April 3, 1991


                                            KUWAIT OIL FIRES
                                     INTERAGENCY INTERIM REPORT
 More than 500 oil weil, storage tank and refinery, and facility fires are currently raging in Kuwait and
 each day produce an enormous amount of smoke and other pollutants.  The quantity and character of
 the smoke plumes are not yet certain, and the fires are expected to continue for some period.  The
 fires originate .in seven oil fields, located both north and south of Kuwait City, with the majority
 centered in the Al Burgan oil field south of the Kuwait City airport. The fires may represent one of the
 most extraordinary manmade environmental disasters in recorded history.

 In response to this situation, the Saudi Government requested U.S. technical assistance on the public
 health and environmental  impact of the fires.   The U.S. Embassy in Saudi Arabia concurred in this
 request, and voiced its additional concerns about the health effects of the fires on the hundreds of
 thousands of U.S. troops in the region as well as the thousands of American citizens residing in Saudi
 Arabia and the other Gulf countries.  Similar concerns were also expressed in Kuwait. An Interagency
 Air Assessment Team consisting of representatives from the Environmental Protection Agency, the
 Nations^Oceanic and Atmospheric Administration, and the Department of Health and Human Services
 was formed and deployed to the Persian Gulf area.   In country, this team was supplemented by
 representatives of the United States  Coast Guard, the Department of Defense, and the Department of

 The specific mission of the team was to assess the conditions through air sampling and monitoring in
 oil fields and other areas to determine the ef.ect on public health; to review the health infrastructure;
 to determine the capability of the region to dea' with the health threat  through air monitoring and
 appropriate corrective action; to provide technical  assistance; and to consider appropriate follow-up
 action.                                                                •

 Meteorological conditions over the past two months have tended to transport the smoke  plume toward
 the southeast, with periodic excursions toward the northeast. March through July are relatively windy
 months; there are normally 30 days of very strong winds from the  northwest in this period, which
 produce sandstorms and rapidly ventilate the smoke. From August to October, the incidence of strong
 winds  should  drop sharply.  The plume is generally below about  12,000  feet.  As the summer
 progresses, it is possible that the height of the plume will increase and that it will then  be evident to
 greater distances.

 Emissions from oil fires may have the potential of causing health effects of both an acute and chronic
 nature, although there is considerable uncertainty as to the extent of the threat.  Chemicals such as
 sulfur dioxide and hydrogen sutfide as well as carbon monoxide  and polycyclic aromatic  hydrocarbons
 are often found along  with paniculate matter in oil  fires. While only  a limited assessment is possible
 at this time, the Team did not detect such chemicals in any significant quantity;  also,  preliminary
 analysis of the substantial amount of paniculate  matter did not reveal  any chemicals at levels of
 concern. Additional testing is needed to better define if other toxic materials may be associated with,
 the high levels of particulates found.

The host nation governments also provided the Team with an abundance  of preexisting  air monitoring
 information covering the past several years in Saudi Arabia, Kuwait, and Bahrain. These data provide
                                                           .HEADQUARTERS LIBRARY
                                                            ENVIRONMENTAL PROTECTION AGENCY
                                                            WASHINGTON, D.C. 20460

  a useful baseline on limited parameters for comparing the conditions that exist and may evolve during
  the next several months. The Team concurred with the Saudi Government's view that the considerable
  Saudi public and private sector competence in air monitoring would need to be supplemented with
  support and technical assistance.  This situation is even more acute in Kuwait City due to the lack of
  power and to the disruption of the governmental and scientific infrastructure.  The Team considered
  it of critical importance that additional technical assistance be made available to the countries, if

 There is need to consider the overall problem from several perspectives so as to ensure  economy as
 well as  success.  First, because  high levels of particulates were found in the air, and prolonged
 exposure to paniculate matter may contribute to respiratory discomfort and perhaps long-term or
 permanent respiratory disorder, it will be necessary to find out what is in the plume and how it varies
 over time and distance.  Second, we should assess the immediate and long-term human health risk.
 Third, we should quantify effects on the environment, especially on crops and climate.
 With the assistance of the Saudi and Kuwaiti Governments, the U.S. Interagency Air Assessment Team
 has been conducting a reconnaissance survey of the fire plumes and their effects in Kuwait and Saudi
 Arabia since March 10,1991. The primary objective of the Team was to obtain preliminary, short-term
 data on the emissions from the smoke emanating from the oil well fires at a variety of locations, in
 order to:

       1,      Determine if there is an acute health threat associated with the Hydrogen Sulfide (H,S)
               and Sulfur Dioxide (S03) and particulates, three pollutants that might be emitted from
               burning oil wells;

       2.      Identify and quantify the gaseous and paniculate byproducts being produced froir the
               burning oil wells; and                                                         .

       3.      Determine if .the  materials  associated with these  fires are  affecting areas  where
               American citizens are located.

       4.      Assess the potential extent of the health effects related to the emissions from the fires
               and the status of the Kuwaiti and Saudi health infrastructure.

Based on these objectives, limited, real-time data was obtained directly from the Kuwait oil fields, as
well as from Kuwait and Saudi Arabia locations where embassy officials, troops, and citizens work and
reside. Additionally, the Team conducted a number of interviews with Health officials to evaluate the
extent of acute respiratory problems related to smoke exposure. While only a cursory assessment is
possible at this point some data obtained by the team were encouraging. The preliminary findings are
as follows:

       1.     Umttad sampling did not reveal the existence of high concentrations of sulfur dioxide
              or hydrogen sutfide near the burning wells or in population areas in the path of tine oil
              well emissions;

       2.     High levels of paniculate were found in the air;

        3.     The  results of the current monitoring findings and health interviews with medical
               personnel in  the affected  areas suggest that at  the  present time susceptible
               subpopulations, such as individuals with asthma and chronic obstructive lung disease,
               may experience exacerbation of their symptoms.  Special health concerns, warnings,
               advisories, and precautions are clearly warranted for these individuals. This situation
               does not appear  to  be  life threatening under current exposure conditions but, if
               meteorological conditions change, i.e., poor air mixing or plume touchdown, there
               could be adverse health effects for these susceptible individuals; and,

        4.     The long-term effects on health are not readily ascertainable at this time  due to
               insufficient data on the populations exposed, the composition of the smoke plume, the
               impact of oil pools) and long-term meteorological patterns. Both the Kuwaiti and Saudi
               health communities have expressed great interest in obtaining training and support
               from the US medical community that can be continued by themselves in future years.
               Aggravating the problem is the severe damage  done to the scientific infrastructure of
               Kuwait thus limiting the current in-country analytic capabilities. Any  response by the
               US would have to include both training and equipment.

 The Team has stressed, however, that their observations represent only a preliminary assessment and
 that considerable follow-up will be necessary to evaluate definitively the nature and magnitude of the
 human health, ecological, and atmospheric effects of the oil fires. Such follow-up activities will need
 to be carefully coordinated with the governments in the region as well as with other governments and
 international organizations, such as WMO, WHO, and UNEP, which are seeking to assist in evaluating
 the situation.
The local populations are being exposed to an increased health risk, the magnitude of which cannot
be estimated with any degree of certainty without further measurements and surveys. The extent to
which conditions may worsen needs to be understood and a forecast capability developed.  Without
such  measurements  and  assessment, and  development of, a  predictive capability, the regional
population  remains exposed to an uncertain risk, and reconstruction of the area may be impeded.
Moreover, without such input, an accurate and defensible quantification of environmental effects will
not be possible.

In addition to providing direct answers to questions regarding the effects of the smoke plumes on the
atmospheric environment intensive studies of the plumes will accelerate progress in understanding
manmade effects on regional and global air quality, meteorology,  and climate. Because the expected
changes in air chemistry, solar radiation, and cloud microstructure are so large, observations of these
processes could circumvent the need for many years of study directed at  much lesser phenomena.

The program proposed below is comprised of three primary elements: human health surveillance and
risk assessment,  air  monitoring, snd development of a forecast  capability.  These elements will be
closely linked to achieve the goals of understanding and predicting the degree of human health risk and
the effects on atmospheric processes.

