BACKGROUND

Memory

14. More than eight years have now elapsed since the Gulf conflict and memories have faded. Moreover, in the intervening period there has been considerable media coverage of Gulf veterans’ illnesses and the events surrounding the Khamisiyah demolitions, which may tend to influence individuals’ recollections. Accordingly, GVIU sought, so far as possible, to obtain corroboration from unconnected sources for all the data presented in this paper.

32 Field Hospital in the Gulf Conflict

15. Field Hospitals are located in the casualty evacuation chain between the Field Ambulance units and the General Hospital. Their role is to provide life and limb saving surgery and resuscitation. During the Gulf conflict 22 and 32 Field Hospitals (FH) fulfilled this role.

16. 32 FH's specific task was to provide third line medical support to 1 (UK) Armd Div, which was going to take part in the coalition attack across the Iraqi border to the northwest of Hafar al-Batin and then swing right into Kuwait itself. It was a regular Army unit which, for service on Op GRANBY, was formed from a core of some 127 regular Army personnel from British Military Hospital Hanover. However, it received significant reinforcements from the other regular Army units, the Royal Navy, Royal Marines, Royal Air Force, Territorial Army (TA) and Reserves in order to bring it up to deployment strength. In total, 32 FH deployed with 529 personnel including: 155 from the Royal Army Medical Corps (RAMC); 108 from the Queen Alexandra’s Royal Army Nursing Corps (QARANC); 146 individual reinforcements from other Corps of the Army; 37 from the RN; 33 from the Royal Marine Commando; 42 from the RAF; 2 civilians from the Red Cross/St John’s Ambulance; and 6 Kuwait interpreters. Of the Army personnel listed above, the nominal roll for 8 January 1991 (before the hospital deployed to the Gulf) records that some 68 were reservists and a further 34 were TA personnel. It is not known whether any further TA personnel or reservists joined the hospital after this date.

17. On 30 December a site located some eight kilometres (km) east of Al Qaysumah in northern Saudi Arabia at grid reference PS181380 was selected for 32 FH, which was planned to be a 200 bed tented hospital facility. This was some five km north east of the Trans-Arabian Pipeline (Tapline) Road, (1) approximately 80km from the Iraqi border and some 45km from the Forward Force Maintenance Area (FFMA) also located on the Tapline Road. (A map showing these locations and other places mentioned in this report is at Annex A). 32 FH remained in this location until after the conflict when it was dismantled and the staff moved back to Al Jubayl prior to their return to Europe. The hospital became operational from 0600 hours 20 January 1991, although a number of facilities, such as the tented accommodation and some of the wards remained to be completed at that time.

18. In common with all other UK and coalition medical facilities, 32 FH treated far fewer casualties than had been expected. The majority of cases treated came into the category of Disease and Non-Battle Injuries (DNBI), i.e. sickness, dental cases or injury through accidents prior to the commencement of the ground war. It received its first patient on 20 January 1991 and some 1250 patients over the period 19 January to 23 March 1991. 32 FH treated the majority of battle casualties to pass through the UK medical chain during the ground war (24-28 February 1991). In total, some 160 battle casualties were treated, the majority of which were Iraqi prisoners of war (PWs). (2)

19. On 31 January, the 32 FH Unit Log notes that two Canadian Advanced Surgical Centres (ASCs) would be attached to 32 FH. In the event, the two Canadian ASCs formed the advanced party of a 100 bed field hospital collocated with 32 FH. (3) This was based on the core of the 1 Canadian Field Surgical Hospital and was guarded by a mechanised infantry company, C Company, 1 Battalion, the Royal Canadian Regiment of London. 1 Canadian FH was co-located with 32 FH throughout its deployment in the Gulf. UK records dated 9 March 1991 state that the total Canadian deployment on the site was 530 (Canadian sources give a total of 536). On 5 March 1991, the 32 FH Commander’s Diary records that all Canadian personnel left the 32 FH site. From here they travelled to Al Jubayl from where they were flown back to Canada. A plan of the 32 FH/1 Canadian FH site is at Annex B.

PROTECTION AGAINST CHEMICAL AND BIOLOGICAL WEAPONS

20. As a result of the threat posed by Iraqi chemical and biological weapons (CBW), UK personnel participating in the Gulf conflict were generally issued with at least one full set of Individual Protective Equipment (IPE) (4) before departing for the theatre. Any shortfalls in this equipment were to be made up from in-theatre stocks. Interviews with 32 FH personnel, including the Quartermaster (QM) and Regimental Quarter Master Sergeant (RQMS), confirmed that this was the case at 32 FH, where most personnel arrived with their own issue IPE. (5) On arrival in theatre, 32 FH personnel were issued with Nerve Agent Pretreatment Sets (NAPS) tablets, and three Combopen auto-injectors for the post-attack treatment of nerve agent exposure. (6)

21. Once 32 FH was fully deployed the resuscitation (resus) department and two theatres of the surgery department were housed in tent-like structures providing permanent Collective Protection (COLPRO) (7) against nuclear, biological and chemical (NBC) warfare. This enabled staff to work without the encumbrance of IPE. Appropriate COLPRO is needed for both patients and staff and hence formed a central part of the hospital's structure. In the case of 32 FH, a signal dated 17 November 1990 from 7 Armoured Brigade to the British Army on the Rhine (BAOR), stated that 32 FH was to have two complete Field Hospital COLPRO sets plus half a set as replacements. These sets would have consisted of Porton Liners, which is the standard COLPRO used to construct Field Hospitals. Further COLPRO in the form of Winterbourne Liners was deployed elsewhere in the hospital (details of the COLPRO equipment known to be available to 32 FH are attached at Annex C).

