Skip directly to search Skip directly to A to Z list Skip directly to site content




The Agency for Toxic Substances and Disease Registry (ATSDR) prepared this consultation inresponse to requests from Fort Wainwright, a U.S. Army post in Fairbanks, Alaska, and theShannon Park Baptist Church, located next to the post. ATSDR was requested to evaluate thepotential health effects from using the church well to water the lawn. The Alaska Department ofEnvironmental Conservation and Fort Wainwright took water samples from the well in the pastand found that the samples contained the following volatile organic compounds:1,1,1-trichloroethane, 1,2-dichloroethane, 1,2,4-trimethylbenzene, isopropylbenzene,n-propylbenzene, sec-butylbenzene, toluene, and xylene. Because the nature and extent ofgroundwater contamination at the nearby areas of Fort Wainwright are under investigation andthe church would like to use the well water for irrigation, both the church and the post authoritiesrequested this consultation. Fort Wainwright is currently suppling drinking water to the churchbecause samples of the well water have shown levels of 1,2-dichloroethane above the maximumcontaminant levels of the U.S. Environmental Protection Agency (EPA).


Shannon Park Baptist Church is located near the corner of Steese Highway and Lazelle Road. The church is adjacent to the northwest corner of Fort Wainwright. This corner of the postincludes three Installation Restoration Program Sites: the Chemical Warfare Disposal Area(Operable Unit 1); the Tank Farm Facility (Operable Unit 3); and the Tank Farm Above GroundStorage Tanks (Operable Unit 5). The closest site, the Tank Farm Facility, is locatedapproximately 1,500 feet from the Shannon Park Baptist Church well. Fort Wainwright is in theprocess of investigating whether on-base contamination has migrated off-base and contaminatedthe Shannon Park Baptist Church well. Other potential private sources of groundwatercontamination are also located nearby.

The Alaska Department of Environmental Conservation began sampling the Shannon ParkBaptist Church well water in 1991. A summary of the sampling results from 1991 through June1997 and 1998 is presented in Table 1. Fort Wainwright took over sampling and analysis of thewater in the latter half of 1997. The highest concentrations detected were 9.75 µg/L forisopropyl benzene and 6.02 µg/L for 1,2-dichloroethane.


ATSDR concluded that using the Shannon Park Baptist Church well water for irrigation will notbe an apparent public health hazard. The scientific calculations on which this conclusion isbased are explained in this section. These calculations evaluated the three most likely wayspeople could be exposed to the compounds in the water: (1) inhalation of the volatile organiccompounds after they volatilized from the water; (2) ingestion of the soil or grass after thecompounds were deposited onto the lawn, and (3) ingestion of the water.

To protect human health to the greatest degree possible, our evaluations are based onassumptions that overestimate human exposure. We assumed that the chemicals of concern werepresent at the highest concentration detected in 1991 to 1998 water samples (see Table 1). Wealso assumed that the lawn was watered at least twice a week with an amount of 2 inches ofwater applied each time meaning that 50,864 gallons of water were used each time. We assumedthis water was spread evenly over the lawn and that it took one hour to apply this water.

For each exposure route analyzed, we evaluated the potential health effects to an individualexposed to the most contamination. This hypothetical location would be the center of the lawnarea. In reality, very few people, if any, would be present at this location. Most of the peoplepresent would be in the church, adjacent to the lawn. Depending on the hours that the lawn iswatered, few people may be present.

Volatilization and Inhalation

Dose Calculation

The predominant fate of the volatile organic compounds detected in the groundwater from thechurch's well is to move to the air. This process is called volatilization. To accurately determinethe rate and amount of compound volatilizing is difficult because many factors are not known orvary within ranges that are not precisely known. Some of the factors involved in determining therate of volatilization include sprinkler type, sprinkler operation, ambient conditions, and waterpressure. U.S. EPA has completed experiments with pivot irrigation experiments systems toclean up groundwater contaminated with 1,1,2-trichloroethene and 1,1,1-trichloroethane. Theresults of these experiments showed that 90% to 96% moved from water to air (U.S. EPA 1996,ATSDR 1998b). These experiments were probably conducted under conditions that wouldmaximize volatilization. Actual volatilization rates from the sprinkler system at the ShannonPark Baptist Church would probably be less. Hence, our assumption of 100% volatilization isanother example of our overly protective approach.