  1.     Air Monitoring

 The objective of the air monitoring program is to collect the necessary data to determine the nature
.and concentration of pollutants associated with the fires,: and demonstrate the recovery of the
 environment as the fires are extinguished.

 The development and deployment of an integrated monitoring network will serve several interrelated
 purposes,  ft will assure data consistency throughout the region.  It will provide better data to assess
 the immediate health risk and potential for long term risks.  It will be used to initiate, test, and refine
 forecast models discussed later in this plan, and thus greatly assist in the development of location and
 condition specific alerts — assisting in issuing special advisories for populations at risk under unique
 conditions. Lastly, it will provide a better basis for scientific  understanding and knowledge of the
 important regional and international issues and will assist in assessing possible extended consequences
 of the fires.

 The U.S. Interagency Air Assessment Team is working with.the Saudi Arabian Meteorology and
 Environmental Protection Agency (MEPA), Saudi ARAMCO, and King Fahd University of Petroleum and
 Minerals to develop an air monitoring plan for the Gulf Region that will provide the air monitoring data
 to assess the impact of the Kuwaiti fires in Saudi Arabia.  While focused primarily on the needs of
 Saudi Arabia, the plan forms the general basis for a regional network that should meet the needs of
 other Guff nations as well.

 The following activities are proposed in cooperation with and support of the host governments:

       •      Immediate steps would be taken to collect and analyze meteorological observations and
              forecasts,  record visual observations of  the  smoke plume,  and review  existing
              monitoring  data.    Plume observations via  satellite would  be obtained  daily,
              supplemented by periodic on-scene aerial transects designed to Characterize the overall
              geometry of the plume.

       •      A ground-based sampling network of portable equipment would be installed by EPA and
              others at approximately 15-20 locations to measure particulate matter less than 10
              microns in diameter (the particle size most likely to  penetrate deeply into the lungs).
              The ratio of the less than 10 micron particles to total particulate load would be
           .   established. Limited organic analysis would be  undertaken.

       •   •  Measurements of carbon monoxide, carbon dioxide, methane, hydrogen suifide, sulfur
              dioxide, particle size distribution, elemental  and organic carbon, metals, polycyclic
              aromatic hydrocarbons, and acid aerosols would be obtained close to  the fires by.
              NASA and MIST. These measurements should attempt to characterize and categorize
              emissions from several specific wells.

       •      Specially equipped aircraft from the University  of Washington (April 15  to May 15),
              NCAR (May 1 to June 1), and NOAA (July 1  to August 1) would be  deployed  to
              measure downwind plume composition and dispersion, radiative properties and climatic
              effects, and effects on clouds and precipitation.  On  the basis of the initial aircraft
              results, a longer-term sampling program would be designed to monitor the relaxation
              of the atmospheric environment as the fires are extinguished.   .

EXHIBT 1: Air Sampling Results
                             U.S. EMBASSY, KUWAIT
Results for 3/16

       none detected

Sulfur Dioxide:7
       none detected

Inorganic Acids:
       HNO, Sppb
                          detection limit: 2-4.6 ppb

                          detection limit: 0.04 mg/m3

                          detection limit: 1-6 ppb
Benzene 0.4 ppb,  n-Heptane 0.13 ppb, Toluene 0.61 ppb, Ethylbenzene 0.-3 ppb,
para-Xylene 0.29 ppb, o-Xylene 0.12 ppb.

Na <  1.0 ug/m3, Mg 2 ug/m3, Fe 2 ug/m3, Ca 8 ug/m3, Al 2 ug/m3.
Results for 3/17

       Naphthalene 0.31 ppb

Sulfur Dioxide:
       < 0.02 mg/m3                        •

Inorganic Acids:
       HCI 3.0 ppb, M2S04 1.0 ppb, HN03 2.0 ppb
       Cyclohexane 1.31 ppb,  Benzene 4.0 ppb, n-Heptane 6.0  ppb.  TCE  0.7 ppb,
       Methyteydohexane 2.0 ppb, Toluene 7.7 ppb, n-Octane 3.0 ppb, Ethylbenzene 1.7
       ppb. p-Xytone 5.4 ppb, Styrene 0.4 ppb. o-Xylene 2.4 ppb, n-Nonane 1.9 ppb. Cumene
       0.2 ppb. Mesttylene 0.6 ppb, D-Umonene 0.1 ppb, n-Decane 1.5 ppb, n-Undecane 1.0
       ppb. n-C120.7ppb, n-C130.4ppb.                      \
       Na 10 ug/m3, Mg 2 ug/m3, Fe 3 ug/m3, Ca 10 ug/m3, Al 2 ug/m3.

 EXHIBT 1: Air Sampling Results
                        U.S. EMBASSY. KUWAIT, continued
Results for 3/18

       Naphthalene 0.16 ppb

Sulfur Dioxide:
       < O.Offmg/mS

Inorganic Acids:
       HCI 6 ppb, HN03 2 ppb, H2S04 1.0 ppb.
       Cyclohexane 1.2 ppb. Benzene 5.2 ppb, n-C7 8.1 ppb, Methylcyclohexane 3.0 ppb,
       Toluene 14.6 ppb, n-C8 4.7 ppb, Ethylbenzene 3.2 ppb, p-Xylene 9.3 ppb, o-Xylene
       4.7 ppb, n-C9 2.7 ppb, Cumene 0.3 ppb, Mesitylene 1.3 ppb, Naphthalene 0.18 ppb,
       n-CI 0 1.7 ppb, n-C11 1.0 ppb, n-C12 0.7 ppb, n-C13 0.3 ppb.
       Na 5 ug/m3, Mg 1 ug/m3, Fe 1 ug/m3, Ca 7 ug/m3, Al 1 ug/m3.

 EXHIBT 1: Air Sampling Results
                                 CAMP FREEDOM
 Results For 3/17
       Naphthalene 0.09 ppb, 2-Methylnaphthalene 0.06 ppb, 1 -Methylnaphthalene 0.04 ppb.
 Sulfur Dioxide:
       0.12rng/m3  0.045 ppm

 inorganic Acids:
       HCI 16 ppb, HF 23 ppb, H2SO4 31 ppb
       Cyclohexane 7 ppb, Benzene 4.6 ppb, n-C7 25.6 ppb, Methylcyclohexane 9.5 ppb.
       Toluene 13 ppb, n-C8 18 ppb, Ethylbenzene 2.7 ppb, p-Xylene.8 ppb, o-Xylene 4.5
       ppb, n-C910.9 ppb, Cumene 0.5 ppb, Mesitylene 1 ppb. Naphthalene 0.18 ppb, n-C10
       6.3 ppb, n-C11 4.1 ppb, n-C12 2.3 ppb, n-C13 0.9 ppb, n-C14 0.5 ppb. These levels
       may also include emissions from vehicles in the area.  •     ..
       No Data.

 Results for 3/18

       Naphthalene 0.28 ppb

Sulfur Dioxide: ,
       <0.04 mg/m3 , < 0.015 ppm

Inorganic Acids:
       HN03 4.0 ppb, H2S04 4 ppb.
       Cyclohexane 2.8 ppb. Benzene 6.9 ppb, n-C7 9.7 ppb, Methylcyclohexane 3.9 ppb,
       Toluene 16 ppb, n-C8 5.4 ppb, Ethylbenzene 3.1 ppb, p-Xylene 9.5 ppb, Styrene 0.3
       ppb, o-Xylene 4.5 ppb, n-C9 3 ppb, Cumene 0.3 ppb, Mesitylene 1.1 ppb, n-CIO 1.5
       ppb, n-Cl 1 0.8 ppb, n-C14 0.2 ppb.
       Al 2 ug/m3, Ca 8 ug/m3, Fe 2 ug/m3, Mg 2 ug/m3, Na 3 ug/m3.

 EXHIBT 1: Air Sampling Results
                         MEPA DHAHRAN, SAUDI ARABIA
Results For 3/t 3

       none detected

Sulfur Dioxide:'
       <0.1 mg/m3, <0.037 ppm

inorganic Acids:
       H2S04 5 ppb.
       Benzene 0.3 ppb, Cyclohexane 0.1  ppb, Toluene 0.5 ppb, Ethylbenzene 0.-J ppb,
       p-Xylene 0.2 ppb, o-Xylene 0.1 ppb, Mesitylene 0.04 ppb.
       Al 2 ug/m3, Ca 5 ug/m3, Fe 2 ug/m3, Mg 2 ug/m3, Na 1 ug/m3.
       Cd o.ol ug/m3
Results For 3/14

       none detected

Sulfur Dioxide:
       <0.08 mg/m3 , < 30 ppb.