Chemical Warfare Agent: Detection and Monitoring

22. All UK Army units are directed to have personnel trained in various nuclear, biological and chemical (NBC) warfare duties. During the Gulf conflict HQ British Forces Middle East (BFME) issued an NBC Directive which set out a concept of operations in accordance with extant doctrine that required all UK units to establish an NBC Control Party capable of co-ordinating all aspects of NBC defence. (8) In accordance with the HQ BFME NBC Directive, 32 FH had established a dedicated NBC cell of two NCOs who were responsible for monitoring and maintaining the alarm systems used to detect a chemical attack. The cell was equipped with Nerve Agent Immobilised enzyme Alarm and Detectors (NAIADs), Residual Vapour Detectors (RVDs) and Chemical Agent Monitors (CAMs). (9) A separate dedicated unit operating at theatre level was responsible for biological warfare agent detection. (10)

32 FH after the ground war

23. At 0500 hours on 28 February 1991, a temporary cease-fire was declared and discussions commenced between the Coalition and the Iraqi Government concerning a permanent cease-fire. The 32 FH NBC cell personnel recall requesting the CO of 32 FH to approve the NBC cell’s stand down at 2015 that evening because it was believed that Iraqi military forces no longer presented a threat. The CO granted this request and all NAIAD alarm systems were switched off. The recollections of various members of 32 FH indicate that from this point on, unit personnel were no longer required to carry their IPE at all times, as had previously been the case. This activity was conducted without approval from HQ 1 (UK) Armoured Division. In fact, a review of contemporary war diaries, unit logs and signal traffic suggests that no official order was issued to units in the Gulf instructing them to cease or modify NBC monitoring activities. The Force Maintenance Area (FMA) (see paragraph 29 below) issued the first signal on this subject on 7 March 1991.

24. Some 32 FH staff believe that all the hospital's COLPRO was removed and packed away on 28 February. However, the Commander's Diary entry for 7 March 1991 states that plans were being made to collapse the hospital's tentage. The unit’s Diary records that by 8 March 1991 a third of the hospital's tentage had been dropped by this date including Wards One and Two, Reception, Resus and part of the main corridor. While no mention of COLPRO is made in this entry, it is probable that the COLPRO for these areas had also been dismantled. The recollections of those who formed the unit Rear Party, who were responsible for the final dismantling of the hospital complex, confirm that at least part of the COLPRO was still in situ on 20 March 1991.

25. On the morning of 1 March 1991, staff from 32 FH NBC cell packed away the unit's NAIADs systems (as noted above, these had been switched off at 2015 the night before) along with other detection and monitoring equipment (RVD and CAM). At 1025 hrs, HQ 1 (UK) Armd Div issued an order which stated that the threat of chemical or biological attack was considered to be low and that personnel were to cease taking NAPS tablets.

26. On 4 March 1991, 32 FH received orders to reduce its size to 100 beds and 4 surgical teams. It is possible that unused/unnecessary items of COLPRO were packed away at this time as one member of 32 FH has suggested. Another member of 32 FH recalls taking two lorry loads of stores, including tentage and COLPRO, to Al Jubayl on 4 and 5 March 1991.

27. On 6 March 1991, 32 FH staff began to prepare equipment at the hospital for the return to Europe. Wards One and Two were emptied and the hospital reception closed. The Treatment and Resus Departments were moved to the Evacuation Department. From this time on, the main thrust of 32 FH’s efforts was focused on preparations for their return to Europe.

28. On 7 March 1991, HQ 1 (UK) Armd Div issued an order to all units under its command reiterating that the threat of chemical or biological attack was low and stated that 1 (UK) Armd Div personnel were no longer required to carry any IPE.

29. The FMA based in Al Jubayl also issued a signal the same day (7 March 1991) to all unit NBC Officers/Senior NCOs and Quarter Masters regarding the recovery of NBC kit from British units in the Gulf. This signal stated that all NBC equipment with the exception of IPE, RVD and Decontamination Apparatus Portable No2 (DAP 2) should be handed over to 62 Ordnance Company (62 Ord Coy) (11) for onward transmission to Europe. Opportunities to hand NBC equipment over to 62 Ord Coy were to be taken when units moved to Al Jubayl, prior to their return to the UK or Germany. Therefore, 32 FH and all other Army units would have handed over their NAIADs and CAMs to 62 Ord Coy on or after this date as part of their move back to the UK.