ATSDR used an air model to calculate air concentrations of the compounds during lawnwatering. We treated the lawn watering as an air emission source. The area modeled includedthe lawn area (0.94 acres) and adjacent land shown in Figure 1. The model we used was the U.S.EPA Industrial Source Complex Short Term Version 3. For this model, we treated the lawnirrigation as an area source (as opposed to single or multiple point sources) releasing thecompounds 0.5 meters above the ground. This means that the volatile organic compoundsmoved from the water to the air across the entire lawn. An area source creates a worst casecondition compared to point sources.

The air model was run to calculate the highest concentration at each of 575 point locations 0.5meters above the lawn (see Figure 1). The model was run using five years of meteorologicaldata from the Fairbanks Airport, the nearest location for which EPA had complete meteorologicdatasets. Additional information on calculations for the air modeling are presented in theAppendix.

The air model was run for a standard groundwater concentration of 0.071 µg/L (emission rate of 1E-9 grams per second per square meter) for a generic volatile organic compound. Because volatile organic compounds in air are dispersed with no deposition or degradation, the model results are independent of the type of chemical and only dependant only on the compound's concentration. Hence, we ran the model for the standard concentration in the water and adjusted the output based on each compound's actual concentration in the water. The results of the air model for the generic volatile organic compound are shown in Figure 1. The air concentrations were then linearly adjusted based on the actual maximum concentration of each chemical compound detected in the groundwater. Table 2 shows the maximum air concentration for each compound is listed. Concentrations ranged from 1.4 micrograms per cubic meter (µg/m3) for 1,2-dichloroethane to 0.06 µg/m3 for toluene.

Public Health Implications and Conclusion

The compounds listed in Table 1 can have noncarcinogenic health effects at elevatedconcentrations. 1,2 Dichloroethane and 1,1,2-trichloroethane are also considered carcinogens. Elevated levels of carcinogens can increase the lifetime risk of getting cancer. To determine ifthe modeled air concentrations could potentially be a health problem, we compared the modeledconcentrations to comparison values (sometimes referred to as screening values).

ATSDR calculates comparison values as an overly protective concentration in air, water, or soil. Ambient concentrations of a chemical that are less than a comparison value are unlikely to pose ahealth threat although the total dose from multiple media or multiple chemical exposure stillneed to be considered. Ambient concentrations above a comparison value do not necessarilyrepresent a health threat, but indicate that further analysis is necessary. The carcinogeniccomparison values are based on the added excess risk of one additional cancer case in apopulation of one million equally exposed people. This one cancer case would be in addition tothose cancer cases that would normally occur in an unexposed population of one million people. This normal rate is approximately 1 in 4 or roughly 250,000 people in the population of onemillion. Hence our cancer risk values are very conservative.

The noncarcinogenic and carcinogenic health screening values for the compounds detected in thegroundwater at the Shannon Park Baptist Church are listed in Table 1. For each chemical,whether carcinogenic or noncarcinogenic, the modeled air concentrations were below thecomparison values indicating that there is no apparent public health hazard. When consideringthe multiple chemicals, we also conclude that no apparent public health hazards are likelybecause the ambient concentrations are 20 to 30 times below the comparison values for all thecompounds.

Deposition and Contamination of Soil

Dose Calculations

In this section, we evaluated Shannon Park Baptist Church's concern about contaminating thelawn with the groundwater. ATSDR defined the lawn as contaminated if eating soil or grassfrom the lawn would present a health hazard to humans.

In considering inhalation, we assumed 100% volatilization of compounds from water to air. Inconsidering ingestion, we assume zero volatilization with complete deposition on the soil. Although zero volatilization is unlikely, we used this scenario to be overly protective of publichealth. For these calculations, we used the Equilibrium Criterion Model (EQC, see Appendix formore information) which calculates concentrations in air, water, soil, and sediment based on auser input into air, water or soil. This input is the amount of the compound applied to the soilthrough watering the lawn. For this model, we assumed direct application of the compounds tothe soil. The amount deposited is a function of the maximum concentration in the groundwater(See Table 2). The amount varies from 1.87 kilograms per hour (kg/hr) for isopropylbenzene to0.05 kg/hr for toluene. Sample calculations for deriving these values are shown in the Appendix. Other model parameters are also shown in the Appendix. We used values provided with themodel except to prohibit transfer of the contaminants to sediments and to water. This creates amore health protective scenario. Another protective assumption is that the model assumes thatthe compounds are applied to the soils 24 hours per day, 365 days per year. Because the lawn isonly watered a few hours a week for about 6 months a year, the modeled results are veryprotective of human health.

Based on the model, the concentration of the compounds in the soil vary from 0.000012milligram per kilogram (mg/kg, also referred to as parts per million) to 0.0000004 mg/kg fortoluene.

