PUBLIC HEALTH ASSESSMENT
AIR FORCE PLANT PJKS
WATERTON, JEFFERSON COUNTY, COLORADO
Comparison values represent media-specific contaminant concentrations that are used to select contaminants for further evaluation to determine the possibility of adverse public health effects. The conclusion that a contaminant exceeds the comparison value does not mean that it will cause adverse health effects.
Cancer Risk Evaluation Guides (CREGs)
CREGs are estimated contaminant concentrations that would be expected to cause no more than one excess cancer in a million (10-6) persons exposed over their lifetime. ATSDR's CREGs are calculated from EPA's cancer potency factors.
Environmental Media Evaluation Guides (EMEGs)
EMEGs are based on ATSDR minimal risk levels (MRLs). An EMEG is an estimate of daily human exposure to a chemical (in mg/kg/day) that is likely to be without noncarcinogenic health effects over a specified duration of exposure.
Maximum Contaminant Level (MCL)
The MCL is the drinking water standard established by EPA. It is the maximum permissible level of a contaminant in water that is delivered to the free-flowing outlet. MCLs are considered protective of public health over a lifetime (70 years) for people consuming 2 liters of water per day.
Reference Media Evaluation Guides (RMEGs)
ATSDR derives RMEGs from EPA's oral reference doses. The RMEG represents the concentration in water or soil at which daily human exposure is unlikely to result in adverse noncarcinogenic effects.
APPENDIX C: Estimated Exposure and Health Effects
Estimates of Human Exposure Doses and Determination of Health Effects
Deriving Exposure Doses
ATSDR estimated the human exposure doses/concentrations for past ingestion of Denver municipal water originating from the Kassler Water Treatment Plant. Deriving exposure doses requires evaluating the concentrations of the contaminants to which people may have been exposed and the frequency and duration of those exposures. Health effects are also related to individual characteristics--such as age, gender, and nutritional status--that influence how a chemical might be absorbed, metabolized, and eliminated by the body. Together, these factors help influence the individual's physiological response to chemical contaminant exposure and potential noncancer (noncarcinogenic) or cancer (carcinogenic) outcomes. In the absence of site-specific exposure information, ATSDR applied several conservative exposure assumptions to define site-specific exposures as accurately as possible for residents near the AFP PJKS site.
Evaluating Potential Health Hazards
The estimated exposure doses are used to evaluate potential noncancer and cancer effects associated with chemicals of concern. When evaluating noncancer effects, ATSDR uses standard health guidelines, including ATSDR's Minimal Risk Levels (MRLs) and EPA's Reference Doses (RfDs), to determine whether adverse effects will occur. The chronic MRLs and RfDs are estimates of daily human exposures to a substance that are unlikely to result in adverse noncancer effects over a specified duration. ATSDR compared estimated exposure doses associated with AFP PJKS exposure scenarios to conservative health guidelines such as MRLs or RfDs for each contaminant. If the exposure dose is greater than the MRL or RfD, then a possibility exists that noncancer effects could occur.
To evaluate cancer effects, ATSDR uses Cancer Potency Factors (CPFs) that define the relationship between oral exposure doses and the increased likelihood of developing cancer over a lifetime. The CPFs are developed using data from animal or human studies and often require extrapolation from high exposure doses administered in animal studies to the lower exposure levels typical of human exposure to environmental contaminants. CPFs represent the upper-bound estimate of the probability of developing cancer at a defined level of exposure; therefore, they tend to be very conservative (i.e., overestimate the actual risk) in order to account for a number of uncertainties in the data used in the extrapolation.
ATSDR estimated the potential for cancer to occur using the following equation. (The estimated exposure doses and CPF values for the contaminants of concern are incorporated into the equation):
Lifetime Cancer Risk = Estimated exposure dose (mg/kg/day) x CPF (mg/kg/day)-1
Although no risk of cancer is considered acceptable, it is impossible to achieve a zero cancer risk. Consequently, ATSDR often uses a range of 10-4 to 10-6 estimated lifetime cancer risk (or 1 new case in 10,000 to 1,000,000 exposed persons), based on conservative assumptions about exposure, to determine whether a concern regarding cancer effects is valid. This range is consistent with values adopted by EPA for evaluating the need for cleanup at hazardous waste sites.
In addition to estimating the likelihood of noncancer and cancer effects, ATSDR reviewed the literature to evaluate possible health effects associated with exposure at the doses/concentrations estimated for the pathway described above.
Evaluation of Potential Exposure Pathways
Estimated Exposure From Ingestion of Municipal Water From the Kassler Water Treatment Plant
The following equation and assumptions were used to estimate exposure to contaminants in drinking water:
where:
| Conc.: | Maximum concentration in the Kassler infiltration gallery water ( parts per million [ppm]) |
| IR: | Ingestion rate: adult=2 liters per day; child=1 liter per day |
| EF: | Exposure frequency, or number of exposure events per year of exposure: 7 days/week x 52 weeks/year |
| ED: | Exposure duration, or the duration over which exposure occurs: adult=27 years; child=10 years |
| BW: | Body weight: adult=70 kg; child=16 kg |
| AT: | Averaging time, or the period over which cumulative exposures are averaged (10 years or 27 years x 365 days/year for noncancer effects; 70 years x 365 days/year for cancer) |
Assumptions for Estimating Exposure Doses
-- ATSDR assumed that an adult drinks 2 liters and a child drinks 1 liter of water a day and that all drinking water came from the affected source.
-- The exposure frequency (EF), or number of exposure events per year, was assumed to be 365 days per year, based on a 7-day-a-week exposure over 52 weeks per year.
-- The duration of exposure (ED) is assumed to have been a 27-year period for an adult. This value is the length of time that AFP PJKS had been in operation before the infiltration galleries at the Kassler Water Treatment Plant were closed. This 27-year figure most likely overestimates the actual duration of exposure, which was likely a small fraction of the 27 year maximum potential exposure period.
-- The averaging time (AT) was assumed to be 365 days/year for either 10 or 27 years for noncancer effects, and 70 years for 365 days/years for cancer effects.
