Nuclear Security and Strategic Analyses Dr Marko Beljac

29Nov/090

Melbourne’s Sydenham Train Line Passengers Fume, Whilst Eastern Suburban Patrons Are Smugly Satisfied

The Age today has a pretty revealing article on Melbourne's public transport system. The paper has obtained customer satisfaction survey results through Freedom of Information. They tell us something about the world's "most liveable city"

...The latest ratings, which are significantly lower than those in 2003, have been fuelled by record lows on parts of the Sydenham and Pakenham lines...

...Most Sydenham passengers on average rated overall satisfaction an abysmal four out of 10, while Pakenham line passengers gave their route a 4.6....

I'm one of those passengers on the Sydenham line. However, not everybody in Melbourne is that pissed off with the level of service according to the report

...The most satisfied passengers on the network are all in the eastern suburbs, with the Lilydale, Glen Waverley and Alamein lines, each recording satisfaction levels above six out of 10...

Of course, of course the uppity classes get the good service whilst the working classes get dished out the crap. This is despite the fact that public transport is supposed to be a public good.

I wouldn't even know how many times the train gets delayed on Platform 5 at Flinders Street. Funny how you don't hear that annoyingly insincere Connex announcement on the platforms that service the Lilydale etc lines.

Yet we are lead to believe that Victoria has a Labor government.

Pig's ass it does.

Filed under: Australia No Comments
28Nov/090

US Army War College Study on Intrawar Deterrence

Hey, how good is the US Army War College! Pretty good in my opinion.

I have not checked them out in a looong time, but they really do have very good strategic working papers. I think one must check out their publications list on a regular basis, which is advice I haven't been exactly following. For example, they have a fascinating study that rips into the notion of the "operational art of war", which the author takes to be an apolitical way of thinking about something that is inherently political; i.e war. A co-author is a one-star.

Anyway, there is a very important study published by the College on intrawar deterrence in the Middle East. It's written by Andrew Terrill. It's biggish, but the summary carries the main conclusion of the paper (really it's a monograph; it comes out to 126 pages!)

...A central conclusion of this monograph is that intrawar deterrence is an inherently fragile concept, and that the nonuse of WMD in both wars was a result of factors that may or may not be repeated in future conflicts. Additionally, the tactics for intrawar deterrence will require constant adjustment as war is waged and develops in unexpected ways. Signaling and political communication is inherently difficult in
such crisis and few unequivocal statements are taken at face value. U.S. planners must never become too comfortable with the elegance of any plan involving intrawar deterrence, and the U.S. leadership must be prepared to accept the possibility that there are always a number of ways such strategies can break down
...

I take that as a bit of a dig against Bush era nuclear strategy. This study should be read very carefully by all those involved in the current Nuclear Posture Review.

27Nov/090

The Press and Lieber Foreign Affairs Article on 21st Century Nuclear Deterrence

I have now read the Press and Lieber article in Foreign Affairs on nuclear deterrence for the 21st Century. It is a pity that this article was not written in a journal like International Security, that way we could have got more meat. Although there is a technical annex at Press' website, which looks a lot like John Mearsheimer's website.

I like that unit on realism that Mearsheimer teaches at Chicago. I'm not a realist, but I would have loved to enroll in a unit like that.

Anyway, back to Press and Lieber. As I expected the article has good detail and the whole thing is based on very careful analysis (especially the technical annex). You wouldn't see this in an Australian university.

In saying that, naturally, I don't agree with the article one little bit.

I'll try and explain my reasons for saying this. At the outset I should stress that I did not buy Foreign Affairs in order to read this. The damn thing costs about $22 in Australia. I don't like paying that much for a magazine, not matter how august.

I have gotten my copy of the article over the database. As such I will not cite from it in this blog post.

The opening assumption is one that I cannot agree with. Press and Lieber are concerned with ensuring that the US maintains escalation control in the context of selective regional nuclear proliferation. This sounds reasonable, but actually it isn't. The purpose of escalation control is to help make sure that the world remains a safe place for US offensive conventional power projection. This is oft put in terms of such phrases as; "defending our allies" and "defending our vital interests". These are technical phrases meaning defending the prerogative to commit aggression.

Making the world safe for continued US aggression is not a world that most of the people of the world want to live in. I do not and cannot accept such an operating assumption.

That's the first point.

Secondly the main argument here is a bit strange for mine, even if you accept the underlying assumption. Most of the discussion is devoted to making a case for keeping/developing low yield nuclear weapons directed towards the counterforce mission. They have an interesting analysis suggesting that highly accurate delivery platforms would enable STRATCOM planners to develop attack options directed towards knocking out missile silos through airburst detonations.

