I dismissed the Sanger/Shanker piece Pakistan Is Rapidly Adding Nuclear Arms, U.S. Says as propaganda for three reasons:

[H]ere are the three points where Sanger manipulates the reader:

• The assertion of "confidential briefings" for which he does not
  name a source and does not explain how he got knowledge of these.
• The "rapidly adding" in the headline and first graph also not sourced at all and not confirmed by the rest of the article.
• Moving Senator Webb's question of control about future money to
  Pakistan into the context of adding nukes when it is much more
  generally asked in the context of Pakistan's military stand versus
  India.

Looking deeper into the issue there is another and much more relevant reason to dismiss the piece. For now Pakistan simply does not have the ability to rapidly add nuclear arms.It lacks the ingredients.

To build nuclear weapons one needs either highly enriched uranium (HEU) or plutonium.

Capacity of  uranium enrichment centrifuges is measured in Separative Work Unit (SWU). One centrifuge running 1 SWU for one year can produce some 6 grams of HEU.

Pakistan's uranium enrichment program at Kahuta consists of some 3,000 centrifuges. There are no reports that this number has increased. These centrifuges are of earlier P-1 and P-2 designs and were copied by A.Q. Khan from a Dutch program. Meanwhile Pakistan has developed centrifuges of P-3 type with 12 SWU per year and may have been working on P-4 models with 20 SWU per year. 3,000 of those could in the very optimal case produce 60,000 SWU per year.

An implosion weapon using U235 would require about 20 kg of 90% U235. Roughly 176 kg of natural uranium would be required per kg of HEU product, and about 230 SWU per kg of HEU, thus requiring a total of about 4,600 SWU per weapon. To enrich natural uranium for one gun-type uranium bomb would requires roughly 14,000 SWUs.

If all the centrifuges Pakistan operates were replaced with more modern P-4 type and if these were up and running without any flaw it could possibly produce at maximum 60,000 SWU per year or enough for some 20 of the best-possible-design uranium bombs per year.

But the above is the very worst case. The real HEU production capacity is likely much less and it is dubious that Pakistan, without any testing, would be able to use the best thinkable bomb design.

Pakistan also has no good reason to increase its number of uranium based weapons. Most of Pakistan's currently 40 to possibly 80 nukes are of this type. Such weapons are by design relative clumsy. The 'Little Boy' gun type uranium bomb the U.S. dropped on Hiroshima used 64 kilogram of highly enriched uranium but had a total weight of some 4,500 kilogram.

The only reasonable way to deliver such big weapons is by airplanes which have to enter enemy airspace and defeat all enemy air defenses to deliver their bombs on a target. Ballistic missiles are much preferable means to deliver nuclear ammunition. But these also need much smaller weapon packages.

Modern nuclear weapons are therefore plutonium fission based or thermonuclear weapons triggered by plutonium which use a quite complicate sphere to compress the nuclear material.

Pakistan has done experimental work on plutonium type of weapon (though not tested any) and has been working over the last years to expand that program to production capacity. A indigenous build (with Chinese help) heavy water and natural uranium research reactor began operating in Khushab in 1998. It has a capacity of some 50 thermal megawatts and may produce enough plutonium for about one weapon per year.

Two new reactors are now getting build next to the old one in Khushab. David Albright's ISIS published (pdf) pictures last month which were taken at the end of January and show that the outer buildings for these reactors now near completion. But they are far from being operational and it is likely to take years before they are up and running at any reasonable capacity. Pakistan is also in the process of expanding production capacity for heavy water needed to run these reactors. The plutonium separation plant (pdf) near Rawalpindi and is a site (pdf) near Dera Ghazi Khan where the chemical uranium processing and (likely) actual weapon production gets done are also adding some buildings.

It seems that Pakistan is expanding its nuclear program in a move away from uranium based weapons to plutonium based weapons which will be deliverable by ballistic missiles. All the above expansion plans are observable on satellite pictures since at least 2002. There is no surprise in this and the expansion of the capacity is moderate. Albright is estimating that the new heavy water reactors will have a capacity of some 100+ thermal megawatts each. The total would thereby go from 50 to 250 MWt and may be enough to produce 5+ additional weapons per year.

The expansion of Pakistan's production of nuclear weapons is also moderate when one looks across its boarder:

As of September 2005, India was estimated to have a stockpile of around 45-95 warheads. In addition, Defense News reported in their November 1, 2004 edition, that an Indian Defence Ministry source quoted that "in the next five to seven years India will have 300–400 nuclear and thermonuclear weapons distributed to air, sea, and land forces." It is estimated that India currently possesses enough separated plutonium to produce and maintain an arsenal of 1,000-2,000 warheads.

Sanger's sensational article about Pakistan 'rapidly adding nuclear arms' seems false to me because nothing in the available material shows that Pakistan currently even has the basic production capacity or stockpiles of nuclear material needed to 'rapidly add' nuclear weapons. The thrust of Pakistan's program is to expand its plutonium production capacity. But it is still in the process of doing this and for now Pakistan simply does can not 'rapidly expand' its nuclear arsenal.

There is a nuclear arms race going on between India and Pakistan. This is not good and should stop. But it needs two sides to agree on that.

India yesterday launched a nuclear capable ballistic missile with a reach of 2,500 kilometers (1,560 miles). Such launches are certainly not an incentive for Pakistan to slow down its programs. Nor is the propaganda campaign currently waged by Washington an incentive for Pakistan to lower its deterrence capabilities.