30. On 8 March 1991, 32 FH was instructed to reduce to a 50 bed capacity. On the same day, the UK Joint Headquarters (UK JHQ) at High Wycombe cancelled its kitting instructions for the operation. This meant that full IPE was no longer to be issued to troops deploying to the Gulf as a matter of course; (12) rather they were to deploy with just their respirators. (13) All other NBC items were to be issued as required from theatre stocks. While 32 FH's QM could not remember the precise date on which he recalled all used and spare IPE, it appears likely that he did so on the 8 March 1991. The recollections of a number of 32 FH personnel suggest that hospital staff retained their respirators and a pre-packed set of IPE, although it is possible that some individuals retained only their respirators. There are varying recollections of the disposal of used IPE. The QM recalls that it was burnt, however, another member of 32 FH states that the charcoal liners of some NBC suits were removed so that the suit could be used to clothe prisoners of war (PWs). As noted earlier, a third of the hospital’s tentage had been dismantled by this date (8 March).

31. Thus by 11 March 1991, in common with most if not all of the other UK troops in theatre, it would appear that 32 FH staff would have retained their respirators and one unused NBC suit, but would not have carried them as a matter of course. They would have ceased taking NAPS tablets. NAIADs, CAMs and RVD would not have been in use. The COLPRO at 32 FH was partially dismantled.

THE KHAMISIYAH INCIDENT

Post-Conflict Demolition Activity

32. At the end of the Gulf conflict, action was taken to ensure that the Iraqi military equipment and facilities that had been overrun were not re-utilised once the coalition forces had withdrawn. Captured Iraqi military equipment that was not to be retained by coalition forces was destroyed and occupied military facilities were demolished. This activity was conducted as quickly as possible in order to meet respective deadlines for coalition forces to return home.

Demolitions at Khamisiyah

33. The Khamisiyah Ammunition Storage Facility was a site of some 25 square kilometres. It consisted of two sections; one of 88 warehouses, the other of 100 hardened concrete bunkers. An earth berm and security fencing surrounded the latter. It was located some 350km south east of Baghdad, some 200km north west of Kuwait City and some 270km north of 32 FH at Al Qaysumah. (14)

34. The demolition activities at the Khamisiyah Ammunition Storage Facility in March 1991 have been the subject of extensive investigations by the US Government. The findings have been published in a number of reports; the most important of which are listed at Annex D.

35. No UK units were involved in the Khamisiyah demolitions, nor were there any operating nearby at the time. (15) The discussion below draws heavily upon the US reports and it is intended only as a summary of the available information. Readers seeking additional detail should consult the US source reports.

36. The US units involved in the demolition activity at Khamisiyah were aware of Iraq's chemical weapons capability and the possibility that chemical munitions might be encountered without warning. On 1 March 1991, US forces captured the Khamisiyah Ammunition Storage Facility. An initial reconnaissance of the facility was conducted which included a cursory inventory of the contents of the permanent concrete bunkers located at the site. US Engineers also destroyed Iraqi anti-aircraft emplacements located around the Khamisiyah facility at this time.

37. During 2 and 3 March, a further reconnaissance of the Khamisiyah Ammunition Storage Facility was conducted by US troops carrying M8 Chemical Alarms. (16) They also undertook a visual check and inventory of the munitions found. Those conducting this task had been instructed that Iraqi chemical munitions were marked with coloured bands, like those used by former Warsaw Pact forces, or with a skull and crossbones. No such markings were found on these munitions and preparations were therefore made for demolishing the bunkers on the basis that only conventional munitions had been found there.

38. In fact, Iraqi chemical munitions were largely unmarked and the average coalition soldier would have been unable to distinguish such chemical munitions from those of a conventional type. The US investigations have established that senior US commanders were aware at the time of the Gulf conflict that Iraqi chemical munitions might be unmarked, but that this information had not been passed down to those conducting the demolitions at Khamisiyah. At this time, the UK was aware that Iraq had historically marked its CW munitions with either one or two coloured bands or they had not marked them at all. However, the UK was not aware that the Khamisiyah Ammunition Storage Facility contained CW munitions. Had the troops undertaking demolitions at Khamisiyah known that they were dealing with chemical weapons, they would have used very different disposal techniques. In the event, they adopted standard procedures for destroying conventional munitions and therefore no special warning of their activities was given to other US forces or their coalition partners.

4 March demolitions

39. At 1400 hours on 4 March 1991, US troops conducted the first demolitions at Khamisiyah, destroying a number of bunkers including what subsequently became known as "Bunker 73". Immediately after these demolitions, US M8 CW alarms deployed at Khamisiyah sounded an alert and in accordance with US Mission Oriented Protective Posture (MOPP) (17) procedures, IPE was donned by those involved. However, follow-up tests conducted with US M256A1 (18) Chemical Agent Detector kits suggested that the M8 had given a false alarm. Therefore, no further action was taken and the demolitions continued without IPE being worn.

10 March demolitions

40. On 9 March 1991, an open pit adjacent to the Khamisiyah complex was found to contain stacks of crated 122mm rockets. As they exhibited none of the markings that the US forces expected on chemical munitions, the rockets were judged to have conventional warheads. At approximately 1600 hrs on 10 March 1991 these rockets, along with most of the remaining warehouses at the site, were destroyed. The US forces conducting the demolitions had already left Khamisiyah and were 30 minutes' travelling time away when the charges went off.