Public Health Implications and Conclusion

The calculated soil concentrations are 100,000 times lower than the carcinogenic ornon-carcinogenic screening values (shown in Table 1). This indicates if the water contained thehighest concentration of the compounds ever detected, the soils would pose no apparent publichealth hazard. Because the concentrations of the compounds in the soils would be so low, thegrass would also pose no apparent public health hazard.

Incidental Ingestion of Water

The third potential exposure scenario is accidental ingestion of the water. Table 1 shows thelevels of contaminants detected in the well water and compares these values with screeningvalues. The screening values used were the EPA regulatory limits for drinking water (maximumcontaminant levels or MCLs). If EPA did not have a MCL for a specific compound, we usedATSDR or EPA screening values (explained in Table 2.) Of the 8 chemicals detected,1,2-dichloroethane is the only chemical that exceeded the MCL. In fourteen samples collectedbetween November 1991 and June 1997 and 1998, the MCL was exceeded 6 times. The highestconcentration was 6 µg/L.

Public Health Implications and Conclusion

ATSDR evaluated the concentrations of the compounds detected in the well water. Theconcentrations of 7 of the 8 compounds detected (1,1,1-trichloroethane, 1,2,4-trimethylbenzene,isopropylbenzene, n-propylbenzene, sec-butylbenzene, toluene, and xylene) were below theMCLs or screening values. Only 1,2-dichloroethane exceeded a MCL.

Because this MCL was exceeded, we conducted a risk assessment for accidental ingestion ofwater containing 1,2-dichloroethane. MCLs assume daily exposure over a lifetime. Since thelawn watering is assumed to occur twice a week for 6 months per year, actual exposure will beless. The amount of water consumed will also be less than the amount assumed by the MCL. Therefore, we calculated a screening value based on this reduced frequency of ingestion. If weassume a child consumes 0.05 liters of the well water per each watering event (based oningestion of water from recreational swimming, EPA 1989), the comparison level concentrationbecomes 178 µg/L (see the Appendix for calculations). The highest concentration of1,2-dichloroethane measured in the well water is 30 times below this value. Hence, no apparentpublic health hazards are expected from incidental ingestion of the groundwater during lawnwater from 1,2-dichloroethane or the other 7 volatile organic compounds detected.


The ATSDR Child Health Initiative recognizes the unique vulnerabilities of infants and childrenexposed to environmental contamination and hazards. As part of this health consultation,ATSDR considered the greater sensitivity of children and concluded that watering the lawn withthe groundwater at the maximum concentrations detected poses no apparent public health hazardto children or adults.


ATSDR has reviewed the future use of groundwater at the Shannon Park Baptist Church for lawnwatering. Using the maximum concentrations detected in the groundwater, ATSDR evaluatedthe potential effects of inhalation of the volatile organics, ingestion of soil, and incidentalingestion of the groundwater. In each case, there are no apparent public health hazardsassociated with individual compounds or the compounds together. Because the health risks areso low and the exposure scenarios are so protective, there is no apparent public health hazardfor an individual exposed to the air, soil, and water at the same time.


Actions Taken and Proposed

  • Fort Wainwright is sampling the Shannon Park Baptist Church well quarterly for volatile organic compounds.

  • Fort Wainwright is supplying the church with alternative drinking water daily.

  • Fort Wainwright is investigating sources of contamination on-base and is required by State and Federal law to take appropriate remedial actions as necessary for these sources.

ATSDR has no recommendations. Using the groundwater to water the lawn does not pose an apparent public health hazard and Fort Wainwright is investigating the source of contamination.


ATSDR 1998a. ATSDR Hazardous Database System (HazDat) Record of Activity (AROA),record of communication regarding Shannon Park Baptist Church, July 1, 1998.

ATSDR 1998b. ATSDR Hazardous Database System (HazDat) Record of Activity (AROA),record of communication regarding spray irrigation of contaminated groundwater, July 7, 1998.

ATSDR 1999. HAZDAT Substance Comparison Value System, April 12, 1999.

EPA 1989. Risk Assessment Guidance for Superfund, Volume 1. Human Health EvaluationManual (Part A), EPA/540/1-89/002. December 1989.

EPA 1996. Fact Sheet - Demonstration of VOC Treatment and Disposal Via Spray Irrigation,Hastings, Nebraska. June 25, 1996.

EPA 1998a. U.S. EPA, Region 9 Preliminary Remediation Goals (PRGs), 1998.

EPA 1998b. U.S. EPA, Region 3 Risk Based Concentrations, April 15, 1998.

EPA 1998c. Federal Register, Volume 63, Number 216, Pages 60332-60343. Notice ofAvailability of Draft RCRA Waste Minimization of Persistent, Bioaccumulative, and ToxicChemicals. November 9, 1998.

