5. Investigation of Iraq’s Alleged Attacks at Al Jubayl

On September 5, 1997, we met with two individuals who, after completing their own research, concluded that Iraq had attacked Al Jubayl causing the loud noise. Using information from GulfLINK and from their own sources, they developed two scenarios for the loud noise. In the first scenario, Iraq conducted a chemical warfare attack using a Styx[54] missile launched from a fast patrol boat. The missile, aimed at Al Jubayl, caused the loud noise, and a dusty mustard chemical warfare agent in the missile’s warhead caused the burning skin sensation experienced by some Seabees. In the second scenario, Iraq conducted a chemical warfare attack using aircraft. In this scenario, Iraq’s aircraft sprayed a dusty mustard chemical warfare agent, which caused the burning skin sensation experienced by some NMCB-24 Seabees; the loud noise occurred when Coalition forces shot down two of Iraq’s fighter aircraft near Al Jubayl.

a. Styx Missile Attack

The first scenario suggested that the detonation of a Styx missile in or near the city of Al Jubayl caused the loud noise. This scenario alleged that an Iraq-sponsored group (military unit or terrorist group) sailed down the coasts of Kuwait and Saudi Arabia in a fast patrol boat and launched a Styx missile with a warhead containing a dusty mustard chemical warfare agent into Al Jubayl. To support this theory, they offered the following information:

Did an Iraq-sponsored terrorist group or military unit launch a Styx missile into Al Jubayl? To answer this question, we contracted an independent review of naval operational records, reviewed classified and unclassified government documents, interviewed key personnel who would have knowledge of the alleged incident, and reviewed existing interviews of Gulf War veterans who reported witnessing the loud noise incident.

Figure 6. OSA fast patrol boat while underway

The first three points supporting the missile attack theory—that Iraq had developed a dusty mustard chemical agent, that Iraq equipped Osa fast attack patrol boats with Styx missiles, and that Styx missiles can be mated with a chemical warhead—do not require an in-depth discussion. US intelligence knew for several years before Iraq invaded Kuwait about Iraq’s capability to aerosolize chemical warfare agents (CWAs).[60] However, an analysis of contemporary information indicates that it was very unlikely that Iraq had dusty mustard in its inventory during the Gulf War.[61] At the start of the Air War (January 17, 1991), Iraq’s navy possessed seven Osa fast attack patrol boats armed with Styx missiles.[62] Before the war, the intelligence community reported that Iraq had the capability to produce a warhead for its Styx missiles that could deliver a chemical warfare agent.[63] However, we have uncovered no evidence to indicate that Iraq had actually armed its Styx missiles with chemical warheads during the war.

(1) Rear Area Operations Center Radio Log

The fourth point presented in support of the Styx missile attack scenario included entries made to the 24th Marines Rear Area Operations Center (RAOC) radio log on January 19, 1991. The two individuals who developed the attack scenario pointed to the terrorist threat warning logged at 2:05 AM and the activity that followed as proof that some type of attack occurred.

The first entry (Figure 7), at 2:05 AM,[64] noted that a Marine captain notified the RAOC of a threat for a possible terrorist attack by sea. The second entry made five minutes later (at 2:10 AM) showed the RAOC notified the Navy commander in charge of port security about the terrorist threat warning. At 2:20 AM, a third entry updated the first entry and identified enemy prisoners of war as the source of the terrorist attack information. The RAOC log contained no other entry about the terrorist threat warning or any other attack warnings. The RAOC log showed that no activity occurred for 92 minutes until 03:32 AM when Camp 5 reported that it was receiving incoming fire.

Figure 7. Terrorist attack warning of January 19, 1991

Camp 5’s notification was the first in a series made to the RAOC that indicated some type of event occurred at Al Jubayl. Camp 15 reported an explosion at 3:33 AM. Next, King Abdul Aziz Naval Base (KAANB) reported a second incident of incoming fire at 3:34 AM. In addition, a report made at 3:34 AM from a unit known only by its call sign, A6B,[65] notified the RAOC of explosions in the harbor area. Although the RAOC asked the reporting units for amplifying information concerning personnel injuries and equipment damage, the RAOC never received casualty reports or damage reports.

