Thursday, April 12, 2007

previous entry | main | next entry | TrackBack (0)

Do not freak out about Iran's "industrial" nuclear program

In TNR Online, Michael Levi explains why Iran's claim of having an "industrial" enrichment program is a crock:

[Iran's] progress is actually much less than meets the eye. It has developed nothing remotely resembling an industrial capacity to enrich uranium, nor is there any evidence that it has made surprising new strides toward a nuclear weapon. And taking the Iranian claims at face value would be worse than error; it would be a strategic miscalculation that could help entrench the Iranian nuclear program and make it even more difficult to oppose.

According to the "Iran Dossier" prepared by the International Institute for Strategic Studies, 3,000 first-generation Iranian centrifuges operating perfectly for approximately one year could produce enough fissile material to fuel one nuclear bomb. That makes the Iranian announcement sound pretty scary. But it's far from clear that Iran can come anywhere close to perfection in operating its machines. David Albright recently estimated, based on data published by the International Atomic Energy Agency (IAEA), that the centrifuges in Iran's 164-machine cascade were operating roughly 20 percent of the time. If the new 3,000-centerfuge plant functions at that level, it would take five years for it to produce enough material for a bomb. There is, of course, an outside chance that Iran has made immense technical leaps in recent years; but we shouldn't let worst-case fears that lack hard evidence dominate our policymaking.

Moreover, as Jeffrey Lewis has noted, Iran has so far used less than one ton of uranium hexafluoride, the form of uranium used in a centrifuge plant. That number has special significance. Iran bought what experts call "hex" from China back in 1991--one ton's worth, enough for the work Tehran has completed so far. But Iran's homemade hex is thought to be of poor quality: If the Iranians fed it into their centrifuges, the machines could break down. So, if Iran has used only Chinese uranium to date, even its shaky performance so far may overstate its capabilities, since, according to my calculations, it would need at least seven tons to make a bomb. It's possible Tehran has acquired more high-quality hex elsewhere, but IAEA investigations suggest that this is unlikely.

Nor would the Iranian facility be industrial scale even if it were functioning perfectly. Common sense demands that an industrial-scale enrichment plant be able to support a nuclear industry. A simple estimate, though, shows that the new facility would take roughly ten years to produce the fuel needed to operate Iran's single nuclear power plant for one year. If the Iranian facility is industrial scale, then my kitchen is a bakery.

posted by Dan on 04.12.07 at 12:38 AM


I'm sure it's all fine, then. After all, everyone knows that Pearl Harbour is too shallow to be vulnerable to attack by torpedo planes.

Carry on, then.

posted by: Philo-Junius on 04.12.07 at 12:38 AM [permalink]

You failed to take into account the learning curve. And the fact that a miscalculation would be irreversible and terminal.

posted by: jaim klein on 04.12.07 at 12:38 AM [permalink]

I think its always wisest to assume you know everything. Particularly when all we hear about it how dispersed and secretive Iran's nuclear research facilities are.

posted by: Mark Buehner on 04.12.07 at 12:38 AM [permalink]

I agree with the above 3 comments, but with slightly less sarcasm. The serious problem is that how do we prevent a false negative prediction (eg, Pakistan) with 100% certainty while minimizing the false positives (eg, Iraq)and hopefully not pissing off the Europeans too much. Yes, I agree that Bush has made this job much more difficult, but knowing who to blame doesn't make the problem easier to solve.

posted by: ArmchairEconomist on 04.12.07 at 12:38 AM [permalink]

The Koran tells Moslems that it is okay to lie, cheat, steal and backstab a non-believer if it advances Islam.

Any non-Moslem who negotiates with a Moslem should understand that. And if one doesn't, or one chooses not to believe that, one deserves what one gets...

I do not hate Moslems. I just expect that they will follow their religion. And Mohammed told them to kill, convert, or enslave all non-Moslems...

posted by: jtb on 04.12.07 at 12:38 AM [permalink]

I went and read the posting at and did not come away with a warm fuzzy happy happy joy joy.

Key line at end of Lewis' post:

Anyway, I donít have any inside information, just a guess.

Even more disturbing was a rather well put together post further down by 'yale', including a very impressive table. Some nut paras:

All parties are working from flyspecks of data. In the past, engaging in augury by observing the roster and relative physical positions of Soviet officials on top of the Kremlin Wall on May Day was considered state-of-the-art. Now we are reduced to determining how soon Iran may have the Bomb by mass extrapolation of extracted partial sentences from IAEA officials.
It is useful to compare the current situation with the US Bomb program in April Ď44. Every necessary technology was in disaster-mode, with production almost non-existent. Yet, over the months everything fell into place, and by April Ď45, Japan was doomed.

Substitute Israel for Japan, if you will.

As my mom used to say, "It's all fun and games until one of you puts an eye/city out".

posted by: jdwill on 04.12.07 at 12:38 AM [permalink]

What i don't get is the criticism of employing "worst-case scenariors." If I can think of one uncertain problem where preparing for a worst-case scenario makes sense, it's the mullahs getting a nuke.

posted by: srp on 04.12.07 at 12:38 AM [permalink]

Examining worst-case scenarios is an important first step in any study of nuclear proliferation problems, but the all-important next step is assigning a reasonable probability of their occurrence.

The trouble with our impression driven thinking is that we do not take the final step of assigning a confidence interval that tells us how sure we are that we know what we know.

In the case of Iran, the worst-case scenario is a nuclear conflict in the Middle East with Israel. The probability of this occurring is quite low, perhaps only 5% over the next few years. But the confidence interval of this prediction is not high. We don't have a high degree of certainty with regard to our prediction.

So what should policy-makers do?

posted by: Erasmus on 04.12.07 at 12:38 AM [permalink]

Looking at the proliferation issue from this standpoint seems to conflate the risk of Iran becoming a major nuclear power capable of producing enough bombs and rockets to threaten the region (or the rest of the world) militarily with the risk of the proliferation of nuclear material to terrorists. The latter risk does not require huge amounts of highly radioactive material. It does not even require a nuclear bomb. So, how much time do we really have? Why am I not relieved?

posted by: Karl B. on 04.12.07 at 12:38 AM [permalink]

I always find it interesting that the crowd that says "we can't bomb Iran, we don't know where all their stuff is" then goes on to confidently make prdictions about timelines, production rates, etc.

posted by: Jean Arrache on 04.12.07 at 12:38 AM [permalink]

It's good to know you don't hate Muslims, but only believe they are a people who "lie, cheat, steal and backstab."

posted by: Comforted on 04.12.07 at 12:38 AM [permalink]

Let me remind you that when Saddam was separating uranium, he was doing it with a giant mass spectrometer or "Calutron", which is exactly how the uranium for the Hiroshima bomb was separated. If it worked for Saddam, why not Iran? This consumes massive amounts of electricity (leading to an ancient joke that we didn't drop an atomic bomb on Hiroshima, we dropped the TVA), however I doubt that would bother them. Saddam hid his facilities quite easily. Somehow this never gets discussed with respect to Iran.

posted by: JRM on 04.12.07 at 12:38 AM [permalink]

Post a Comment:


Email Address:



Remember your info?