PRELIMINARY PUBLIC HEALTH ASSESSMENT
(a/k/a GALAXY INCORPORATED)
ELKTON, CECIL COUNTY, MARYLAND
In subsections A and B of this section (On-site and Off-site Contamination), all availableinformation on chemical contaminants at or near the Spectron/Galaxy site is discussed. Some ofthe chemicals present may not pose a threat to human health. The purpose of this discussion is toidentify those chemicals which may pose a threat and need to be carefully evaluated as to theirpotential to impact public health. That evaluation is described in the Public Health Implicationssection (Toxicological Evaluation subsection). Factors important to the identification of thesechemicals include the frequency of detection and the amount of chemical detected, the quality ofthe sample collection and analysis and comparison of the levels measured in air, water, and soilto health-based comparison values.
Comparison values are concentrations of chemicals in various environmental media to which aperson could be exposed for a long time period with the expectation that no adverse healtheffects would result. Comparison values are generally set well below levels which might causehealth effects following exposure. Consequently, exceedance of a comparison value does notmean that a person exposed to this contaminant is expected to experience health effects; it meansthat a more comprehensive evaluation is necessary. The comparison values used in thisdocument were developed by the U.S. EPA and the U.S. Agency for Toxic Substances andDisease Registry (ATSDR). With the exception of Maximum Contaminant Levels (MCLs) indrinking water, which are legal standards, the comparison values are screening concentrations to assist in identifying chemicals which should be more completely evaluated.
The following health-based comparison values were used in evaluating the availableenvironmental sampling data or are included discussions of these data.
CREG (Cancer Risk Evaluation Guide): The contaminant concentration identified by ATSDR,below which it is expected that the increased risk of cancer from a lifetime exposure to thatcontaminant is less than one-in-one million (generally considered an insignificant risk level).
EMEG (Environmental Media Evaluation Guide): The contaminant concentration, identifiedby ATSDR, which is unlikely to cause adverse non-cancer health effects following a specifiedperiod of exposure (a lifetime for chronic EMEGs, a year or less for intermediate or acuteEMEGs).
MCL (Maximum Contaminant Level): The contaminant concentration in drinking water thatEPA, after considering health effects data as well as the availability and cost of water treatmenttechnology, deems protective of public health over a lifetime of exposure. MCLs are legalstandards which must be met by public drinking water systems.
RFC (Reference Concentration): An estimate developed by EPA of a concentration of asubstance in air which is unlikely to cause adverse non-cancer health effects following long term(lifetime) exposure.
RfD (Reference Dose): An estimate developed by EPA of a daily dose of a substance (amount ofsubstance per weight of exposed individual) that is unlikely to cause non-cancer adversenoncancer health effects following long term (lifetime) exposure.
RMEG (Reference Dose Media Evaluation Guide): The concentration of a chemical in wateror soil identified by EPA which is unlikely to result in adverse noncancer effects, following alifetime of exposure.
As Little Elk Creek is not used as a source of drinking water and no comparison values arecurrently available for evaluating contaminant levels in surface water used for recreationalactivities (e.g., swimming and wading), no comparison values are included in the discussion ofsurface water data.
The only on-site sampling data that could be located were results of groundwater sampling.
No record of on-site air monitoring could be located for the Spectron/Galaxy site. It would bereasonable to conclude, however, that the same contaminants identified in off-site air monitoringwere also present on site, likely at higher concentrations.
The site is covered by asphalt, below which is approximately 11 feet of soil and fill material(e.g., bricks, old foundations). Below this is a layer of weathered rock, and deeper still isfractured bedrock. The average depth to groundwater on the site is about 6 feet (10). Thedifferent zones of underground material mentioned above can all serve as aquifers because theyare porous and can store and transport groundwater. There are no confined aquifers under thesite because groundwater can move between the different underground layers (for example, fromsoil into weathered bedrock, and then into fractured bedrock).
Shallow groundwater under the site flows toward and discharges into Little Elk Creek. Contaminated groundwater seeps along the west bank of the creek, adjacent to the site, provideevidence of this. Contaminant data for these seeps are discussed in the section of this documentdescribing off-site contamination of surface water. It is also expected that groundwater in deeperrock layers would flow toward the creek; however, this deeper groundwater may flow in otherdirections (for example, parallel to the creek) (10).
The first seven groundwater monitoring wells were installed on site at the Spectron/Galaxyfacility in May 1980. These wells range in depth from nine to twenty-nine feet (1). Samplingfrom these wells in 1980 detected elevated levels of volatile and semi-volatile organiccompounds (1). A final report from this first sampling project is not available. These wells weresampled again in January 1981 (Table 1, Appendix B) by Ecology and Environment, Inc. Thechemicals 1,2-dichloroethane (120,000 ppb), 1,1,1-trichloroethane (80,000 ppb), toluene (42,000ppb), and methylene chloride (32,000 ppb) were detected at levels which greatly exceed theircomparison values for drinking water (2).
In August 1983, a groundwater treatment system was installed by Griffen Technics Corporationas a result of an EPA Settlement Decree signed in March 1982. The influent (contaminatedwater) and effluent (cleaned water) water streams were sampled each month. Although thesystem operated for four years at above 90% efficiency, and at 99% efficiency for somechemicals, it was abandoned in 1987 when it became obvious that the system was not reducingthe levels of contaminants found in the groundwater (1, 10). That is, the contaminantconcentrations in the influent remained relatively constant over the four year period. Maximumcontaminant levels in on-site groundwater samples taken in 1987 are presented in Table 1, Appendix B.
In 1991, six additional on-site monitoring wells were installed. Currently, there are twelveon-site monitoring wells. The most recent data, however, only include samples from six of thewells. Monitoring wells 3 and 10 have consistently shown the highest levels of contamination. These wells are located between the abandoned buildings and the creek, where the lagoon andsludge pits were believed to have been located. Monitoring well 2 was installed for backgroundlevel comparison purposes. This well was also found to be contaminated from the site; however,contaminant levels were hundreds and thousands of times lower than monitoring wells 3 and 10(13).