The first indication that aircraft, rather than the Styx missile discussed in this scenario, may have caused the loud noise appeared in an entry at 3:39 AM when a reporting location, identified as Gateway, informed the RAOC that they thought the explosion was a sonic boom. The next two entries at 3:45 and 3:47 AM recorded notifications from I MEF and the 7th Service Support Group, respectively; they believed the loud noise was a sonic boom. The RAOC then notified all stations on the network at 3:58 AM that no one discovered damage, injuries, or evidence of any explosion(s) and that all stations should resume normal operations.

Five minutes after what seemed to be the end of the loud noise event, the UK’s Force Maintenance Area (FMA) NBC cell notified the RAOC of the first of three chemical warfare agent detections by British forces personnel. The location (Figure 8) of the first detection, recorded at 4:03 AM, was at grid coordinates Universal Transverse Mercator (UTM) grid UK 652832 (latitude 26�57'51" north, longitude 049�38'31" east)—west of King Abdul Aziz Naval Base adjacent to a major eight-lane southeast-northwest highway and not near one of the industrial complexes or a built-up area of the city. The RAOC responded by notifying higher headquarters and other units in the Al Jubayl area of the detection. The FMA informed the RAOC at 4:19 AM that it confirmed the detection as blister agent. At 4:35 AM, the FMA reported a second blister agent detection at UTM grid coordinates UK 644845 (latitude 26�58'33" north, longitude 049�38'01" east), a different location from the first positive detection, and a second positive reading for blister agent at the site of the first detection. The second detection occurred about a mile north from the first detection, along a north-south highway, near the intersection with the highway where the FMA unit made its first detection. While the first detection occurred along the side of the northwest-southeast highway, the second detection occurred about one-half mile from the north-south highway in an open storage area. At 5:15 AM, the FMA reported a third positive CWA detection at UTM grid coordinates UK 545935 (latitude 27�3'22" north, longitude 049�31'58" east)—located near a cement factory approximately 14 miles northwest of the first detection, 13 miles northwest of the second detection, and one mile west of Camp 13.

Figure 8.  Location of reported CWA detections by British forces in Al Jubayl

As might be expected, the British chemical warfare agent detections caused a flurry of activity in the Al Jubayl area as units responded to the perceived threats. One location, Camp 4, notified the RAOC that the camp was receiving incoming fire from the direction of Camp 5. In addition, some confusion existed over the validity of the chemical warfare agent detections reported by the British. Consequently, the USCENTCOM NBC cell dispatched one of its own teams to verify the detections. At 05:53 AM, the FMA NBC cell reported that petrochemical fumes probably caused all the alarms and positive readings. At 06:30 AM, all stations reverted to MOPP level 0+. Two hours later, at 7:48 AM, the USCENTCOM NBC team reported that they could not confirm any of the British detections.[66]

The activity portrayed in the RAOC log does indicate that some type of event occurred at Al Jubayl. However, the RAOC log contains no information that confirms that an attack actually occurred. There is no mention of the RAOC receiving casualty reports or damage reports. The log also contains no information regarding the location of a missile impact point or detonation. If an actual attack occurred, the log certainly would contain some confirming information.

(2) Interviews of Key Personnel

The RAOC radio log indicates a high level of activity in Al Jubayl during the early morning hours of January 19, 1991. However, the log does not contain sufficient information to tell a complete story and needs supplementary interviews of individuals who were at Al Jubayl and participated in the events in question. Five people who, because of their positions, would have had first-hand knowledge of the events are:

The Marine chief warrant officer on duty in the RAOC during the early hours of January 19, 1991, remembered receiving the terrorist threat warning and making follow-up notifications. He also remembered hearing the loud noise shortly after the terrorist warning, but never heard what caused the noise. He did not believe that a terrorist attack caused the noise. He explained that a high level of confusion existed at the time of the incident—especially after the FMA NBC cell in Al Jubayl broadcast over the radio that it had made a positive detection for a chemical warfare agent. Every unit on the radio net heard and reacted to the broadcast.[67]

