PETITIONED PUBLIC HEALTH ASSESSMENT
GREAT LAKES CHEMICAL CORPORATION
EL DORADO, UNION COUNTY, ARKANSAS
The following sections contain an evaluation of the environmental data available for GLCC. In preparing this evaluation, ATSDR uses established methodologies for determining how people may be exposed to potential contamination related to GLCC and what harmful effects, if any, may result from such exposure. Chemical exposure pathways (or routes of physical contact with chemicals) that ATSDR evaluates are ingestion, inhalation, and skin contact. ATSDR uses comparison values (CVs), which are screening tools used to evaluate environmental data that are relevant to the exposure pathways. Comparison values are concentrations of contaminants that are considered safe levels of exposure. Chemicals detected below CVs are not likely to represent a health concern; chemicals detected above CVs require a more detailed evaluation of site specific exposure conditions. For a complete discussion of these criteria (quality assurance considerations, human exposure pathway analyses, ATSDR health comparison values, and the methods of selecting contaminants above comparison values), refer to Appendix D.
The Environmental Protection Agency (EPA) requires that airborne emissions from GLCC be reported in the Toxic Release Inventory each year for emission permitting . Methyl bromide is the predominant contaminant in GLCC air emissions reported to EPA. To fulfill ADEQ permitting requirements, methyl bromide was monitored from 1990 to 1991 in ambient air at the perimeter of GLCC property. The quality of the air monitoring results during the entire year is questionable because of monitoring equipment malfunctions. After the technical problems were corrected, the maximum methyl bromide concentration was around 0.05 ppm . This concentration is about 27 times below EPA's risk-based concentration for ambient air of 1.33 ppm . ADEQ monitored ambient air at the GLCC property line and Parker's Chapel (a nearby residential community) weekly from 1995 to 1996. During this sampling event, the highest methyl bromide detection was 95 times below EPA's risk-based concentration. Also in 1995, ATSDR collected air samples in the crawl space of a home nearest GLCC property to determine potential human exposure to air contaminants. No contaminants (including methyl bromide) were detected at levels of health concern . On-site ambient air monitoring has not resumed to date. According to these air monitoring results, no adverse health effects are expected to occur from breathing ambient air near GLCC.
There was no ambient air monitoring around GLCC before 1990. ATSDR cannot determine if the GLCC air emissions during this time are related to the respiratory irritations reported by community members.
Several aquifers are within the geological formations beneath GLCC. In order of increasing depth from the surface, the formations include the Cockfield formation, the Cook Mountain formation, and the Sparta formation [1, 2]. The Cockfield formation includes two groundwater aquifers: the interbedded unit is 30 to 80 feet below the surface and the main sands of the Cockfield is 100 to 180 feet below the surface . Results from soil borings and pumping tests have shown these aquifers are hydraulically connected in a downward direction . The Cook Mountain formation acts as a confining space between the Cockfield formation aquifers and the deeper aquifers of the Sparta formation are more than 200 feet below the surface and is over 200 feet thick in some places . The public water supply in this area draws groundwater from the Lower Sparta aquifer in the Sparta formation.
Since 1987, GLCC has collected more than 70,000 groundwater samples and analyzed them for various chemical constituents. On-site groundwater is clearly contaminated with several chemicals listed in Appendix C, Table 1; however, the extent of the on-site groundwater contamination plume has not been completely determined . Many detected chemicals have maximum concentrations exceeding ATSDR's screening values(1) for drinking water, however, this groundwater is not used as drinking water. Although ethylene dibromide (EDB) and ethylene dichloride (EDC) are the two most frequently detected contaminants, the maximum concentrations of EDB and EDC are below concentrations known to cause adverse health effects in animals and humans  . Sampling results from the perimeter of GLCC indicate that EDB and EDC contaminated groundwater remain on-site. No human exposure to contaminated groundwater occurs on-site, groundwater monitoring at the perimeter of GLCC should continue to determine whether site-related contaminants migrate off-site.
Very few studies have been conducted to measure contaminants in groundwater beyond GLCC boundaries. The Environmental Protection Agency (EPA) conducted a study in 1988 of private groundwater wells near GLCC and describing private wells not being affected by GLCC industrial activities . EPA conducted a more extensive off-site groundwater study in 1991 that identified 54 private wells within 1.5 miles of GLCC . Although most wells were not being used for drinking water, 24 private wells were tested for chemical contamination. No site-related contaminants were detected in the private wells. However, polycyclic aromatic hydrocarbons (PAHs) were detected in 1991 above screening values in one well that went out of service in 1968 (Appendix C, Table 2) . It is not known if and how much of these PAHs were in well water before it went out of service. PAHs are not water soluble and were probably bound to sediments in the water which would alter its aesthetic quality (making the water cloudy). Although water with such aesthetic quality was probably not consumed often, the maximum concentrations of PAHs detected in this private well are below levels known to cause health effects in animals and humans .