41. Some time between 11 and 13 March 1991, two separate groups comprising two US servicemen each returned to Khamisiyah to take photographs. Both groups stood in the pit containing the 122mm rockets without wearing any IPE. Both groups noted that some of the bunkers at Khamisiyah were still intact due to failed demolition charges. At the time, none of these individuals showed any immediate symptoms that might be associated with exposure to nerve agent. When interviewed by the US Department of the Army Inspector General in 1997, these individuals had not developed any subsequent health problems.

43. On 18 April 1991, Iraq made its first declaration concerning weapons of mass destruction to the United Nations (UN). This did not mention CW munitions being stored at the Khamisiyah Ammunition Storage Facility.

44. The first indications that CW munitions had been located at Khamisiyah appeared in Iraq's declaration to the UN on 16 May 1991, in which Iraq listed 2,160 122mm artillery rockets as having been destroyed at Khamisiyah. They also listed some 6,240 intact 155mm artillery shells as being stored at "Khamisiyah Stores (Nasiriyah)".

45. During the UNSCOM 9 inspection from 15 to 22 August 1991, Iraq stated that Coalition troops still occupied Khamisiyah on 18 April 1991 and that Iraq was unable to account for the CW munitions at this site until after the Coalition forces had departed (in fact US forces did not depart from the Khamisiyah area until late April).

46. On 26 October 1991, UNSCOM 20 inspected the Khamisiyah Ammunition Storage Facility. Whilst there, their Iraqi guides stated that the facility had previously housed CW munitions that had been destroyed by Coalition forces. On 27 October UNSCOM 20 recovered an intact 122mm artillery rocket from the area known as "the pit". This was found to contain sarin/cyclosarin nerve agent. UNSCOM 20 also conducted tests with a Chemical Agent Monitor (CAM) at Bunker 73, but failed to detect the presence of nerve agent. UNSCOM 20 was also shown the storage site of over 6000 intact 155mm artillery rounds containing Mustard gas. These were located some 2km to the west of the Khamisiyah Ammunition Storage Facility. This site had not been found by US forces when they had occupied the area.

47. A further inspection of Khamisiyah on 14 May 1996 by UNSCOM representatives confirmed the presence of CW munitions in Bunker 73. This was based on an examination of munition debris found within the Bunker.

US INVESTIGATIONS

48. On 21 June 1996 the US Government announced that it had been established that Iraqi chemical weapons had been present at Khamisiyah in March 1991 and destroyed by US troops during demolition operations. The initial US assessment was that the area of possible exposure to nerve agent as a result of the demolitions was within a radius of 50km of the Khamisiyah depot.

49. The US authorities subsequently undertook a more detailed investigation of the circumstances of the demolitions, including experimental destruction of simulated Iraqi 122mm rockets to measure the possible dispersion of chemical warfare agent. This found that, as a result of the demolitions, a portion of the nerve agent present in the munitions would have been instantly burned and destroyed by the explosive charges. Another portion would have been aerosolised and released into the atmosphere. The remaining agent would have spilled out onto the surrounding area, and been partially absorbed by the remains of the wooden crates and the soil. That portion of the agent released into the atmosphere by the force of the demolition explosions and that subsequently released from the wood and the soil by evaporation would become subject to the dynamics of the wind and other meteorological conditions.

Modelling the Plumes

50. Firstly, it is important to note that all references to nerve agent exposures are entirely theoretical and based on a mathematical computer model. Exposures might have happened, but there is no evidence that any actually did happen.

51. In order to understand the extent of potential exposure to nerve agent caused by the Khamisiyah demolitions, the US Department of Defense (DoD) generated models of the possible dispersion of nerve agent over time. For 4 March 1991, when Bunker 73 was blown up, the DoD modelling suggested that the likely movement of the resulting plume of nerve agent would have been to the east and north east, away from US troops in the area. This would also have been away from other coalition forces, including those such as 32 FH at Al Qaysumah in Northern Saudi Arabia.

52. The DoD modelling of the demolitions which took place on 10 March 1991 suggested that a larger modelled plume containing low concentrations of nerve agent could have been driven by the prevailing wind some 300km south, into Saudi Arabia by the end of that day (10 March). Wind changes on 11 March 1991 result in the plume shifting towards the west, but by 12 March 1991 the DoD modelling suggests that a modelled plume would have been travelling north west, along the Euphrates valley. The modelled plume did not cover Kuwait City, nor did it cover Al Jubayl or Dhahran in Saudi Arabia. However, it appears that an area of northern Saudi Arabia (including Hafar al Batin, the King Khalid Military City (KKMC), and Al Qaysumah in Saudi Arabia) could have been within the possible modelled plume on 11 March 1991. This area included the location of 32 FH and other UK forces.

53. The results of the DoD modelling, including maps showing the modelled plume were published in September 1997 (19) and are reproduced, with permission, at Annex E with the positions of UK units shown (where known).

Limitations of the DoD modelling

54. By its nature, the process of modelling weather systems is far from exact; the dynamic and often unpredictable nature of weather systems means that there are inherent limitations in atmospheric modelling. The availability of only a relatively small amount of contemporary meteorological data at the time that the modelling was undertaken also reduced the level of refinement that the US DoD was able to achieve.