The Marine captain in the headquarters of the 3d Battalion, 24th Marines, admitted that he notified the RAOC about the terrorist threat, but he did not recall where he obtained the information. He explained that he might have gotten the information from one of the many intelligence reports routinely circulated to all command levels. Like the CWO, he also did not know what caused the loud noise, but was sure that it was not a terrorist attack. He thought that the British CWA detections caused some of the confusion, because the FMA NBC cell was responsible for the civil defense warnings in the Al Jubayl area and, therefore, controlled or had direct access to the siren and loudspeaker systems. When the FMA NBC cell received notification of the first positive test result, it broadcast this information at about the same time as the first reports of the loud noise—thereby increasing the level of confusion. By the time it determined that the original positive report was false, the damage had been done. Most of the units in the area were responding to what they believed to be an attack.[68]

The Marine colonel was the regimental commander in the Al Jubayl area, the RAOC commander, and commander of the land-based security forces for Al Jubayl.[69] When asked if he remembered the loud noise, the colonel answered that once the air war started, loud noises or sonic booms occurred regularly. When told of the log entry that warned of a possible terrorist attack, he stated that no terrorist attack occurred in Al Jubayl on January 19 or at any other time during the Gulf War. In his opinion, the cause of the confusion during the early days of the war was a rumor mill that was running rampant, the constant air raid alarms that turned out to be false, and several individuals who reported positive CWA detections. He finished the interview by stating that all reports of positive CWA detections in the area proved false.[70]

The Navy commander in charge of the port security defense forces in Al Jubayl told us that he did not remember hearing the loud noise, but remembered hearing reports of a positive detection and later learning that diesel engine exhaust caused a false detection. The only incident (terrorist or otherwise) he recalled was the Scud missile that impacted in the harbor waters on February 16, 1991. The commander assured us that if an attack had occurred on January 19, he and his counterpart, the head of the land-based security forces, would have been notified immediately, and that details of the event would have been briefed at the general's (the commander of the rear area in Al Jubayl) morning briefing. He said that he attended every briefing and an attack on Al Jubayl (other than the February Scud incident) was never briefed. We asked if he thought a vessel (either Iraq’s or terrorist) capable of launching a Styx missile could have approached the coast undetected to fire a missile into Al Jubayl. Without hesitation, he responded, "No." He explained that there were too many security measures in place for that to have happened. To emphasize the point, he said that early in the Desert Shield period, two individuals attempted to penetrate the harbor in a small boat, but security forces observed and apprehended the individuals before they could get very far. Security forces turned both individuals over to Saudi officials. He believed that the two individuals were not involved in any terrorist activity because Saudi officials never contacted Coalition authorities about any terrorist activity in that case.[71]

Finally, the Marine general who commanded the rear area forces in Al Jubayl had no recollection of the loud noise, nor did he believe that an attack occurred at Al Jubayl on January 19. He said that he received no notification of an attack but would have if any attack occurred. He did say that he vaguely remembered the numerous alarms initiated by the FMA NBC cell in Al Jubayl.[72]

(3) White Cloud

At 3:38 AM, shortly after the loud noise in Al Jubayl, a security guard at Camp 13 reported that he saw a white cloud south of the camp and that it was moving northward (Figure 9).[73] Seven minutes later, the guard reported that the cloud continued to move northward. The security log contained no other information about the cloud, nor did it mention whether the camp’s security forces acted in response to the cloud sighting. Although the NMCB-24 command post log did not mention the cloud, one person in the command bunker during this period remembered hearing a report of a cloud-sighting southeast of the camp.[74]