Likelihood of Health Effects From Ingestion of Municipal Water From the Kassler Water Treatment Plant
The highest TCE, 1,1,1-TCA, and hexavalent chromium levels detected in the infiltration galleries of the Kassler Water Treatment Plant before its closure in 1985 were below the MCLs. Nonetheless, community members are concerned about harmful effects from drinking water from the Kassler Water Treatment Plant. In response to the concern, ATSDR evaluated exposure to drinking water containing the maximum detected levels of TCE (3.1 ppb), 1,1,1-TCA (0.54 ppb), and hexavalent chromium (30 ppb) to determine whether a public health hazard existed.
Noncancer: Using the equation to calculate exposure dose, ATSDR estimated exposure doses from ingestion of municipal water with TCE (adult: 0.00009 mg/kg/day; child: 0.0002 mg/kg/day), 1,1,1-TCA (adult: 0.00002 mg/kg/day; child: 0.00003 mg/kg/day), and hexavalent chromium (adult: 0.0009 mg/kg/day; child: 0.002 mg/kg/day (see Table C-1). The resulting exposure doses for TCE and hexavalent chromium are lower than their respective health guidelines (MRLs) of 0.002 mg/kg/day and 0.003 mg/kg/day.(3) No chronic oral MRL is currently available for 1,1,1-TCA. For comparison, ATSDR reviewed the available toxicological literature to determine possible adverse effects associated with exposure at doses estimated for this pathway. On the basis of this review, the exposure doses estimated for 1,1,1-TCA by ATSDR are several orders of magnitude lower than the lowest doses reported in the toxicological literature capable of producing noncancer effects in experimental animals administered oral doses of 1,1,1-TCA (ATSDR, 1995a). Therefore, drinking water containing the highest detected levels of TCE 1,1,1-TCA, and hexavalent chromium reported in the Kassler Water Treatment Plant before its closure is not likely to result in adverse noncancer effect.
To account for dermal and inhalation exposures that might occur during domestic use of the water, ATSDR assumed that these exposures each equal the exposure from drinking the water. In applying this assumption, ATSDR multiplied the estimated oral dose by three (to account for all three routes: ingestion, inhalation, and dermal exposure) (see Table C-1). The resulting overall estimated doses are less than the MRLs for an adult and child exposed to TCE and for an adult exposed to hexavalent chromium. The estimated exposure dose for a child exposed to hexavalent chromium is slightly higher than the MRL. The slight difference does not indicate a public health hazard, however. It should be noted that the very conservative assumptions (e.g., exposure to the maximum concentration) used in the evaluation allowed ATSDR to estimate the highest possible exposure dose, even though ATSDR does not expect that children ever drank the most contaminated water for the 10-year potential exposure duration evaluated. Therefore, use of water for bathing, washing, and/or drinking would not have resulted in adverse health effects.
Cancer: TCE has been shown to cause cancer in laboratory animals given large doses. The link between TCE and cancer in humans drinking water is controversial, however. Available studies for TCE are inconclusive and the data are inadequate to establish a link. EPA is currently reviewing the scientific literature pertaining to the carcinogenicity of TCE to determine its cancer classification. As a conservative measure, ATSDR used the current CPF for TCE to estimate the excess lifetime cancer cases resulting from exposure to water containing the maximum concentration of TCE (3.1 ppb). ATSDR estimated 1 new cancer case per 1,000,000 persons could be expected if people are exposed over an extended period of time (see Table C-2). On the basis of these results, ATSDR concludes that ingestion of TCE at the levels detected in the infiltration gallery water, or even dermal contact or inhalation of its vapors, is not expected to result in an increased likelihood of developing cancer.
TABLE C-1. Estimated Exposure Doses--Noncancer Effects
from Ingestion of Water from the Kassler Water Treatment Plant
| Contaminant |
Maximum Contaminant Concentration (ppm) |
Estimated Exposure Dose (mg/kg/day)a | Total Estimated Exposure Dose (mg/kg/day)b | Health Guideline Oral (mg/kg/day) | Basis for Health Guideline | ||
| Adult | Child | Adult | Child | ||||
| Trichloroethylene
|
0.0031 | 0.00009 | 0.0002 | 0.0003 | 0.0006 | 0.002 | MRL |
| 1,1,1-Trichloroethane | 0.00054 | 0.00002 | 0.00003 | 0.00006 | 0.00009 | not available | |
| Chromium VI | 0.0304c | 0.0009 | 0.002 | 0.003 | 0.006 | 0.003 | RfD |
Conc. = Maximum contaminant concentration measured in the Kassler
Water Treatment Plant infiltration galleries, before closure (ppm)
IR = Ingestion rate: adult = 2 liters per day, child
= 1 liter per day
EF = Exposure frequency, or the number of exposure events
(1 event x 7 days x 52 weeks or 365 days per year)
ED = Exposure duration, or the duration over which exposure
occurs: adult = 27 years; child = 10 years
BW = Body weight (kg): adult = 70 kg (154 pounds); child
= 16 kg (35.2 pounds)
AT = Averaging time, or the period over which cumulative
exposures are averaged (10 years or 27 years x 365 days)
b Total Estimated Exposure Dose = estimated
exposure dose x 3 (to account for ingestion, dermal, and inhalation exposure
routes)
c One detection of 0.07 ppm Cr VI was recorded
during the evening of February 13, 1982. That detection represents an outlier
in the available data set; the remainder of the detections falling in the range
from 0.01 to 0.30 ppm Cr VI. For the purposes of this analysis, given the identified
factors which mitigate any level used in these calculations, we selected 0.304
ppm as the representative maximum value represented in the data set.
Key: ppm = parts per million; mg/kg/day = milligrams contaminant per kilogram body weight per day; MRL = Minimal Risk Level ; RfD= Reference Dose
TABLE C-2. Estimated Exposure Doses--Cancer Effects
Ingestion of Water From the Kassler Water Treatment Plant
| Contaminant |
Maximum Contaminant Concentration (ppm) |
Estimated Exposure Dose-Cancer (mg/kg/day) a | Total Estimated Exposure Dose-Cancer (mg/kg/day) b | Cancer Potency Factor | Lifetime Cancer Risk c |
| Trichloroethylene | 0.0031 | 0.00003
|
0.00009 | 0.011d | 1 x 10-6 |
Conc. = Maximum contaminant concentration in the Kassler Water
Treatment Plant infiltration galleries, before closure (ppm)
IR = Ingestion rate: 2 liters per day
EF = Exposure frequency, or the number of exposure events
(1 event x 7 days x 52 weeks or 350 days per year)
ED = Exposure duration, or the duration over which exposure
occurs = 27 years
BW = Body weight (kg): 70 kg (154 pounds)
AT = Average time or the time over which cumulative exposures
are averaged (70 years x 365 days)
b Total Estimated Exposure Dose = Estimated Exposure Dose x 3 (to account for ingestion, dermal, and inhalation exposure routes)
c Lifetime Cancer Risk = Total Estimated Exposure Dose (Cancer) x Cancer Potency Factor
d EPA is currently reviewing the CPF for TCE. ATSDR uses the previously derived CPF for screening purposes.