They argue that highly accurate US delivery platforms would enable even low yield warheads to generate enough blast overpressure to knock out missile silos if airbursted at a given height. Let us, for arguments sake, simply accept this point. This means the US could have a low casualty counterforce option, the holy grail of nuclear war fighting.

What makes this argument very odd is that for an analysis about 21st deterrence it looks a lot like cold war era analysis. Silo busting, silo counting etc were all regular features of cold war era analysis. Look at Press and Lieber's opening assumption again; they are concerned about the impact nuclear proliferation would have on US escalation control post cold war.

How many proliferant states are building silos for their missiles? Does Pakistan, India, North Korea and Iran maintain missile forces housed in silos?

Not really.

Proliferant states (we still can't put Iran 100% into this category) are heading towards road mobile missiles, and even Russia and China maintain road mobile missiles. To land a knock out counterforce blow in the context of regional "regime change" operations would require destroying road mobile missiles, not missiles housed in silos. To do this you need not just accurate delivery platforms but super reliable real time intelligence.

One would think that there would always be some inherent uncertainty about the intelligence information gathered purporting to show the precise location of road mobile missiles. This is especially the case for solid fuelled missiles, one would feel. The targeting task here would involve developing enough overpressure to knock out a missile within a certain circular footprint or area. That's different to the pin point targeting of a silo, although even here accuracy is measured in terms of circular error probability. However, the silos aren't moving.

The bigger the uncertainty in intelligence the bigger the footprint that needs to be covered in a counterforce attack. That is, the bigger the uncertainty the greater yield that is needed to meet the damage expectancy criteria of the STRATCOM war plan. For a counterforce strike these are very high.

It is the presence of such uncertainty that would prevent the US from developing escalation control even if it embarks upon something like the Reliable Replacement Warhead, which is what Press and Lieber are making a case for. This is done in a sort of unspoken way, which must be done lest the reliability cover for RRW is blown. If the US arms control community reads this article and keeps sticking to reliability in their analysis on RRW, then they are hopeless.

But there is something else here that is at issue. The authors don't mention it, but surely they are aware of it. What they are really offering is a type of window of vulnerability only this time in reverse. Remember the window of vulnerability scare of the late 1970s? Hawks in the US supposed that the Soviet heavy ICBMs, esp the SS-19 and the SS-18, gave Moscow the ability to knock out US missile silos in a limited first strike during a crisis.

The US would then have its bombers and boomers left over, which would not be able to knock out remnant Soviet nuclear forces given their lax counterforce potential. The US could only launch an attack against urban-industrial targets in a countervalue attack. But because the USSR would retain an ability to mount a countervalue attack in response, the US would be deterred from escalating to all out nuclear war. Washington would then be forced to back down and concede to Moscow.

Imagine if the US were indeed to develop low yield airbursting counterforce capabilities against Russian and Chinese silos. Imagine this can be achieved with low Russian and Chinese civilian deaths. Moscow and Beijing would now have a window of vulnerability type problem; escalate to general nuclear war (and be assured of destruction) or back down and let the US have its way in a crisis.

I think this is about escalation control by reversing the old window of vulnerability argument in order to achieve escalation control with respect to Russia and China. I have had a sneaking suspicion that this was an essential part of Bush era planning for quite some time.

However, notice a key point here. For the scheme to work the US would still need the ability to destroy Russian and Chinese society in large urban-industrial attacks. Escalation control is meant to be a means to escape MAD, but the whole thing must ultimately rely upon MAD.

That has always been my position with respect to strategies of escalation control and limited nuclear strikes. You can't escape MAD. Period.

26Nov/090

Could the Fordow Enrichment Plant be a form of Minimal Covert Latency?

There are a lot of implications that one could draw from the revelation of the Fordow Fuel Enrichment Plant (formerly known as the Qom enrichment plant) and the subsequent IAEA safeguards report, following detailed Agency inspection of the facility.

The first set would focus on safeguards itself. For the purposes of this particular blog entry I am going to leave that aside.

I want to concentrate here on the size of the plant and what this might tell us about the purpose of the FFEP and how it might fit into Iran's broader nuclear program.

When the secret plant was revealed by the United States, Washington's knowledge of the plant was not limited to just non-human intelligence gathering, many quickly jumped to the conclusion that Fordow was meant to be part of a secret parallel nuclear bomb program.