55. The DoD modelling provides the only information to suggest that staff at 32 FH (and possibly other UK units in the Gulf) may have been exposed to a very low concentration of nerve agent between 10/11 March 1991. This DoD work was a major effort aimed at assessing the maximum possible size of the population potentially at risk. Whilst every effort was made by DoD to ensure that this modelling was rigorous, it was nevertheless subject to certain limitations (as described above), meaning that there is no conclusive evidence that exposure to nerve agent ever occurred at any given time or place beyond the immediate environs of Khamisiyah itself.

56. The DoD modelling does not represent a cloud of material that was actually observed at the time of the demolitions, nor is there any evidence that such a plume in fact ever existed. So far as can be discovered, no chemical alarm was sounded by any British unit after 26 February 1991. However, since UK units had been given instructions on how to pack away their CW detection and monitoring equipment in preparation for their return to Europe on 7 March 1991, it is most unlikely that any UK units would have had an operating NBC Cell with active CW agent alarm systems by 10/11 March 1991. Nor were any UK personnel reported as having symptoms of nerve agent poisoning at that time (10/11 March).

57. The DoD used mathematically derived computer models to generate a theoretical picture of what might have occurred during and after the demolitions on 10 March 1991. As noted in the DoD report itself, the maps of the modelled plume which were published in September 1997 represented a composite of five different meteorological/dispersion simulations, so that the shape of the modelled plume on the map represents the outermost perimeter of all the models combined. The report makes it clear that the DoD would not expect everyone who was present in the areas under the modelled composite plume footprint, as depicted by the report, to have been exposed to nerve agent. Additionally, it should be noted that the DoD modelling does not take into account the environmental degradation of the nerve agent in question. Sarin is relatively unstable and degrades rapidly once exposed to the open air. Hence by presenting the modelling in the form shown on the published maps, DoD made available the sum of a number of worst case projections. On the one hand this ensured that the risk was not underestimated, but at the same time it did present a highly unlikely version of events. The model has been criticised for these reasons in the US.

58. On 1 September 1998, the US Senate Committee on Veterans Affairs published a Report of the Special Investigation Unit into Gulf War Illnesses. This report concluded that the DoD/CIA model was likely to have considerably over estimated the area in which the sarin and cyclosarin that may have been released by the Khamisiyah demolitions could have been dispersed. This was attributed to flaws in the modelling of the demolitions, and a failure to consider all available meteorological data and modelling expertise. The report included an alternative model for the plume, prepared by the US Air Force Technical Assistance Centre. This modelled plume’s footprint covers a much smaller area, extends further east into Kuwait, and far less south into Saudi Arabia than the DoD model and is reproduced at Annex F. (20)

59. However, the US Air Force modelling represents a single approach to the depiction of weather and agent dispersion; the US DoD model represents a combination of five independently derived models and as such is less a model than an aggregation. The aggregation of the five models created a far larger composite plume footprint than is supported by the data used to build its individual component models. A scientific comparison between the results of the two differing approaches to the modelling is therefore not possible.

60. The US DoD has responded to these criticisms of its modelling effort. In its Second Annual Report published in November 1998, the Office of the Special Assistant for Gulf War Illnesses (OSAGWI) stated that, because of differing assumptions and methodologies, differences were expected between different models. It was important to note that modelling was not reality and that no model could assure absolute certainty. The important point was whether the results of the modelling were scientifically credible. The US DoD modelling had been reviewed by a panel of experts from both within and without the US Government and their view was that the results of the modelling were credible. That being the case, OSAGWI had decided to concentrate on refining the assumptions and methodology of the US DoD modelling rather than incorporate new and different models and their results.

61. In evidence to the Presidential Special Oversight Board on 13 July 1999, DoD and the CIA acknowledged that re-modelling of the demolitions and subsequent plume footprint was underway. More information is available to reduce uncertainties and allow a move away from the combination of worst case assumptions which underlay the 1997 paper. Personnel involved in the demolitions have stated that the demolition charges were less optimally placed; UNSCOM accounting and photography indicated that a much smaller number of rockets were destroyed; and photographs taken by soldiers at the time also show a smaller percentage of damaged rockets. Therefore, the current US DoD assessment is that less nerve agent was released than previously assumed, and environmental degradation of the volatile agent was now calculable. Probably fewer coalition troops were under the modelled plume footprint and hence potentially exposed. In addition, detailed meteorological data supplied by the UK Meteorological Office will also be incorporated into the re-modelling. We await details of the new model with interest. Once it is available, MOD will re-examine the findings in this paper to confirm which UK units might theoretically have been exposed.

62. More recently, on 1 September 1999, the American Journal of Epidemiology published a scientific paper titled "The Postwar Hospitalization Experience of Gulf War Veterans Possibly Exposed to Chemical Munitions Destruction at Khamisiyah, Iraq". (21) This paper examined the post-war hospitalisation of US Gulf veterans who were possibly exposed to nerve agent as a result of the Khamisiyah demolitions in order to establish whether they suffered any greater incidence of ill-health (see paragraph 64 below). While the paper relies in part on the DoD modelling of September 1997, it is noted that those responsible for the paper also commissioned their own "epidemiological plume" model which was constructed from a combination of the best meteorological and dispersion models in order to establish unit-specific dose estimation (the dispersion model actually used was one of the group of five used by the US DoD). Once completed, the "epidemiological plume" modelling differed considerably from the US DoD modelling in that the plumes generated on the specific days in question (10/11 March) generally covered a smaller area and were located further to the west, away from UK units. The paper notes that this was because an updated regional meteorological model was used with one of the three dispersion models. However, the paper also states that in future the US DoD plans to redefine the 1997 modelling to include the "epidemiological plume". The results of the "epidemiological plume" modelling cannot be reproduced here for reasons of copyright.