Key: ppm = parts per million; mg/kg/day = milligrams contaminant per kilogram body weight per day
APPENDIX D: Fate and Transport of Trichloroethylene, Hydrazine, n-Nitrosodimethylamine, and Hexavalent Chromium
At AFP PJKS, past operations and disposal practices have released contaminants to soil and surface water. Some of these contaminants have also migrated to groundwater. Contaminant fate and transport discussions focus on the factors that influence how contaminants move and change after they have been released to the environment. Contaminants may migrate away from the point of release in an unchanged form and/or may be changed (degraded) as they migrate. The following provides a brief overview of the physical and chemical properties that can affect how contaminants move through the environment, a description of several decrease contaminant concentrations in the environment, and a discussion of how several site-specific contaminants would be expected to move and possibly degrade at AFP PJKS.
Physical and Chemical Properties
The most important factors that impact contaminant movement through the environment are the contaminant's chemical and physical properties. Chemical and physical properties describe, for example, how a contaminant interacts with water or how likely it is to enter the atmosphere. Each contaminant or group of contaminants have unique chemical and physical properties that influence its movement. A description of several chemical and physical properties that strongly influence contaminant migration is provided below. Table D-1 provides the chemical and physical properties for the site-specific contaminants trichloroethylene (TCE), hydrazine, n-nitrosodimethylamine (NDMA), and hexavalent chromium.
Table D-1: Chemical and Physical Properties of Contaminants Detected at AFP PJKS
| Contaminant | Water Solubility (ppm) | Vapor Pressure (mm Hg) | Henry's Law Constant (atm-m3/mole) |
Specific Gravity (unitless) | Organic Carbon Partition Coefficient (mL/g) |
| TCE | 1,100 | 69 | 0.0103 | 1.5 | 0.2 |
| Hydrazine | miscible | 10 | no data | 1.01 | no data |
| NDMA | >100,000 | 0.1 | no data | 1.0 | no data |
| Hexavalent Chromium | insoluble | negligible | not available | 7.1 | not available |
Migration and Degradation Processes
As contaminants move through the environment, contaminant concentrations decrease. This decrease results from the physical process occurring during migration or chemical interactions and biological processes that change or degrade the contaminant into new forms of chemicals. Physical migration and chemical degradation processes that may be occurring at AFP PJKS include:
Physical Migration Processes
Chemical Degradation Processes
Fate and Transport of Site-Specific Contaminants
Understanding the chemical and physical properties for site-specific contaminants and potential degradation processes provides a basis for evaluating the overall fate and transport of contaminants at AFP PJKS, as discussed below. Site-specific conditions, such as soil characteristics and groundwater depth, will also impact contaminant fate and transport. The 1990 Remedial Investigation and the 1999 Supplemental Remedial Investigation provide a more detailed evaluation of the fate and transport of contaminants at AFP PJKS.
Trichloroethylene
TCE has been detected in on-site groundwater (to a maximum of 11,000 parts per billion [ppb]) and surface water (to 22 ppb) at PJKS. Lower levels were measured off site. TCE is a colorless liquid with a sweet odor and is mainly used as a solvent to remove grease from machinery parts. The water solubility value of 1,000 ppm indicates that once released, TCE will dissolve readily in surface water or groundwater. Once in surface water or groundwater, two factors (Henry's Law Constant and the organic carbon partition coefficient) influence how TCE will migrate with surface water or groundwater or migrate to other media, such as air or soil. The Henry's Law Constant is above 10-3, and indicates that TCE easily volatilizes to the air. TCE will volatilize to air from surface water faster than it will from groundwater because TCE in surface water is in direct contact with the atmosphere. Interactions between TCE and natural components of the atmosphere quickly degrade TCE into other chemicals. Studies have found that once in the atmosphere, approximately one half of the TCE is degraded within one week. This degradation time, however, varies depending on the season and temperature. TCE in groundwater will volatilize to the air found in soil and eventually to the atmosphere. The organic carbon partition coefficient is low, indicating that TCE will only adhere to soil to a small degree. TCE, therefore, would be expected to be transported with water along the groundwater or surface water flow pathways.
Chemical reactions in the air will quickly degrade TCE. Studies have found that hydrolysis and photolysis do not substantially contribute to TCE degradation. Biodegradation does occur, but is minimal.
Hydrazine
At PJKS, hydrazine (to 2,000 ppb on site) has been detected in the groundwater in the alluvium near the D-1 Test Stands. Hydrazine is a clear, colorless liquid with an ammonia-like odor and was often used as a component of rocket fuel or an additive to boiler water to reduce corrosion. Hydrazine's water solubility is considered miscible, which means that an unlimited amount of hydrazine can dissolve in water. As such, hydrazine released to either surface water or groundwater is expected to completely dissolve. Because hydrazine is very soluble in water, a high vapor pressure and Henry's Law Constant are needed for volatilization from water to occur. Hydrazine has a moderately high vapor pressure, indicating that volatilization from soil and water is likely. Although there are not enough data to determine the Henry's Law Constant, laboratory studies show that hydrazine in water does readily volatilize. There are not enough data to determine the organic carbon partition coefficient for hydrazine, therefore, it is unknown how adherence to soil may influence hydrazine movement in groundwater.