The samples taken from the on-site monitoring wells identify eighteen volatile organiccompounds, three semi-volatile organic compounds, and ten inorganic substances (metals)elevated above drinking water comparison values. Table 1 (Appendix B) shows the highestconcentrations of contaminants detected in on-site groundwater.
Surface Soils and Sediments
The entire surface of the site was paved in 1983 (1). No sampling data are available for surfacesoils or sediment contamination prior to capping. Considering the history of the site (leakingdrums, an open lagoon, sludge pits), as well as the fact that on-site groundwater is heavilycontaminated, it is reasonable to expect that the soil underlying the pavement is heavilycontaminated.
The Spectron/Galaxy site has been paved since 1983. There are reports that there were at leastfive springs on the facility grounds which were buried under the cap (10). EPA sampled thesprings prior to paving, and the analyses revealed that the spring water contained elevatedconcentrations of at least eight contaminants (10). The specific data from this sampling are not included in this report because they are not available.
The off-site media that have been sampled include air, groundwater, surface water, sediments,and fish tissue. Investigations are currently underway to better characterize the extent of off-sitegroundwater and surface water contamination.
Six separate air monitoring surveys have been conducted near the Spectron/Galaxy site, with theearliest survey conducted in 1970 and the most recent in August 1995. Comparing the results ofthese different surveys is made difficult because of factors such as differences in sampling andanalytical methodologies, and differences in sampling locations and weather conditions. Theability to conduct such surveys (both in terms of the number of different chemicals that can bemeasured and the levels at which they can be detected) has improved greatly during this timeperiod. Also, air monitoring gives a "snap shot" of conditions at the time of sample collection. Emissions from the plant during its period of operation would have varied over time because ofprocess changes at the facility, as well as differences in the make-up of the material that wasbeing processed or stored on site at the time air samples were taken.
The geography of the site area is important when considering the off-site transport ofcontaminants in air. At times, conditions in the valley exist that would tend to keep airbornecontaminants from being dispersed from the area. Inversions can occur in the valley (cool air atground level with warmer air above), and as a result, contaminants would tend to be trappedalong the valley floor (3). Also, chemical vapors can be denser than the surrounding air, resulting in a tendency for the vapors to settle along the valley floor.
The earliest air sampling conducted near the site was performed by Thiokol Corporation for theMaryland Department of Health and Mental Hygiene (DHMH) in May and June 1970. Thesampling was conducted off site over the course of 4 days (14). Samples were analyzed by gaschromatograph, using a flame ionization detector. The instrument was calibrated toquantitatively measure levels of acetone, methylene chloride, methyl ethyl ketone, benzene, andtoluene. A mass spectrometer was also used to identify other airborne contaminants. Duringeach day of the four sampling days, between eleven and fifteen samples were collected at twofixed sites (upwind and downwind of the site) and a roving station. A total of 51 samples werecollected over the four sampling days.
As can be seen in the first column of Table 2A (Appendix B), concentrations exceedingcomparison values were reported for seven compounds. The percent of the 51 samples collectedin which these compounds were detected is as follows: acetone (64%), methylene chloride(40%), methyl ethyl ketone (36%), benzene (28%), toluene (8.5%), methyl isobutyl ketone(10%), and tetrachloroethane (1.9%). The levels of methyl isobutyl ketone and tetrachloroethanewere estimated (i.e., they were detected but the precise concentration is not certain). It was alsoreported that three different unidentified compounds (each containing 2 carbon atoms) werereported in two thirds of the samples taken.
The investigators reported that the highest observed concentrations were those collected at lowelevations (several feet above the ground near the stream) by the roving sampler. They alsonoted that air concentrations tended to be highest at lower valley elevations in the evening, whenthe air was still. The sampling team recorded smelling both highly disagreeable, nauseatingodors, as well as fragrant odors.
Ambient air sampling was also conducted in 1970 by Dr. Pietro Capurro, a physician who livedin the valley about 1.5 miles south of the site. Using a gas chromatograph, Capurro reporteddetecting up to 140 parts per million (equivalent to approximately 895,000 g/m3) of carbon tetrachloride in the air near his home. He noted that when he moved to higher ground, 80 to 100 meters from the initial sample, no chemical could be detected (6).
A second round of off-site air monitoring was conducted by government agencies at varioustimes near the site during the period from August 1976 to January 1977. This sampling wasconducted to provide information to a task force that was convened to investigate allegations ofillness caused by site emissions (3). The investigation was conducted by the Maryland DHMHand the Cecil County Health Department, with assistance from the U.S. EnvironmentalProtection Agency. Continuous monitors were placed at two off-site locations, one (station 1)described as "relatively close to the plant" and the other location (station 2) as "a few hundredyards downstream" from the site. Another sampling station had been planned upstream from thesite but could not be established because of equipment malfunctions.
Samples were analyzed using a gas chromatograph with a flame ionization detector. Thedetection limit for all of the chemicals is not known; however, for benzene the detection limit islisted at 0.1 ppm (325 µg/m3). The instruments at sites 1 and 2 were calibrated to measure 20 compounds described as being typically associated with the plant operation (3). A list of thesesubstances could not be located. Air sampling was conducted at station 1 from August 26, 1976through September 14, 1976. Most of the contaminants detected were in the concentration rangeof 1 - 5 parts per million (ppm), with the exception of methyl isobutyl ketone, which wasidentified at a concentration of 63 ppm (15,000 µg/m3) (Table 2B, Appendix B). Air samplingwas conducted at station 2 on 31 days from September 29, 1976 through November 17, 1976,and again from December 21, 1976, through January 13, 1977. Similar contaminants weredetected during both sampling periods at concentrations ranging from less than 1 ppm toapproximately 5 ppm.
To be consistent with the comparison values, the concentrations of airborne chemicals listed inTable 2 are presented in the measurement units of micrograms chemical per cubic meter air(µg/m3). In converting air concentrations from ppm to µg/m3, the ppm value is multiplied by afactor unique to each chemical. Thus, chemicals detected at similar ppm concentrations mayhave very different µg/m3 concentrations.
