Geneva blogger appear

John Borrie at UNIDIR has started a new blog (on blogspot), 'aimed at negotiators, policy wonks, activists, researchers and anyone curious about disarmament and human security'. Welcome to the blogging community, John.

North Korea’s test site

I'm posting a placemark of North Korea's suspected test site, and I have highlighted bits of the road network, as well as the location of the nearest train stations, and airstrips. I find it intriguing that the area has no less than two mainline stations, less than an hour's car-ride from the suspected site, especially since there is relatively little noticeable mining activity in the near area. I've also added the USGS, IDC and IPC ground zero coordinates.

The area itself is oddly laid out, if one compares it with the Pakistan's Ros Kah range. For instance, the support buildings is sited only some 400 meters from the suspected tunnel entrance, whereas the Pakistani test director had placed his observation point at a 10 kilometer distance. The North Korean support buildings would have been in danger of serious structural damage, if the aim was to set off a 20 kiloton device. Indeed, they could have suffered some damage after the actual half-kiloton explosion.

The test tunnel may have been horizontal. Pakistan opted to excavate a one kilometer long tunnel, with several "zero rooms". At its furthest point, the tunnel had approximately 700 meters of mountain above it. Pakistan's site had been selected and prepared in the late 1970s, and then sealed up for potential future use. The site itself contained 24 cold test sites, 46 short tunnels and 35 underground accommodations for troops and command, control and monitoring. The mountain was "dry as a bone". We know little of ground water conditions at this particular site, although the mountain itself is predominantly granite.

There are rumours about a second test site on the western side of the mountain. However, there seem to be little in terms of suitable roads or infrastructure on that side, suggesting that a second site is sited closer to the first. However, it must be at some distance, to safeguard its structural integrity from the shock of the first test.

Warm standby redux?

We may perhaps see something come out from tomorrow’s Larijani – Solana meeting after all. A couple of hours ago, George Jahn reported that the “United States and other world powers … could try to sidestep the deadlock over enrichment by trying to agree on a new definition of the term”. Well, okay…

There is no agreed definition of what constitutes “enrichment”, or, for that matter, “enrichment related” activities. The Paris Agreement collapsed because the Iranians and the EU-3 had different views on this. The disagreement back then was if the resumption of activities at the uranium conversion facility in Esfahan fell under the term. The Europeans tried to point out that uranium conversion was mentioned in the agreement, but to no avail. What Jahn probably stumbled upon is an idea to agree on a definition of “uranium enrichment related activities”.

Maybe the idea of a warm stand-by will resurface? Jeffrey Lewis has written quite a bit about this. If the EU and Iran moves in this direction, we may actually see the end of the present trench warfare.

And what could possibly illustrate what it means to put the centrifuges in standby better than the Robot Economist's old photoshop magic?

The mother of centrifuge proliferation... almost

I must comment on these new items from Russia. According to Russian Deputy Prime Minister Sergei Ivanov, the Federation is developing new ninth generation centrifuges. This has been known for some while. The Gen-9 centrifuge is Russia's first supercritical rotor design and, according to Oleg Bukharin, the country plans to it in operation around 2010. So, this isn't really that interesting...

However, Mr. Alexander Samordosky, the Chief Designer at Kovrov Mechanical Works (KEMP), seemed to think that this meant that older generation centrifuges could be sold wholesale. Interfax quoted him as saying that sixth generation centrifuges "are virtually a passed stage and we're not disclosing any secrets here".

Sergei Ivanonv quickly modified the director's statement, saying that "exports to Iran are completely ruled out".

What a relief, we may be spared the "A. Q. Samordosky" network after all.

VERTIC forthcoming publications

Thought I would make Jane, VERTIC's networker, happy, and promote my employer on the blog. Need to say this: This does not mean that the blog is affiliated with VERTIC or any other organization. My views, my responsibility.

For a year, VERTIC has had a project on Iran. The lead researchers have been James Acton, our science and technology researcher, and yours truly. We've been collecting information, polishing our papers, collecting information, had our papers reviewed, polished, and reviewed again. Today, I actually laid down my pen and sent my humble contribution to the subeditor. We are talking about two "verification matters" reports.

My report (VM7) is entitled "Thinking inside the box: exploring legal approaches to build confidence in Iran's nuclear programme". It has been reviewed by one premier thinker on CBM's, one high flying professor in international law, and no less than two senior legal officers working somewhere within the UN system. It's the first briefing paper I've written that contains 20 pages of analysis and 10 pages of footnotes. Literally. The lawyers may like this paper. Well, hopefully, since I am bidding for membership in the International Law Association's new Committee on International Arms Control Law.