Once in air, hydrazine is completely degraded within a few minutes or hours by chemical reactions with components of the atmosphere. In water and soil, chemical reactions with oxygen and biodegradation will breakdown hydrazine. The primary chemical reaction in water produces nitrogen gas and water. In oxygen deficient waters, ammonia may also be produced. The rate at which the processes occur depend on site-specific conditions such as water quality and soil characteristics.
n-Nitrosodimethylamine
N-Nitrosodimethylamine (NDMA) (to 42 ppb) was detected in groundwater in the alluvium and bedrock along the East Fork area of AFP PJKS. NDMA, which is a breakdown product of hydrazine, is a yellow, oily liquid with a faint odor. Similar to hydrazine, NDMA has a high water solubility indicating that NDMA detected in groundwater is likely completely dissolved. Volatilization to air from water is expected because NDMA has a moderately high vapor pressure. Currently, there are not enough data to determine the Henry's Law Constant or organic carbon partition coefficient. The tendency for NDMA to volatilize from water or adhere to soil, therefore, is unclear. NDMA, however, could be expected to move with groundwater flow based on its high water solubility.
Hexavalent Chromium
Hexavalent chromium has been detected in groundwater downgradient of the Inactive Site at the Lockheed Martin Astronautics property surrounding AFP PJKS. Chromium, usually as trivalent chromium, is a naturally occurring element in the environment. Hexavalent chromium is generally produced by industrial processes. Hexavalent chromium is not found in the environment as an individual molecule but as part of a compound, such as calcium chromate or potassium chromate. Hexavalent chromium is considered insoluble in water. If released to water, only very small amounts of the hexavalent chromium will dissolved; most will bind with soil, sediment, or other particles in the water to form chromium compounds. As indicated by the vapor pressure and Henry's Law Constant, hexavalent chromium will not volatilize. When detected in water samples, hexavalent chromium is usually present adhered to suspended particles versus dissolved in the water. Based on hexavalent chromium's tendency to adhere to soil and its inability to volatilize, it is often considered immobile in the environment and most likely migrates with windblow dust or with particles suspended in surface water or groundwater flow.
Chromium is a naturally occurring element in the environment, and therefore, will not degrade. Chromium, however, may be transformed from the hexavalent form to the trivalent form through chemical reactions with organic matter in water.
References:
ATSDR. 1997. Toxicological Profile for Hydrazines. Agency for Toxic Substances and Disease Registry. U.S. Department of Health and Human Services. September 1997.
ATSDR. 1997. Toxicological Profile for Trichloroethylene (Update). Agency for Toxic Substances and Disease Registry. U.S. Department of Health and Human Services. September 1997.
ATSDR. 1998. Toxicological Profile for Chromium (Draft). Agency for Toxic Substances and Disease Registry. U.S. Department of Health and Human Services. August 1998.
Hemond, H. and E. Fechner. 1994. Chemical Fate and Transport in the Environment. Academic Press. San Diego, California.
APPENDIX E: Responses to Public Comments
The Agency for Toxic Substances and Disease Registry (ATSDR) received the following comments/questions during the public comment period (April 9 to July 1, 1998) for the AFP PJKS Public Health Assessment (PHA) (April 8, 1998). Comments were received from a total of seven respondents, including four from government or business entities and three from community members or groups. Comments from the latter group were highly critical of the PHA and resulted in an extended fact-finding effort conducted by the ATSDR Ombudsman which concluded late in 1999. Interested parties raised numerous issues during this fact-finding effort and were encouraged to supply or identify factual materials or information to ensure that the scope and content of this assessment was accurate. In this section ATSDR documents our reply to the issue or question posed in the comment and indicates if changes or additions were made to the text.
For comments that questioned the validity of statements made in the PHA, ATSDR verified or corrected the statements. The list of comments does not include editorial comments concerning such things as word spelling or sentence syntax. ATSDR has not addressed requests for information to be included in the PHA, unless the party who filed the request provided the supporting documentation.
Response: ATSDR's assessment of potential contamination of drinking water in Deer Creek Mesa wells is based on information presented in the environmental investigation documents and well monitoring results (Earth Tech, 1996; Engineering Science, 1988, 1989, 1990; Geraghty & Miller, Inc. 1990; Parsons Engineering, 1999). ATSDR recognizes that sampling of the Deer Creek Mesa wells occurred over 10 years ago. Results from off-post groundwater monitoring located on the Lockheed Martin Astronautics (LMA) property surrounding the AFP PJKS site indicate that contamination has not migrated beyond the LMA property. In addition, the observed elevations of the watertable in areas of concern and the area hydrogeology strongly suggest that TCE could not migrate to the Deer Creek Mesa wells. The extent of off-site migration is limited because the bedrock and alluvial groundwater discharges to a tributary of the Lariat Gulch where the contaminants are rapidly volatilized. The tributary is approximately 3,500 feet southwest (and upstream) of the point where Lariat Gulch exits the LMA property. Together, these findings provide compelling evidence that contamination has not reached the Deer Creek Mesa wells.
Response: Thank you. The text has been modified as suggested.
Response: Methylene chloride and acetone were detected in 1987 samples collected from Deer Creek Mesa wells. These chemicals were also found at similar concentrations in both field and trip blanks for sampled wells. Field blanks are prepared to detect errors resulting from the sampling process; while trip blanks detect possible laboratory contamination. Given that methylene chloride and acetone were detected in the blanks at levels similar to those in the well samples, it seems highly unlikely that the aquifer or private well water is actually contaminated with either of these two chemicals. A note has been added to the text indicating that methylene chloride and acetone were detected in the samples and that they are probably unrelated to the site.
Response: ATSDR reviewed the U.S. 1990 census data because it is the most recent census data available. The area surrounding AFP PJKS may be more populated than it was in 1990, but the public health assessment identified no off-site environmental hazards. For this reason, ATSDR believes that a more elaborate review of census data was not warranted. ATSDR will use the year 2000 U.S. census data when it becomes available.
Response: To date, several rounds of monitoring of strategically-placed groundwater wells indicate that none of the suspected contaminants associated with the AFP PJKS site pose a threat to off-site areas. AFP PJKS-related contaminated groundwater is either contained within the AFP PJKS property or within the surrounding LMA property. Some pathways for off-site migration of LMA contaminants detected in alluvial groundwater have been identified within the South Central Valley, including Filter Gulch and the west branch of Brush Creek. To date, the AFP PJKS plumes do not appear to have migrated near Filter Gulch or toward Brush Creek in significant quantities.