Seven of the 20 contaminants monitored were detected (Table 2B, Appendix B). Comparisonvalues were exceeded for benzene, ethylbenzene, methyl ethyl ketone, tetrahydrofuran, andtoluene. It appears that air samples were not analyzed for methylene chloride, a contaminantwhich was detected on other occasions at relatively high concentrations. Some contaminantswere detected at station 2, but not at station 1, which is closer to the site. This likely occurredbecause the stations did not always operate on the same days, and emissions from the site variedon a day-to-day basis. It is possible that more contaminants were present in the air at this timebut could not be measured because of the limited capability of the instruments to detectchemicals present in low concentrations.
Air monitoring was again conducted during the period from July 31, 1984, to August 23, 1984,by the Maryland DHMH mobile air monitoring laboratory (15). The laboratory was stationedsouth of the site on Kirk Rd., about 100 yards from the intersection with Little Elk Creek Road. Samples were analyzed by two gas chromatographs, with the actual sampling time determined bythe operator. The system was calibrated to measure the concentrations of 29 differentcompounds. Samples were initially collected for 20 minute periods; however, this waseventually reduced to 1 minute (17 times per day) on August 8th, because it was found that thecapillary columns were being overloaded by methylene chloride. This reduced the sensitivity ofthe procedure to the "low parts per billion for chlorinated compounds," and 10-200 ppb fornonchlorinated compounds. Concentrations of nonchlorinated compounds were not reportedafter August 8th.
Concentrations of 12 different VOCs were measured during July - August 1984 (Table 2C). Table 2C lists five of the contaminants measured by the mobile laboratory during this period. The table shows 24-hour average concentrations (based on short term samples) and the minimumand maximum 24-hour averages. Two of the chemicals reported in both the 1976/77 and 1984monitoring, benzene and toluene, showed a considerable decline in average concentration overthe seven year period. It can be seen from Table 2C that overall average concentrations formethylene chloride, benzene, and tetrachloroethylene exceeded comparison values, with 24-houraverage concentrations exceeding comparison values for toluene and 1,1,1-trichloroethane.
A fourth off-site air monitoring survey was conducted near the site by Maryland Department ofthe Environment's Air Management Administration (formerly with DHMH) from March 1987through November 1989. Samples were collected over approximately 24-hour periods from theexterior of 3 residences near the site (16). Sample frequency was changed from "almost daily" toapproximately weekly in late April 1987. The residences were to the north, northeast and southof the site, with the closest residence approximately 250 feet from the site, and the farthest about1,500 feet away (personal communication with Walter Cooney, MDE Air ManagementAdministration). Air samples were analyzed for 24 different VOCs, including both chlorinatedand nonchlorinated compounds. Annual averages were calculated for a total of 19 differentVOCs.
Table 2D shows the range of annual average concentrations of 8 different contaminants for eachof the three sites for 1987, 1988, and 1989. These contaminants were listed because theircomparison values were exceeded (no comparison value exists for 1,1,1-trichloroethane). Forcomparison purposes, Table 2D includes annual average concentrations for 1993 for six of thecontaminants at a MDE air monitoring station in Baltimore City (Fort McHenry). Concentrationsof airborne contaminants were generally highest for residence II, located about 500 feet south of the site.
At the request of MDE, a fifth round of off-site air sampling was conducted by a contractor forthe PRP group in November 1993. Samples were collected using charcoal tube samplers placedat established stations around the perimeter of the site. Although some VOCs were detected inthe sampled air, the results were considered unreliable because of problems with the samplingmethodology. Specifically, there was evidence that chemicals had "broken through" some of thesampling tubes, thus invalidating the results for these contaminants.
The most recent off-site air sampling was conducted by a contractor under EPA supervisionduring the week of August 7th, 1995, at the request of MDE and ATSDR (41). Five samplingstations were established on private property around the site, including two of the three locationswhere samples were collected during 1987-1989. Air samples were collected over a three dayperiod using both canister and charcoal tube samplers. Samples were analyzed for a total of 27different VOCs. During this sampling, no contaminants were detected at levels exceedingcomparison values.
A residential well survey conducted at the time wells were sampled in 1992-1993 indicated that the well depths ranged from 20 to 250 feet and that several of the wells had been hand dug. Some residents obtained their drinking water from springs located on their property. No springs were sampled. If a well had a treatment system, an attempt was made to obtain the samples before treatment.
Residential well sampling conducted in March 1987 and August 1989 by MDE did not show anylevels of contaminants above Maximum Contaminant Levels (MCLs) or Cancer Risk EvaluationGuides (CREGs) for drinking water (18). MCLs are maximum allowable levels of chemicalcontaminants in public drinking water supplies. A CREG is a concentration of a contaminant inwater below which it is expected that the increased risk of cancer from lifetime exposure to thatcontaminant is less than one-in-one million (generally considered an insignificant risk level).
MDE and EPA sampled twenty residential wells again in November 1992 (Appendix B, Table3). Tetrachloroethylene (PCE) contamination was detected in eleven of the residential wellssampled. According to a March 1993 EPA report which summarized the November 1992 wellsampling data (19), five wells exceeded the MCL for PCE of 5 parts per billion (ppb, equal toµg/L or micrograms per liter). The levels which exceeded the MCLs ranged from 6 ppb to 14ppb. For three of the samples which exceeded the MCL, the reported concentrations wereestimates, i.e., the chemical was detected but the precise concentration is not certain. Anotherchemical, 1,1-dichloroethene was detected in two wells at 2 and 5 ppb, both of which wereestimated values, and below the MCL for 1,1-dichloroethene of 7 ppb, but above the ATSDRcomparison value of 0.06 ppb. Some samples taken from the same residential wells were splitwith MDE. Split samples are samples that are divided and analyzed by different laboratories forcomparison purposes. In the samples going to the second lab, no contaminants were detectedabove the analytical detection limit, which was 5 ppb. Because of this discrepancy, it isuncertain whether or not the contaminants were actually present in the sampled wells. It ispossible for low level contamination to occur during the collection and analysis of samples. As a precautionary measure, five residents whose well samples showed possible contamination atlevels above the legal drinking water standard (MCL) were supplied with bottled water.