Public drinking water for El Dorado residents is provided by the El Dorado Water Department and Parker's Chapel Water Association from groundwater within the Sparta formation at a depth of approximately 600 feet . The municipal water is tested at various frequencies for more than 150 chemicals including pesticides, herbicides, metals, volatile organic compounds (VOCs), and semi-volatile organic compounds (SVOCs) under the Safe Drinking Water Act. Ten years of sampling results for potable water were reviewed from both municipalities. Neither of the two public water municipalities ever detected ethylene dibromide or ethylene dichloride . Thus, these site-related contaminants do not appear to have migrated to the local municipal wells.
Two streams, Bayou de Loutre and Little Cornie Bayou, on GLCC property eventually flow into the Ouachita River and are not used for drinking water. GLCC releases stormwater and non-contact cooling water to the Bayou de Loutre and Little Cornie Bayou through four permitted outfalls . In fulfillment of the National Pollutant Discharge Elimination System permit requirements, GLCC monitors each outfall four times a year for more than 40 chemicals. Samples are collected directly from the GLCC effluent.
Several chemicals were measured in the surface water from GLCC outfalls from 1988 to 1996. Appendix C, Table 3 lists the maximum concentration of chemicals detected in effluent from the four outfalls. These concentrations characterize water quality of GLCC effluent only; the concentrations of chemicals in surface water will be lower as outfall water mixes with water in receiving streams. The state of Arkansas has approved the Bayou de Loutre and Little Cornie Bayou for fishing and recreation, and as industrial and agricultural water supplies . Therefore, human exposure to the surface water near GLCC is limited to recreational use because it is not being used for drinking water.
The two receiving streams of GLCC outfalls flow on or nearby several residential properties. Surface water samples were collected in 1988 from one residential property nearest GLCC. The monitoring results are limited to analysis for one chemical that is produced at GLCC, ethylene dibromide (EDB). The maximum EDB concentration detected at the residence was in a drainage ditch at 0.8 ppb. ATSDR resampled the surface water at this residence in 1995; no contaminants were detected at levels of health concern . The only site-related chemicals detected were total bromides at a concentration of 1.85 ppm. Assuming that all of the total bromides detected are EDB (one of the more toxic compounds), this level is below concentrations known to cause adverse health effects in animals and humans .
Sediment and sludge samples were collected and analyzed for chemical contamination from on-site SWMUs such as waste lagoons, acid pits, tail brine ponds, and aeration ponds. Bromides (16,000 parts per million [ppm]); chlorides (4,200 ppm), and ethylene dibromide (0.86 ppm) were detected in the sediment and sludge. The sediment in on-site SWMU is not accessible to the public or GLCC workers; therefore, no human exposure to these contaminated sediments is expected to occur.
Sediments from creeks and drainage ditches near GLCC outfalls on two residential properties were analyzed for chemical contamination in 1988 and 1991. Sediments move and the rate depends on numerous factors including flow rate, depth of water, and type of sediment. Consequently, substances detected will vary at a given location and depth. The only confirmed site-related chemicals detected during 1988 and 1991 in the sediment were EDB and bromides. The maximum concentration of EDB (7 ppb) found in off-site sediments is below comparison values for residential soil(2). The maximum concentration of bromides detected in off-site sediment was 665,000 ppb. Although no comparison values exist for bromides in sediment or soil, exposure to the maximum concentration from incidental ingestion is below levels known to cause health adverse health effects in animals and humans   . No adverse health effects are expected to occur from exposure to off-site sediments.
Since construction in 1965, GLCC has disposed of wastes into several surface impoundments, aeration ponds, and landfills. In 1987 and 1988, a series of soil boring samples detected contaminants that leaked from several of these SWMUs through the soil and into the groundwater. On-site soils near the SWMUs have been completely characterized at various depths by extensive soil sampling. There were 88 chemicals detected at various depths near the SWMUs . Most sampling efforts targeted EDB and EDC because they are both waste products of GLCC manufacturing processes. EDB and EDC were detected in nearly every soil sample and at different depths near SWMUs (Appendix C, Tables 4 and 5) . Other brominated compounds were detected such as tribromophenol, bromides, and bromochloroethane. Soil sample results show an apparent increasing concentration gradient with increasing soil depth. If the concentration gradient increases with increasing soil depth, soil gas may be released with construction or within existing structures. Soil gas substances may migrate into structures, particularly basements or underground utilities. Personal protective equipment should be worn because GLCC workers may have incidental contact with contaminated soil or soil gas during excavation projects. Security measures at GLCC prevent public access to contaminated soil on-site.