In addition to the logs from Camp 13, we also reviewed the logs from other units and command elements for the same period (Table 1). Of these, the RAOC log was the only other log containing any information on a cloud. The log showed that Camp 13 notified the RAOC about the cloud sighting, but not until 5:17 AM—99 minutes after the security guard first saw the cloud. There was no explanation for the delayed report, but the RAOC log entry offered more information about the cloud’s possible location—"approximately 5 SE" [southeast] of Camp 13’s north post. The Camp 13 log gave no distance and located the cloud south of the camp. Because there was no unit of measurement after the number 5, the RAOC entry could have meant five kilometers or five miles. However, because the weather report stated that the visibility throughout the area was reportedly near zero during the period in question, we are not sure how someone could spot a white cloud at a distance of five kilometers or five miles. Another explanation for the number 5 could have been the guard’s use of the number when referring to the position of the cloud, as in the five o’clock position. If the location of the guard post in the RAOC log was correct, the five o’clock position would have placed the cloud to the southeast of the guard post. Of course, other explanations for the 5 are also possible.

Table 1. Logs reviewed for white cloud information

Unit Identification

Type of Log


I MEF/G-3[75]

Operations Log

Log contains no reference to the cloud.


Desk Log

Log contains no reference to the cloud.


Staff Journal

Document is heavily redacted; could find no reference to the cloud.
1ST Surveillance Recon Intel Group[78]


Document contains no reference to the cloud.

To identify possible cloud sources, we studied the area surrounding Camp 13. Since the cloud was reportedly to the south or southeast of the camp, we studied that area (Figure 9). Because the exact location of the north post was unknown, and to ensure that no potential cloud source was eliminated from consideration, the quadrant and distances were calculated from the southern-most point of Camp 13.

As the map indicates, several possible cloud sources existed in the southeast quadrant and within five kilometers of the camp. These sources include open-air aqueducts, methanol plants, a gas company, an iron and steel plant, a water filtration plant, and a petrochemical plant. Past the five-kilometer zone into the five-mile zone, we find another petrochemical company, an oil refinery, more aqueducts, several unidentified buildings, open storage areas, and a pumping station combined with a gas-oil separation facility. The suspicious cloud could have come from any one of these plants. Many of these types of facilities throughout the world commonly emit steam and other gases. The venting of cloud-forming gases could have occurred as part of routine operations or in response to perceived emergencies.

Although there were a number of facilities in the quadrant to the southeast of Camp 13 that could have produced the cloud, it could have come from other locations and been carried on the wind to the location where it was first sighted (Figure 10). Except for a housing area and a few isolated buildings and water towers, the southwest quadrant was virtually empty and therefore an unlikely origin of the cloud. The northwest quadrant contained cement and steel fabrication plants, but such facilities generally do not emit gases. The northeast quadrant, however, contained more gas-producing facilities than any quadrant—three fertilizer plants, three methanol plants, and four petrochemical plants. Two of the three fertilizer plants were within 1,000 meters of Camp 13’s northeastern boundary. One plant was as close as 500 meters from the camp.

Figure 9. Camp 13’s southeast quadrant

Even though the RAOC log and Camp 13 log placed the cloud to the southeast and south of Camp 13, the industrial plants in the northeast quadrant might have produced the white cloud. Seabees at Camp 13 reported ammonia as the prominent odor during the time of the loud noise incident[79] and many local industries used ammonia. These industries included fertilizer production, poultry farming, food processing, refrigeration, and chemical production. It is possible the ammonia odor could have come from one or both of the fertilizer plants and spread over the area within a cloud. This would have agreed with some reports that there was precipitation in the area in the form of a mist.[80] The perception that something was in the air during the morning of January 19, 1991, was not confined to the area around Camp 13. One individual, who was at the King Abdul Aziz Naval Base, reported something in the air (not a mist), which caused a metallic taste in his mouth.[81]

Figure 10. Camp 13’s northeast, northwest and southwest quadrants

(4) Burning Skin

Some people who witnessed the loud noise event reported they experienced (or knew someone who experienced) a burning sensation on the skin—primarily on the hands and lips.[82] Some people also experienced a numbing sensation on the hands and lips.[83] Still others reported feeling no unusual sensations. Reports of burning skin were not limited to the people at Camp 13. Individuals assigned to the NMCB-24 detachment at the King Abdul Aziz Naval Base also reported this phenomenon.[84] However, although reports of burning skin came from two different locations, all reports of burning skin came from a small number of Seabees from NMCB-24. We found no similar reports that any other unit, even those that were collocated with NMCB-24, experienced the burning skin phenomenon.