Eight of the residential wells initially sampled in November 1992 were resampled in February1993, and the samples were split with MDE (Appendix B, Table 3). The results of both labs didnot reveal any contaminants at levels above detection limits, which were 5 ppb (19). Residentialwell sampling conducted in November 1993 showed no contamination in any samples, and thesupply of bottled water to area residents was discontinued.
Twenty-one residential wells located in the vicinity of the Spectron site were sampled inSeptember 1995 by a contractor to the responsible party group (24). Included in this samplingwere wells in which contaminants were detected in 1992. Five VOCs were detected in one wellat levels well below comparison values (Table 3). As a precaution, the PRP group has arrangedfor a carbon filtration unit to be attached to this well. Starting with the September 1995sampling, these wells will be monitored at a frequency of approximately every six months.
Sediments and Soils
To date no off-site sediment or soil sampling has been conducted. There is no reason to expectany significant contamination of off-site soils because the asphalt cap is expected to inhibitsurface soil migration from the site. In May 1993 a total of nine sediment samples were collectedfrom Little Elk Creek, immediately adjacent to the site (21). The samples were collected with asampling tube at depths ranging from 0.5 feet to 4 feet. The sediments were analyzed in the fieldfor the VOCs methylene chloride, 1,1,1-trichloroethane, and tetrachloroethylene. Thesecontaminants were detected in all of the samples in maximum estimated concentrations of 429,280, and 159 parts per million (ppm), respectively.
The Spectron/Galaxy Site is bordered by Little Elk Creek, a tributary to the Elk River. Little ElkCreek flows approximately 12.5 miles from north of the Maryland-Pennsylvania line to itsconfluence with the Elk River below Elkton, Maryland. Little Elk Creek drains a watershed ofapproximately 42 square miles. Approximately 29.7 square miles of the watershed is inMaryland. The average annual flow is approximately 32 cfs (cubic feet of water per second) withminimum and maximum flows of 3.3 cfs and 5,400 cfs. The width of the stream in the vicinityof the Galaxy/Spectron site ranges from 10 to 30 feet (22).
Surface water sampling was performed by MDE in 1979 and 1980 and showed extensivecontamination of Little Elk Creek by volatile organic compounds (22). Water samples weretaken about 30 yards upstream of the plant (reference station), and approximately 100 yards, 500yards, and 3 miles downstream of the site. A total of 10 organic compounds were detected in thesamples, including methylene chloride, 1,1,1-trichloroethane, and 1,2-dichloroethane atmaximum concentrations of 361 ppb, 158 ppb, and 115 ppb, respectively (Table 4, Appendix B). Benzene was also detected at a maximum concentration of 4 ppb. The maximum concentrationswere detected in the sample from the closest station, approximately 100 yards downstream of theplant (station 3). No contaminants were detected in the reference water sample collectedupstream of the site.
In November 1989, samples from the Little Elk Creek and groundwater seeping out of the creekbanks adjacent to the site were analyzed to determine the possible migration of contaminationfrom groundwater at the site into the creek (23). The creek samples were taken a short distancedownstream from the site, just downstream from the Providence Road bridge. The resultsshowed a number of chlorinated organic compounds, including methylene chloride,1,1,1-trichloroethane, and 1,2-dichloroethane at 548 ppb, 100 ppb, and 52 ppb, respectively (ppb= parts per billion, equivalent to µg/L). No contaminants were detected in creek samples upstreamof the site (see Table 5, Appendix B). Seep samples (samples of water seeping into creek fromcreek banks) showed maximum levels of methylene chloride, 1,1,1-trichloroethane, and1,2-dichloroethane at 10,800 ppm, 605 ppm, and 418 ppm, (ppm = parts per million, 1 ppm =1,000 ppb) respectively.
Both Little Elk Creek and groundwater seeps were again sampled in December 1991 (24). Theorganic contaminants detected in 1991, as well as their concentrations, were similar to the resultsfrom 1989 sampling (Tables 5 and 6, Appendix B). Metals were also detected in creek samples atlevels consistent with levels found in other Maryland surface waters, and below levels whichmight pose a threat to people who contact the creek water during recreational activities.
Surface water from Little Elk Creek was most recently sampled in September 1995 by acontractor for the PRP group. Samples were collected under the supervision of the EPA fromfour stream locations, including an upstream location and three downstream locations, locatedapproximately 100, 300, and 2,000 feet downstream from the Providence Road bridge (adjacentto the site) (25). A total of 21 VOCs were detected in the sample collected 100 feet downstreamfrom the bridge, in a total concentration of approximately 366 ppb, with similar findings for thesample collected 300 feet downstream. The total VOC concentration in the sample collectedfarthest downstream was 55 ppb, with 15 VOCs identified. Starting with the September 1995sampling, surface water from the creek will be monitored approximately every six months.
The 1991/1992 sampling identified dense non-aqueous phase liquid (DNAPL), a pocket ofchemicals that are heavier than water, under the creek. DNAPL samples were taken fromunderneath Little Elk Creek in June 1992. The majority (52%) of each sample was comprised ofthe VOCs methylene chloride, 1,1,1-trichloroethane, tetrachloroethylene, and1,1,2-trichloro-1,2-trifluoroethane (Freon) (21). This was the first finding of Freon among thesite contaminants.
The groundwater just below the creek bed has been sampled on three occasions usingpiezometers. Piezometers are small diameter steel pipes driven 2 to 4 feet into the stream bed in order to evaluate the contribution of groundwater to the stream. Six samples were collected in 1991, twelve in 1992, and nine in 1993 (see Table 7, Appendix B).
In 1979 and 1981 the Maryland DNR and U.S.EPA cooperated in sampling fish from Little ElkCreek, both downstream and 1/2 mile upstream from the Spectron site (more details on samplingsites could not be located) (22). Site-related VOCs were detected in the fish collected from thedownstream location on both occasions. The fish sampled in 1979 included white sucker, greensunfish and cutlips minnow, and in 1981 included white sucker, eel, dace and sculpin. A total ofthree different VOCs were identified in the fish sampled in 1979. In addition, eight other VOCswere identified in the 1981 fish samples. The 11 VOCs detected were: carbon tetrachloride,chlorobenzene, 1,2-dichloroethane, 1,1-dichloroethane, ethylbenzene, methylene chloride,toluene, 1,1,2,2-tetrachloroethane, tetrachloroethylene, trichloroethylene, and1,1,1-trichloroethane. No VOCs were detected in the fish collected upstream of the Spectronsite.