James's report (VM8) is entitled "The use of voluntary transparency measures to increase trust in states' nuclear programmes: the case of Iran" and it is absolutely massive. It contains all the information one would need on Iran in one document - from complete fuel cycle analysis to pages after pages of complex equations. This paper was reviewed by, I would argue, the very finest minds within safeguards and verification (they're also great persons), with decades of experience from both international safeguards authorities and, of course, national authorities.

Together, the two reports contain nearly 100 pages of information and analysis. Some of this information will be discussed at a joint VERTIC - BASIC seminar held during the NPT PrepCom. Anyone who'd like to come is welcome. It will be chaired by Dutch Ambassador Johannes Landman.

I may miss this project now when its almost over... nah, probably not...

NB. I wish to register that, as a former economist (I did that before law and I surely do not miss the equations), I objected against the term "stable equilibrium" in the title. All economists know that equilibrium is a state where supply meets demand, and where the market is at rest. In other words, the equilibrium is always stable. But my humble views were duly noted, and dismissed.

KRL

I have been staring at this place for way too long now. Welcome to Dr. A.Q. Khan Research Laboratories. Note that I have not bothered to annotate the part of the centre which is suspected to play a part in the development of Pakistan's missile program.

The 3,000 centrifuge fairytale (continued)

As pointed out by both Jeffery Lewis and Paul Kerr, Iran has made fairly limited progress towards industrial scale uranium enrichment. According to an Agency letter published by ISIS, IAEA inspectors visited the Fuel Enrichment Facility in Natanz on 15 and 16 April 2007. During this visit, they were informed by the Iranian operator that it had installed, and is operating eight cascades at the plant. Note that this information does not seem to have been independently verified by the inspectors. Eight cascades equal 1,312 machines.

A week ago, I guessed that Iran could install approximately 5o machines per week. Now it is possible to verify how that guesstimate held up. 58 days elapsed between 17 February (the date of the previous inspection) and 16 April 2007. In those 58 days, Iran seems to have installed 656 centrifuges (or four cascades). In other words, Iran has installed 11 centrifuges per day (or 79 centrifuges per week). If they keep up the pace, the operator will have installed a full module (2,952 machines) by early September 2007.

However, if the people of Natanz manage to pick up the pace, which isn't completely unreasonable, they may finish the first module by mid or early August. I doubt, however, that they will be able to run the module smoothly and without incidents by that date. The Iranian technicians, lacking any experience in running a large number of interconnected cascades, are likely to learn by doing, and by failing.

Of greater concern is the contempt Iran shows the International Atomic Energy Agency. One might recall that the Agency informed Iran that the quick fix of "frequent inspector access" would be in force only as long as the number of centrifuges in the FEP did not exceed 500 (GOV/2007/8, para 8). After that, Iran would need to allow for remote monitoring. This does not seem to have happened.

Also, Iran seems to have cut the seal on the feed autoclave without any inspectors present. Depending on the type of seal applied, this may or may not be serious (if the Agency had applied an electronic seal, it would have registered the date, time of duration of it being opened, allowing for some reconstruction of events).

For those interested in a viewpoint on the effectiveness of seals, I recommend this critical (and quite amusing) article in the Non-Proliferation Review, written by Roger Johnson.

Enter AtomEnergoProm

I cannot say that I've followed the negotiation of Russia's 123 agreement with the United States that closely, but there are exiciting developments going on in Russia. Those interested should check this post at the ACW and have a look at this issue-brief, produced by the Monterey Institute of International Studies.

One important aspect of these agreements is that the "recipient government [pledge] that material and equipment transferred under the agreement will be used exclusively for peaceful purposes". One way of effectuating this pledge is to separate the military and the civilian fuel cycle. Consequently, this is obviously a very significant item from World Nuclear News (my emphasis below).

On 17 April a decree to create the vertically integrated giant AtomEnergoProm (AEP) was been approved by Russian President Vladimir Putin, following a gathering of senior nuclear industry figures at his Novo-Ogarevo residence. The new firm could operate by the end of this year. As Russia gains acceptance as a market economy, its nuclear industry remains a significant resource not yet fully market oriented. Putin seeks to solve this through the creation of AEP as the amalgamation of the country's most significant nuclear enterprises, principally:

- AtomEnergoMash (AEM), nuclear power plant construction.
- AtomStroyExport (ASE), nuclear power plant construction abroad.
- Rosenergoatom, nuclear power operations.
- Technabsexport (Tenex), uranium extraction and enrichment.
- TVEL, uranium extraction and nuclear fuel production.