63. Again, as with the US Air Force modelling (see paragraph 59), the "epidemiological plume" represents a single approach to the depiction of weather and agent dispersion whereas the US DoD model represents a combination of five independently derived models. A scientific comparison between the results of the two differing approaches to the modelling is therefore not possible.

POTENTIAL EXPOSURE EFFECTS BASED ON THE DOD MODEL

64. The UK requested from the US detailed models of potential exposure for five specific sites chosen because they represented the locations where British units were, or may have been on 10/11 March (see map at Annex E). For example, the 3rd Battalion, Royal Regiment of Fusiliers was chosen as the most westerly location of the 4 Armoured Brigade concentration in central Kuwait, and 91 Ordnance Company as representative of units in the Forward Force Maintenance Area (FFMA). Graphs showing the modelled sarin concentration for each site are at Annex G. These graphs show the modelled concentration in milligrams (mg) of sarin per cubic metre (m-3) over time (min). By multiplying the level of concentration and its duration (the area under the solid line) one arrives at a total dosage, expressed as milligram minutes per cubic metre (mg.min.m-3). Based on the information provided by the US DoD from their modelling studies, the best current estimate of the dosages were:

Unit 5b
Elements of 16/5 Lancers (Medium Recce)
0.01950 mg.min.m-3

Unit 44a
3rd Battalion Royal Regiment of Fusiliers
0.00161 mg.min.m-3

Unit 81b
91 Ordnance Company FFMA detachment
0.00000052 mg.min.m-3

Unit 104
32 Field Hospital
0.02527 mg.min.m-3

Unit 113f
174 Company Royal Military Police, TP4
0.1327 mg.min.m-3

65. The dosage of sarin that it is estimated would kill 50% of the human population (known as the LCt50) is between 50 and 70 mg.min.m-3. The first noticeable effects of exposure to sarin can be expected between 0.5 and 2 mg.min.m-3. In the maps at Annex E the US have adopted 1 mg.min.m-3 as the contour for first noticeable effects and 0.1296 mg.min.m-3 as the "General Population Limit". This represents a dosage of sarin/cyclosarin which the general population (including babies and old people) could be expected to remain exposed to for 72 hours with no effects. At 0.5 mg.min.m-3 the only effect that would be produced is a barely discernible narrowing of the pupil of the eye, called miosis. Only a trained opthamologist could detect this change, and below this dosage there is no detectable effect.

66. This 0.5 mg.min.m-3 dosage for the first noticeable effect is 3.6 times higher than the highest modelled dosage for a representative UK unit location (i.e. 0.1327 mg.min.m-3 for 174 Coy, Royal Military Police), and almost 20 times higher than the dosage modelled for 32 FH. For 16/5 Lancers it is nearly 30 times higher.

67. The actual dose (or quantity) which is inhaled by an individual exposed to a given dosage will depend on the breathing rate and level of activity. A man walking at 2.5mph would have a breathing rate of approximately 30 litres per minute. A man walking through a lethal dosage of sarin at this rate (assuming that the entire dose he inhaled was retained in his body) would inhale a dose of 1.5 to 2.25 mg of sarin. A man walking at 2.5mph through the highest modelled dosage level of sarin (0.137 mg.min.m-3), would inhale a dose of 0.004mg (4 micrograms (4m g)) of sarin (one four hundredth of the lethal dose). The modelled dosage of sarin at the site of 32 FH is 0.02527 mg.min.m-3 (the equivalent of a 0.76 m g dose). This is less than 1/2000 of the lethal dose.

68. Cyclosarin (GF) is similar to sarin, but less volatile (boiling point 239oC compared to 158oC for sarin). Toxicity estimates for cyclosarin are less robust than for sarin as less work has been done on this agent. However, it is estimated that the LCt50 for cyclosarin would be higher than sarin, between 113 and 235 mg.min.m-3. Therefore, the possible presence of cyclosarin in the modelled plume does not have a bearing on the calculations above.

Health impact of Sarin (GB)

69. Like other nerve agents, sarin acts by binding to and inhibiting the enzyme acetylcholinesterase (AChE) which is present throughout the human body. AChE is critical to the transmission of impulses between nerves within and outside the brain and from nerves to muscles, glands and other organs. When sarin is absorbed into the body, it is transported by the blood to all parts of the body, inhibits available AChE and prevents that enzyme from performing its normal function and causes the nervous system to become over-stimulated. This over-stimulation results from the excess of the transmitter substance acetylcholine (ACh) giving rise to the characteristic symptoms and signs of nerve agent poisoning. This is known as cholinesterase inhibition.

70. The characteristic symptoms of nerve agent poisoning are: tightness of the chest; headache; runny nose; excessive salivation; dimness of vision (miosis of the eye); nausea; excessive contraction and seizing up of the muscles; and collapse. Death may result from oxygen deprivation following bronchial constriction, accumulation of secretions, and paralysis of the respiratory muscles or inhibition of the central respiratory centres, aggravated by severe cardiovascular failure.

