Chapter Two:
Environmental Measurements

The magnitude of the fires and the extensive publicity about the amount of smoke sparked concern around the world regarding effects on global processes such as ozone depletion, acid rain, global warming, and other atmospheric phenomena. For example, some worried that the smoke could disrupt the monsoon season in southern Asia and trigger a drought affecting crops in India and Pakistan (Horgan, 1991a).

Although no global effects were noted, the passage of the smoke was detected worldwide where very sensitive equipment was located. Beginning in February 1991, the National Oceanic and Atmospheric Administration (NOAA) recorded numerous "soot spikes" in air-sampling data at its observatory located 4000m above sea level in Hawaii. Wind-pattern records indicate that the sooty air had left the Persian Gulf 7-10 days earlier. The March levels at Mauna Loa were five times higher than during March in the previous three years; these levels were still low enough to pose negligible health risks. No global disaster occurred, as some predicted immediately after the fires began (Horgan, 1991a, 1991b).

Firefighters started to cap the flaming oil wells in April, expecting that they would need up to two years to extinguish all fires. In the meantime, a coordinated international research effort began in the spring of 1991 to assess the impact of the fires on health and the environment.

U.S. Standards and Recommendations

SOURCE: ATSDR, 1997.

The American Conference of Governmental Industrial Hygienists (ACGIH), a professional society--not a government agency--devoted to the administrative and technical aspects of occupational and environmental health, develops Threshold Limit Values (TLVs) as recommendations or guidelines for use in the practice of industrial hygiene in the workplace. TLVs are TWA concentrations that should not be exceeded during any 8-hour workshift of a 40-hour workweek. All these concentration limits are set at levels believed to be without adverse effects for nearly all workers during repeated daily exposure. Table 2.3 lists NIOSH, OSHA, and ACGIH recommended occupational limits for the noted pollutants.

SOURCES: NIOSH, 1997; ACGIH, 1997.

NOTE: NA = Not applicable.

^aComponents of coal tar pitch volatiles.
^bConcentration limits are for their respective TWAs except for NO ₂.
^cShort-term exposure limit, a 15-minute TWA.
^dPM₁₀ not otherwise regulated.

As discussed above, in the United States there are different standards of acceptable levels of contaminants set according to which population needs to be protected. The EPA's NAAQS, for the ubiquitous criteria pollutants, are targeted for the general population for continuous ambient exposures. ATSDR-MRLs give the concentrations of hazardous substances that will likely not affect the general population through daily exposure. NIOSH, OSHA, and ACGIH standards are for occupational exposures during 40-hour workweeks lasting the worker's lifetime.

The armed forces deployed to the Gulf theater are a unique segment of the population. They are predominately young (18-40), fit, and healthy. They stayed in the region after the onset of the Kuwait oil fires for (at most) 6-8 months. There are no studies in the scientific literature that focus on the possible health effects due to exposures to contaminants in this population segment. What standards should be considered appropriate for this group, exposed for 24 hours over a limited period of time? The NAAQS and MRL ambient levels are unrealistic because of the margins of safety included to protect the most vulnerable members of the population. The NIOSH, OSHA, and ACGIH occupational standards, although seemingly more appropriate, assume a step-function exposure with periods of low levels in-between but lasting for many years. Perhaps the occupational standards combined with an empirical factor to account for the continuous exposure would give a reasonable exposure-limit estimate.

Measurements Taken Before the Oil Well Fires

SOURCE: Rowe, 1991.
NOTE: NA = not applicable.

Even though only limited environmental data are available, high particulate levels have been documented throughout the Persian Gulf region: 5080 µg/m³ in Riyadh, Saudi Arabia in 1982; 630 µg/m³ in Kuwait in 1983; and 674 µg/m³ in Bahrain in 1987 (Madany and Raveendran, 1992). A study from Dhahran, Saudi Arabia, reports TSP[1] measurements for 100 days from November 1980 to March 1981. The average maximum daily value for the period was 737 µg/m³, and the geometric mean for the period was 339 µg/m³. But when the Shammal winds were blowing, peak daily concentrations reached the extraordinary values of 2923 µg/m³ (Khattak, 1982). These values are about 20 times the 24-hour U.S. ambient standard for PM₁₀ of 150 µg/m³. The daily variability in PM₁₀ concentration is illustrated in Figure 4.2 for Doha, Kuwait, from June to December 1991.