Inorganic Acids:
       HN03 2 ppb, H2S04 6 ppb.
       Cyclohexane 0.2 ppb. Benzene 0.5 ppb, n-C7 0.8 ppb. Methylcyciohexane 0.3 ppb.
       Toluene 0.7 ppb, n-C8 0.4 ppb, Ethylbenzene 0.1 ppb, p-Xylene 0.3 ppb, o-Xylene
       0.1 ppb, n-C9 0.3 ppb.
      Al 3 ug/m3, Ca 14 ug/m3, Fe 3 ug/m3, Mg 4 ug/m3, Na 8 ug/m3.

 EXHIBT 1: Air Sampling Results
                      U.S. EMBASSY RIYADH, SAUDI ARABIA
Results For 3/28

       none detected

Sulfur Dioxide: •
       < 0.08 mg/m3 , < 30 ppb

Inorganic Acids:
       N03 3 ppb, H2SO4 < 2 ppb.
       Benzene 0.3 ppb, Toluene 0.6 ppb, Ethylbenzene 1.0 ppb, p-Xylene 0.3 ppb.-
       At 8 ug/m3, Ca 4 ug/m3, Fe 1 ug/m3.
                               PORT SHUAYBAH
Results For 3/17

       no data

Sulfur Dioxide:
       < 0.05 mg/m3, , 19 ppb

inorganic Acids:
       H2S04 19 ppb.
       Benzene 4.2 ppb, n-C7 13 ppb, Methyteyclohexane 5.3 ppb, Toluene 15 ppb. n-C8 5.6
       ppb, p-Xytone 6.9 ppb, Ethylbenzene  2.2 ppb,  o-Xylene 2.9 ppb, n-C9  3.8 ppb,
       Cumene 0.2 ppb, n-CIO 2.9 ppb, n-Cl 1 1.9 ppb, n-C12 1.4 ppb, n-C13 0.8 ppb.
       Ca 4 ug/m3, Na 7 ug/m3, AL 2 ug/m3.

EXHIBT 1: Air Sampling Results
Results for 3/1 5
       no data
Sulfur Dioxide:
inorganic Acids:
       no data
                               Al MAQUA OIL RELO
                        .1 ppm
       Cyciohexane 0.6 ppb. Benzene 1.8 ppb, n-C7 3.2 ppb, Methylcyclohexane 1.1 ppb,
       Toluene 2 ppb, Ethylbenzene 0.4 ppb, p-Xylene  1.4 ppb, o-Xyiene 0.8 ppb, n-€9 2.8
       ppb, Mestilylene 0.3 ppb, n-C10 3.3 ppb, n-C11 3.5 ppb. n-Cl2 3.3 ppb, n-C13 1.8
       ppb, n-C16 1.1 ppb.
       Al 8 ug/m3, Ca 6 ug/m3, Fe 6 ug/m3.

 EXHIBT 1: Air Sampling Results
                               Al AHMADI OIL HELD

Results For 3/16

       none detected

Sulfur Dioxide:
       0.45 rrtg/m3 , 0.17 ppm

Inorganic Acids:
       H2S0427 ppb, HN03  10 ppb, HCI 9 ppb.
       Cyclohexane 0.4 ppb, Benzene 3.9 ppb, n-C7 2.5 ppb,  Methylcyciohexane 3 ppb,
       Toluene 2 ppb, n-C8 2.3 ppb, Ethylbenzene 0.5 ppb, p-Xylene 1.5 ppb, o-Xylene 0.9
       ppb, n-C9 2.9 ppb, Mesitylene 0.4 ppb. Naphthalene 0.5 ppb, n-C10 3.4 ppb, n-C11
       3.8 ppb, n-C12 4 ppb, n-C13 2.9 ppb, n-C14 2.9 ppb, n-C15 2 ppb, n-C16 1.7 ppb.
       Al 7 uQ/m3, Ca 4 ug/m3, Fe 20 ug/m3.
Hydrogen Sulfide: ND 0.3 ppm - detection limited .

       SO2 0.2 ppm; CO 1.9 ppm
                                      A TO

EXHIBT 1: Air Sampling Results
                              Al BURGAN OIL FIELD

Results For 3/17

       none detected; detection limit 50 ppb

Sulfur Dioxide:
       1.8 mgtmS , 0.67 ppm

Inorganic Acids:
       H2S04 30 ppb, HN03 32 ppb, HCi 15 ppb.
       Benzene 8.7 ppb, n-C7 4.6 ppb, Methylcyclohexane 2.5 ppb, Toluene 4.31 ppb, n-C8
       5.1 ppb, Ethylbenzene 1.3 ppb, p-Xylene 4.2 ppb, o-Xylene 2.4 ppb, n-C9 7r1 ppb.
       Naphthalene 1.6 ppb, n-C10 9.1 ppb, n-CH 10.4 ppb, n-C12 11.3 ppb, n-C13 7.8
       ppb, n-C14 7.4 ppb, n-C15 5.4 ppb, n-C16 4.6 ppb.
      Al 20 ug/m3, Ca 120 uo/m3, Fe 20 ug/m3, Mg 30 ug/m3.

      20 ppb;

      S02 0.23 ppm, CO 1.6 ppm, H2S none detected (0.1 ppm detection limit.)

 EXHIBT 1: Air Sampling Results
                               Al WAFRA OIL FIELD
 Resufts for 3/19

       no data

 Sulfur Dioxide:
       none detected 0.3 mg/m3 , 0.11 ppm.

 Inorganic Acids:
       none detected 10 ppb detection limit
       Benzene 2.3 ppb, n-C7 0.6 ppb, Methylcyclohexane 1.6 ppb, Ethylbenzene 0.3 ppb,
       o-Xylene 0.4 ppb, Mesrtylene 0.4 ppb                                  -
       no data

Hydrogen Sulfide:
       NO 0.16 ppm dection limited.
                              SAB1RJYAH OIL FIELD

Results For 3/17


Ground Level Sample:
       H2S none detected (0.1 ppm); S02 0.13 ppm; CO 1.1 ppm; NO and N02 none

       Benzene 9.8 ppb. Toluene 8.7 ppb, Ethylbenzene 10  ppb, m&p-Xylene 27.8 ppb
       o-Xylene 24 ppb, m-Ethyltoluene 15.8 ppb, n-C6 83 ppb, n-C7 60 ppb, n-C8 91 ppb,
       n-C991 ppb, n-C1089 ppb, n-C11 65 ppb.

3000 ft Sample:                                                        .
       H2S none detected; S02 0.08 ppm, CO none detected; NO and N02 none detected
       0.5 ppm.

1000 ft Sample:
       H2S, S02, CO, NO, and N02 none detected.