71. Sarin is the most volatile of the classic nerve agents and resembles petroleum in its vaporisation. It is categorised as a non-persistent agent. To produce any biological effect, sarin would either have to be absorbed through the skin (dermal exposure) or inhaled. In field conditions its dermal toxicity is low and can be disregarded in this case because the US model assumes a vapour cloud without liquid droplets.

72. One of the most sensitive target for the effects of sarin is the eye because of the thinness of the cornea and the ease with which the vapour may be absorbed via this route. If sarin is inhaled it will cross the air/blood interface in the respiratory tract and be transported to all parts of the body. Along the way, the molecules of sarin are subject to a range of natural processes both within the blood and in other body organs (lung, liver and kidneys for example) which would destroy them. Therefore, for sarin to reach the key sites of action in the body (the nervous system), the affected individual would have to be exposed to a sufficient dose to overwhelm these processes.

73. Historical information from experiments at CBD Porton Down shows that when humans were exposed to a dosage of 2 mg.min.m-3 (14.6 times greater than the highest US modelled exposure for UK units and 79.1 times greater than that modelled for the site of 32 FH), it produced a 3.8% inhibition of the enzyme AChE present in the blood stream. The normal variation in human blood of AChE is between + 8-10% therefore an exposure to a dosage of 0.137 mg.min.m-3 would not produce any measurable inhibition of AChE that could be distinguished from normal variation.

74. Given the above, MOD believes that the highest concentrations of sarin that may theoretically have been present in the modelled plume where UK units were located were well below that which would be expected to produce any biologically detectable effects in humans. There are no contemporary reports of any personnel serving during the Gulf War or its aftermath having any signs or effects of exposure to low doses of sarin. If such an exposure did occur, it must have been at such a low level that no biological effects were produced.

Human Toxicity of Sarin and Cyclosarin

75. The scientific evidence indicates that long term heath problems would be unlikely to result from exposure to the very low levels of nerve agent suggested by the US modelling. These levels are too low to elicit any measurable effect at the time of exposure and humans have been exposed to much higher levels of nerve agents without showing any excess of ill-health twenty years later. No biological mechanism has yet been identified which would explain longer-term ill health in instances where the acute effects of cholinesterase inhibition were much less than the natural variations which occur in the human body.

76. Moreover, the recent study published in the American Journal of Epidemiology has found that there is no evidence that US veterans possibly exposed to nerve agent as a result of the Khamisiyah demolitions have experienced unusual ill health. The study examined the hospital records of 124,487 US Gulf veterans who were identified as possibly being exposed to very low levels of nerve agent using its own "epidemiological plume" (see paragraph 58 above). These records were compared with those of 224,804 US Gulf veterans who were identified as not being exposed to nerve agent. During 54 months of observation, it was found that the rate of hospitalisations and the type of diagnoses did not vary significantly between the two groups. It was the study’s conclusion that the data do not support the hypothesis that Gulf veterans who were possibly exposed to very low levels of nerve agent as a result of the Khamisiyah demolitions experience unusual post war morbidity.

77. However, because exposure to very low levels of nerve agent is neither an observed nor an inherently plausible cause of ill health, there has been very little research into the long term effects of such an exposure. The US authorities have embarked on an extensive research programme to investigate this area further. The Ministry of Defence will monitor this work closely.

Detection Capabilities based on the DoD Modelling

78. UK alarms, detectors and monitoring equipment for CW agents operate in relation to specific concentrations of nerve agent. The crucial factor in answering the theoretical question: "Could UK equipment have detected the modelled release of very low levels of nerve agent from Khamisiyah?" is not dose or dosage, but the maximum concentration at any given time. This is represented by the highest point of the solid line on the graphs at Annex G. The highest modelled concentration is about 0.0012 mg.m-3 for 174 Company RMP. This was below the minimum concentration which the most sensitive UK device could have been expected to detect. (22) At the site of 32 FH, the modelled concentration was well below the UK’s 1991 detection capability. Although, as has already been noted above, by this date the majority of the UK’s CW agent detection and monitoring equipment had been switched off and packed away, even had it been deployed, it would not have detected the modelled concentration of nerve agent.

WHO WAS PRESENT AT 32 FH ON 11 MARCH 1991?

79. During most of March 1991, the staff of 32 FH were packing up and preparing to go home. The withdrawal of the unit's staff from theatre was a phased process and therefore it has not been possible to determine how many were still present on 11 March 1991. However, it has been possible to identify one group of staff who had left the day before and another that were still present later in the month.

80. Although the Commander’s Diary records that the provisional recovery plan required the TA members and other reservists to leave 32 FH for Al Jubayl on 11 March 1991, the recollection of a number of 32 FH staff (including members of the TA) is that the TA members and other reservists serving with 32 FH departed the hospital for Al Jubayl on the morning of 10 March 1991. They boarded coaches that drove down the Tapline Road (otherwise known as Main Supply Route DODGE) to Al Jubayl, an eight hour journey. We therefore believe that these individuals were not present at 32 FH on 11 March 1991, but were located in Al Jubayl, an area not covered by the DoD modelled plume. The implication is that most of the full time members of the unit were still present when the TA and other reservists left and were there the following day, 11 March. This would have amounted to about 400 personnel.

