We assess that Iraq did not use chemical or biological weapons against Coalition troops based on our thorough review of intelligence reporting and on the lack of casualties that was a signature of chemical use during the Iran-Iraq war. We assess that Iraq probably did not use these weapons because of a perceived threat of overwhelming Coalition retaliation.

We assess that Iraq had chemical weapons at two sites (see figure 1) in Iraq--the An Nasiriyah Ammunition Storage Depot SW and the Khamisiyah (US name Tall al Lahm) Ammunition Storage Area--within the Kuwait Theater of Operations (KTO) (1)  during Desert Storm. Both of these sites were large rear-area depots near the northern boundary of the KTO that stored mostly conventional ammunition. UNSCOM reporting and other information indicate that Coalition bombing did not destroy the bunker containing the chemical agents temporarily stored at An Nasiriyah. We have recently determined US troops were near a release of chemical agents at Khamisiyah, and DOD is assessing potential exposure.

According to Iraqi statements to UNSCOM in May 1996, An Nasiriyah stored 6,000 155-mm mustard rounds from early January until they were moved to Khamisiyah after Feb. 15, 1991. Iraq stored the munitions starting just before the air war at one bunker--called Bunker 8 by Iraq--at An Nasiriyah Ammunition Storage Area SW. According to Iraq, these mustard rounds were moved to Khamisiyah because of fear of additional Coalition bombing.

The Coalition bombing of An Nasiriyah on Jan. 17, 1991 did not cause a release of chemical agent because the bunkers that were bombed on that date did not contain chemical agents. In May 1996, UNSCOM inspectors examined the rubble surrounding the bunkers at An Nasiriyah that were bombed on Jan. 17, 1991 and determined that the bunkers contained only conventional weapons. Although mustard rounds were in Bunker 8 at An Nasiriyah on 17 January, UNSCOM information indicates they were not damaged. No other agents were known to be at An Nasiriyah.

UNSCOM inspected chemical munitions at or near Khamisiyah in October 1991 and identified 122-mm sarin/cyclo-sarin (GB/GF) nerve-agent-filled rockets and 155-mm mustard rounds. At that time it was not clear whether these chemical weapons had been present during the Gulf war or whether, as was suspected at other locations, the Iraqis moved the munitions there shortly before the 1991 UNSCOM inspection.

During its October 1991 inspection of the Khamisiyah facility, the Iraqis told UNSCOM that Coalition troops had destroyed chemical weapons at a bunker earlier that year.(2), and UNSCOM found chemical munitions at two open sites (see figure 2):

• Remnants of 122-mm rockets were identified at a single bunker among 100 bunkers, called "Bunker 73" by Iraq. It was unclear whether the munitions in Bunker 73 were chemical because there was no sampling or positive chemical agent monitors (CAM) readings and inspectors did not document characteristic features of chemical munitions.
• Several hundred mostly intact 122-mm rockets containing nerve agent--detected by sampling and with CAMs--were found at a pit area about 1 km south of the main storage area.
• Over 6,000 intact 155-mm rounds containing mustard agent, as indicated by CAMs, were found in an open area several kilometers west of Khamisiyah.

The recent comprehensive review of all information enabled us to determine that US troops--not Iraq--destroyed the rockets in Bunker 73. In March 1996, in conjunction with DOD investigators, we determined that the US 37th Engineering Battalion had destroyed that bunker along with over 30 other bunkers on March 4, 1991.

However, it was not until UNSCOM's May 1996 inspection at Khamisiyah that it was determined that Bunker 73 contained remnants of 122-mm chemical rockets. During this inspection, inspectors documented the presence of high-density polyethylene inserts, burster tubes, fill plugs, and other features characteristic of Iraqi chemical munitions. Analysis of the contents of the rockets that UNSCOM found in 1991 in the pit area just outside the Khamisiyah Storage Area shows that the identical rockets in Bunker 73 had been filled with a combination of the agents sarin and GF. Therefore, we conclude that US troops destroyed chemical rockets in Bunker 73.

During the May 1996 UNSCOM inspection, Iraq claimed that some of the rockets located in the pit area had been destroyed by occupying forces. On the basis of very recent interviews of 37th Engineering Battalion personnel, DOD now believes that demolition personnel did set charges on stacks of rockets in the pit on 10 March 1991 at 1630 local time.