To identify active facilities that release chemicals into the environment near the Spectron/Galaxysite, we searched the 1987, 1988, and 1989 Toxic Chemical Release Inventory (TRI). TRI is adata base developed by EPA from the chemical release (to air, water, soil) information providedby certain manufacturing industries. A federal law requires that facilities of a certain size report on the releases of over 300 listed chemicals and chemical categories.
No facilities listed in the TRI data base for Cecil County would be expected to have an impact onthe site area. The closest TRI facilities are located in Elkton, six to seven miles from the site.
Much of the analysis done prior to the pre-design study was done by the Maryland HazardousWaste Laboratory which follows strict guidelines for quality assurance and quality control(QA/QC). Some data were collected that did not follow the same QA/QC guidelines that existtoday. Those data are nonetheless valuable in that they identify the chemicals that were present in the sampled media and give some indication of the levels that existed at the time of sampling.
The data collected during the pre-design study, conducted in 1991-1992, underwent strictscrutiny for QA/QC requirements. Among the parameters verified were: holding times, matrixspike recoveries, surrogate compound, instrument calibration, and duplicate sampling recoveries.
For the Focused Remedial Investigation (FRI) in 1993, most of the samples taken were analyzedby field screening methods since the purpose of the FRI is to characterize the DNAPL (densenon-aqueous phase liquids) and not to determine the exact concentrations of contaminants at thesite.
During the August 1993 site visit, holes in the fence were visible which would allow trespassersto gain access to the site. Beer bottles and graffiti are evidence that the site has been accessed. Many old chemical storage tanks, where the chemicals were stored and distilled, remain on-site,as well as several abandoned vehicles. The storage tanks were cut open during the removalaction which provides access to the interior of the tanks. This equipment could potentially pose aphysical hazard to trespassers who could injure themselves climbing and playing in the tanks. Efforts have been made to restrict access to the site, and the area is well marked with warningsigns as a hazard.
In a May 1995 site visit, it was noted that the holes in the perimeter fence had been repaired andthe main gate had been reinforced to make it more difficult to force open.
Exposure pathways are analyzed in order to determine the means by which individuals (primarilylocal residents) may be exposed to site contaminants, either currently or in the past. Possiblefuture exposure pathways can also be identified. An exposure pathway consists of five elements: a source of contamination; transport through an environmental medium; a point of exposure; aroute of human exposure; and an exposed population.
Exposure pathways can be described as completed, potential, or eliminated. In a completedexposure pathway, the five elements exist and exposure to a contaminant has occurred in the pastor is occurring. In a "potential" exposure pathway, at least one of the five elements is missing. Itis possible that the missing element exists (e.g., an exposed population) but has not beenidentified. Potential pathways indicate that exposure to a contaminant could have occurred in thepast, could be occurring now, or could occur in the future.
Ambient Air Pathway
During the air monitoring that was conducted near the Spectron/Galaxy site in August, 1995, noairborne contaminants were detected above "background" levels (i.e., levels of these VOCs thatare commonly found in residential areas). This recent sampling is likely indicative of conditions that have existed at the site since EPA completed an emergency removal of chemical wastes from the site in March 1990.
Air monitoring that was conducted previous to the removal action, from 1970 through 1989,detected a total of 13 different VOCs in concentrations which exceeded comparison values (26). The Cecil County Health Department and DHMH Air Management Administration received oversix hundred odor complaints from the Providence Valley area during the period of 1961 through1976 (3).
The valley residents were exposed to high levels of volatile contaminants in the air during muchof the 1960s and 1970s. Although no air monitoring was conducted in the 1960s, samplingconducted in 1970 is likely indicative of conditions in earlier years. Workers were likelyexposed to the highest levels of VOCs in the air; however, local residents would have beenexposed for longer periods of time. The greatest exposure would be expected to have occurredduring the period of operation of the open lagoons. These lagoons were discontinued in 1970;however, several smaller sludge pits were reported to have been in use for some time after 1970.
There are 95 people living within 0.25 miles of the site. Those who live closest to the site,approximately 20 people, would have likely inhaled greater concentrations of the contaminantsthan those living further away. However, because chemical vapors have been observed toconcentrate along the valley floor (instead of being quickly dispersed), it is likely that peopleliving in the valley a considerable distance from the site could also have been exposed to highlevels of contaminants. Dr. Pietro Capurro, a former pathologist at the Union Memorial Hospitalin Elkton, Maryland, reportedly identified carbon tetrachloride in the blood of three valleyresidents who were hospitalized with similar symptoms in 1970. He also reported identifyingsimilar chemicals in the air and in the blood of three people living near the site in November1970 (6).
People who worked on the site would have likely been consistently exposed to elevated levels ofvolatile organic compounds. From 1961 to 1977, 166 people are reported to have worked at thesite. It was known that of the 166 employees who had ever worked at Galaxy Chemical duringthis period, 131 or 78.9% had been employed for a period of one year or less (3). During stagnant air conditions people living a considerable distance from the site in the valley area could have been exposed to the contaminants through inhalation.
As previously noted, on-site groundwater monitoring analyses indicate that the on-sitegroundwater is highly contaminated with volatile and semi-volatile organic compounds andinorganic compounds. The most recent residential well sampling (September 1995) resulted inthe detection of 5 VOCs in one well in concentrations that would not pose a threat to residentsusing the water (i.e., levels were below comparison values).
Residential well sampling conducted in November 1992 resulted in the possible detection of theVOC tetrachloroethylene in 11 wells and 1,1-dichloroethene in two wells (see previousdiscussion of residential wells under Off-site Contamination). No contaminants were detected inany of these wells when they were resampled on two occasions in 1993. Therefore, the supplyingof bottled water to residences where some sampling rounds had indicated the presence ofcontaminants at levels above MCLs, has been discontinued.