Apart from Rosenergoatom, all the above are Joint Stock Companies (JSCs), which means they operate as a market oriented company. The state owns 100% of the shares of all but ASE, of which it owns over 50% through various means. Before inclusion in AEP, Rosenergoatom will be corporatized.

Alexey Grigoriev of Tenex explained to World Nuclear News that AEP will be an integrated vertical holding company for the various 'daughters', the top level of which would be the companies listed above. At the level below would be the daughters' individual subsidiaries, which over time will also be converted to JSC status. AEP would also be a JSC rather than a traditional state-owned enterprise. This distinction, Grigoriev said, would be very strongly written in the decree.

Under this arrangement investment should be more attractive to other Russian businesses as well as those abroad, and officials hope this will help spur AEP to achieve Putin's targets for nuclear's share of electricity to grow to 25% or 30% by 2030.

The head company will fully belong to the state, as well as its 'daughters' and 'grand-daughters'. "For example, private investors may well be involved in uranium extraction," said Sergei Kiriyenko, head of the Federal Atomic Energy Agency (Rosatom). He added that the first example of a private company cooperating profitably with an AEP daughter is the joint venture deal between AEM and Alstom to supply the non-nuclear turbine generator part of future nuclear plants.

Legislation was changed in February to allow certain entities other that the Russian state to own nuclear materials. One of the next steps in the development of AEP will be the compilation of a list of companies able to own those materials for civilian nuclear power purposes. Another thing very strongly written in the decree is that the President would have control over the list.

Much work lies ahead for officials in checking the completeness, fairness and necessity of companies listed, as this list will be a major part of a simultaneous change occurring in the Russian nuclear industry: The clear separation of the civilian nuclear power sector and the military defence complex.

Within a few days, Grigoriev said, executives could be named for the top positions in AEP and by the end of the year, or early in 2008, the state should have completed paperwork to allow AEP to "work in full swing."

The creation of such a giant will certainly mean changes in the global market balance. AEP will have the distinction of encompassing the full nuclear fuel cycle, nuclear design and construction and engineering, and the key business of power generation itself.

Grigoriev concluded that AEP would not have the pretense of ambitions for a very big share of the world's competitive nuclear market, simply an equal position - in line with the other main market players.

Yongbyon Dreams

On 14 April 2007, the deadline to complete initial actions under the so-called Denuclearization Action Plan will expire. To recap, North Korea was supposed to "shut down and seal for the purpose of eventual abandonment the Yongbyon nuclear facility, including the reprocessing facility and invite back IAEA personnel to conduct all necessary monitoring and verifications as agreed between IAEA and the DPRK".

However, the process seems to have crashed, not over an insurmountable technical problem, or over whether North Korea has accounted for its suspected uranium enrichment program. Nope. The problem is that cash is king for the rulers in Pyongyang. Since funds weren't transferred, the DPRK decided not to play ball. And to be fair, the plan calls for "coordinated steps" to be taken along the principle of "action for action". Show us the money, and we'll suspend activities at our nuclear complex.

The expiry of the deadline does not mean that the deal is dead. According to New Mexico Governor Bill Richardson, "The North Korean government … [will] move promptly – within a day after receiving the funds – to invite the International Atomic Energy Agency to Pyongyang to draw up the terms for shutting down the Yongbyon reactor". A glimmer of hope, perhaps, for the people at the International Atomic Energy Agency who are probably sitting in their offices at the Vienna International Centre, eager to get to work.

According to The Korea Times, a new IAEA delegation has been formed to travel to North Korea. This time around, it's reported that the delegation will be headed by the DDG Safeguards, Olli Heinonen. His agenda is rumored to be quite sensible: to discuss inspection procedures and the schedule for an eventual dismantlement of North Korea's nuclear facilities. While these issues may sound mundane and boring, they're essential for the proper implementation of the Action Plan. Presently, there is no comprehensive safeguards agreement to guide the inspectors. North Korea isn't a member of the Agency (it revoked its membership on 13 June 1994). Neither is it a party to the 1968 Nuclear Non-Proliferation Treaty (it announced its withdrawal on 10 January 2003). Consequently, North Korea's safeguards agreement lapsed when the country withdrew from the NPT. It is quite unclear what, if any, impact UN Security Council Resolution 1718, has on this situation. This resolution demanded that
the DPRK return to the NPT and IAEA safeguards, but it also decided that the DPRK should act "strictly in accordance" with the treaty, and with its safeguards agreement. It also contains a broad mandate for the Agency, since the North Korean's are required by law to give "such access to individuals, documentation, equipments and facilities as may be required and deemed necessary" for the Agency to verify compliance with the resolution. Okay… that's what the resolution states, but I would eat my hat (with a lot of salt) if Kim Kye Gwan would be even moderately intimidated by the Agency wagging a Council document in his face. Isn't it quite intriguing, by the way, that the United States, a permanent member of the Security Council, agreed to a Denuclearization Action Plan which leaves verification to be agreed by the Agency and the DPRK? Talk about direct normative conflict. And what does article 103 of the Charter say about those? Hmmm …