At the time, and subsequent to that, I suggested that a possibility for Fordow was that the plant might be a type of strategic hedge against a US led air campaign against Iran's nuclear facilities.

I specifically suggested that Fordow might be a facility designed to maintain continuity of knowledge and expertise in case that Iran's main enrichment facilities, the Pilot Fuel Enrichment Plant (PFEP) and Fuel Enrichment Plant (FEP) at Natanz, are destroyed through air strikes.

Critical to this view was the size of the Fordow plant. The initial US intelligence briefing stated that Fordow was designed to house 3000 centrifuges. I had surmised that Fordow could well be designed to house much less than 3000, on grounds that a covert pilot plant would suffice for continuity of knowledge and expertise. A few small cascades would suffice.

The latest IAEA safeguards report confirms that Fordow was designed to house 3000 centrifuges.

I am of the view that this empirically rules out the possibility that I had initially broached. As the noted physicist Julian Schwinger noted after the unexpected results of an experiment were conveyed to him; "Gentleman, we must bow to nature."

So, why 3000 centrifuges?

One obvious possibility is the one initially raised by US officials; namely, that Fordow is part of a covert parallel Iranian fuel cycle devoted to producing fissile materials for a nuclear weapons program.

3000 centrifuges rings alarm bells because that is how many centrifuges are spinning at Pakistan's AQ Khan labs in Kahuta. As Global Security Org's primer on Kahuta points out,

...About 1991, Pakistan installed additional centrifuges, raising its HEU production capacity roughly threefold. By 1991 about 3000 machines were thought to be operating with a production capacity of 30-50 kg U-235/year, enough for 2-3 implosion weapons a year...

These figures are based on a implosion bomb fuelled by 12.5-15 Kg of weapons grade uranium.

An important point to consider here is that the Kahuta facility would have more advanced centrifuges than Fordow. That is, they would have a greater separative capacity. According to the IAEA Iran has stated that it will install 3000 IR-1 centrifuges at Fordow, which are modelled on the "vintage" Pakistani P-1 centrifuge.

Now Iran, as the IAEA has revealed, has not installed these centrifuges at Fordow yet. However, the Agency has seen all the ancillary facilities at Fordow and surely knows enough about Iranian enrichment activities to verify that Fordow will indeed house 3000 IR-1 centrifuges.

If Iran has a parallel military fuel cycle then 3000 IR-1 centrifuges just doesn't cut it. This is because a parallel fuel cycle for military purposes must involve covert actions across the entire nuclear fuel cycle, from uranium mining and milling to uranium conversion.

As the best analyst on Fordow outside of government, Ivanka Barzashka, pointed out after the initial revelations were made

...If the 3,000 machines are IR-1s with a separative capacity of 0.5 kg SWU/yr, it would take them about 90 years to get one year’s fuel load. This of course makes no sense. However, if they want to get one bomb’s worth of HEU (from natural uranium), they need 6,320 kg SWU/yr and this would take you a little over 4 years...

One bomb's worth over 4 years is simply not rational. It should be stressed that Barzashka might be making a small error. She bases her figures on a IR-1 SWU/yr of 0.5 kg. This does not necessarily reflect the actual separative capacity of the IR-1. Iran is spinning its IR-1 machines at less than full capacity. If we take actual capacity on board then we might have to really use ~1-2 SWU/yr for the IR-1, which more closely resembles the figure for the P-1.

I don't believe that this affects her conclusion too badly. That's because her initial calculation is based on one bomb's worth. That bomb would need to be tested. To have one bomb over one year you really need enough WgU for two bombs; one to test and one for an actual warhead.

In short, the numbers by no means stack up for the parallel fuel cycle theory either. It must be stressed, however, that Iran has reserved the right to install more advanced centrifuges at Fordow in future. That would change things, but not too greatly now that Fordow has been revealed; Fordow will now be safeguarded by the IAEA just like Natanz.

We must note that these are theories based on idealised assumptions. Iran seems to have a problem with impurities, especially molybdenum, which makes enriching to weapons grade challenging. For example, more than a few of these 3000 machines would crash if they attempted to enrich impure UF6 to weapons grade. That should not be neglected in analysis.

Minimal Latency?

So if my initial possibility can be ruled out and the numbers look wobbly for the covert parallel fuel cycle theory then what gives?

Perhaps we need to consider what I shall dub "minimal latency".

Could Iran be trying to develop a covert latent or breakout capacity? Any nuclear power program with all the associated fuel cycle activities (either from the front or back end), including all those that are safeguarded, give a state what is called by non-proliferation analysts as "latency" or potential "breakout capacity."