The studies available for review did not discuss the potential for residential well contaminationlevels fluctuating over time. There is, however, continued concern over the potential for futurecontamination of residential wells from site chemicals. There are no continuous confining layers(e.g., clay) to prevent the movement of contaminants into the bedrock aquifer (1). Contaminantsin groundwater under the Spectron/Galaxy site will tend to follow the topography of the valleyand flow towards the adjacent Little Elk Creek. Underground pockets of pure chemicalcontaminants may move in different directions than the contaminated groundwater as they movedownward in the aquifer.
Individuals using contaminated well water can be exposed to contaminants through directingestion of the water and through inhalation of volatile chemicals and dermal contact by usingthe water for bathing, cooking, and other domestic activities.
COMPLETED EXPOSURE PATHWAYS
|POINT OF EXPOSURE||ROUTE OF|
|Private Wells||Groundwater||Residences||Ingestion, inhalation,skin contact||Users of waterfrom privatewells||Past|
|AmbientAir||Air||On-site and nearby|
workers, local residents
Exposure During On-Site and Off-Site Work and Remediation
Workers could potentially be exposed to site contaminants while engaged in site sampling andsite remediation activities. Workers could be exposed by directly inhaling airbornecontaminants, through direct contact with contaminated media, or by incidentally ingestingcontaminated soil.
Surface Soils and Sediments
The entire surface of the Spectron/Galaxy site was paved in 1983, and prior to that, no soilsampling was done (1). It is reasonable to assume that the soil beneath the pavement is heavilycontaminated with volatile and semi-volatile organic compounds and inorganic compounds. Thepavement is cracked in many places and could potentially expose remedial workers or trespassersto the contaminants beneath the pavement by incidental ingestion or inhalation.
It is probable that people would have been exposed to contaminated surface soils and sediments before the site was paved. Site trespassers and workers at the site would have been at greatest risk for exposure by this pathway.
No off-site soil or sediment sampling has been completed. It is reasonable to assume that thestream sediment and soil along the bank of the stream near the site is likely contaminated. Contact with contaminated soils and sediment along this section of the stream represents apotential source of exposure to local residents, especially children who play in and along thestream.
Surface water samples from Little Elk Creek contained elevated levels of volatile organiccompounds (VOCs), with the highest levels found in samples from leachate seeps taken from thebanks of Little Elk Creek near the site. Individuals could potentially be exposed to off-sitesurface water contaminants through direct contact. Children can swallow creek water whileswimming and can also inhale volatile chemicals. EPA and MDE have posted signs whichrecommend no fishing or swimming in the creek.
One local resident reported that children occasionally fish near the site and another reported thatchildren swim about a mile downstream of the site. It is possible that exposures to sitecontaminants occur during these activities. It is not known whether elevated levels of chemicalspersist one mile downstream of the site. This potential pathway is of concern because access to the stream is not restricted.
Fish tissue data are only available from a 1979/1980 collection. As has been observed with othermedia, it would be expected that tissue levels have declined since then. The potential forexposure to site contaminants via fish ingestion would exist only with the consumption of fishcollected immediately downstream of the site. There is no evidence that edible game fish arecaught from the affected section of Little Elk Creek. A local resident reported that childrenoccasionally catch and release fish from the creek near the site.
|PATHWAY||ENVIRONMENTALMEDIA||POINT OF EXPOSURE||ROUTE OF EXPOSURE||POTENTIALLY EXPOSED POPULATION|
|Adjacent to sitein Little ElkCreek||Ingestion, skincontact||Local residents especially children||Past|
|Soil||Soil||banks of LittleElk Creek whereseeps enter creek||ingestion, skincontact||local residents||Past|
|Surface Water||Water||Off-site in LittleElk Creek||Ingestion, skin contact||Workers, children||Past|
|Users of Waterfrom private wells||Future|
In this section of the health assessment the major contaminants of concern associated with theSpectron/Galaxy site are discussed. For each of these contaminants there is a discussion of thepotential for illness based on assumed exposure to the levels which have been measured in the airnear the site. This discussion is based on contaminant levels in air because this is the pathway(inhalation) through which people are known to have been exposed to relatively high levels ofcontaminants for an extended period of time. It is difficult to estimate the amount of a particularcontaminant to which people living near the site may have been exposed over the years. Levelsof contaminants in the air varied both on a short term (between and within days) and long term(year-to-year) basis. Air sampling data, odor complaints, and site history, indicate that off-site airconcentrations were the highest from the mid-1960s through the early 1970s. This is the periodduring which open lagoons of chemical waste were found on the site, and additionalcontaminants were emitted directly from the facility. During this period it has been reported thatpeople were hospitalized as a result of exposure to airborne contaminants from the site (5,6).
Estimating to what people living near the site may have been exposed is also made more difficultbecause there are long periods of time during which no air measurements were made. Forexample, no air sampling was conducted near the site from 1970 to 1976. Also, not all potentialcontaminants were measured in the air each time samples were taken. This can be seen with thechemical methylene chloride, which was detected at elevated levels in all but one of the roundsof air monitoring which were conducted over the years, starting in 1970. Off-site air monitoringconducted in 1976/1977 did not include analysis for methylene chloride, resulting in a data gapwith respect to levels of this contaminant to which local residents may have been exposed from1970 to 1984.
Estimating Health Risks From Chemical Exposures
The process of estimating the health risk associated with exposure to a chemical contaminantinvolves two major steps: 1) estimate the amount of chemical to which a person is or wasexposed and the length of exposure; and 2) compare the exposure level with a level for thatchemical which would not be expected to cause illness. In general, people can be exposed to ahigher concentration of a chemical for a short period of time (referred to as an acute exposure)without adverse effects as compared to a longer time period (referred to as a chronic exposure).
Exposure to a contaminant can be estimated by directly using concentrations of the substancewhich have been measured in air or water or by converting these levels to an internal "dose." Adose is the amount of a chemical which is taken into the body through ingestion, inhalation, orabsorption through the skin. Chemical doses are commonly expressed in terms of milligrams ofchemical per kilogram of body weight per day (mg/kg/day). Dose estimates are needed in orderto estimate health risks from chemical exposure.