Leaving legal issues aside, the inspection protocol must be robust. According to Frenchman Robert Mogavero, the time needed to dismantle of French reactor of comparable type and power as Yongbyon is ten years. In addition, unless they want to slam the SCRAM switch and then just seal the place up, inspectors are looking at a shutdown period of at least a week. Most likely, a couple of inspectors would initially need to be stationed in North Korea.

Oh joy, two nuke guys, living in Spartan accommodations, driving around Yongbyon in a white UN-issued Toyota Jeep. That, if anything, is sacrificing oneself for world peace…

The 3,000 centrifuge fairytale

On 2 April 2007, it was reported that Iran has installed 1,000 centrifuges in the underground Fuel Enrichment Plant at Natanz. Then, yesterday, reports surfaced that Iran has completed an unbelievable Persian variant of shock and awe by installing 3,000 machines! Curiously enough, in the flood of media reporting on the matter, I haven't found one direct quote by any Iranian official confirming this figure. I can see where the figure is coming from, however, since Iran has said for more than a year that they are going to proceed with the installation of one module. In Iranian terminology one "module" comprises 18 cascades of 164 centrifuges (i.e. 2,952 machines). Natanz is dimensioned to be able to hold 18 of these modules, or some 53,000 centrifuges.

Dr. Jeffery Lewis of the Arms Control Wonk called Iran's statement "a stunt" and Paul Kerr held that "Iran almost certainly does not have 3,000 centrifuges installed". Even states seem fairly confused by the information, and Russia's Foreign Minister Sergei Lavrov has said that he is seeking clarification from the International Atomic Energy Agency as to how much kit the Iranians have amassed. France and Australia has joined the line of confused states.

Has Iran installed 3,000 centrifuges? If not, how many might they have in their underground plant? Paul Kerr points out that a "western diplomat" in late March 2007 told Michael Adler that Iran had installed six cascades. Of these cascades, four were "running" (presumably under vacuum). This figure makes sense. On 15 March 2007, David Albright testified that the "Iranians have been installing the necessary infrastructure for the entire module. As of late February, only the 18th cascade was composed only of its concrete foundation and pedestals that will eventually hold the centrifuges. All the other cascade positions had some amount of equipment already installed". If Albright's correct, construction has been steady but slow.

Let's do some calculations of our own. Construction of the module began, as announced, in the final quarter of 2006. One Agency inspection was carried out in the underground facility on 5 November 2006 during which the Agency reported that "construction was ongoing". The IAEA did not mention centrifuges in its report. The next inspection was carried out on 17 February 2007. During this inspection, Agency inspectors were informed that two cascades had been installed and were operating under vacuum and that another two 164-machine cascades were in the final stages of installation. Assuming that the Iranians started installing on 6 November 2006 (which is a pretty big assumption), they managed to install nearly four cascades of 164 machines (656 machines) in 14 weeks, or approximately 50 machines per week.

If the claim that Iran has installed 3,000 centrifuges is true, it would amount to a huge increase in construction activity at the site and a remarkable feat of Iranian project management. Indeed, that would be some 2,300 machines in seven weeks (or two cascades per week)! Not bad. However, if they maintained their previous rate of some 50 machines per week, they should have installed some 1,000 machines by now (i.e. 672 + (50x7)). In light of such a linear analysis, the report that Iran installed six cascades at the site is not unreasonable.

However, expect the numbers to change. After all, learning by doing is the best way to get good at what you're doing. The Iranians may have managed to mount, test and run slightly more machines per week, simple because construction is getting routine. It remains to be seen, however, if the technicians at Natanz manage to bring eighteen cascades on-stream by the end of May 2007. With 50 machines per week, construction would be completed by the end of 2007.

So, what's happening from a safeguards perspective? Well, on 31 January 2007, approximately 8.7 tons of UF6 was transferred to the FEP and connected to the feed autoclave. The autoclave, however, is under Agency seal. If the Iran's pronouncement that it has fed gas into the centrifuges is correct, and the agency weren't informed of that, the Iranians have removed seals without Agency inspectors present, something that will amount to a safeguards violation. Moreover, the operation of more than 500 centrifuges by Iran would amount to quite an insult to the Agency, which rightly insists that remote monitoring must be implemented before more than 500 centrifuges are powered up.