81. Members of the unit's Rear Party, that is, the group responsible for remaining behind and packing up the last of the unit's equipment before vacating the unit's location, recall that they were still at 32 FH on 20 March 1991. Hence those members of 32 FH who formed the Rear Party would also have been present on 11 March 1991. (23)

CONCLUSION

82. All references to nerve agent exposures are entirely theoretical and based on a mathematical computer model. Exposure might have happened, but there is no evidence at all that any actually did happen. Even if troops had been exposed to nerve agent at the exceptionally low levels modelled, the impact of this on their levels of acetylcholinesterase would have been much less than the impact of normal bodily variations. There would have been no biologically detectable effect. There is no evidence to suggest that there are adverse long term health effects from temporary exposure to the levels of sarin postulated in the model. However, there has to date been very little research in this field, and the UK is monitoring closely a substantial programme of US research which is looking at the effects of chronic and low dose exposure to chemical warfare agents.

Ministry of Defence
London
December 1999

Map - Location of 32 Field Hospital

Annex B

Sketch of the site of 32 Field Hospital

Annex C

COLLECTIVE PROTECTION EQUIPMENT KNOWN TO BE AVAILABLE TO 32 FH

1. A signal dated 17 November 1990 from 7 Armoured Brigade to the British Army on the Rhine (BAOR), where BMH Hanover was based, states that 32 FH was to have two complete Field Hospital COLPRO sets plus half a set as replacements. These sets would have comprised Porton Liners, which is the standard COLPRO used to construct Field Hospitals. Contemporary documentation shows that the following Collective Protection (COLPRO) equipment was also issued to 32 FH:

3 x Winterbourne Liners to NBC Adviser, 32 FH, for delivery care of 62 Ord Coy1 by 19 December 1990;

9 x Winterbourne Liners to NBC Adviser, 32 FH, for delivery care of 62 Ord Coy by 4 January 1990;

28 x Air Filtration Units & Spare Filters to NBC Adviser 32 FH, for delivery care of 62 Ord Coy by 25 January 1991;

7 x ARIES [product name] Air Cooling Trolleys2 (for the provision of filtered dry conditioned air) to NBC Adviser, 32 FH, for delivery care of 62 Ord Coy by 23 January 1991;

1 x ARIES Air Cooling Trolleys to NBC Adviser, 32 FH, for delivery care of 62 Ord Coy by 28 January 1991;

6 x ARIES Air Cooling Trolleys to NBC Adviser, 32 FH, for delivery care of 62 Ord Coy by mid-February 1991; and

5 x EBAC [company name] Portable Air Conditioning unit (PAC) 20 Air Conditioning Units3 to NBC Adviser, for delivery care of 62 Ord Coy by 11 January 1991.

2. In addition to COLPRO, 32 FH was also issued with 1000 full sized casualty bags on 23 February 1991. These bags were made of the same NBC resistant material as IPE and contained a built in respirator. A number of half-sized bags designed to protect just the upper body were also available (references to half sized bags unfortunately only record their delivery, not the number delivered).

Annex D

US GOVERNMENT PUBLICATIONS ON THE KHAMISIYAH INCIDENT

CIA Report on Intelligence Related to Gulf War Illnesses published by the CIA published on 2 August 1996.

Khamisiyah: A Historical Perspective on Related Intelligence published by the Director of Central Intelligence Persian Gulf War Illnesses Task Force published on 9 April 1997.

US Demolition Operations at the Khamisiyah Ammunition Storage Point published by the Special Assistant for Gulf War Illnesses published on 14 April 1997.

Reducing Uncertainties in Modeling Demolition Activities in the "Pit" at Khamisiyah published by the Director of Central Intelligence Persian Gulf War Illnesses Task Force published on 17 July 1997.

Highlights of Intelligence Warnings About Chemical Weapons at Khamisiyah published by the Director of Central Intelligence Persian Gulf War Illnesses Task Force published on 21 July 1997.

Modeling the Chemical Weapons Agent Release at the Khamisiyah Pit published by the Director of Central Intelligence Persian Gulf War Illnesses Task Force published on 24 July 1997.

Modeling the Chemical Warfare agent Release at the Khamisiyah Pit published by the CIA and DoD published on 4 September 1997.

Department of the Army Inspector General Inquiry into Demolition of Iraq Ammunition published by the Department of the Army published on 10 October 1997.

Second Annual Report of the Office of the Special Assistant for Gulf War Illnesses November 1997-November 1998 published in November 1998

OTHER RELEVANT PUBLICATIONS

The Special Investigation Unit for the US Senate Committee on Veterans Affairs report on ‘Gulf War Illnesses’ published on 27 August 19981.

American Journal of Epidemiology, Volume 150, Number Six, "The Postwar Hospitalization Experience of Gulf War Veterans Possibly Exposed to Chemical Munitions Destruction at Khamisiyah, Iraq" published on 1 September 1999.

Annex E

Map 1 - 10 March 1991

Map 2 - 11 March 1991

Map 3 - 12 March 1991