For a chemical warfare agent to cause burning skin, the person experiencing the symptom would have to absorb a certain dose of chemical warfare agent, i.e., the amount of a CWA absorbed through the skin, inhaled, or ingested over a given period. Although each person responds in a predictable way (e.g., develops burns and blisters on the skin, develops blurred vision, or loses consciousness, etc.) after becoming exposed to a CWA, the time it takes for symptoms to first appear may vary slightly from individual to individual. In addition, the dose that an individual must absorb to cause a reaction can also vary slightly from individual to individual. Established dosage levels will predictably cause a reaction in a defined percentage of a population. Although there are several dosage designations such as incapacitating dose (prevents a person from performing duties) and lethal dose (causes death), in this case narrative we are concerned with the effective dose—that which causes an effect in a defined percentage of a population.[85, 86] Table 2 summarizes the effects of chemical warfare agents and the times that normally elapse before the effects appear.

Descriptions of the incident at Al Jubayl indicate that the onset of the burning sensation began shortly after the loud noise. As shown in Table 2, the only CWA that can produce a burning sensation on the skin within seconds of receiving an effective dose is lewisite. This eliminates nerve agents and blood agents as possible causes because immediate pain is not associated with these agents. This eliminates mustard agents (including dusty mustard) as a possible cause for the same reason. We also eliminate lewisite from consideration, not because of the injuries it causes, but because a post-war analysis indicates that Iraq did not have lewisite in its chemical warfare agent arsenal during the war.[87]

Although we eliminated blister agents as possible causes for the burning skin, there was still one other chemical warfare agent that causes pain upon contact. That agent was phosgene oxime. However, after the Gulf War when the United Nations Special Commission on Iraq inspected, inventoried, and destroyed Iraq’s chemical warfare agents and munitions filled with chemical warfare agents, it never found phosgene oxime in Iraq’s chemical warfare agent inventory.[88] Therefore, we eliminated phosgene oxime as a possible cause for the burning skin.

In addition to knowing the effects of chemical warfare agents, what constitutes an effective dose is important (Table 3). The effective dose is the minimum amount of agent that will cause an effect, e.g., burning skin. The amounts listed for the vapor form of agent are the minimum concentration needed to receive an effective dose. Factors such as weather and topography have an impact on the vapor cloud in terms of how long it lasts before evaporating, the speed in which it moves, and the direction of movement. Therefore, it can be expected that when an enemy uses a chemical warfare agent in an attack, the concentration level will be many times more than that needed for an effective dose. The amounts shown in the liquid column are for the pure form of the agent before exposing it to other elements that may reduce its effectiveness.

Table 2. Chemical warfare agents and their effects

  Agent Organ Severity of the Exposure Effects Onset of
(H & HD)
Eyes Mild Tearing, Itchy, Burning, Gritty feeling 4-12 hours
Moderate Same as for a mild exposure plus; Reddening, Lid swelling (edema), Moderate pain 3-6 hours
Severe Marked lid swelling (edema), Possible corneal damage, Severe pain 1-2 hours
Airways Mild Runny nose (Rhinorrhea), sneezing, nosebleed (Epistaxis), Hoarseness, Hacking cough 6-24 hours
Severe Same as for a mild exposure plus: Productive cough, mild-to-severe difficulty in breathing 2-6 hours
Skin Mild Redness of the skin (Erythema) 2-24 hours
Severe Blister (Vesication)
Lewisite [90]
Eyes Pain and Irritation; swelling (Edema) to eyelids, conjunctiva, cornea, and iris. Immediately
Airways Extreme irritation to nose and lower airways Immediately
Skin Immediate pain and irritation followed by blister development ( Vesication) 15 to 30 minutes
Redness (Erythema)
Tabun (GA)
Sarin (GB)
Soman (GD)
The effects (symptoms) that nerve agents have on humans are very different from blister agents. Nerve agents are absorbed into the body through breathing, by injection, or absorption through the skin. The severity of the effects and the speed at which symptoms first appear depend on factors such as whether the agent is in liquid or gaseous form and the way it is absorbed into the body. Regardless of its form or how ingested, one can expect symptoms to appear within minutes of exposure. Of the many symptoms of nerve agent exposure, the most common effects are constriction of the pupil (miosis), twitching muscles, difficulty breathing, heavy salivation, and heavy perspiration.
Hydrogen Cyanide(AC)
Cyanogen Chloride (CK)
Although not as toxic as nerve agents, exposure to high amounts of agent (greater than 100 mg on the skin) can result in death within 5-8 minutes from exposure. Likely battlefield exposure would be to gaseous forms of AC or CK. Early signs of exposure can include rapid and deep breathing (hyperpnea), headache, anxiety, personality changes, agitation, and seizures.