               Table 1
Kuwait OO Well Fires Real Time Monitoring
          March 13-20,  1991
Site, Date, Time
1) MEPA Faciltiy, Dhahran, SA
3/131 100 hrs
MEPA Facility
3/14 1300 hrs
2) AIDhuba :
3/15 1430 hrs
3) Umn Al Haiman
3/15 1500 hrs
4) Mina Abdulla
3/1 5 1530 hrs
5) Near A! Maqwa Oil Reid 3/1 5
1630 hrs
6) In Al Maqwa Well Plume
3/1 5 1700 hrs
7) U.S. Embassy
3/06 0900 hrs
U.S. Embassy .
3/1 8 1230 hrs
8) In Al Ahmadi Well Plume
3/1 6 1230 hrs
In Al Ahmadi OH Field
3/1 6 1300 hrs
9) In Af Burgan Well Plume
3/17 11 00 hrs
10) In Sabiriyan Well Plume,
Pooled Oil
3/18 1530 hrs
1 1 Mn Al Wifra Well Plume
3/19 1200 hrs
12) Freedom City
13) U.S. Embassy Riyadh, SA
3/20 1300 hrs
' • - ' .055

               Table 2
Kuwait Oa Well Fires Real Time Monitoring
          March 24-27. 1991
I Site, Date, Time

1) Al Safer Motorway
and Wafra Road
3/24 1440hrs
2) Al Ahmadi Gathering
Center #22
3/24 1530hrs
3) Al Ahmadi Hospital
3/24 1530 hrs
4) 1 mi. NW of Station
2 in Ahmadi OH Reid
3/24 1730 hrs
5) Al Maga Oil Field,
.5 mi south of 7th
Ring Road near oil pool
3/25 1400 hrs
6) Al Ahmadi Oil field
(sam * as Station 4)
3/25 1500 hrs
7) Al Ahmadi Hospital
(same as Station 3)
3/25 1545 hrs
8) Al Safer and Wafra
3/25 1615 hrs
9) Al Ahmadi Hospital
(same as Station 3 and
3/27 1020 hrs
101 Al Ahmadi Hospital
(same as Stations .3, 7,
and 9)
3/27 1040 hr»
11) Al Ahmadi Hospital
(same as Stations 3. 7,
9. and 10)
3/27 11 00 hrs
12) Shuaiba Port
3/27 121 5 hrs

(1 5 min.
(15 min.
(32 min.
(10 min.
(17 min.

(13 min.
(15 min.
(16 min.
(20 min.

(20 min.

(20 min.

(15 min.



























' -







' —



Site, Date, Time

i '•
13) Al Safer Motorway
and Wafra Road
3/27 1300hrs
14) 16 Kilo-meters SE of
Al Safer and Wafra
3/27 1330 hrs
15) 27 Kilo-meters SE of
Al Safer and Wafra
3/27 1350 hrs
16) 5 Kilometers South
of Khafji, Saudi Arabia
3/27 1510 hrs
(12 min.
(12 min.

(15 min.

(14 min.


— ;





— .


0.0 .

— . •


                Table 3
   Sulfur Dioxide Bubbler Measurements
           (Acidimetric Method)
Temporary Hospital Locations in Kuwait City
              March 13-24


Mubarek Al Kabeer

Al Farwaniya


Al Jahra

. .3/13/91
3/24/91 .
3/16/91 -
' 3/23/91
3/24/91- '
Concentration (ug/m3)
23.72 J

                                      APPENDIX B
                         A REGIONAL GULF AIR MONITORING PLAN
                                  IN RESPONSE TO THE
                              1991 KUWAITI OIL FIELD FIRES
              Purpose of the Plan
              Team Membership
              Concerns and Questions
              Basic Understandings
Short and Longer Term
              Nature of the Fires and Oil Reid Locations
              Qualitative Meteorological Observations
              Preliminary Air Quality Information
              Health Issues and Concerns
              Data Bases and Information Management
             Preliminary Findings and Conclusions to Date
             Phased Steps for Expanding the Network
             Measurement Parameters at each Site
             Health Monitoring Surveys

 IV.    SUMMARY   •
             Preliminary Findings and Conclusions to Date
             Basic Network Requirements
             Urgency and Timing
             Phased Approach to Implementation
             Unresolved Issue*

       List of existing and proposed air monitoring sites
       Locations of existing and proposed air monitoring sites

                          A REGIONAL GULF AIR MONITORING PLAN
                                   IN RESPONSE TO THE
                               1991  KUWAITI OIL FIELD FIRES
        The U.S. Interagancy Air Monitoring Team is working with the Saudi Arabian Meteorology
 and Environmental Protection Agency (MEPA) to develop an air monitoring plan for the Gulf region
 that will provide information to assess the impact of the (Kuwaiti] fires in Saudi Arabia. This plan
 is being discussed and developed with the King Fahd University of Petroleum and Minerals in
 Dhahran, the Cooperation Council for the Arab States of the Gulf (CCG), and the Saudi Arabian Oil
 Company (ARAMCO). The plan is being developed at the request of Dr. Tawfiq, Vice President of
 the Saudi Arabian MEPA. Meetings have been held with officials from all of the above mentioned
 organizations.                                  ' •          ,  .

        1..1  Approach                     ,

        The Team is gathering information on the existing air monitoring networks in the region
 operated by MEPA, ARAMCO, Kuwait Bahrain, and the Royal Commission of Jubayl and Yunbo.
 The spatial distribution of the existing network in the Region is being reviewed as to the location of
 site *, the air pollutants and meteorological variables that are monitored at each of these sites and
 the quality of existing data.  That review is to determine if the existing network needs to be
 expanded in terms of the air pollutants and meteorological variables mo litored and additional air
 monitoring stations to determine the impact from the oil fires.  The capabilities of the existing
 agencies and governments to deal with a more complete network is also being investigated.

        1.2  Objectives of Air Monitoring in the Gulf Region

       Air monitoring data is needed for the following reasons:

a.     To provide an Earty Warning Health Advisory System for the Gulf Region to respond to the
air pollution resulting from the Kuwaiti oil fires.  The proposed Early Warning System could be
based on an adaptation of the U.S. Air Quality Index, the Pollutant Standards Index (PSD, which
can be modified to use Saudi air quality standards.  The index would provide for health advisories
to the affected populations so they can minimize their exposure to high pollution levels.

b.     To track the air pollution from the Kuwaiti oil field fires over time to asses the potential long
term health and ecological effects. The air monitoring network proposal being developed is being
coordinated with a parallel effort to develop a health monitoring information system.

c.     To collect samples of airborne particles to perform toxicrty testing and dose response
assessment utilizing in-vivo animal models.

d.      To facilitate evaluations of models which are use to
behavior of the oil field emissions. Data from the expanded GULF REGIONAL AIR MONITORING
NETWORK of ambient airquality and meteorological data will be important for those evaluations.

        2.1 Preliminary Investigation of the Existing Monitoring Networks

        The air quality monitoring sites listed in Table 1 have been identified for each of the Gulf
 nations that could contribute monitoring information within the sphere of influence of airborne
 effluents from the Kuwait oil fires.

        Based upon our review of the existing  networks, the principal pollutants which are missing
 are in the Saudi network and in the present Kuwaiti network are PM10, which represents paniculate
 matter with particles less than 10 microns in diameter, poiycyclic aromatic hydrocarbons (PAH) and
 volatile organic compounds (VOC). -With respect to PM10, these  are the particles which are most
"likely to penetrate deeply into the lung.  It should be noted that Kuwait has collected paniculate
 data using an Anderson Cascade Impactor, with limited size distributions below seven microns
 within a total suspended  paniculate sample. Because of the importance of this particular pollutant
 and the extensive paniculate resulting from the oil fires, the Team is recommending that special
 efforts be initiated to gather PM10 data and if possible to determine its constituents - trace metals
 and hydrocarbons. The
 PM10 data collection effort should be supplemented with the collections of PAH samples and if
 possible, grab samples for VOC analysis.

       An ongoing effort is being conducted to examine the analytical laboratory support for air
 monitoring in Kuwait,  MEPA, KFUPM, and Saudi ARAMCO. The Kuwaiti laboratory capability to
 analyze air and paniculate samples has been left largely intact. As of March 27,  1991, two of the
 three continuous monitoring stations have been activated and are collecting data. The remaining
 site has its continuous sampling equipment however,  like the analytical laboratory, is without
 electrical power.  There is no projected date for power at these locations.  The analytic il laboratory
 has had experience with polyurethane foam (PUR extractions for PAH measurements 
adequate equipment, standards, and training, they all could support the sampling and analysis
required for this program.



The Distribution of Air Pollutant and Meteorological Monitoring Sites by Gulf Nation.

