risks involved, and develop intelligent response policies. Right now we are flying blind.
Dear Jack: Based on the analysis of M. Herschel Specter, the likelihood of any plume event from an attack on al-Barakah is extremely low. Central to his analysis are the extreme safety margins that have been built into modern nuclear power reactors with large dry containments. We are *not* talking about Chernobyl, which lacked a containment and generated a radioactive plume.
Please see Herschel's 2019 analysis "Become a Nuclear Safety Expert, Rev. 2" at https://www.nuceng.ca/refer/Specter-2019.pdf. The physics of modern light water reactors argue against plume dispersal because the radioactive material of concern tends to "plate out" while inside the containment, based on Sandia Labs analysis.
This plating out behavior of radioisotopes was observed at both Three Mile Island in 1979 and Fukushima dai-ichi in 2011. Nuclear fuel is a dense ceramic-like solid. Its density is twice that of lead.
All,
Obviously, it depends on the weapon. The USA GBU-57 can penetrate 18 meters of concrete. It will bust a big hole in just about anything. And a sophisticated attacker can use sequences of these things as we did in the earlier attack. Such an attack will result in a Chernobyl sized plume, but probably more spread out which is a good thing. We need the capability I asked for, if only to protect ourselves against an adversary which has essentially the same capability that we have.
I don't know what weapons the Iranians have. But nuclear plants have other vulnerabilities. For light water reactors these include the spent fuel pool and a station black out producigna Fukushimalike situation
The SOARCA study that Specter cited was contradicted before it got off the press.
Yes most of the cesium willplate out ina TMI like casualty, but there are other ways of releasing iodine and cesium as Fukushima proved. The best defense in a SBO is early and frequent venting. Kind a nice to have an idea of where the plume will go and how much damage it will do. Likke I said, right now we are flying b,ind.
To Ken's comment, without any guidance at all, people will assume the worse and panic. A good plume model combined with SNT will make the point that even a Chernobyl is can be handled.
Jack: For an Iranian weapon to be able to harm al-Barakah, there must be a delivery means and a warhead. Iran's air force has been essentially eliminated. The warheads on Iranian missiles can damage apartment buildings and flammable fossil fuel infrastructure. Their effectiveness against a hardened target like al-Barakah is questionable.
Gene,
We are veering off topic. Iran may or may not be able to produce a big release at Barakah. The spent fuel pools are much more exposed than the reactor itself and I'm pretty sure they use dense packing. Pls see
https://jackdevanney.substack.com/p/the-dense-packing-fiasco
But the Russians almost certainly have that capability in Ukraine.
But the issue on the table is: do we need a good plume model and a good radiation harm model so that we both prepare for such a release (or any release) and intelligently react if one happens?
Plume modelling is already mature and well demonstrated for wildfires and refineries.
Bunker bombs will not achieve atomic scale dispersion of heavy elements.
We know from tbe detailed Cernobyl stieds that their are no detectable cancer signals in the communites under the plume or ecological signals.
I accept fears from films and media outlers are difficult to overcome.
Thank you for your summary of my comments, Adrian.
Adrian,
The problem is NRC's plume program MACC2 is a simple Gaussian
plume model. It can't model real weather pattern and can only produce completely misleading dose rates. Worse, the NRC uses LNT to convert the resulting does rate profiles to cancer incidence. The combined result is panicked evacuations which nearly always will cause far more harm than the radiation.
We must correct that situation.
How thick is the concrete containment? Could a missile penetrate it?
How thick is the steel on their reactor vessels? Would a missile break it open? How much radioactive material could they push out of the building? If it is just steam from a rapid decompression, is that a worry?
Lacking a good plume model, could they just assume worst case (Fukushima) and tell everyone to close their windows and let it blow over. Looks like nothing but dessert for 20 miles in any direction.
If the Iranians attack, they could go for economic damage, not radiation dispersal. The damage could take years to repair.
As always the fear of radiation is more important than the actual risk. I agree with some comments below that releases from a missile strike are unlikely. The problem is if someone actually tries to do a conservative analysis and the public hears of it.
BTW, I worked on air pollution models decades ago. We stored actual weather data so that we could run our Atmospheric Boundary Layer model with real weather data.
Good discussion
Based on my experience, modern containment buildings are the most robust structures in the world. 3-4 feet of reinforced concrete with a rebar system that was designed to withstand an internal steam explosion and contain about 50 psig of pressure with minimal leakage.
Obviously a bunker buster would likely penetrate it.
But what about damaging the reactor coolant system and pressure vessel? Another layer of protective steel.,
Worst case scenario, we have a LOCA with a breach of containment The containment spray CS system would probably still function unless there were multiple missiles.
I believe that the UAE was required to install post Fukushima FLEX equipment. Portable equipment designed to keep the core cool and minimize the spread of contamination.
That’s definitely a beyond Design basis accident.
I hope they’re operators are as well trained as ours
But, as discussed, the fear and widespread panic, resulting from such an event would be quite damaging to the nuclear industry.
Jack is correct, the actual radiological impact and health effects would be minimal.
Let’s just pray it doesn’t happen.
Ken,
Prayers aren't enough. The probability of the next big release is 1.00. It is only a question of when.
Parts of this thread have gotten dangerously close to the Negligible Probability Lie
May I remind everybody of the Two Lies that killed nuclear power.
The Intolerable Harm Lie: Any big release of radioactive material would be so catastrophic that it cannot be allowed to happen.
The Negligible Probability Lie: The probability of a big release of radioactive material from a nuclear power plant is so low that we can just assume it won't happen.
The Intolerable Harm Lie begat the Negligible Probability Lie. We can't achieve zero big releases, and attempting to do the impossible has driven nuclear's does-cost to ten or more times its can, did and should-cost, to the tragic detriment of humanity. We would be far better off with more releases and should-cost nuclear, than without both,
But that does not mean we can just assume the harm from a big release will be "minimal" or "not detectable". We must do an efficient job of preparing for the next big release and reacting to it. That requires a good plume model and a radiation harm model that is consistent with undisputed biology.
Is that too much to ask from a nuclear establishment that eats up scores of billions of dollars per year of taxpayer money with little or nothing to show for it?
Jack,
Prayers never hurt.
I believe in prayer.
I pray every day! 🙏
But I also fully support science and having a good plume model and plans.
Diablo Canyon Power Plant DCPP, excelled in those two area.
On March 22, 2026, Erik Townshend's Substack published a detailed article, "Worst-Case Scenarios if Nuclear Power Plants are Targeted in the Iran Conflict - Sobering thoughts on what's possible if involved governments resort to the unthinkable... "
https://eriktownsend.substack.com/p/worst-case-scenarios-if-nuclear-power
Erik's article explores some very chilling scenarios.
Gene,
You are making my point. The only way to counter vague claims of "catastrophic consequences" is to have the capability the GKG is asking for. But if tha harm model used is LNT, the results will be catastrophic and totally misleading.
I agree with your conclusion regarding LNT.