Response: Likewise, ATSDR prefers to evaluate indoor air monitoring data when assessing potential environmental exposures. According to occupational safety personnel at the site, air monitoring data are not available, but they indicate that site conditions have not prompted concerns about indoor air quality. In the absence of these data, however, ATSDR took a two-step approach to evaluating the potential for harmful exposure to indoor air contaminants. First, ATSDR identified on-site buildings located in the path of the plumes. The only buildings identified in the path of a plume are the T-6 Block House and the Propulsion Research Laboratory in located in the northern section of the site. Second, ATSDR reviewed soil gas concentrations (1.4 ppm) collected near these buildings and estimated indoor air concentrations values based on a conservative or "safe" screening approach. Using this screening approach, it appears that the estimated indoor air concentration is safely below levels associated with public health hazards. Furthermore, on the basis of this screening evaluation, indoor air monitoring is not deemed necessary.
Response: Whenever possible, ATSDR uses data collected directly from drinking water wells or groundwater collections system, rather than data from monitoring wells. There are many factors that can result in differences in contaminant levels measured in monitoring wells and those actually detected in the private or municipal water well or infiltration system. ATSDR believes that the contaminant levels detected in the water in the Kassler galleries or the Five-Sided Well more accurately represent the raw-groundwater component supplied as a component of the water processed by the Kassler Water Treatment Plant. For this reason, ATSDR used contaminant concentrations detected in the infiltration gallery system in our exposure dose assessment. It should be noted that contaminant concentrations in the galleries would have been greatly diluted within the treatment and distribution system before reaching any residential taps.
Response: Thank you. ATSDR has added that information to Tables C-1 and C-2.
Response: Magnesium-thorium alloy was used by the Air Force in the production of missiles. In 1971, 25 kilograms (or 55 pounds) of magnesium-thorium alloy shavings were disposed of in the D-1 Landfill. This waste contained a total of 55.5 microcuries of radioactivity. In 1986, the drums containing the waste were removed from the landfill and transported to an approved off-site radioactive material handling facility. At the time of removal, the drums were intact and rust-free. Monitoring also confirmed that the drums emitted only background levels of radiation. Based on this information, radioactive waste from the drums could not have entered the surrounding soil or groundwater at levels above background.
Response: TCE was the only site-related contaminant detected in the creek (surface water or sediment) at elevated levels. Together, TCE's experimentally measured bioconcentration factor--a measure of a chemical's tendency to partition to the tissue of organisms--and scientific literature indicate that TCE does not readily accumulate in fish tissue. For this reason, ATSDR considers exposure to site-related contaminants from ingestion of Brush Creek fish to be minimal.
Response: ATSDR conducted its evaluation and made conclusions and recommendations based on environmental data and exposure information most pertinent to the site that have been provided by AFP PJKS; LMA; local, state, and federal agencies; and the public. If anyone has additional information on the pathways addressed in the PHA, or evidence of other potential exposures, ATSDR encourages that person to share the information with us so that we can determine whether a health hazard exists.
In regards to our evaluation of on-site chemicals, it must be remembered that this public health assessment is an evaluation of human exposure to contamination. ATSDR considers the potential for on-site exposure to be low, even though high concentrations of certain contaminants have been measured on site. ATSDR suggests that the high on-site soil contaminant concentrations present more of a remediation concern than a human health concern.
Response: The situation described herein by Lockheed Martin Astronautics is an accurate characterization of past conditions. ATSDR's exposure scenario, however, represents a hypothetical event (considering worst case conditions) using the hypothetical maximum concentration--the actual concentrations at the taps would have been much lower than the maximum detected in the infiltration galleries. This hypothetical maximum exposure scenario assessment is not intended to describe actual past exposures to specific contaminant levels, rather, it is designed to illustrate that, even if these past exposure levels and exposure scenarios existed, no adverse health effects would have resulted.
Response: ATSDR has reviewed the summary of the Renaud case in the Ferrara citations. Accordingly, we have modified the exposure discussion to more accurately reflect the two widows of time during which Friendly Hills could have received water from the Kassler Water Treatment Plant distribution lines. The information in the Ferrara citations indicates, however, that the upper bound on the second window of time is December 1984, not February 1983 as suggested. Based on this information, we have modified our discussion to reflect the time periods (June 1997 through December 1980 and June 1982 through December 1984) for when Kassler Water Treatment Plant water could possibly have reached Friendly Hills.
Response: No evidence exists to suggest that hydrazines has been detected in surface water or groundwater beyond AFP PJKS/LMA site boundaries, or that they reached the Kassler area. ATSDR contacted scientists at Wright-Patterson AFB and Brooks AFB to gather more information about the uses of hydrazines at the AFP PJKS site. Hydrazines (referring collectively at this site to hydrazine, monomethylhydrazine, and unsymmetrical dimethyl hydrazine [UDMH]) are clear, colorless liquids used for rocket fuels. Testing of rocket engines with hydrazine fuels resulted in inadvertent releases of hydrazines on site and generation of hydrazine-containing waste/rinse water, which was collected, treated, and recycled for subsequent rocket engine testing. Any inadvertent release of those rinse waters to Lower Brush Creek from the T-8A pond (the downgradient component of the collection system) would have been treated rinse waters.
Any hydrazines that may have been released from PJKS are unlikely to have migrated to the Kassler area for several reasons. First, hydrazines break down readily in the environment. Most of the hydrazines were released to the air, where they quickly dissipated. Smaller amounts of hydrazines entered surrounding soil or surface water. Hydrazines released to on-site soil or surface water either evaporated into the air, dissolved in surface water, or bound to soil, but most eventually dissipated. Second, the waste/rinse water was treated in a two step process using hydrogen peroxide and an acid base neutralizing tank. This treatment reduced the concentrations or removed hydrazines from the waste/rinse water before it was released to the T-8A pond. Third, a substantial amount of reduction-known as biodegradation-occurs when hydrazines move through an environment rich in organic mater. The organic material in the stream bed and in the muddy water of Brush Creek would have further adsorbed and degraded any residual hydrazines that entered the creek. Fourth, and most importantly, monitoring conducted by the Air Force found no evidence of hydrazines in off-site surface water or groundwater, further supporting that hydrazines could not have reached the Kassler area.
