Postol also said the surveillance and targeting capabilities do not currently exist to make a global strike ballistic missile effective. Instead, he argues so-called targets-of-opportunity -- short-notice, high-value targets like terrorist leaders or nuclear weapons being smuggled -- can be attacked with existing platforms and weapons, like the Predator unmanned aerial vehicle.

U.S. Strategic Command, the military command responsible for U.S. nuclear forces, has briefed to Congress its desire to develop a global strike capability separate from the fleet of aircraft bombers, ship-launched cruise missiles, and UAVs, as well as the nuclear intercontinental ballistic missiles. This strike capability would be able to hit any target on Earth within 60 minutes of it being acquired.

Marine Gen. James Cartwright, the commander of STRATCOM, told Congress in March one of his first actions toward that end is to deploy within two years conventional ICBMs on Trident submarines, which now carry 24 nuclear missiles each.

Under Cartwright's plan, within two years there will be 12 Trident submarines carrying 22 nuclear and two conventional ballistic missiles each.

Postol warned Thursday that mixing conventional and nuclear warheads -- which look the same on radar -- on a Trident submarines risks triggering Russia's early warning system. That in turn could lead to a mistaken, retaliatory launch on the part of Russia.

According to Postol, one of the known vulnerabilities of Russia's early warning radar system is its vulnerability to a nuclear blackout. A single nuclear missile detonated high in the atmosphere can blind all its early warning radars below, rendering it unable to monitor subsequent launches. Were the United States to launch a nuclear attack against Russia, the first step might well be a high-altitude detonation to blind its warning system. Russia, then, would be faced with a decision: either wait to see if the Trident missile fired explodes and blinds its radars, or launch a retaliatory strike immediately.

"The global strike system would pose a problem every time it was launched within the line of sight of Russian radars," Postol said.

To prevent Russia from having to make such an awful choice, Washington and Moscow have agreed to share information about routine missile launches. That arrangement, however, calls for at least 24-hours advance notice. In scenarios envisioned for the global strike mission, there would be less than 60 minutes to notify Moscow.

Moreover, a 2000 agreement between U.S. President Bill Clinton and Russian President Vladimir Putin to build a Joint Data Exchange Center in Moscow to speed and ease communications about missile and rocket launches in Moscow has yielded no such center. Construction of the JDEC has been mired in political and bureaucratic red tape, said Pavel Podvig, a Russian military expert with Stanford University's Center for International Security and Cooperation.

Assistant Secretary of Defense for International Security Policy Peter C. W. Flory told the Senate Armed Services Committee in March that he had complete confidence Russia would know the trajectory of a global strike conventional weapon, and would not mistake it for a nuclear missile aimed at its own territory.

Postol suggested that might be giving the Russians too much credit: 11 years ago, Russian radar operators mistook a scientific rocket launch out of Norway for a surprise U.S. nuclear attack that would hit within five minutes. Norway had given Moscow prior warning of the launch, but that word had not been passed down range.

Given Trident submarines' primary mission -- a nuclear deterrent to Russia -- Tridents armed with conventional ballistic missiles would likely always be at sea north or east of Russia, and therefore within view of Russian early warning radars, setting the stage for a mistaken retaliatory launch.

Postol suggests if a global strike conventional missile is necessary, it might better be launched, with necessary modifications, from Navy ships in the Indian Ocean, or from land-based silos on the island of Diego Garcia. Neither would come close enough to trigger Russian early warning radars if their targets were in the Middle East or southern Asia.

But Postol doubts scenarios exist that would require an ICBM, sub-launched or otherwise. The very nature of 60-minute targets of opportunity means the U.S. military or intelligence community would already have "eyes on" the target and would require the ability to make minute adjustments in the trajectory of the weapon to hit the target accurately. ICBMs travel at such great speeds -- and are relatively incommunicado during reentry -- that those adjustments would be difficult to make before impact.

Postol also asserts no U.S. spy satellites are even capable of finding a time-sensitive target without first being cued where to look by human intelligence operators or tactical systems like UAVs already in the area.

And if those tracking capabilities are already on the scene, they would likely already have a strike weapon in place to use.

Postol offers the attack on Jordanian terrorist Abu Musab al-Zarqawi in Iraq this year as an example. He was trailed for months, but when his exact location became clear to human intelligence operatives, they called on F-16s already in the area for the bombing mission.

It's possible to imagine that special forces could observe a target in a place where fighters or Predator strike aircraft are not immediately available, Postol said, but those same forces could also track the target for the hours required to get an aircraft on the scene.

Given safety, cost and technological arguments against the Trident conversion, Postol believes enthusiasm for the program stems from quarters in the Navy, government and industry seeking new missions for the submarine force in the post-Cold War world. Postol is not looking to undermine the Trident in its nuclear deterrence role. The problem arises when the warheads are mixed, he said.

Related Links
Learn about nuclear weapons doctrine and defense at SpaceWar.com