                     Saudi ARAMCO
                     Royal Comm.
                     for Jubayl
Number of Air antfl Met Sites

       3 fixed sites - meteorology and air pollution at
       different locations of the same city

1 mobile site • meteorology and air pollution

1 fixed meteorology site

8 air quality sites with meteorology

6 additional meteorology sites

Cluster of Five Stations in
                           3 sites with continuous monitors (2 with power, 1
                           withoit power).              x

                           4 sites with Anderson samplers, high vols. and
                           dustfall buckets,  (no power)

                           1 additional TSP site (no power)

                           6 additional dustfal' sites

                           4 temporary S02 bubbler sites located at 4 hospitals
                           (some power)
                           2 sites

                                 3 mobile monitoring sites

        A substantial number of meteorological measurement sites exist within the eastern
 provence of Saudi Arabia and the nations of the Gulf Region. Most of those sites are along the
 shoreline of the Gulf.  ARAMCO operates 14 sites with meteorological data. Eight of those sites
 have collocated air quality measurements.  Three of those sites are over the Gulf waters on
 platforms or on an island.  MEPA (Saudi Arabia) has five sites with collocated air quality
 measurements, in addition, there are surface observations collected at many of the airports
 throughout the kingdom. Surface meteorological data are being collected at other Gulf region
 locations. Their locations may be identified through the WMO publications. Previously, there were
 other surface measurement sites within Kuwait but their operational status remain unknown at this

       Two upper air balloon sounding sites are operating within Saudi Arabia.  One site is at
 Dhahran and the other is about 115 miles to the SW of Kuwait City at At Qaysumah. Twice daily
 soundings are collected at those locations,  at 0000 and 1200 GMT.  Prior to the war in Kuwait,
 twice daily upper air soundings were made  at the Kuwait international Airport, the resumption of
 those soundings could be of substantial benefit to describing the airflow across the areas of
 Kuwaiti oil fires.
       2.2  Meteorological Observations

       The following summarization are based upon first hand observations of the smoke plumes
and fires.. Those observations were made during overflights and during vehicle traverses both
within the oil fields and along roads outside of the burning oil fields.

       For any given day, the prevailing large-scale meteorological pattern will be the main driving
feature which determines where the smoke plumes will be located and how dense they will be.

       Indi iduai smoke plumes appear to act in manners typical of buoyant plumes from ground
level sources or plumes from short chimneys.  Plume rise, the development of a bent-over plume
geometry, etc., seem to apply to the individual well-head fires; some have jets of fire and others
are nearly surface based burnings of the more combustible fractions of crude oil spread across the
ground in the vicinity of the well.  Most of the fire plumes rise to between 500 to 1000 feet above
ground level before becoming mostly bent over, although some plumes have a significant amount
of smoke remaining within a few hundred feet of the ground.

       Collectively, as groups of multiple fires within oil fields with a high density of burning wells
(particularly Greater Burgan), they  assert a meteorological influence of their own. It is suggested
that the grouping of fires with a horizontal diameter of 15 to 25 miles provides enough of an
intense "heat-island* that significant additional vertical rise of the smoke occurs inside the area.
That additional plume rise lifts smoke to elevations often 3000, 4000, to 5000 feet above ground
within the initial few miles downwind.  Eventually, portions of the smoke rise even more, with
multiple layers often forming at heights up to 8000, 9000. even 12000 to 13000 feet. Between
such layers and at the tops of layers many tens of miles downwind, a generally diffuse and
homogenous zone of smoke hes been observed.  The eventual smoke height limits are bounded by
the regional vertical temperature structure and synoptic weather characteristics.  Information
reviewed to date suggest that those maximum heights are mostly 8000 to 12000 feet within the
initial 100 to 200 mites downwind from the KuwaW oil field fires.

       With the creation of a local heat-island, a distinct inflow of near surface air has been
observed within the initial 500 to 1000 feet above ground level. At times, that inflow of wind is.

 estimated to be 5 to 15 m/sec in strength.  Smoke plumes at the peripheral bounds of the burn
 area tend to slant inward toward the center of the burning field instead of pointing downwind with
 direction of the expected ambient wind..

        Local variations in daily wind flow, along with the fire-storm like winds, are likely to
 produce preferred locations and times of day at which more concentrated smoke plume exposures
 reach ground level. Prevailing winds are from the northwest throughout tine year.  During the
 daytime a sea breeze can be expected to develop at the Gulf shoreline and progress inland as the
 day progresses. That inflow of air can readily clear out the smoke plumes and yield substantially
 cleaner air at ground level on the Gulf side.  Along the leading edge of the sea breeze front there
 likely may be a zone of extended and elevated exposures to fire effluent. That zone may well
 extend down the shoreline from  Kuwait City some 100 km. The area of greatest susceptibility
 appears to be to the southeast of fires in the Al Ahmadi oil field (part of the greater Burgan field).
        2.3 Data Base Management

        In addition to evaluating the existing air monitoring networks, the Team made an initial •
 review of existing date systems to handle the air monitoring data.  The previous meteorological, air
 quality, and visual observations of the oil field fire plumes should be archived, along with data to be
 obtained during the period of on-going Kuwaiti fires.  A dedicated facility for the performance of
 that archiving does not appear to exist.  The data management task is likely to be a sizeable task
 and extended of a period of a year and more.

       Some of tine general functional needs of the data management system include the
 following.  The data which will need to be assembled into the data base will likely come from many
 different sources and exist in diverse format* and media.  One role of the data base management
 activity will be to assemble all information into a common, uniform structure.  Tie second and   .
 equally important part of the data archiving is the provisions of a uniform and consistent
 mechanism for the retrieval of data by participating agencies.. The degree to which that data base
 is well .formulated, will significantly affect the efforts of users of the  data as they attempt to study
 and interpret the measurements.

       A number of possible methods exist for the set-up of a computerized data base.
 Commercial software and hardware of various degrees and complexity  and cost exist which would
 satisfy the data management needs. Before the choices of system software and hardware art
 made, the functionality of the overall system  and the manners in which users work with the data
 sets should be considered. For example, h may be required that the  data base be a  •relational"
 data base.  Other sources of software that might be used to handle a large volume of data would
be the U.S. Environmental Agency's (USEPA) Aerometric Information and Retrieval System. That
system can handle hourly data and has considerable software available to both summarize  and
analyze the ambient air data.

       MEPA has asked the Team to recommend the type of computer and associated software
needed to manage the data collected in order to implement the monitoring plan. For now,  it is
more appropriate to defer specific recommendations.  A number of general performance
characteristics may be stated but specific details should be formulated in conjunction with data
base specialists at a later date.

        While preliminary, the Team believes that the following recommendations should be
 implemented based on our initial data gathering exercise.  A general objective is listed as a
 recommended goal and below that objective,are listed several needed items or activities to facilitate
 the achievement of the overall goal.

 1.  Objective:   Provide a framework for an early warning advisory capability for areas expected to
               be impacted by effluents from Kuwaiti oil field fires.

  Needs:      a.      Meteorological data observations and forecasts
               b.      Visual observations from key receptor areas
              . c.      Review existing 'monitoring data
               d.      Calculate smoke trajectories and concentrations

 2. Objective:   Provide a cursory wide-area indication of the distribution and composition of the
               Kuwaiti oil field fire effluents.

  Needs:      a.      Establish 10 to 15 PM10 monitoring locations using portable monitors.
               b.     Train personnel in the operation of the portable PM-10 monitors and develop
                     the analytical support capability within Saudi Arabia and Kuwait.
               c.     Define the PM10 to TSP ratios.
               d.     Define the composition of the plume by XRF analysis for limited organic
              e.     Establish a central media preparation  and analysis location.
              f.      Define the baseline contribution of the ambient aerosol from the surrounding
3. Objective:  Characterize the ?jirial smoke plume.

       Needs: a.     Collect many of the same plume measurements recommended for the
                     ground monitoring array
              b.     Collect descriptions of the width and vertical extent of the smoke plume at
                     several downwind distances
              c.     Characterize the regional background by samples outside of the smoke

4. Objective:   Develop a more complete profile of the smoke plume constituents

       Needs: a.     Procure equipment  for a limited number of comprehensive air quality
                     monitoring stations to collect:  TSP, PMto, organic, and inorganic
              b.     Establish a limited number of comprehensive air quality monitoring stations
                     to collect: TSP, PM10/ organic, and inorganic constituents. These should be
                     collocated with the continuous monitors wherever possible.
              c.     Train individuals to  operate and maintain the sampling instrumentation in
                     support of tiie monitoring program
              d.     Procure the necessary analytical laboratory equipment required for analyses
                     of the samples collected under this objective
              e.     Train laboratory personnel in the preparation of the sampling media, QA/QC
                     procedures required and the subsequent sample analytical procedures.