Off-site soil sampling was conducted by individual residents and ATSDR. Soil samples were taken from three homes near GLCC. Although not related to GLCC industrial activities, benzo[a]pyrene was the only chemical detected above ATSDR's screening values. Benzo[a]pyrene was detected at one home west of GLCC at a maximum concentration of 0.43 ppm . Occasional exposure to benzo[a]pyrene in soil at this home is not expected to cause adverse health effects. GLCC does not manufacture or process benzo[a]pyrene, so this chemical most likely came from a different source such as roadway runoff or a grease or oil spill.
ATSDR conducted an Exposure Investigation (EI) in 1995 that entailed the collection of blood samples from a family of four residents living near GLCC to determine whether environmental exposure to brominated compounds was occurring at levels of health concern. Ambient air was identified as one of the possible sources of exposure to brominated compounds . The blood serum bromide levels ranged from 10 to 18 milligrams per deciliter (mg/dl) or from 100 to 180 ppm. These bromide concentrations in serum are about 10 times above what would normally be expected (5 mg/dl or 50 ppm) . These levels of bromides in blood serum indicated exposure to brominated compounds, however toxic effects are not associated with bromide levels less than 500 ppm .
ATSDR conducted a second EI in 1996 to retest the family of four and to determine bromide exposure in other area residents. Bromide levels in blood of 21 residents within one mile of GLCC ranged from 2.5 mg/dl (25 ppm) to 7.2 mg/dl (72 ppm). These bromide levels are within normal background range. During the first and second EI, the laboratory method used during the blood analyses predicted short-term, high level exposure to brominated compounds. However, this method was inadequate to estimate long-term, low level exposure to brominated compounds. In order to ensure that residents were not being exposed to low levels of brominated compounds, a third round of blood testing was conducted on 19 of the 21 persons who participated in the second EI. The level of blood serum bromide found in all 19 samples was within the normal average range . No evidence was found of a correlation between these blood bromide levels and adverse health effects.
Children are at greater risk than are adults from certain kinds of exposure to hazardous substances emitted from waste sites and emergency events. They are more likely to be exposed for several reasons; children play outside more often than adults, increasing the likelihood that they will come into contact with chemicals in the environment. Since they are shorter than adults, they breathe more dust and heavy vapors close to the ground. Children are also smaller, and thus receive higher doses of chemical exposure per body weight. Finally, the developing body systems of children can sustain damage if toxic exposures occur during certain growth stages.
ATSDR closely reviewed possible exposure situations for these children while evaluating this site. ATSDR also used the Environmental Media Evaluation Guidelines (EMEG) for children, who are considered the most sensitive segment of the population. ATSDR did not identify any chemical contaminants at levels of health concern to children living near GLCC.
Residents in El Dorado expressed concern that potential exposures from industrial activities at GLCC would result in increased incidence of ALS in the community. Arkansas has no disease registry, however, ATSDR reviewed death certificates for the disease category of ALS to determine if occurrence rates were unusual . ATSDR obtained raw data regarding Motor Neuron Disease (MND) deaths in El Dorado from Arkansas Department of Health (ADOH) and the National Center for Health Statistics. ADOH also performed its own review of MND data and provided a copy of their review to ATSDR. ATSDR reviewed both ADOH's assessment and the raw data on MND mortality .
Review of Motor Neuron Disease Mortality Data
Summary data provided by ADOH indicate 13 MND deaths occurred in Union County, Arkansas, in 1980-1993 . This number is in agreement with the National Center for Health Statistics' database. ATSDR performed three comparisons to evaluate the MND mortality rate in the El Dorado area:
County: ATSDR reviewed the rate of MND deaths in all of Union County (13 cases in the study periods of 1980-1993) against the expected number of cases for a county-wide average population of approximately 47,000 during the same 1980-1993 time period. ATSDR did not find a statistically significant (95% confidence interval) difference in the number of actual MND deaths as compared with the expected number of MND deaths.
City: ATSDR compared the six deaths known to be within the El Dorado city limits to the number of cases expected to occur during the same time frame in an average population of about 24,000. There was no statistically significant increase found in the number of deaths known to be within the city limits.
Zip Code: ATSDR located a database containing a 1993 population estimate of 35,000 for the zip code 71730, which includes the city of El Dorado and a portion of the rural area surrounding the city. The MND mortality rate was not significantly higher than would be expected for an average population of 35,000 over the same time frame.
Comparison with Arkansas Department of Health Analysis
An ATSDR review of the ADOH evaluation of the same data indicated their results to be in agreement. ADOH's report found neither the county-wide MND mortality rate nor the city-wide MND mortality rate was significantly elevated. The zip code comparison calculations were not performed.
During 1980-1993, no statistically significant elevation in the rate of MND deaths exists for Union County, the city of El Dorado, or zip code area 71730. There is no known environmental cause of ALS . Since specific health risk factors are not known, ATSDR cannot conclude any association between contaminants from any source (including GLCC) and the MND mortality in El Dorado, Arkansas.
No physical hazards were identified during the site visit or by ADEP's records.
1. For a complete discussion of screening values, refer to Appendix D