Fort Wainwright 1999. Fax of data. June 7, 1999.

Hart Crowser, Inc., 1998. Birch Hill Tank Farm Groundwater Investigation Operable Unit 3,Fort Wainwright, Alaska, Prepared for Department of Army, U.S. Army Engineer, Alaska Unit,July 22, 1988.

Mackay, 1996. Evaluating the Environmental Fate of A Variety of Types of Chemicals Usingthe EQC Model, Environmental Toxicology and Chemistry, Vol 15, No. 9. pp. 1627-1637.


Brian M. Kaplan
Environmental Health Scientist
Federal Facilities Assessment Branch
Division of Health Assessment and Consultation
(404) 639-6001


Carole Hossom
Environmental Health Scientist
Federal Facilities Assessment Branch
Division of Health Assessment and Consultation

Gary Campbell
Environmental Health Scientist
Federal Facilities Assessment Branch
Division of Health Assessment and Consultation

Table 1.

DATE Analyte (µg/L)
Toluene Xylene 1,2-Dichloro-ethane 1,2,4-Trimethly-benzene 1,1,1-Trichloro-ethane 1,1,2-Trichloro-ethane Isopropyl-benzene (cumene) o-propyl-benzene sec-butyl-benzene
11/13/91 <MDL <MDL 2.7 <MDL <MDL <MDL 4.4 <MDL <MDL
12/18/91 ND(1.0) ND(1.0) 2.9 ND(1.0) ND(1.0) ND(0.20) 6 ND(1.0) ND(1.0)
01/26/94 ND(0.20) ND(0.20) 5.04 0.25 ND(0.20) ND(0.20) 7.98 0.46 0.67
02/10/94 ND(0.20) ND(0.20) 4.4 0.4 ND(0.20) ND(0.20) 7.9 0.4 0.7
06/06/94 ND(0.20) ND(0.20) 5.35 0.3 ND(0.20) ND(0.20) 7.27 ND(0.20) 0.63
07/21/94 ND(0.20) ND(0.20) 5.38 0.32 ND(0.20) ND(0.20) 8.13 ND(0.20) 0.65
09/23/94 ND(0.20) ND(0.20) 4.13 ND(0.20) ND(0.20) 0.43 6.65 ND(0.20) 0.54
11/22/94 ND(0.20) ND(0.20) 3.99 ND(0.20) ND(0.20) 0.45 4.43 ND(0.20) 0.35
1/24/95 ND(0.20) ND(0.20) 4.53 ND(0.20) ND(0.20) 0.6 ND(0.20) ND(0.20) ND(0.20)
04/04/95 ND(0.20) ND(0.20) 5.86 0.24 ND(0.20) ND(0.20) 9.75 ND(0.20) 0.65
06/26/97 0.26 0.47 5.6 0.20 ND(0.20) ND(0.20) 4.0 ND(0.20) 0.31
06/12/982 ND(1) Not reported ND(1) ND(1) Not reported Not reported Not reported Not reported Not reported
8/20/982 ND(1) Not reported 6.02 ND(1) Not reported Not reported Not reported Not reported Not reported
12/18/982 ND(1) Not reported 4.26 ND(1) Not reported Not reported Not reported Not reported Not reported

1. Source: Hart Crowser, 1998, Samples were analyzed for VOCs using EPA Method 502.2/524.
2. Source: Fort Wainwright, 1999, Samples were analyzed for VOCs using EPA Method 8260.
3. The June 1997 samples were collected by Hart Crowser. All other samples were collected by the Alaska Department of Environmental Protection.
4. ND(0.30) means the compound was not detected at the detection limit of 0.30 ug/L
5. MDL means below minimum detection limit where the MDL was not reported.

Table 2.



Soil Ingestion

Incidental Water Ingestion

Max Modeled Air Concentration (µg/m3)

Lowest Screening Value-Non Cancer (µg/m3)

Lowest Screening Value-Cancer (µg/m3)

Deposition Rates (kg/hr)

Modeled Concentration (mg/kg)

Lowest Screening Value-Non Cancer (mg/kg)

Lowest Screening Value-Cancer (mg/kg)

Maximum Detected Concentration in Groundwater (µg/L)