Measurements Taken During the Oil Well Fires

May 6-June 12 Measurements

SOURCE: Ferek, 1992.

^aTOC = Total nonmethane organic carbon in the vapor phase.
^bPercentage of carbon specie contributing to the total carbon in the plume is included in the parentheses. It indicates combustion efficiency; a higher percentage means more efficient combustion.
^cSalts = Sum of all salts measured.

July-August Measurements

The ground-level sampling in Kuwait City was taken during an abrupt change in weather conditions on August 7. Until August 5, winds with speeds of 35-55 km/hr (10-15 m/sec) transported the smoke from the oil fields toward the Persian Gulf, with little visual evidence of a plume across Kuwait City. On August 5, a strong, low-lying inversion developed, winds calmed, and visibility declined sharply. Winds then returned from the northwest, and, by August 9, dispersed the material that had accumulated over the city.

In the August 7 sample, the dominant component was sand dust. Sulfur, as SO₂, increased by about 50 percent during the inversion. Concentrations of Cl, Pb, Br, and Zn increased at least ten-fold. The ratio of Pb to Br of about three remained constant throughout the sampling, suggesting that their primary source was local automotive traffic and not the oil well fires. Concentrations of Ca, K, Ti, Fe, and Mn increased by less than 50 percent during the inversion period. The ratios of Si, Fe, K, Ti, and Mn to Al were within 30 percent of the values for local desert soils and similar to soils in Arizona. PAH levels in Kuwait City were typically less than 0.001 µg/m³ as shown in Table 2.11 (Stevens et al., 1993).

July 31-August 4 Measurements

The daily PM₁₀ concentrations ranged from 139 to 673 µg/m³ but, although quite high, cannot be attributed only to the oil fires. Comparison with available data on particulate matter from before the Gulf War shows that these levels are common for the region.

The mean daily concentrations of Ni and V ranged from 7 to 42 ng/m³ and 11 to 42 ng/m³, respectively. They were strongly correlated, indicating that they had a common source: the oil fires (Madany, 1992).

Table 2.6 depicts the average concentrations of individual components of the PAHs in Bahrain for the study period as well as what is considered background, rural, and urban values for cities around the world for comparison (Madany, 1992). It is important to note that the PAH levels in Bahrain, which is downwind from Kuwait, were at least an order of magnitude lower than the European cities with high PAH concentrations.

SOURCE: Madany, 1992.

NOTE: NA = not available.

USAEHA Environmental Measurements, May-December

Permanent ambient air monitoring stations were established at four locations in Saudi Arabia and six in Kuwait, although two in Kuwait had to be abandoned quickly due to logistical difficulties (see Figures 2.1 and 2.2). The sites selected were locations where large concentrations of U.S. troops and DoD civilians were stationed for long periods of time. The two mostly civilian sites were the U.S. Embassy in Kuwait and the Al Ahmadi Hospital; the others were mostly military sites.

Figure 2.1--Monitoring Sites in Kuwait

Figure 2.2--Monitoring Sites in Saudi Arabia

Based on the hazardous substances to be monitored--crude oil, by-products of incomplete combustion, breakdown products--scientists from USAEHA de-cided on a list of pollutants of concern that needed to be monitored in the Gulf region affected by the oil well fires (Table 2.7).

SOURCE: USAEHA, 1994.

Air Sampling. The USAEHA sampling campaign measured little change in the general air quality during the monitoring period. Table 2.8 lists the highest mean values of the contaminants listed in Table 2.7 as measured by the USAEHA during their 1991 campaign. Although considerable increases were noted in particulate matter, these concentrations were considered to be within the range common to this area.

NOTES: NA: Standard or recommended concentration not available.

^aUSAEHA, 1994.
^bThere were only a few samples above the detection limit.
^cTLV performing heavy work.
^dPM₁₀ not otherwise regulated.

NOTE: Toluene measurement for Los Angeles not available.

High levels of airborne particulate matter (sand and soot) were observed at several monitoring sites. Analysis of the samples indicated that the particles were mostly sand-based materials; high levels of airborne sand particulates are typical for this region of the world (Kirkpatrick, 1997). Within the PM₁₀ samples of particulate matter, levels of PAHs and toxic metals were low.