NB. The picture highlights the central part of the new 50,000-rial bank note issued by Iran's central bank. The quote, by the Prophet Mohammed reads, "if the science exists in this constellation, men from Persia will reach it". They are definitely trying.

Cameras and the FEP

In yesterday's edition, Global Security Newswire wrote that, "Iran also has yet to respond to an IAEA request to place additional remote cameras at the Natanz uranium enrichment plant. Enrichment could begin in a matter of days, diplomats said Friday, and the cameras would strengthen the agency's capability to detect any efforts to produce weapon-grade uranium". According to an anonymous diplomat, "centrifuges are in the lowest level of the three-tiered Natanz facility and are blocked from view by a wall. The agency has only been able to mount cameras outside that wall". This would suggest that cameras are mounted at a location outside the FEP centrifuge hall.

It would appear that this arrangement lapsed when Iran started to install centrifuges at the underground FEP. When that happened, "the Agency requested that arrangements be made for the relocation of cameras into the cascade hall during the Agency's next visit to FEP, which is scheduled to take place between 3 and 5 March 2007" (GOV/2007/8, paragraph 9). If the anonymous sources are correct, it would appear that this is more than a question of remote monitoring, which is very interesting news indeed.

More on Natanz and LFUAs

Reportedly, Olli Heinonen, the Finnish deputy director-general of the International Atomic Energy Agency (IAEA), has written a letter to the Iranians, again demanding that the Iranians allow for Agency remote monitoring at their two uranium enrichment plants at Natanz (see an earlier post about this here). Contrary to what's sometimes reported, this has nothing to do with the installation of cameras, but with whether those cameras will be hooked up to secure communications, so that imagery can be transmitted off-site in real time. The agency wants to do this because a cascade area inspection scheme, known as Limited Frequency Unannounced Access (or LFUA), have been deemed unfeasible at the Iranian site.

So, why's that? To answer that question one first has to figure out what LFUA's are. It is all about inspection, we know that, but what is the difference between a LFUA inspection and a routine inspection. Or is it so that a LFU inspection is a variant of a routine inspection? In the report Multilateral Approaches to the Nuclear Fuel Cycle, the Agency elaborates a bit on what makes the LFUA scheme special (see paragraph 142). It is because it secures access to the cascade area "with short notice", without specifying how short that notice is. What are inspectors doing once they get access to the area? Well, the report is not a model of clarity. For instance, activities to be performed and the frequency of access depend on the "design and operation specifics" of the facility. However, activities may include:

• visual observation;
  


• radiation monitoring
  


• non-destructive assay measurements;
  


• environmental sampling; and the
  


• application and verification of seals.
  

So why can't the Agency do this at Natanz? A couple of weeks ago, a reader of this blog pointed out to me that it is "a simple matter of physical location". He then continued,

"at a URENCO plant, you're in countries where inspectors have permanent visas and can show up at the plant gate without the operator having any idea they are coming ahead of time. At Natanz, Iran knows long before the inspectors arrive, just because of visas and suchlike to get into the country, and the time it then takes to get to Natanz. So by the time you get there it's by no means 'unannounced'."

So, why is it important that the access is "unannounced", i.e. that the operator doesn't know that inspectors are coming? Well, it seems to be because the Agency attaches great significance to the deterrent effect that unannounced access to the cascade hall would have on a potential proliferator. Wolfgang Fischer and Gotthard Stein explains that the LFUA concept is all about access to the cascade hall, since observation ensures that the cascade sequence has not been modified (which would mean the plant has been set up to enrich uranium above five per cent uranium-235). It is a complicated affair to change the sequence, and to carry out the necessary reconnection of piping et cetera, and this is why it is important for inspectors to show up in a truly unannounced manner. If they could not, the deterrent effect of random inspections would be lost.

Therefore, travel conditions may disqualify a facility from the LFUA scheme. For instance, China's Shaanxi facility employs a tailored safeguards scheme, since:

"…travel conditions to the Shaanxi Plant at least at present are not compatible with requirements for unannounced access at the facility for the performance of unannounced inspections in the cascade area".

The Hexapartite approach (of which LFUA forms part) is now more than 20 years old, and in possible need of updating. Therefore, it was reviewed in April 2005 by a technical meeting on techniques for IAEA verification of enrichment activities. The meeting was organized jointly by the Agency divisions of Technical Support and Concepts and Planning. Its recommendations are not yet public.