Comparison Value

Source of Value--Screening Value or MCL

























































































Screening values for cancer for inhalation are based ATSDR's cancer risk evaluation guides (CREGS) modified for an exposure scenario of 1 hour of exposure, 2 times a week, and 6 months a year for 70 years. Default CREG values are based on 24 hours per day, 365 days per year and represent a concentration at which a person's additional risk of cancer after exposure for 24 hours a day, 365 days a year, and for 70 years is one in one million. All other screening values are derived from default values from one of the following three sources: (1) ATSDR's comparison values including environmental media evaluation guides (EMEGs) and cancer risk evaluation guides (CREGs, ATSDR, 1999) , (2) U.S. EPA Region 9 preliminary remediation goals (EPA, 1998a), and (3) U.S. EPA Region 3 risk based concentrations (EPA, 1998b). When these sources disagreed, the lowest value of the three was selected. For ingestion, U.S. EPA values for drinking water maximum contaminant levels (MCLs) were used. When EPA did not have a MCL for a particular compound, a screening value was used.

Modeled Air Concentrations, Base Emission Rate
Figure 1. Modeled Air Concentrations, Base Emission Rate




  1. Lawn area is 240 feet by 170 feet or 40,800 ft2. This also equals 0.94 acres or 3,790meters2.

  2. Each lawn water episode applies 2 inches of water over the 40,800 ft2


  1. Amount of water applied during each watering event is:
  2. (2 inch) x (40,800 ft2) (1 ft/12 inches) = 6,800 ft3

    With 7.48 gallons/ft3, this equals to 50,864 gals

    With 3.78 L/gal, this equals 192,266 L per watering event.

  3. At 6 µg/L, amount of DCA available for fate, transport, and exposure is:
  4. (6 µg/L) x (192,266 L) = 1,153,595 ug. Similar calculations for the other organics


  1. The lawn watering is similar to an area source with a release height of 0.5 meters. Thisheight is approximately the height of the breathing zone of a child.

  2. The 2 inches of water are applied in the span of 1 hour.

  3. We assumed all the contaminants in the water volatilized.

  4. 1,153,595 ug is emitted during the span of 1 hour. Converting this value to g/s/m2, we get
    (1,153,595 ug) * (1 g/1,000,000 ug) / (3,790m2) / (3600 seconds) =
    = 8.45 x 10-8 g/s/m2

  5. The receptor grid is shown in Figure 1. The grid between nodes in the x direction is 11meters and in the y direction it is 10 meters.

  6. We used the Industrial Source Complex Short Term Version 3 model developed by EPAand modified by Trinity Consultants (Version 1.07) to allow a greater number of grid pointsand sources to be modeled.

  7. Data for meteorology was obtained from the U.S. EPA Support Center for RegulatoryModels at The meteorology station used was Fairbanks,Alaska for both the lower and upper atmospheric data. We used years 1986, 1987, 1989,1990, and 1991. This was the latest data available for a 5 year period. The meteorologydata for 1988 was not available. A model run was completed for each year. The 1-hourmaximum results at each grid node was calculated for each year. The 1-hour maximumsfrom each year were then added and divided by 5 to create and average.


  1. The Equilibrium Criterion Model (EQC) (Version 1.01) was used . This model wasdeveloped by Mackay et. al. (Mackay, 1996) and is available from the Trent UniversityEnvironmental Modeling Centre in Peterborough, Ontario, Canada( This model is wellaccepted by academia and industry (U.S. EPA, 1998c)

  2. From before, 6800 ft3 of water per watering event or 192266 Liters With 1,2-dichloroethane (DCA) at 6 µg/L, the total amount of DCA applied to the lawn is 1153596 ug or 1.154 kg.


  1. Amount consumed during each watering event is 0.05 L for a child or an adult (EPA, 1989)

  2. Exposure is for 30 years, 6 years as a child and 24 years as an adult.

  3. Tap water ingestion calculated using equations 2 and 4 from EPA 1998a. Equation 4 wasmodified to not include volatilization.

  4. Exposure duration was 2 times per week for 6 months.

EQC Model Parameters
EQC Model Parameters

Sample Emission Rate Input Screen for EQC Model
Sample Emission Rate Input Screen for EQC Model

Example of Chemical Properties Used in EQC Model
Example of Chemical Properties Used in EQC Model

Table of Contents The U.S. Government's Official Web PortalDepartment of Health and Human Services
Agency for Toxic Substances and Disease Registry, 4770 Buford Hwy NE, Atlanta, GA 30341
Contact CDC: 800-232-4636 / TTY: 888-232-6348

A-Z Index

  1. A
  2. B
  3. C
  4. D
  5. E
  6. F
  7. G
  8. H
  9. I
  10. J
  11. K
  12. L
  13. M
  14. N
  15. O
  16. P
  17. Q
  18. R
  19. S
  20. T
  21. U
  22. V
  23. W
  24. X
  25. Y
  26. Z
  27. #