5. Objective:  Determine the need for expansion of the monitoring network to a wider regional

       Needs: a.     Review the data developed from the limited network.
              b.     Assess the current and projected status of control of the oil field fires and
              c.     Review the population health survey statistics.
              d.     Review the suitability of the sampling strategy, and modify where needed
              e.     Expand the limited network as the situation requires, data analysis indicates
                     an additional need for data, the response of the affected populations
                     indicate, or the model requires additional parameters.

       Many of the same considerations listed for air quality above also apply to meteorological
considerations.                                                        •                  ,,

6. Objective:  Provide a meteorological data stream to facilitate the modeling and prediction of
              areas expected to be impacted by effluents from Kuwaiti oil field fires.

   Needs:     a.     Upper air balloon sounding data representative  of the Kuwaiti oil field fire
                     area and Gulf region plume transport.

              b.     Supplemental surface based measurements of wind speed and direction,
                     temperature, moisture content of the air (dew point, relative  humidity, etc.),
                     solar radiation, atmospheric pressure, precipitation.

7. Objective:  Provide a meteorological data set to investigate the areas of climate modifications
              occurring due to effluents from Kuwaiti oil field fires.

   Needs:     a.     Supplemental surface based measur intents diffuse and direct solar radiation.

              b.     Special collections of precipitation throughout the region to examine the pH
                     and chemistry of the rains.

             ' e.     Aircraft soundings and profiles of smoke, winds and temperatures, air
                     quality related measurements of plume compositions and concentrations
                     representative of the Kuwaiti oil field fire area  and Gulf region transported


        This section of the report discusses a prioritized plan of stepwise incremental actions for
 the phased implementation of the recommendations discussed.above. Five phases for
 implementing the plan follow.                                '

        4.1 Phase 1.  Provide a Framework for an Earlv Warning Advisory

        In order to accomplish this task, the following action items need to be initiated or
 incorporated into the task framework.
        1.      Gather daily weather forecasts to predict meteorological conditions which would
               effect  pollution potential in both Saudi Arabia and Kuwait.

        2.      Use visual observations from key receptor sites to determine possible pollution

        3.      Gather existing air monitoring data from fixed and mobile sites operated by MEPA,
               Saudi ARAMCO, and the governments of Kuwait and Bahrain to develop a data
               base of existing data.

        4.      Establish a daily briefing for representatives of the many entities concerned with
               behaviors and fate of the aerial effluents from the oil field fires in  Kuwait.

        5.      Develop a daily map depiction of the aerial distribution of the smoke plumes across
               the region using satellite imagery, for each day since initiation of the oil field fires.

        6.      Issue a dairy general statement about the expected behavior(s) of the oil fire
               plumes. Areas of p*1 tent' illy adverse conditions could be treated as locations for
               which  advisories would be issued.

        7.      Provide forecast meteorological conditions across the region for the next 2 to 3  day
               period, including the expected location(s) of the smoke.

        8.      Obtain data from the MEPA network throughout the eastern province.

        9.      Direct the craws of the SLAR aircraft (USCG Falcon Jet), to continue on a regular
               basis the present visual observation and mapped notations concerning the horizontal
              extent  of the smoke plumes and the estimations of altitudes of layer bases and

        10.    The above information could be compiled initially in hard copy form. Later it should
              be stored on electronic media in a way that an existing PC data management
              system could readily incorporate it

       4.2    Phase 2. Establish a PM.T Monitoring Network Using Portable PM.« Monitors.

       Working in conjunction with the Saudi Arabian Meteorology and Environmental Protection
Agency (MEPA), the proposed plan has been developed to collect information on PM10, which

represents paniculate matter with participates less than 10 microns in diameter.  At the present
time there is no PMto monitoring in the Gulf region.


       The objectives for this effort are as follows:

       1)  Determine the magnitude or the health threat to residents of population centers and
       field-based military personnel impacted by the oil field fires and typical sources (windblown
       desert soils).

       2)  Establish a scientifically based capability to alert these affected populations prior to the
       onset of the potential health threats from real-time measurements.

       3)  Establish a technical basis for executing predictive air quality dispersion models which
       simulate the oil well fire emissions, background sources, and consequent impacts over
       space and time.

       4)  Establish a regional network of PM10 stations using a consistent monitoring methodology
       across the countries of Saudi Arabia, Kuwait, and Bahrain.

       5) Train personnel to operate and analyze the media produced by the network from each of
      the participating countries.

      6) Develop  a regional data base and encourage the sharing of data developed from the
      network with all participants.

      The above objectives convert the foregoing goals into discrete actions:

      1) Determine the spatial temporal frequency, and severity of the impact t j the resident
      populations and military centers affected through the application of saturation sampling  .
      techniques with portable PM,0 samplers.     _

      2) Establish the  correlation of real-time surrogate monitoring data to data generated from
      direct sampling methods through the collocation and simultaneous operation of both
      methods over time.                                                 '  •

      3)  Where possible use impact data collected from samplers as an input to the dispersion
      model, run tile model 'backwards* to develop a better estimate of the emission rates of the
      fires and produce a higher level of confidence in the predictive modeling results.

      4)  Obtain from the I'rterature and/or develop from source sampling analysis chemical
      profiles of all major pollutant sources in order to:
                ,'                          '                     •
             a) identify those contaminants that pose the greatest health concerns and to
             develop an estimate of acceptable ambient levels (AALs) prior to the conduct of
             field  work.

             b) identify the chemical "signature* of the major contaminants and other tracers
             characteristic of the primary sources,

               c)  enable apportionment of these contaminants and other tracer compounds from a
               simple total mass concentration measured by the ambient samp'es, and

               d)  attempt to relate these levels in turn to the surrogate real-time monitoring
               methods for use in issuing timely health risk alerts.

        5) Ensure that the data generated  by the network are of a demonstrably high quality
        (precision & accuracy), completeness, representativeness, and comparability.
TOPS 070/030
SMOKE 060/030
SMOKE 070/030
SMOKE 070/030
THIN SMOKE 050/030
THIN SMOKE 090/080
SMOKE 090/060
SMOKE/HAZE 080/060

       Conduct a PM,0 saturation sampling study for the determination of the temporal and spatial
features of the impact of the oil well fires and attempt to reconcile the data with existing model
estimates.  PM10 mass concentrations would be available within 24-48 hours following sampling.
No on-site meteorological, gas, or aerosol monitoring or chemical analysis it required (chemistry
could be attempted later on the preserved media).  One or more portable nephelometers would be
collocated at several sampling sites to develop correlations between manual and continuous (real-
time) methods for alert advisories.

       A total of 15*20  portable  PM,0 sampler* equipped with quartz filters would be run
simultaneously on a daily basis or "triggered* (impact forecast) basis throughout the study area.
Network design would involve a "nested* approach to address the objectives:

        1) samplers sited at background locations (not impacted) by the smoke plume and
        samplers in populated areas.
        2} samplers in populated areas impacted by smoke.

        Samplers could be "ganged* (2 or more) and programmed to run consecutively at individual
 sites if filter clogging problems occur because of high loading.  Further,  multiple samplers could be
 collocated at certain sites to collect fine participates (less than 2.5 microns) and coarse particulates
 (2.5-10 microns) on teflon sample filters (facilitating XRF elemental analysis).  One of the 10-12
 sampling sites would be equipped with duplicate samplers in an effort to develop sampling   ,
 precision estimates.           .                        '

       This comprehensive program will yield the following:

        t)  short turnaround PM,0 concentrations,

       2)  gross estimates of the fire-specific contributions to total mass could be derived by
       subtracting background concentrations from the impact site concentrations,

       3)  applying assumptions on the source profiles to pollutant loading attributable, to the welt
       fires, estimates of individual target compound loadings  could be computed and a
       comparison to AALs made,

       4). correlations factors can  be determined between real-time surrogate methods and manual
       methods, and

       5)  impacted sampling media would be available for subsequent  intensive chemical analysis
       in an attempt to reconcile assumed source signatures and extracts can b j used to perform
       any other analytical tests (mutagenicity). Special  precautions may be needed to preserve
       tine sample integrity during storage and transfer.

       limitations:  no orvsite meteorological data to calculate emissions rates, no on-stte
chemistry (unless developed) to confirm critical assumptions, and no concurrent gas or acid aerosol
measurements to evaluate or correlate with the paniculate data.

