The CIA only modeled a preliminary hazard area for the possible release on the first day after the February 13/14th strike. The direction and size of this hazard area differ from the direction and size of the hazard area produced by DoDs follow-on modeling for the same time frame. (See Figures 20 and 21.) It is important to note that neither hazard area approaches US troop locations along the Iraqi-Saudi Arabian border.
Figure 20. CIA's preliminary modeling results for the February 13/14th airstrike
Figure 21. DoD's modeling results for the February 13/14th airstrike, Day 1
The differences in the shape and direction of the hazard areas derive from the fact that the CIA used only a single weather model and a single transport/diffusion model, rather than an ensemble of models, as well as the differences in the meteorological information used in the preliminary and follow-on modeling. The CIA intended the preliminary modeling to be a quick turnaround analysis providing a first look at the potential exposure area. As such, the preliminary meteorological models used only unclassified operational weather data.
As discussed in the narrative, the follow-on modeling used more comprehensive weather data. In addition to the unclassified information used in the preliminary modeling, the follow-on modeling used more detailed declassified information from US Air Force archives and took into account the local weather effects of the nearby lakes and the terrain around Ukhaydir.
The follow-on modeling also used a more recent version of the OMEGA meteorological model. At the time of the preliminary modeling (August 1997), OMEGA was undergoing substantive enhancements. The preliminary modeling used version 2.0 of the OMEGA model, while the follow-on modeling used the updated and improved version 3.5 as part of its ensemble of weather models. The differences between the preliminary and follow-on hazard areas can be attributed, in part, to the enhancements in the OMEGA meteorological model as well as to the more complete and detailed weather information in the follow-on modeling.
| First Page | Prev Page | Next Page |