It should be noted that monitoring sponsored by the National Toxics Campaign (NTC) reports findings of hydrazines in off-site surface water samples collected from Lower Brush Creek. The NTC data, however, have largely been discredited because of a number of suspect practices and errors associated with sample collection, analysis, and handling. Critical review of the NTC sampling, analysis, and data discloses the following problems: use of a method prone to false positive results (ASTM D 1385); use of inadequate analytical equipment; lax quality control and assurance practices (i.e., poor record keeping [no record or chain of command for samples; no record of a sampling plan or a sampling procedure, no record of sampling date]); and lax laboratory procedures (i.e., failure to calibrate the analytical equipment for hydrazine, failure to perform a duplicate sample for hydrazine, failure to determine a detection limit for hydrazine). Because of these problems, the data are of questionable quality and have little usefulness for evaluating the potential for human health hazards in our public health assessment.
Response: ATSDR also asked Wright-Patterson AFB and Brooks AFB scientists about hydrazines' "interactions" with chlorine. Chlorine interacts with hydrazine to initiate the oxidation process. As a result of the oxidation process, hydrazines are broken down, resulting in greatly reduced concentrations and the release of breakdown byproducts. One primary breakdown product of UDMH is nitrosodimethylamine (NDMA), a compound shown to cause cancer in laboratory animals but not in human. Site information does not suggest that AFP PJKS used chlorine in its treatment process. Rather, as noted in Response 14, AFP PJKS treated rinse water containing hydrazine with hydrogen peroxide and an acid-base neutralizing solution. Through these treatment processes, hydrazines were reduced to nitrogen and water (by its interaction with hydrogen peroxide). The Air Force monitored for hydrazine and NDMA levels in surface water and groundwater, but they found no levels above detection levels beyond site boundaries.
Kassler Water Treatment Plant used chlorine at their facility. Water entering the Kassler Water Treatment Plant was chlorinated at the well head and then pumped into storage tanks, where it was blended with ammoniated water and chlorinated again. In the unlikely event that water containing hydrazines reached Kassler Water Treatment Plant, chlorination would have further reduced hydrazines in water.
Response: The report in question is the Miller et al. 1989 report that was prepared and submitted on behalf of the plaintiffs in the Friendly Hills Renaud litigation. The report evaluated and attempted to support the assertion that water entering and leaving the Five-sided well at the Kassler Water Treatment Plant between 1975 and 1985 was contaminated with harmful levels of chemicals from the AFP PJKS and the Lockheed Martin Astronautics facilities. Based on an evaluation of contaminants originating at Tanks 6032 and 6034 and released from the T-8A pond (or Lockheed Martin Astronautics aeration ponds) into Brush Creek, the report concluded that 88 part per billion (ppb) total hydrazine and 7.6 ppb trichloroethylene (TCE) could have entered the Five-sided well at the Kassler facility.
The report was critically evaluated by a neutral expert (a geochemist/chemical hydrogeologist) appointed by the court. The expert found that the conclusions drawn in the report could not be reasonably relied upon because of a number weaknesses in the evaluation, including the use of questionable assumptions, misrepresented facts, factual omissions, poor documentation, and flawed methodology.
Following the expert review of the report, as well as testimony by the report's lead author, the court ruled that the report was "inadmissable as a matter of law because the opinions were based on fundamentally flawed methodologies and incorrect assumptions." Without credible evidence of contamination in the Five-sided well at the Kassler Water Treatment Plant, there was no basis for proceeding with the plaintiff's claim that contaminants originating at AFP PJKS or Lockheed Martin Astronautics had reached residential taps.
ATSDR also reviewed the report and the expert's evaluation. While the report used a standard approach to evaluating contaminant transport, ATSDR found that the authors relied upon assumptions at various steps that oversimplified the analysis of contaminant concentration/migration, ultimately leading to overestimates of the contaminant concentrations entering Kassler's system. Because of the problems in the evaluation, ATSDR was unable to rely upon the data presented in the report for its assessment of potential human health hazards. Summaries of a few of the inappropriate assumptions or misinterpretation of data are provided below:
Estimating source (contaminant) loading.The source loading concentration is the initial concentration of hydrazine, UDMH, and NDMA (and TCE) present in rinse water from Tanks 6032 and 6034. This concentration was assumed to have been released to Pond T-8A.
Estimating discharge/flow from Pond T-8A/aeration pond into Brush Creek. This step attempted to predict the amount of rinse water, and ultimately the amount of hydrazine and other contaminants, that was released from Pond T-8A and the Lockheed Martin Astronautics aeration ponds into Brush Creek.
Estimating concentrations entering and leaving Kassler system/Five-sided well . Through this step, the authors attempted to predict contaminant concentrations that would reach and be intercepted by the Kassler infiltration gallery and eventually enter the Five-sided well.
For the reasons given above, and because hydrazine, UDMA, or NDMA was never detected in off-site groundwater, surface water, or raw water in the Five-sided well, ATSDR cannot base its findings of potential past human health hazards upon data presented in the Miller et al.1989 report.
Response: The commenter may have been referring to the results of a 1986 sampling in which 1,1,1-TCA (1.5 to 1.6 ppb) and 1,2-dichloroethylene (1.6 ppb) were detected in one hot water tank in a home at Friendly Hills. The levels detected in the hot water tank were safely below EPA's maximum contaminant levels (MCLs) for drinking water. No VOCs were detected in one occupied or two additional unoccupied homes tested at Friendly Hills. Although the source of the low levels of VOCs in the one hot water tank is still unknown, it would be reasonable to assume that contamination occurred during manufacturing, operating, or installation of the individual tank rather than through the water supply that served all tested homes.
Response: ATSDR contacted Kirkland Air Force Base for information about radioactive/nuclear material uses at AFP PJKS. Kirkland AFB oversees and maintains all records on closure at former Atomic Energy Commission (AEC) munition sites that were transferred to the Air Force in 1965. Based on records searches and discussion with staff no evidence exist suggesting that AFP PJKS conducted AEC operations. ATSDR also contacted LMA to gather more information about the use and occurrence of radioactive materials at their Waterton facility. According to the LMA safety officer, LMA has never been involved with the design and production of radioactive/nuclear materials or weapons. They acknowledge that their Waterton facility used small amounts of radioactive materials in finished products. The material used by LMA either: 1) required a license and storage in a "sealed source" (designed and approved by state or federal regulatory agencies), or 2) required minimal control or were exempt from licensing due to low hazard potential (e.g., smoke detectors, laboratory sample standards). In either case, the material had only low radioactive emissions. Any unused material was either returned to the original suppliers or disposed of in a permitted low-level radioactive waste site. Some small amounts of radioactive materials were disposed of in containers in on-site landfills; these containers have been removed and shipped off site to a permitted disposal facility.