       Resources: 1-2 professionals, 1 field technician per site, if H must be operated individually
(actual number contingent upon the network logistics and potential  •dogging' implications),
portable PM,0  saturation  samplers, portable nephetometers, tap-top computer, microbalance,
expendables and sundry support gear.               .

       In summary, this approach is PM10 mass data-rich, and  assumption rich and in contrast to
being reconciliation and broad pollutant characterization-poor.

      •* ,

       The alternate approaches identified above involve  the use of a variety of sampling and
analytical methods summarized below:                                 .
       Portable PM10 Saturation Samplers
              segregate  and capture of filter, particulates of 10 micron size (respirable particles)
         '•'. •  and smaller.                         .             .

               battery-operated, lightweight, rugged, inexpensive, small, and quiet.
               easily deployed and operated.
               programmable timer for unattended on and off.
               rechargeable battery packs.
               continuous operation up to 30 hours on a single charge.
               precise and accurate.
               low detection limit of approximately 5 ug/m3
               sustained operation under high paniculate  loadings, e.g. 100 ug/m3 or more.
               electronic sample flow regulation.
               electronic sample flow totalizer.
               low flow shutoff/warning.
               can accept a variety of other pollutant sampling media (e.g. PDF, DNPH, charcoal,
               denuders, etc.) or take whole air samples (Tedlar bags) with tittle or no
               modification.                                                        "

       Portable nephelometers
               many of the same attributes of the PM,0
               battery operated
               effectively measure particulates of .1 micron  diameter or smaller.
               continuous reading, storing five minute averages.
               continuous operation from 2-48 hours on a single battery  charge.
               internal storage for up to nine days of sampling data.
               data download to a portable lap-top computer through RS232 serial port.,
               operationally equivalent to standard nephelometers.

               five to six place balance.
               rugged, transportable while precise and accurate.

       Reid XRF Unit
               similar MOLS with situ laboratory units.

       Support Gear                                                    .
              calibration and audit gear, tools  and diagnostic equipment, etc.


       The recommendation for siting of the portable PM,0 samplers is predicated on providing a
large area of coverage for developing a better estimate of  the areas impacted by the plume, a
cross-sampling of population and troop centers, and to a capability to provide a technology transfer
to Saudi Arabia, Kuwait and Bahrain.

              collocated site at K.F.U.P.M. (2  sites)
       -two sites in Riyadh <1 U.S. Embassy & 1 MEPA location)
      • -   •    one site Royal Commission at Jubayl
              one site Saudi ARAMCO at Tanajib
              one site MEPA at Khafji

               Total of seven (7) sites.
               three sites (3) located at the two operational continuous monitoring sites within
               Kuwait City (one site collocated)
               one site at Camp Freedom
               One site U.S. Embassy
               One site Al Hamadi (Kuwait Oil Company Hospital)

               Total of six (6) sites.
               two sites at selected troop locations

               Total of two (2) shes.
              one site to be determined

              Total of one (1) site.

        A total of sixteen samplers are committed to field sampling with the remainder as spares or
 as changes to the sampling plan requires.

 Phase 3: Characterize the Aerial Plume.    .   .

        This phase should follow closely with Phase 2, in order to characterize the 3-dimensional
 nature of the smoke plumes from the fires in Kuwaiti  oil fields.  To achieve that goal, many of the
 same plume measurements collected by the existing and proposed ground level measurement
 locations should be provided by the aerial sampling platform. Obviously, the longer time integrated
 samples (e.g., 24 hour total values, averages, etc.) cannot be reproduced by aircraft borne devices.
 Short-term and across plume integrated measurement descriptions may be obtained to characterize
 the special extent and details of actual constituents of the elevated portions of the oil fire plumes.

       The aerial sampling activities may be separated into measurements which address the three
 general zones of plume characteristics, from an a meteorological sense.  Those zones are the  1)
 close-in zone, 2)intermediate or transition zone and 3)extended or distant zone.  Measurements
 very near the wellhead are difficult to impossible to obtain due to excessive heat and great levels of
 turbulence.  Measurements at intermediate distances  will be difficult and many locations within the
 clustered groups of burning wells may be unsafe for aerial traverses due to the extremely dense
 smoke and hidden turbulent plumes.  Measurements  at the longer distances, a few miles
downwind of the burning wells, should be possible.   Measurements from a few to several hundred
miles downwind of the fire area should be feasible. Within that distance range the approximate
concentrations,  plume dimensions, and estimated mass flux in the downwind directions may be
approximated. The aerial sampling strategies should concentrate on the obtaining of those types of

         4.4    Phase 4. Develop a Complete Profile of the Smoke Plume Constituents.

         Obtain additional-equipment to expand existing continuous monitoring high priority sites
  Kuwait and Saudi Arabia
        The survey conducted by the Team during Phase 1 of this plan indicated that within the
  region, respirable paniculate sampling technology, aerosol and total paniculate sampling and
  analysis for volatile organic compounds (VOC), semi-volatile organic compounds (SVOC), and PAHs
  were either not available or insufficient to properly characterize the effects of the oil well fires on
  the population centers and the troop concentrations within the region.

        This phase of the plan proposes to bring into the region several new technologies and to •.
  train personnel within the region to operate samplers, to condition, and analyze the several new,,
  media necessary to support this expanded  network.  The objective of this process is to develop a
  stand-along capability within each participating country for aerosol and paniculate monitoring
  which will support the Gulf regional air quality characterization .and index plan outlined in Phase V:

        During this phase of the plan the Team proposes to expand the continuous air and
 meteorological monitoring currently being conducted  within the region at six sites. These sites are
 recommended based on  the need to jointly develop the sampling and analytical capability within the
 region to ensure that it becomes self-sufficient and sustainable. The particular technologies
 outlined below are not currently operateJ within the region nor are the analytical procedures
 required to  support them currently being  utilized. However, the Phase I survey indicates that with
 sufficient training, additional equipment,  and some experience with actual field samples the transfer
 should be relatively smooth.

        Initially,  the Team's recommendation of six sites strategically placed along the axis  of the
 area of greatest impact by the oil will fire plume wilt generate sufficient samples for the required
 training, while also providing critical data not currently being collected by the existing  networks or
 available through the portable Pwl,0 network proposed in Phase II.  As this data base develops it
. can be used to better define the constituents of the plume and thus permit a more accurate
 assessment of the potential long-term health risk.

       The equipment listed below should be collocated with the full compliment of continuous air
 and meteorological monitors described in Phase V at these six proposed sites. A brief description
 of the equipment is provided below:

       TSP High Volume Sample - used to collect • 24-hour sample of the total suspended
       participates, operated nominally at 50 OFM, and typically uses 8x10 inch glass fiber,
       quartz, or teflon filter media.

       PM,0 High Volume Sampler • used to collect a 24-hour sample of the 10 micron and smaller
       sized fraction of the total suspended paniculate sample collected by the TSP sampler
       above, ssmpler is typically operated at  40 CFM, and utilizes an 8x10 inch Quartz or teflon
       filter'media. Note: glass fiber media should not be used rf there is concern for a known
       sulfate artifact formation problem.
       PM10 Manual Dichotomous Sampler - used to collect a 24-hour sample of the 10 micron
       particular size fraction of the TSP, sampler operates at 16.7 liters/minute, and utilizes two
       (2) 37  mm diameter Teflon filters to collect a  fractionated sample with a cut point of 0-2.5
       micron (fine fraction) and 2.5-10 micron (coarse fraction).

         VOC Canister Sampler * used to collect up to a 24- hour integrated whole air sample in six-
         liter evacuated stainless steel canisters, interior walls are passivated to minimize sample
         degradation, samples volume can be regulated.by either limiting the volume to ambient
         pressure or pumping in addition sample to an approximate volume of 16 liters, these
         samples can be used for the determination of total hydrocarbons or analyzed for specific
         hydrocarbons, multiple analysis are available from a single pressurized canister sample.
       -  Note: An extensive canister cleanup process is required prior to the collection of additional
         air samples.