We are still trying to determine the number of rockets US forces could have destroyed. Once we determine the number, we will model the likely hazardous area caused by the destruction. Iraq told the May 1996 UNSCOM inspectors that it moved about 1,100 rockets out of Bunker 73 to the pit 2 km away to avoid chemical contamination of the bunker facility. The Iraqis claimed the rockets started leaking immediately after they were transferred from the Al Muthanna CW Production and Storage Facility just before the air war.

As discussed previously, more than 6,000 mustard rounds were moved from An Nasiriyah to an open area several kilometers west of the main facility at Khamisiyah. These munitions were found undamaged by UNSCOM in October 1991. They were later moved to and destroyed at UNSCOM's Al Muthanna destruction facility.

Modeling of the potential hazard caused by destruction of Bunker 73 indicates that an area around the bunker at least 2 km in all directions and 4 km downwind could have been contaminated at or above the level for causing acute symptoms including runny nose, headache, and miosis. An area up to 25 km downwind and 8 km wide could have been contaminated at or above the much lower general population dosage limit.(3) From wind models and observations of a video of destruction activity at Khamisiyah, we determined that the downwind direction was northeast to east.

Modeling Assumptions About Bunker 73

Some of the following modeling assumptions were based on data from US testing in 1966 that involved destruction of a bunker filled with 1,850 GB rockets with maximum range similar to that of Iraqi rockets found in Bunker 73:

• 1,060 rockets as indicated by Iraq.
• Rockets filled with 8 kg of a 2:1 ratio of GB to GF (contents assumed to be 100 percent agent) based on UNSCOM information and sampling from the pit.
• Ten percent of rockets ejected from the bunker, half of which randomly fall within a 200-meter circle, the other half falling within a 2-km circle based on US testing.(a)
• Ejected rockets released agents on impact.
• A 15-meter mean agent release height was chosen to be conservative when determining ground hazard.
• All but 2.5 percent of agent in the bunker degraded by heat from explosion and motor/crate burning based on US tests.
• Winds slow to the northeast to east, based on modeling and analysis of a videotape of the destruction activity at Khamisiyah.
• Our models do not include the effect of the reported 32 to 37 conventional ordnance bunkers detonating and burning simultaneously with the chemical bunker. The added thermal energy created by explosions and fires in the other bunkers and solar heating caused by the increased amounts of smoke would tend to degrade agent as well as more quickly disperse the agent between the ground up to the maximum altitude of 800 to 1,200 meters.(b) This more rapid vertical spreading would tend to lower ground contamination in the area.

(a) DOD documents and multiple veterans reported that munition "cook-offs"--munitions that ignite and are ejected from their storage due to the demolition fire--sent ordnance as far as 10 km or more from the bunker facility. Nonetheless, we did not model this phenomena because we have been unable to determine whether any of the cook-offs involved chemical rockets, and if so, the number of rockets and how far they went.

(b) This altitude represents the estimated height of the mixing layer--the lower turbulent part of the atmosphere above which agent transport is inhibited due to a laminar boundary layer. This layer can often be seen from aircraft while landing in cities with polluted air.

Iraq declared that 200 mustard-filled and 12 sarin-filled aerial bombs at the Muhammadiyat (US geographic name Qubaysah) Storage Area were damaged or destroyed by Coalition bombing. We have modeled the contaminated area resulting from bombing of Muhammadiyat, a site at least 410 km from US troops stationed at Rafha and even further from the bulk of U.S. troops. Bombing of this facility began on 19 January and continued throughout the air war. We have been unable to determine exactly when the chemical bombs were destroyed. On the basis of recent Iraqi declarations, we have modeled a release of 2.9 metric tons of sarin and 15 metric tons of mustard on all possible bombing dates to find the largest most southerly hazardous area. Southerly winds occurred for only a few of the days the site was bombed. For general population limit dosages (above 0.013 mg-min/m3), downwind dispersions in the general southerly direction for sarin and mustard fall below this level at about 300 and 130 km, respectively.

Neither the first effects nor the general population limit levels would have reached US troops that were stationed in Saudi Arabia.(4)

Iraq declared that 2,500 chemical rockets containing about 17 metric tons of sarin nerve agent at Al Muthanna (U.S. geographic name Samarra), the primary Iraqi CW production and storage facility, had been destroyed by Coalition bombing. UNSCOM inspectors were unable to verify the exact number because of damage to the rockets. We have modeled possible bombing dates for this bunker and determined that the most southerly dispersal for reaching the general population limit dosage is 160 km, well short of U.S. troops.