Table 3. Effective doses for blister agents


Mustard (H and HD)[93]

Lewisite (L)[94]






Eyes 10 to 70 mg�min/m3 Less than 10 �g[95] 150 mg�min/m3 Less than .001 ml
Airway 100 to 500 mg�min/m3 Less than 10 �g [96] 500 mg�min/m3 Less than .001 ml 90
Skin 200 to 2,000 mg�min/m3 10 �g >1500 mg�min/m3 [97] 14 �g or less

Finally, Table 4 summarizes the physical properties of blister agents—properties that would also help determine if a blister agent that could have caused the burning skin was present at Camp 13.

Table 4. Physical properties of blister agents


(H and HD)


Color Oily liquid, from colorless to amber. Colorless to brownish liquid

Like garlic, or mustard

Freezing Point 14.45 degrees Celsius (58 degrees Fahrenheit) -18.2 degrees Celsius to 0.1 degrees Celsius (depending on purity). (-76 degrees Fahrenheit to 32.18 degrees Fahrenheit)

The symptoms reported by the Seabees are important factors in identifying a cause for the burning skin. Each person who reported experiencing the burning or numbing sensation said they felt the sensation on their hands and lips. Tables 2 and 3 indicate that the eyes are the most sensitive organs to blister agents, and the airway is more sensitive than the skin. Is it possible to develop symptoms on the skin and not develop symptoms in the eyes or airways? We have speculated that those who experienced the burning sensation put on their protective masks first, which prevented eye, lip, and airway exposure to the agent. However, we cannot explain why symptoms developed on the Seabees’ lips and not their eyes or airway even though the eyes and airway are more sensitive to CWAs than the skin. The mask protects all three parts of the body. If a CWA in vapor form was present, all three body parts should have been exposed to the same, or nearly the same, amount of agent, yet not one person complained of eye, or respiratory problems. In addition, not everyone present at the time of the incident who put on their protective equipment developed the reported symptoms, and some veterans who were present and presumably exposed to the same substance stated during interviews that they did not wear their protective equipment and did not develop symptoms. Lastly, the symptoms themselves are puzzling. Blister agents cause pain. That some people claim they were exposed and then developed pain-like symptoms is not extraordinary. What is unexplained, however, is the fact that the symptoms never progressed beyond a burning sensation to blister development. For this to occur, everyone who experienced the burning sensation would have had to receive nearly the same dose—one large enough to cause burning, but too small to cause blisters. For that to have happened, Iraq, or whoever carried out the alleged attack, would have had to control the release of agent so that the concentration of agent never reached the effective dose for causing blisters or more serious injuries to the eyes and airway. An aircraft spraying agent or an exploding missile warhead filled with agent does not produce such a controlled release.

Although we may never know the exact cause for the burning skin, other chemical substances are possible sources for the Seabee’s burning skin. As mentioned earlier, Al Jubayl was an industrial center and Camp 13 was located near several industrial plants. Any of these plants could have emitted a chemical substance that caused some of the Seabees’ skin to burn. The release could have occurred as part of normal operations or as part of emergency procedures taken in response to attack warnings. Ammonia is one possible substance and several Seabees reported smelling ammonia during the loud noise of January 19. Ammonia can burn a person’s skin.[100]

| First Page | Prev Page | Next Page |