I believe that Fordow was meant to be covert. Tehran got caught red-handed. I have believed that from the get-go.

Consider another aspect of Barzashka's analysis

...The third option is to take LEU from Natanz and enrich it to a bomb’s worth of HEU. This would take about a year, depending on how much material they are willing to waste. So, if they are trying to divert LEU from an existing facility such as the one at Natanz, the numbers add up perfectly (almost too perfectly). However, diversion of nuclear material from the enrichment plant at Natanz or the conversion plant at Isfahan is near impossible to go undetected if the facilities are under IAEA safeguards. Although uranium mines and mills are not under safeguards, so far there is no sign of a clandestine conversion plant in Iran. There is always the option that the Iranians could just kick the inspectors out and have breakout in one year or less...

The problem with this position was always pretty clear. This is pointed out by post inspection analysis by Barzashka and Oelrich at the Bulletin of the Atomic Scientists (I hate going there, but alas I have little choice in the matter).

...This scenario, a favorite among nonproliferation experts, would require enriched uranium to be sneaked out of the Natanz facility, which has been under safeguards since it became operational in 2003. The IAEA's goal for detection of a "significant quantity" of uranium, that is, enough to be of worry for making a bomb, is one month--much shorter than the time it would take the cascades at Fordow to enrich such an amount. If Iran did want to enrich uranium for a weapon, it would be far easier to oust inspectors from Natanz and proceed with much more rapid HEU enrichment there...

Consider.

If Iran kicked out the IAEA then Natanz would be the better breakout plant. However, one month is too long. The US would be able to mass its air power deployed in the region and bomb Natanz. The easy part in any anti-nuclear plant air strikes would be bombing Iran's physical infrastructure. The reactor, the enrichment plant at Natanz and all the rest.

The hard part is hitting the actual LEU or lightly enriched uranium itself. Iran would be able to hide this pretty quickly after kicking out the IAEA. It could be shifted to anywhere in Iran.

So Fordow could be a minimal latency capability. If the US bombs Natanz then the LEU could be diverted to a covert Fordow plant and used to build a nuclear deterrent.

Then Iran could use the umbrella of power provided by a strategic nuclear deterrent to start again on its nuclear energy program.

I dub this "minimal latency" because it does not require Iran to have a full parallel nuclear fuel cycle; just a covert enrichment plant big enough for potential breakout. Notice that this still fits with Iran's way of viewing its status under the Code 3.1 subsidiary arrangement.

This is because under minimal latency Iran need not introduce any nuclear material at Fordow. Under the old Code 3.1 Iran would not be required to; it need only inform the IAEA of Fordow 180 days prior to the projected introduction of nuclear material.

Let's go back to the 2007 Iran NIE. Recall that this stated that, with good confidence, that Iran halted a bomb program in the fall of 2003. Good analysts such as Paul Kerr have surmised that this occurred when Iran consolidated all of its nuclear activities under the aegis of the AEOI.

Hardliners doubtless were not too happy about that. The Iranian elite is deeply divided. Perhaps hardliners extracted a price for this; "Ok, we will ditch the bomb program but we want to maintain a covert breakout capability just in case".

Minimal latency is a possibility. Iran has been dragging its feet on the LEU deal recently stitched together. Tehran has stated that it is concerned about assurance of supply.

True, assurance of supply is an important issue. But no LEU means no minimal latency as well.

Minimal latency is a theory that can be tested. If Iran has other covert fuel cycle plants devoted to other tasks, such as a covert uranium conversion plant, then minimal latency would be pretty much dented.

Oelrich and Barzashka make an interesting point, which suggests something else

...The Iranian Revolution Document Center [in Farsi], a pro-government Iranian research institute dedicated to documenting and analyzing the Islamic Revolution, suggests that Fordow's strategic purpose is to deter an attack on Natanz. Even though Fordow is much smaller than Natanz, this makes some sense. Such an attack would not be meant to cripple Iran's commercial nuclear fuel production, but to stop a weapons program. If Fordow is invulnerable and could alone carry a weapons program forward, the value of attacking Natanz is greatly diminished--although Fordow's limited enrichment capability is still an issue...

But, wait! Let's not be too hasty.

I am thinking here of the movie Dr Strangelove. Remember that scene when the Soviet ambassador said that the USSR has a doomsday device in order to enhance deterrence? The mad scientist in the wheelchair looks on super suspiciously. He said something like; "if that is a deterrent, then why keep it secret? Deterrence only works by overt signalling".

