CONFIRMED -- Iran allowed to collect own samples at nuclear military base -- Verification expert on self-inspection deal: "You need the eyes and the brain to look where to sample... video camera opens up additional methods of deceiving"

The Israel Project's Omri Ceren reports that it has now been confirmed that Iran was in fact allowed to 'inspect' its own nuclear military facility at Parchin. I received this by email.

Reuters this morning conveyed Iranian media reports establishing that the Iranians recently took their own environmental samples at their Parchin military facility, where they conducted tests relevant to the detonation of nuclear warheads, in lieu of having IAEA inspectors take the samples (Reuters story below; original IRNA story here [a]).

The IAEA has long sought access to Parchin: the agency needs to clarify the nature and scope of Iran's past nuclear weapons work - the possible military dimensions (PMDs) of Iran's atomic program - to establish what the Iranians did and how far they got, which are the prerequisites to setting up a verification regime against future violations. The Obama administration had promised lawmakers that IAEA inspectors would be able to inspect Parchin and resolve all PMD issues before any final deal was inked [b][c][d][e][f].

Instead the JCPOA allowed Iran to sign a secret side deal with the IAEA permitting the Iranians to self-inspect the facility rather than grant IAEA inspector robust access.

That side deal was subsequently revealed and published by the AP: the Iranians would get to collect their own samples, those samples would have to come from mutually agreed upon areas under overlapping photo and video surveillance, and the number of the samples would be limited [g][h]. An Iranian statement this morning confirmed that the Iranians collected their own samples: "Iranian experts took samples from specific locations in Parchin facilities this week without IAEA inspectors being present" [i]. An IAEA statement confirmed the sampling was done from mutually agreed upon areas under overlapping photo and video surveillance: "the determination of the spots where the samples are taken is a separate, important, careful activity…. [that] have to satisfy our requirements… the actual swiping or other sample taking [place] under redundant continuous surveillance" [j]. It's not yet clear whether the AP was also correct about the number of samples being limited.

The arrangement means that the IAEA will not be able to establish what happened at Parchin. David Albright, president of the Institute for Science and International Security, explained at a panel hosted earlier this month by the Hudson Institute that self-sampling under surveillance is inadequate. Inspectors need to be on the ground to identify dusty nooks and corners where violators forgot to dust; the mutually agreed upon areas are by definition the ones that violators know have been sanitized [k]:

What you have is, is the situation where there'll be videotaping of the potential locations where sampling would take place. Then the IAEA would direct the Iranians to take the samples. And that's not the normal way to do things.

If I could give the example in Iran of Kalaya Electric, a secret centrifuge research and development facility that Iran denied was such a thing. The IAEA got access and it brought in a very top level centrifuge expert with that access, who looked around. And when they did the sampling finally they didn't find any trace of enriched uranium in the areas that had been heavily modified. But in a another, a secondary building they found in a ventilation duct - which had not been modified - they found traces of enriched uranium...

You need the eyes and the brain to look where to sample.

I brought an example of sampling in North Korea... they sampled in the Yongbyon reprocessing plant in the early 90s... you can see in the sampling they're looking behind this box... Look for where it's dusty. The idea is that it's not been disturbed. In the case of Parchin, it would be look for where the paint doesn't look solid. And so, that's very hard to do with a video camera. So I think the video camera opens up additional methods of deceiving the IAEA. And it's not the normal way they've been doing it. And so I think that's a problem...

The sampling would be done, and then the IAEA access would follow. And so the access is coming at a point where it's not as useful... You want it to drive the inspection effort and the environmental sampling effort, not be done at the end of the process [7:29].

The arrangement was also read more broadly as kneecapping the IAEA. On the experts side, CNN got analysis from Olli Heinonen, former director of the IAEA's verification shop, as Albright, president of the Institute for Science and International Security: "It is very unusual... I find it really hard to understand why you would let someone else take the samples and only see through the camera" and "It's really not normal... I don't know why they accepted it. I think the IAEA is probably getting a little desperate to settle this" [l]. On the Congressional side, a visit by Amano to the Hill on the side deals left Senators fuming [m].

The IAEA reacted to this morning's leak by issuing more assurances about the adequacy of Iranian self-inspections. White House validators have already picked up the "redundant continuous surveillance" theme and you're likely to hear more of it [n].

The problem is that the IAEA assurances read a little bit like a hostage note: lawmakers, experts, and journalists know that the arrangement is unprecedented and that inspectors need to be on the ground, so the IAEA statements may be read as evidence that the agency has bent to political pressure.

What could go wrong?