A small subset of the ambient air samples collected in Kuwait and Saudi Arabia were analyzed to determine particle-type class and particle-size distribution of the PM₁₀ data. The assigned classes fell into the following main categories: earth crustal (silica-rich, e.g., quartz), calcium bearing (calcium-rich, e.g., dolomite, gypsum), salt particles, carbon rich (e.g., soot), and miscellaneous (USAEHA, 1992). Figure 2.4 shows the percentile composition of the particulate matter in the air of Kuwait and Saudi Arabia (USAEHA, 1992). Considering that the sand of the Arabian Peninsula is rich in calcium and silica, it indicates that most of the PM₁₀ is of sand origin and that in Kuwait about 23 percent of the total PM₁₀ mass was contributed by soot from the oil well fires.

SOURCES: Kirkpatrick, 1997; USAEHA, 1992.

SOURCES: Kirkpatrick, 1997; USAEHA, 1992.

USAEHA Industrial Hygiene Measurements, May-June

The results are summarized in Table 2.9. The table provides means and standard deviations, maximums measured, and ACGIH's TLVs. Most measured pollutant concentrations were well below the occupational TLVs.

SOURCE USAEHA, 1994; ACGIH, 1997.

NOTES: ^amean (standard deviation). BDL: Below Detection Limit.

SOURCES: 1991 data: USAEHA, 1994; 1993 data: Kirkpatrick, 1997.

NOTE: NA = Not available.

Measurements Taken After the Oil Well Fires

The results indicate that the metals and VOC readings at Khobar Towers in 1993 were about the same as those in 1991, but PAH and PM₁₀ levels were lower. The November 1991 VOC data for Khobar Towers are similar to the November 1993 levels, except for toluene. The Khobar Towers VOC levels are a factor of three higher than Camp Thunderock. The PAH levels detected in November 1993 were 2 to 8 times lower than in November 1991. For PM₁₀ the November 1993 mean levels are lower than in November 1991, decreasing from a mean concentration of 62 µg/m³ to 52 µg/m³.

At Camp Thunderock, the readings for all four categories were substantially lower in 1993 than they were in 1991. The metal levels from November 1993 are a factor of 2 lower than those in November 1991. In the November 1993 PAH data, most of the compounds are below detection limit and the rest are about three times lower than those in November 1991. PM₁₀ levels decrease from 84 µg/m³ to 44 µg/m³.

Conclusions

Table 2.11 displays the maximum or 95th-percentile concentrations in the air for the pollutants of concern as measured by the USAEHA from May to December 1991, and the ACGIH-recommended exposure limits for hazardous substances in the workplace. These results show that, except for ozone, the maximum concentrations for all pollutants in the Gulf were several orders of magnitude lower than ACGIH occupational standards. And ozone, which is ubiquitous in the U.S., was still lower than in many U.S. urban areas where summer episodes reach twice the NAAQS.

A possible objection to the above comparison is that U.S. personnel in the Gulf were being exposed to air pollution 168 hours per week for up to 70 weeks, while occupational exposures are spread over 40 hours per week during an adult working lifetime. Nevertheless, if the mean pollutant concentrations due to the Kuwait oil well fires are compared to the NAAQS and to the ATSDR's MRL levels, the Gulf levels were lower than the U.S. standards set for the general population (except for PM₁₀), and lower than the daily exposures of millions living in major U.S. cities. However, PM₁₀ levels, unrelated to the oil well fires, were much higher than ambient levels in the U.S., but within the ACGIH's TLV levels. Gulf War veterans, like people living in urban areas in the U.S., were exposed to multiple pollutants simultaneously for which there are no standards.

^aFor particulates, metals, and VOCs, this column gives the 95th percentile [C]; for PAHs and Criteria Pollutants it gives the maximum [C] (USAEHA, 1994).
^bStevens et al., 1993.
^cThere were only a few samples above the detection limit. The ACGIH TLV is for coal tar pitch volatiles.

[2]The cut-off value of 3.5 mm, PM_3.5, was arbitrarily set in the particle analyzer. These measurements were taken before the PM_2.5 standard was promulgated and is the closest available data above PM_2.5.