Report examines possible release of mustard agent during Gulf War

WASHINGTON, July 27, 2000 (GulfLINK) – The Department of Defense has issued a new report, "Possible Mustard Release at Ukhaydir Ammunition Storage Depot," the latest in its series of case narratives regarding possible chemical warfare agent exposures during the Gulf War. The narrative examines evidence concerning the possible release of mustard agent as a result of two Coalition air strikes on the Ukhaydir ammunition storage depot located approximately 100 kilometers west of Baghdad.

Based on Iraqi disclosure statements, United Nations inspection teams’ findings, and intelligence community assessments, investigators were unable to determine if mustard agent was actually released from the site. In several instances the available evidence was contradictory with no means to conclusively decide which was correct.

"It is unlikely that any U.S. servicemembers could have been exposed to any possible release of chemical warfare agent," said Margaret Graf, one of the investigators who worked on the report. "Computer simulations show that the potential hazard area from such an event would not have extended beyond 125 kilometers from Ukhaydir. American forces were several hundred kilometers away."

Part of the initial confusion of evidence resulted from discrepancies in Iraq’s declarations regarding its weapons of mass destruction programs, the report said. In 1991, Iraq first declared 6,394 mustard-filled 155mm artillery rounds stored at the Fallujah Proving Ground. Iraq made no mention in this disclosure of the Ukhaydir Ammunition Storage Depot as a site where chemical rounds were deployed during the war. That fall, the United Nations Special Commission chemical warfare team inspected the rounds and accounted for 6,380 rounds, 14 fewer than declared by Iraq.

Iraq’s second full disclosure did not mention either Fallujah or Ukhaydir. However, in the summer of 1996, responding to questions from U.N. inspectors, Iraq submitted a third declaration that contained additional information on the history of the production, filling and deployment of 155mm mustard-filled rounds, including the fact that 6,394 rounds had been deployed to Ukhaydir in January 1991.

United Nations inspectors visited the Ukhaydir facility twice in 1997 and found four intact mustard-filled 155mm artillery rounds in the rubble surrounding a bomb crater. They found no evidence of chemical agent rounds inside the on-site bunkers. This correlated with Iraq’s claims that chemical rounds were piled in stacks in the open to avoid damage in the event the bunkers were targeted. The U.N. team and the CIA both assessed the mustard-filled rounds counted at Fallujah were the same rounds deployed to Ukhaydir during the war, in part because the totals given for each site are identical. After the U.N. inspections at Ukhaydir, there remained only 10 unaccounted rounds. Iraq later claimed to have found 12 additional rounds in the same place U.N. inspectors searched, but this can’t be verified because no U.N. team members were present at Iraq’s discovery.

Working from the 1996 disclosure and information developed by the U.N. inspectors, investigators from the Office of the Special Assistant for Gulf War Illnesses and the CIA undertook the task of assessing the possibility of mustard agent release. Using imagery and strike reports, the CIA assessed that the damage to the rounds could have occurred during two separate Coalition airstrikes, the first on January 20, 1991, and the second on the night of February 13/14, 1991.

In the January strike, the bombing started a large fire that was indicated by a massive soot and debris footprint. The CIA assessed that this fire could have caused the damage to the 104 burned rounds the United Nations found at Fallujah.

In the February strike, the CIA assessed that a bomb may have struck one of the shell stacks, punched through and detonated in the ground beneath it. Some rounds were possibly destroyed, but little or no burn damage was caused. According to CIA tests, approximately 560 rounds may have fallen from the stack into the bomb crater, possibly damaging some shells and causing the mustard agent to leak.

Using the premise that leaks did occur, the special assistant’s office and the CIA initiated modeling efforts to determine any possible exposure hazard. The CIA developed the source characteristics for the models by establishing what agent was released, how, how much, and at what rate. Modelers deliberately chose numbers that assumed the highest possible amount of released agent consistent with U.N. discoveries at Fallujah and Ukhaydir. In doing this, they ensured that it would be extremely unlikely for future inspections to discover more mustard was released than what was modeled. Another aspect of this conservative approach was to ignore the effects that normal decay or exposure to light might have in minimizing the extent of the exposure area.

"We employed four weather models and two dispersion models and accounted for wind as a factor in chemical warfare agent movement," said Graf.

Four different weather models were used to derive detailed information about conditions around Ukhaydir during the periods in question. This information was plugged into two different transport/diffusion models to assess the possible spread of the mustard agent. In addition to the weather data, the transport/diffusion models used the source characterization data developed by the CIA and dosage information.

Dosage is defined as concentration integrated over a specific period of exposure. The DoD used the "general population limit" to model the area of potential exposure. The general population limit is the concentration below which the general population, including children and the elderly, could remain indefinitely with no effects. The general population limit dosage for mustard is 0.432 milligrams/minute per cubic meter of air.

The hazard areas for Ukhaydir were created by running three different combinations of weather and diffusion simulations. The results of the three runs were overlaid to create a composite hazard area. The outer perimeter of this hazard area represented the possibility for exposures above the general population limit. This perimeter did not exceed 125 kilometers distance from Ukhaydir, making it unlikely that U.S. service members, located several hundred kilometers away, were exposed.

In 1999, the CIA revisited the evidence and modified their assessments. They no longer considered the January 20, 1991, airstrike and bunker fire a case for mustard release and included a description of five possible locations and times as possible explanations for the burned rounds found at Fallujah. They further assessed that the February attack was unlikely to be responsible for damage to shells due to the lack of damage evidence collected during UNSCOM inspections. However, as they still assess that the damaged rounds at Fallujah were stored at Ukhaydir during the two coalition airstrikes, it is impossible to say that the damage didn’t occur at Ukhaydir.

Without additional evidence, the case narrative concludes, it is not possible to resolve the issue of mustard release from Ukhaydir to a certainty. However, even if a mustard release occurred, the modeling conducted in 1997 demonstrates that the hazard area was located far from U.S. forces.

"This investigation remains open," said Graf. "Anyone with new or additional information should call the direct veterans’ hotline at (800) 497-6261 or write via e-mail at"