LMA representatives also state that LMA had never mined or processed radioactive materials at their Waterton facility. During the mid-1970s, LMA was approached with a proposal to explore approximately 80 acres of the southwest portion of the site (and beyond the site boundaries) for geological radiologic materials. The exploration proposal did not, however, result in mining activities at the Waterton facility.
Response: Much attention has been given to the occurrence of cancer or other health problems in areas of Denver that received water from Kassler Water Treatment Plant. Some community members believe that materials from LMA/AFP PJKS entered the drinking water supply (at the Kassler Water Treatment Plant). They also believe the suspected contaminants in their drinking water contributed to alleged high rates of childhood cancer and other health problems (low birth weight, birth defects) among residents of the Friendly Hills/Harriman Park neighborhoods of Denver.
In 1983, a community group conducted an informal, self-reported investigation of health problems in these neighborhoods. Although the community-organized survey reported an elevated number of unexplained illnesses and childhood cancer, a subsequent review by the Colorado Department of Public Health and Environment (CDPHE) found neither frequent reports of chronic health problems (the most common problems were headaches and diarrhea) nor a viable means of exposure to chemicals at LMA or AFP PJKS.
Continued public health concern prompted CDPHE to conduct an investigation into the reported childhood cancer cases. An analysis of all childhood malignancies (ages 0-14 years) reported during 1979-1982 in a selected census tract (120.17) in the neighborhood of concern (population of 4,000) showed there were five cases of childhood cancer; this rate was statistically significant and was approximately 2.5 times the expected rate of 2. An extensive field investigation of ground, surface, and drinking water, basement sumps, soils, sediments, and air conducted by EPA in 1984 showed all detected substances to be within normal background ranges. In the fall of 1984, a second cancer incidence investigation was conducted in 1984 for the original census tract (120.17) and an adjoining tract (120.16 ;CDPHE, 1985). The childhood cancer rates from 1979 through 1983 were not found to be statistically significantly different from the rest of Jefferson County. The data, methods, and results of this study were submitted for review and confirmation to the Cancer Branch, Center for Environmental Health, Centers for Disease Control. In December, 1984, the consensus reported to the residents of Friendly Hills/Harriman Park were told that the cancer "cluster" was most likely due to chance. While the results of the EPA environmental investigation and the CDPHE (1985) cancer rate investigation satisfied many of the Friendly Hills/Harriman Park residents, concern remained high among others. At the request of those residents, in March, 1985 a summary of the CDPHE review the incidence of non-cancer outcomes in the previously studied census tracts was made available to residents and made an addendum to the final report (CDPHE, 1985). Rates of neonatal deaths, fetal loss, birth defects, and low birth weight were found to be within statistically expected ranges for Jefferson County. Again at the request of area residents, CDPHE conducted a follow-up review of the health statistics for 1984 to update the previously reported 1979-1983 findings. Those data continued to show that the incidence of childhood cancer, birth defects, fetal loss, neonatal deaths, and low birthweights were within expected limits.
ATSDR's Division of Health Studies (DHS) evaluated the methodology and findings of the CDPHE (1985) "Investigation into alleged health problems in the Friendly Hills/Harriman Park Area" and the John Hopkins Oncology Center study entitled "Statistical report to the Denver Water Board concerning childhood cancer incidence and neonatal deaths in the Denver metropolitan area, 1979-1987" (Piantodosi, 1990). Together, the weight of all the studies conducted to date suggest that there is no evidence of high rates of childhood cancer or health problems affecting infants or children that might be attributed to living in these neighborhoods or from drinking water from the Kassler Water Treatment Plant. A summary of each study and ATSDR's DHS's conclusions are described below.
In response to a 1984 Citizen's Advisory Committee request for a study on alleged health problems in the Friendly Hills/Harriman Park area, the CDPHE (at the time known as the Colorado Department of Health) completed a descriptive epidemiologic study of childhood cancer, birth defects, fetal loss, neonatal deaths, and low birth weight infants for the period from 1979 through 1983. Jefferson County was divided into 11 census tracts grouped according to demographic similarity, population size, and if the tracts were contiguous. All data are reported as census tract aggregate incidence rates.
Information on cancer in children age 14 and under was obtained from the Colorado Central Cancer Registry. The Friendly Hills/Harriman Park census tract had the second highest cancer incidence rate; however, rates for all census tracts were within the 95% confidence intervals calculated for the expected number of cases for those areas. Comparisons of age-specific cancer incidence rates for children aged 14 and under for the years 1979 through 1983, showed that rates in Jefferson County were not significantly different than rates in the state of Colorado and the United States as a whole. Scientists from the Cancer Branch at the Centers for Disease Control and Prevention reviewed the data and concluded that the rates in the Friendly Hills/Harriman Park area were not significantly higher than expected. However, they note that the wide confidence interval for cancer cases in Friendly Hills/Harriman Park (8.29 to 59.58 cases per 100,00 person-years) "reflects the uncertainties of an estimate based on only 5 cases."
Information on birth defects, fetal loss (miscarriage at 20 or more weeks of gestation), neonatal deaths (deaths within the first 28 days of life), and low birth weight was obtained from the Health Statistics Section of CDPHE. For birth defects, both the number of cases and rates per 1,000 live births in all 11 census tracts did not exceed what was expected for the time period. For fetal loss and neonatal deaths, the number of cases and rates per 1,000 live births plus fetal losses and 1,000 live births, respectively, did not exceed the expected numbers for the time period; additionally, one census tract had a significantly lower number of cases and rate. For low birth weight, nine census tracts (including Friendly Hills/Harriman Park) had case counts and rates within the expected numbers, one census tract was statistically significantly lower than expected, and one census tract was statistically significantly higher than expected.