         Polyurethane Foam (PUF) Low Volume Sampler • used to collect 24-hour aerosol samples
         utilizing small AC or battery operated pumps at flow rates less than five liters per minute on
         relatively small glass cartridges containing a PUF plug, these samples can be extracted and
         analyzed for PAHs, or other SVOCs. Note: Both  the glass cartridge and PUG plug require
         an extensive cleanup procedure prior to re-use.                                 •

         Optional Tenex/Charcoal/XAD-2 tubes - these media can be used with same type of low
         volume pumps described above to collect additional samples for further definition of the
         constituents of the plume for SVOCs.

        Organic compounds will be present in all three phase distributions (particle bound, SVOC,
 and VOC) and each phase will have to be sampled and then a determination will have made as to
 importance of each.

        The particle bound is phase can be collected  for extractive organic analysis from both
 quartz, teflon impregnated  glass fiber filters, or teflon.

        The SVOC phase can be collected on PUF and within the canister.  The VOC phase can be
 collected with canisters and charcoal tubes. Employment of Tenex and S/.D-2 sampling tubes in
 conjunction with PUF,  charcoal tubes, and canisters in an overlapping sampling matrix, can be used
 to confirm of the presence or absence of compounds which could be missed by a less complex
 sampling matrix.

        Proposed locations for the initial six expanded sites:

        The Team recommends .that the six locations follow the general axis of plume drift from
 Kuwait City south into Saudi Arabia, it is further suggested that the operation of these sites be
 divided amongst the key network managers within the two countries:  Kuwait MEPA, and Saudi
 ARAMCO. This division of responsibility supports the philosophy to jointly develop both the
 sampling and analytical capability within all three entities.
        Kuwaiti Locations

*'   .     The Team recommends that two (2) of the sites be located at the existing operational
continuous monitoring stations located in Kuwait City.  A third site should be established in Al-
Ahmadi at the Kuwaiti Oil Company Hospital.  This location is situated within 300 - 400 meters of
several burning wells and is adjacent to the closest residential area associated with any of the oil
fields.                                ,

        Saudi Arabian Locations

        The Team recommends that a site be established at King Fahd University of Petroleum and
 Miners (K.F.U.P.M.) in Dhahran, one site to be collocated with Saudi ARAMCO site in Tanajib and
 the last site should be collocated with the MEPA site in Khafji which currently is only collecting
 meteorology data.

        4.5    Phase 5. Develop Regional Air Monitoring Network to Track the Impact of the
               Kuwaiti Oil Fires on the Gulf Reoioq

        This phase of the proposal will be developed in a series of stages.  The network is directed
 at long term fixed site monitoring, which will collect a battery of both air quality and meteorological
 data that would satisfy the objectives identified in this report. Particular attention will be given to
 providing an early warning system advising the effected public on what cautionary steps should be
 taken to minimize the impact, of air pollution on health.  It is proposed that the early warning
 system will be modeled after the air quality index used in the United States. The second focus will
 be directed at tracking the air pollution problem over time to be used in determining both long term
 health and ecological impacts due to the Kuwait oil ftres. The data base will provide information to
 develop key policy decisions which could minimize the possible health and ecological impacts.

        Recommended Measurement Parameters at Each Site

        The basic plan is for each site within the regional monitoring network to consist of the
 same types of measurement devices. Deviations for that general consistency will be addressed on
 an individual basis.  Two general categories of information are proposed for every location.  Data  <
 measurement are grouped into either meteorological or air quality categories.

        a.  Meteorological  variables

        The nominal set of meteorological measurements to be collected at each site are the

              Wind speed and direction (at 10m)
              Moisture content of the air
              Dew point, relative humidity' or wet bulb
              Solar radiation
              Precipitation (saved for chemical analysis, pH, etc.)
        b. Air quality variables

       The nominal set of air quality measurements to be collected  at each site are the following.
Two general categories of collection needs exist. The first  set of needs relate to the onset of
episodic levels of airborne concentrations.  In  those situation a need may be developing for an early
warning advisory for downwind population centers. The second set of needs ralata to the need to
monitor for longer term, more subtle risks.
Episodic/EARLY WARNING measurements

Longer term monitoring

       Acid aerosoia
       BaP, other PAHs

               NOx, NO
Trace elements
Fine particles
        c. Health Monitoring Survey

        Air monitoring data collected through the proposed air monitoring network will provide basis
 for interpreting the results of health surveys of the populations and ecosystems potentially effected
 by the effluents from the oil fires in Kuwait. The kinds of health data that could be collected

               Health questionnaires
               Blood samples
               Hair samples
               In-vivo animal studies
               Forced expiratory volumes                                  '
               Other morbidity parameters

        Distribution of Proposed and Existing Air Monitoring Sites
                                                                         .   ' /.
       Table 2 shows the location of the existing and proposed air monitoring sites, while Table 3
 lists tine locations of the proposed air monitoring sites. In order to complete the network in an
 orderly manner, it is proposed that the network be developed in several stages. The first order of
 business would be to upgrade the existing monitoring locations so that there exist a full
 complement of air and meteorological monitoring equipment, as well as add new critical air
 monitoring stages (Stage 1).  The second stage would be to establish those sites that would satisfy
 the minimum requirements or tracking the plume caused by the oil fire and to provide an early
 warning system for Saudi Arabia, Kuwait, and Bahrain (Stage 2). The third and final stage would
 be to complete the final network following a review of the quality and quantity of Indings to date
 (Stage).                                                              .   .
       The initial measurements made by the Team suggest that there is not an imminent threat
from SO2 and H2S to the urban populations, while short term measurements of particulate are
frequently high.  Historically, this region has high particulate levels due to wind blown dust.  The
particulate  measurements that were collected by the Team reflect total particulate, as opposed to
respirable particulates, that is PM,0. There has not been a principal focus in the Region on total
particulate, PM1010, and organics up to now. Therefore, the Team developed the five phased
monitoring plan with sn emphasis on better understanding particulates and the aerosol organics
associated  with the oil fires in Kuwait. Particulates and organics could be a source of concern for
both health and ecological effects.

       The air monitoring proposals presented in this report represent the Team's collective
judgement on what needs to be done.  Those judgements are based upon sn on-sfte evaluation of
the situation in Kuwait discussions with officials from the Saudi Arabian MEPA, Kuwait Saudi
ARAMCO, and the King Fahd University of Petroleum and Minerals. Needless to say more work is
needed regarding data management statistical design, date analysis snd quality assurance.
Because of the complexity and immediacy of this problem, an extended  time commitment will be
needed on the part of all Gulf nation agencies to achieve the objectives outlined in this report.

       Cost estimates for the various types of air and meteorological monitors are attached to this
plan, along with documentation on the U.S. Environmental Protection Agency's Pollutant Standards
Index (PSD.     '.'••"

             TABLE 2. Location of existing air and meteorology monitoring stations
Saudi Arabia



Saudi Arabia

Tha Royal Comm.
for Jubyayt and

Tanagib (Mobil* Site)
. Riyadh
Shad gum
Safaniyah Oil Raid
(mat only)
Marjan Oil Raid (mat only!
Abutafaha Raid (mat only)
Quatar of S ttatiora
in Jubyayl
3 Mobila Units, unknown loeationa

TABLE 3.  Location of proposed air and meteorology monitoringstations
Saudi Arabia
. '
• '



1 .


Unit*! Arab Emirates




A Sarrer
Hafar al Batin
28 dag 6 min latitude,
47 deg 51 min longitude
28 deg 30 min latitude,
48 deg 1 min longitude
28 dag 55 min latitude.
47 deg 32 min longitude
29 d«fl 7 min latitude,
46 deg 39 min longitude
Mine Saud
U.S. Embaeey
Al Ahmad)
. Intematk tal
29 deg 23 min latitude,
49 deg 55 min longitude
29 deg SO min latitude.
47 deg 1 S min longitude
30 dag 4 min latitude,
47 dag 42 min longitude
29 deg 33 min latitude. ,
47 deg 50 min longitude
Mine Manama
Bahrain University :
Abu Dhabi
x *
2 -
, a
* 3