How does a covert plant develop an overt deterrent signal? One might argue that it is precisely the covert nature of a secret plant that generates deterrence, but really the deterrent value here lies in the very possibility of a covert plant not necessarily in its actual existence.

Filed under: Proliferation No Comments
25Nov/090

A Strategic Rationale for B61 Modification?

The Federation of American Scientists has got a very good blog entry on the JASON LEP study.

They also have a link to STRATCOM briefing slides on something that looks a lot like the Reliable Replacement Warhead. There is much that can be said here. Even concerns about nuclear terrorism are being used to prop up the case for RRW!

I love that bit.

However, in going through the slides I got interested in what it had to say about the B61 "modification". It's another late night, so beware.

The slide at page 6 states that many features of US nuclear weapons cannot incorporate up to date requirements for reliability and surety. Notice that the nifty little picture depicts the B61 nuclear bomb.

One area singled out on the B61 schematic is the tail section.

OK.

Take a look at page 10. This depicts the B61 program in greater detail. Firstly, notice the silly nonsense about improving the assurance of NATO's extended deterrent. The Germans have recently made noises about wanting US nuclear weapons (i.e. the B61) to be taken out of the country, yet here STRATCOM would have us believe that the B61 needs to be upgraded to ensure the credibility of extended deterrence in order to assure NATO allies!!!

Let's put that aside and concentrate on the B61 tail. Notice the "no smoking type" picture over the B61 tail kit.

At page 19 "actions needed to be taken now" are listed. The list focuses on the W76 and the B61 in so far as warheads go. This is the slide that got me interested. On the B61 it states that an action that needs to be taken now is "support B61 tail kit modification program."

This is interesting in light of previous B61 modification programs, which also focused on the tail kit. According to the Nuclear Weapon Archive primer on the B61 the tail kit was modified for the B61 Mod 11 in order to develop an earth penetrating warhead

...Mod 11: Tactical or strategic bomb with multiple yield options presumably ranging from 10 Kt (and possibly lower yields) to 340 Kt. This is a modified Mod 7 with a one-piece case hardened steel center case, and a new nose piece and rear subassembly to provide ground penetration capability for defeating buried targets ("bunker busting"). The parachute assembly has also been removed, and new aerodynamic fins added for high-velocity, accurate delivery. The B61-11 buries itself 3-6 meters underground before detonation, transfering a much higher proportion of the explosion energy to ground shock, compared to surface bursts. The actual warhead itself is identical to the Mod 7. This is the first new model of a U.S. warhead to go into service since warhead production was suspended in 1989. It is being produced by field modification of existing Mod 7s....

It also has some interesting information on B61 modifications that were previously planned, but did not get up

...Efforts are currently underway at Sandia to develop a new weapon using the B61 warhead package. This is the BIOS (Bomb Impact Optimization System), a guided glide bomb that would permit release of nuclear weapons at a greater range from the target, and by using the GPS satellite system would permit delivery with 1 meter accuracy....

Could the current push for B61 modification be also an attempt to rejuvenate the previous BIOS program? Could we be talking here of a glided earth penetrating warhead with GPS accuracy?

The Global Security Org primer on the B61 is also interesting

...The B61-11 is a new Mod of the B61 being converted from existing B61 Mod 7s. The basic differences in the two Mods are in the nose and tail configurations, and the elimination of the B61-7 parachute and gas generator. The similarities are that all B61-11 internal case hardware and components, including the IHE physics package and warhead electrical system, are from the B61-7...

...The new modification and proposed design of the B61 nose assembly incorporates new radar hardware and sophisticated structured elements to withstand the high-shock environments. Sandia’s work involved repackaging the B61-7 nuclear and electrical systems into an earth penetrator case. The aft portion of the bomb was outfitted with ballast and a drag flare....

It might well be the case that STRATCOM wants to modify the tail kit of the B61 in order to enhance its capacity to act as an earth penetrating warhead. There is precedent for this, clearly.

If so, this would be important. This would show, if true, that the current drive on warhead component replacement is partly driven by the desire to enhance military effectiveness; concerns which follow on from post cold war changes to US conceptions of nuclear deterrence that are focused on regional contingencies. Glide bombs are for high air defence battlespaces, i.e. Russia and China, it might also be added.

The B61 tail kit modification might be the first empirical hint that RRW is driven by strategic considerations, not reliability and the like. This looks a bit like the RNEP, does it not?