The epidemiologic analyses concluded that there is no evidence of excess childhood cancer, birth defects, fetal loss, neonatal deaths, or low birth weight infants in the Friendly Hills/Harriman Park area during 1979-1984. The methods used in the analyses of childhood cancer cases and adverse reproductive health effects seem appropriate given the available resources. DHS agrees with CDPHE's interpretation of the results.
A study conducted by the Johns Hopkins Oncology Center investigated the incidence of childhood cancer and neonatal deaths in the Friendly Hills/Harriman Park area of Denver, Colorado during 1979-1987. The study proposed to determine (1) what factors explain the rates of childhood cancer and neonatal deaths in the Denver area, and (2) if water source is associated with these rates. Because individual data about environmental exposures, biological factors, and demographic characteristics were not available, data were analyzed using census tract specific number of childhood cancers and neonatal deaths and aggregate characteristics of census tracts. Poisson regression was used to estimate the rate ratios.
Analysis of the total number of childhood cancer cases in each census tract as a function of potential explanatory factors showed that residence in Friendly Hills/Harriman Park was not significantly associated with high rate of childhood cancer relative to other areas (RR=1.5, p value p=0.39). Additionally, drinking water from Kassler Water Treatment Plant was not significantly associated with high rate of childhood cancer relative to other water treatment plants (RR=0.73, p=0.28). Residence in Friendly Hills/Harriman Park was associated with a marginally higher risk of non-leukemic childhood cancers (RR=2.4, p=0.05); however, this association disappears when other factors are added to the model indicating that residence in Friendly Hills/Harriman Park by itself is not responsible for the increased risk of non-leukemic cancers.
Neither residence in Friendly Hills/Harriman Park nor drinking water from Kassler Water Treatment Plant was associated with a significantly elevated risk of neonatal deaths (RR=1.0, p=0.93 and RR=0.83, p=0.26, respectively). However, other sociodemographic factors such as non-white race, some occupations, and income were associated with elevated neonatal death rates.
The methods used in the analyses of childhood cancer cases and neonatal deaths seem appropriate given the available resources. DHS agrees with the interpretation of the results given in the Piantadosi report.
In summary, ATSDR's DHS agrees with the CDPHE and Piantadosi interpretations of the available data. At this time, ATSDR's DHS does not recommend conducting a large scale epidemiologic study, ATSDR's DHS would only recommend conducting a new large scale epidemiologic study if a review of more recent data showed statistically significantly elevated rates of childhood cancer or adverse reproductive health effects among Friendly Hills/Harriman Park residents and the environmental data showed a completed exposure pathway at a level that could cause health effects.
References used in responding to public comments:
Anderson, A. 1990. Affidavit of Adrienne Anderson, U.S. Dist. Court, Dist. of Colorado: Civil Action No. 87-Z-42, James Earl Renaud, et al. vs. Martin Marietta Corp., et al..May 31, 1990.
ATSDR. 1999. Agency for Toxic Substances and Disease Registry, Division of Health Studies, Review of health studies in the Friendly Hills/Harriman Park neighborhoods. December 1999.
CDPHE. 1985. Colorado Department of Public Health and Environment (formerly Colorado Department of Health), Final report of investigations into alleged health problems in the Friendly Hills/Harriman Park area. May 20, 1985
French American Metals Corporation (FRAMCO). 1975. Personal communication between James Hiatt, Landman, FRAMCO, and Morris McLain, Martin-Marietta Aerospace. RE: Colorado, Jefferson County, Bonzo area. May 20, 1975.
French American Metals Corporation (FRAMCO). 1975. Personal communication between James Hiatt, Landman, FRAMCO, and Morris McLain, Martin-Marietta Aerospace. RE: Colorado, Jefferson County, Bonzo area. June 9, 1975
Hauser Laboratories. 1987. Personal communications between Hauser Laboratories and Walter Gerash, Attorney. Gerash, Robinson,, Miller & Miranda, P.C. RE: Friendly Hills Water/TCE in a water tank. October 27, 1987.
Hauser. 1990. Affidavit of Ray L. Hauser filed by Denver Water Board, U.S. Dist. Court, Dist. of Colorado: Civil Action No. 87-Z-42, James Earl Renaud, et al. vs. Martin Marietta Corp., et al. May 30, 1990.
Kaltofen, Marco. 1990. Affidavit of Marco Kaltofen. U.S. Dist. Court, Dist. of Colorado: Civil Action No. 87-Z-42, James Earl Renaud, et al. vs. Martin Marietta Corp., et al.
LMA. 1988. Interoffice memo between W.R. Haas, Ph.D., Manager, Environmental Management, Martin Marietta, and Ms. Kelcey Yarbrough Land. RE: Information regarding radioactivity at Martin Marietta Astronautics. March 1988.
LMA. 1988. Interoffice memo between B.Bath and T. Mueller and D.A. Clark. RE: Radioactive material use at Waterton. March 10,1988.
LMA. 1999. Personal communication between ATSDR and William W. Bath, Environmental Management, LMA, RE: Review of the Miller report. October 28, 1999.
Miller, W.L. , Huber, W.C., 1989. Report of water contamination at the Martin Marietta Waterton Facility and the Kassler Water Treatment Plant. October 18, 1989.
Parson Engineering. 2000. Personal Communication between Tom Larson, Parsons Engineering, and ATSDR. February 2000.
Pavlik, H.F. 1990. Comments on reports and testimony provided in the case of Renaud, et. Al. Versus Martin Marietta Corporation, et al., Civil Action Number 87-Z-42. Prepared for the united States District Court District of Colorado, United States Courthouse, Denver Colorado. Prepared by Hannah F. Pavlik, Ph.D. EBASCO Services Incorporated, Lakewood Colorado. August 29, 1990.
Piantadosi, S. 1990. Statistical report to the Denver Water Board concerning childhood cancer incidence and neonatal deaths in the Denver Metropolitan area, 1979 - 1987. Prepared by Steven Piantadosi, M.D., Ph.D., Director, Pncology Biostatistics, John Hopkins Oncology Center. April 24, 1990.
Weinshienk, J. 1990. Order of Dismissal .U.S. Dist. Court, Dist. of Colorado: Civil Action No. 87-2-42, James Earl Renaud, et al. vs. Martin Marietta Corp., et al..May 30, 1990.
