Dr. Lashof, Members of the Committee, I am pleased to appear before you today to discuss our efforts related to reports that US troops may have been exposed to chemical warfare agents in the Persian Gulf. We strongly believe that the committee is making an important contribution to the public's understanding of this issue, and we have given high priority to your requests for support. We know how important this issue is to the Gulf War Veterans, and that our intelligence is essential to understanding what occurred during the war.
This afternoon, I will review our past modeling efforts, provide a report of recent intelligence support to efforts to model the release from the Khamisiyah pit area, and explain our plans for the future.
Last year we identified and modeled chemical releases from three locations. We were able to model these events largely because we had test data that established a reasonable upper bound for the release of agent during the events, as well as other contributing information:
The modeling for the pit poses far more difficult challenges. Let me lead with a review of the areas of greatest uncertainty. As we briefed them to you in November, they are:
We have been aggressively analyzing any thread of information related to these uncertainties in order to more accurately identify the extent of the release. I will discuss each individually.
Number of events. We remain uncertain about many of the activities surrounding the demolition of chemical rockets in the pit area at Khamisiyah. Nonetheless we have constructed a scenario that is most consistent with the information we have acquired so far. Late last year we reported the possibility that there were two demolition events in the pit; we now believe this to be more likely than a single event. Let me explain why.
We have spoken with two of the soldiers who performed demolition activity in the pit area. Both reported performing demolition on March 10. However, they recall seeing different numbers of stacks. One states that there were nine stacks of rockets and the other states that there were three. We have determined there were actually a total of 13. Interestingly, SPOT imagery confirms that demolition activity occurred at 12 of the 13 stacks. A video of the area that UNSCOM has recently made available to us confirms that demolition occurred at both ends of the pit.
A log entry states that 840 rockets (which corresponds with nine stacks) were detonated on 12 March 1991. The soldier who reported he rigged three stacks left the Khamisiyah area after the 10th, so he could not have been involved in the demolition of 840 rockets on the 12th. Both soldiers indicated that they saw no other stacks of munitions and no other individuals performing demolition activities.
The pit is 400 meters long. It is unlikely that the soldiers could have been performing demolition activities at the same time and not have seen each other. We therefore currently assess that they performed demolition activities in two different areas of the pit, on two different sets of stacks, and most likely on two different days. While there is still a possibility of a single or even additional events, the assessment of two events is most consistent with currently available information. If additional veterans recall participating in this activity, and can supply further information, we would be able to increase our confidence in this assessment.
In fact, in conjunction with this presentation, we have prepared an information paper in coordination with DOD's Office of the Special Assistant for Gulf War Illnesses. This paper briefly describes what we know and assess about activities at the Khamisiyah pit area. It also provides photographs in an effort to help locate any veterans who can help shed light on the activities there in March 1991. We have numerous copies here available for the committee, veterans, media and public.
Number of rockets involved. Based on imagery and a log entry, we have also reassessed the total number of rockets in the pit to be 1400, vice 1300. This new assessment increases our estimate of the maximum number of rockets destroyed or damaged in the pit to 650.
Recalling the uncertainties I mentioned earlier, and using the previously mentioned assessments as a guide, we estimate that the maximum number of rockets destroyed on March 10 and 12 were 260 and 390, respectively. We are working hard to use the UNSCOM video and photographs to refine these estimates further. A number of factors support our assessment that some of the rockets involved in the demolitions did not release agent. They include photographs of post-demolition rockets showing little or no damage, burial of rockets due to the demolition, and comments by the demolition experts that there were not enough charges to complete the demolition even with the expected secondary explosion of the rockets' high-explosive warheads.
Weather conditions. The Institute for Defense Analysis (IDA) is managing the effort to determine the weather conditions at the time of the pit demolition operations. To support their efforts, we have sought to accurately estimate one key variable--the wind vector at detonation. Our exhaustive search of available photography has resulted in one promising lead, which we will analyze thoroughly. Recently acquired photographs from April 1991 show soot marks near the destroyed bunkers at Khamisiyah. Since earlier studies identified the dates when these bunkers were destroyed, we soon may be able to confidently estimate the initial wind vector for the 10 March event.
No relevant testing. We do not know how chemical rockets would have been affected by an open-pit demolition. This is perhaps the greatest uncertainty we have. It affects our ability to determine how many rockets would have released agent, how much agent would have been released, the immediate dispersion patterns, the duration of the release, and any degradation factors directly related to the demolition rather than to subsequent atmospheric effects. For example, leakage and soil effects alone could affect the projection of the amount of agent released by a factor of a hundred.
Given the importance of significantly reducing this uncertainty in order to complete a meaningful model, we are currently helping the Department of Defense develop tests to destroy rockets containing CW agent simulants. We expect such testing to provide us with some of the ground truth confidence we were able to apply to modeling the chemical release at the bunker demolition.
We will continue working to reduce the uncertainties in the inputs for modeling the demolition activities at the Khamisiyah pit area. We now expect that our current analytic effort will enable us to refine input parameters and to make modeling possible, although some uncertainties will still remain. These efforts to develop meaningful inputs for a model will take time. We and the Department of Defense are expediting the effort. We will be providing our inputs to DOD and the Institute for Defense Analysis. We will also keep your Committee informed of our progress.
Let me assure you that we are doing all we can to characterize the events at Khamisiyah. We will continue to pursue this analysis on a priority basis.
I would like to close by emphasizing that the Acting Director of Central Intelligence has underscored his commitment to the American public to leave no stone unturned. The efforts I have discussed here are but a part of his initiative.