This process of estimating health risk from contaminant exposure is complicated at the Spectronsite by the fact that people have been exposed to a mixture of different chemicals. Exposure to agiven chemical can sometimes result in an increase or decrease in the toxicity (the ability of achemical to cause adverse health effects) of a second chemical. This includes chemicalexposures which are related to personal habits such as smoking or drinking alcoholic beverages. A person's diet can also be important in determining the degree to which the individual would beaffected by exposure to some chemicals.
Another important consideration when estimating health risks from chemical exposure is thatpeople differ in their susceptibility (sensitivity) to the adverse effects of chemicals. For example,children might be more susceptible to airborne chemicals because they tend to spend more timeoutdoors and have faster rates of respiration (breathing) as compared to adults. Also, people withpre-existing health conditions such as heart disease and asthma can be more susceptible tochemical exposures.
Health Effects From Exposure to Chemical Solvents
Most of the chemical contaminants found in the air near the Spectron site are volatile (evaporatequickly at room temperature) chemical solvents which are commonly used in industry. Some ofthe toxic effects from solvents are common among the different chemicals within this groupwhereas other effects are unique to individual chemicals. Most of the chemicals in this group candepress the central nervous system, causing symptoms which are similar to alcohol intoxication. These symptoms include the following: fatigue, incoordination, confusion/disorientation,irritability, and disturbances of vision and hearing (27). Sufficient exposure to many of thesesolvents can also cause damage to the liver, and some have also shown the ability to damage thekidneys. Some chemicals in this group have also been shown to cause cancer in humans(benzene, vinyl chloride), or are considered probable human carcinogens based on animal testing(e.g., methylene chloride, 1,2-dichloroethane, trichloroethene). Other chemicals in this group cancause unique forms of toxicity. For example, benzene can cause damage to the blood-formingtissues of the body (28).
Drinking ethanol (the alcohol in alcoholic beverages) can increase the toxic effects of somechemical solvents on the liver. For example, it has been shown through animal studies thatethanol exposure increased the toxic effects of the chemical solvents carbon tetrachloride,1,2-dichloroethene, trichloroethene, and trichloroethane (both 1,1,1- and 1,1,2-) on the liver (27, 29). Also, exposure to some chemical solvents can increase the toxic effects caused by exposureto other chemicals in this group. An example of this is exposure of laboratory animals to acetone increasing the toxic effects of subsequent exposure to carbon tetrachloride, trichloroethene, and 1,1,1-trichloroethane on the liver (29).
Illness Reported in Residents of Providence Valley
Dr. Pietro Capurro, a physician who lived about 1.5 miles south of the site, was the author ofthree papers in which he described cases of illness in valley residents which he suggested weredue to exposure to chemicals from Spectron/Galaxy plant (5,6,7).
The cases were observed in 1969 and 1970. The most common symptoms reported by 43 valleyresidents to Dr. Capurro in 1969 included fatigue, irritability, indecisiveness, headaches, andburning eyes and throats (5). He also reported what he interpreted as symptoms of damage to theliver and pancreas in some valley residents admitted to the hospital during this time period. Other symptoms he reported in cases admitted in 1970 included nausea, vomiting, and abdominalpain (6). Dr. Capurro also reported what he believed to have been an elevated number of cancercases, particularly malignant lymphoma (a type of cancer of the lymphatic system) in thepopulation of Providence Valley (7).
This section provides a brief description of the potential health effects associated with exposureto the major volatile organic chemical contaminants which have been repeatedly measured in theair near the Spectron site at levels of public health concern. For each chemical, a description isprovided of the health effects which might be expected in people exposed to levels of thechemical measured in off-site air. For some contaminants, levels measured in the air arecompared to Minimal Risk Levels (MRLs) which have been developed by the ATSDR. A MRLis an estimate of the level of a chemical to which a person could be exposed for a given period oftime without suffering adverse (noncancer) health effects. MRLs can be developed forshort-term (acute) exposures (<14 days), intermediate length exposures (15-364 days), orlong-term (chronic exposures) (< 365 days).
Not all of the substances listed in Appendix B, Tables 1 through 7 are discussed. An emphasishas been placed on contaminants that have been found at consistently high concentrations whichexceed comparison values in several different environmental media. Other contaminants whichhave been detected in the air near the Spectron site, but which are not discussed in this section,include acetone, butanol, carbon tetrachloride, chlorobenzene, chloroform, 1,2-dichloroethene,1,2-dichloropropane, ethylbenzene, ethyl ether, formaldehyde, hexane, isoamyl alcohol, isopropylalcohol, methyl isobutyl ketone, pentane, tetrahydrofuran, and xylene (6, 16). Relatively highlevels of carbon tetrachloride (well above comparison values) were measured in air within thevalley by Dr. Capurro in 1970. Air monitoring conducted since 1984 has not detected carbontetrachloride. Carbon tetrachloride can cause damage to the liver and kidneys and is classified by EPA as a probable human carcinogen (30).
Benzene is a naturally occurring substance produced by volcanoes and forest fires. It is alsopresent naturally in small amounts in many plants and animals (28). The primary use of benzeneis as an industrial chemical. Benzene is used to make other chemicals and products, such aspesticides and some plastics, and is a component of gasoline. Benzene has an aromatic odor andcan be smelled in the air at about 5 ppm (five parts benzene per one million parts of air). Peoplecan be exposed to benzene at low levels in the home and at work through common environmentalsources such as tobacco smoke and automobile exhaust. Short term exposure to high levels ofbenzene in the air can result in central nervous system depression characterized by dizziness,incoordination, headaches, and vertigo (28). If exposure to sufficiently high levels of airbornebenzene occurs for an extended period of time, other adverse health effects such as changes inthe immune system, spontaneous abortion, menstrual disturbances, ovarian decay, and permanentliver damage may be experienced (28). Benzene exposure can also adversely affect the production of blood cells by the bone marrow, resulting in a reduction in the number of red and white blood cells.
Long term exposure to benzene has also been associated with leukemia in humans. Leukemia isa cancer of the tissues that form the white blood cells. Long-term exposure (from five to thirtyyears) to sufficiently high levels of benzene in the work place has been found to increase the riskof leukemia in exposed workers (28). Because of this finding, the Environmental Protection Agency (EPA), the National Toxicology Program (NTP) and the International Agency for Research on Cancer (IARC) have categorized benzene as a known human carcinogen.
Benzene was detected at a maximum concentration of 73,370 µg/m3 (23,000 ppb) in 1970. Although this sample was taken in an area where extended human exposure was unlikely (nearthe bank of Little Elk Creek on a still summer night), half of this concentration would pose asignificant increase in the risk of adverse health effects, including cancer, to people exposed forsufficient periods of time. The levels of benzene in the air decreased significantly between 1970and 1987. During the 1976/1977 air monitoring, benzene was detected in about ten percent ofthe samples at a maximum concentration of 2,958 µg/m3 (Table 2B). Extended exposure tobenzene at this level would be of concern because of cancer risk, but because benzene wasdetected in only ten percent of the samples, the average level of benzene to which people wouldhave been exposed during this period would be considerably less than this maximum value. In1984, the mean 24-hour average benzene level was 7.3 µg/m3. During the 1987-1989 sampling(Table 2D), the average annual level detected at the 3 off-site sampling stations was 6.2 µg/m3(Table 2D). Exposure to benzene at the levels detected in 1984 and 1987-1989 would not be expected to cause illness. The highest level of benzene detected in the 1987 sampling round is similar to the levels found in urban areas (28).
Benzene was also detected in off-site seep samples and on-site monitoring wells. Currentexposure to benzene in seep samples would be expected to occur infrequently; consequently thisis not considered a significant public health hazard. Benzene has not been detected in off-site residential wells.
1,2-Dichloroethane (1,2-DCA) is a man made industrial solvent used in the production of manychemicals. In the past, it was used in consumer products such as cleaning solutions, pesticides,and adhesives (31). 1,2-DCA in soil or water will evaporate quickly into the air. It has apleasant odor and can be smelled at about 6 ppm (parts per million). Exposure to sufficientamounts of 1,2-DCA has been shown to cause toxic effects in humans and laboratory animals,including cancer in laboratory animals. Short, high level exposure to 1,2-DCA can cause centralnervous system depression (similar to alcohol intoxication). Additionally, short term, high levelexposure to 1,2-DCA can adversely affect function of the liver, kidneys, brain, and lungs (31). Relatively little is known about the effects of long-term exposure to 1,2-DCA in humans. Laboratory animals exposed to high concentrations of 1,2-DCA in air over a long period of timedeveloped kidney and liver damage. There was also a study in rats which showed a largeincrease in death of offspring as a result of 1,2-DCA exposure during pregnancy. Based onextensive liver damage in experimental animals exposed to 1,2-DCA in air, a chronic MRL of0.2 ppm (824 µg/m3) has been developed by ATSDR (31).
1,2-DCA has also been shown to cause genetic changes in exposed animals. Mice breathing highlevels of 1,2-DCA for a short time experienced irreversible breaks in DNA strands (geneticmaterial) in their liver cells. This is significant because damage to DNA can result in healtheffects such as cancer and birth defects. Although it has not been conclusively shown in exposedhuman populations, lifetime oral exposure of laboratory animals (rats and mice) to relatively highconcentrations of 1,2-DCA caused cancer. Because of this, EPA, NTP and IARC havecategorized 1,2-DCA as a probable human carcinogen (Group B2) (31).
1,2-DCA was detected during 1987-1989 at a maximum annual concentration in off-site air of4.1 µg/m3 (1 ppb). The highest level detected in the two years following the initial sampling was0.8 µg/m3. No illness is expected to have been caused in people exposed to this level of1,2-DCA. Additionally, the cancer risk of exposed populations would not have beensignificantly increased assuming inhalation of 1,2-DCA at a level of 4.1 µg/m3 over a long periodof time. Air samples taken previous to 1987 were not analyzed for 1,2-DCA, but it is likely thatambient air levels were higher than the levels detected in 1987-1989. It is uncertain if prior to1987, surrounding residents were exposed to levels of 1,2-DCA which would be expected tocause illness.
In addition to ambient air contamination, 1,2-DCA has been detected in off-site seeps, off-sitesurface water, on-site groundwater, and off-site groundwater. Because of infrequent exposure tothese sources, they are not considered to represent significant public health hazards. 1,2-DCAhas not been detected in any of the sampled residential wells to date.
1,1-dichloroethene (1,1-DCE) is a man made industrial chemical which is used primarily as anadditive to plastics to make them flexible. It is also a breakdown product of other chemicals. 1,1-DCE will evaporate quickly into the air. It has a mildly sweet odor and can be smelled in theair at about 500 ppm (parts per million) (32). Humans occupationally exposed to highconcentrations of 1,1-DCE in the air have experienced toxic effects including liver damage anddepression of the central nervous system (i.e., signs of inebriation). The human health effects oflong-term exposure to lower levels of 1,1-DCE are not known.
Exposure of laboratory animals to high concentrations of airborne DCE has resulted in adverseeffects on the central nervous system, kidney, lungs, and especially the liver. Guinea pigs whobreathed high concentrations of 1,1-DCE continuously for 90 days developed liver damage. Based on these laboratory findings an intermediate (up to one year) inhalation MRL (MinimalRisk Level) of 0.02 ppm (81 µg/m3) has been developed by ATSDR. In addition to inhalation,oral exposure (e.g., drinking water) of laboratory animals to 1,1-DCE has been observed toprimarily affect the liver and kidney, with some evidence of adverse effects on the respiratoryand gastrointestinal systems (32). Soft tissue and skeletal defects have also been observed insome laboratory animals exposed to 1,1-DCE before birth (32).
There is no conclusive evidence that 1,1-DCE can cause cancer in humans. 1,1-DCE has beenshown to cause cancer in one experimental animal study. A large percentage of exposed micewho breathed 1,1-DCE for a long period of time developed lung cancer. A number of otheranimal studies did not show evidence of 1,1-DCE causing cancer. Based on this animal study,EPA has categorized 1,1-DCE as a possible human carcinogen (Group C). Other groups such asIARC and NTP have not been able to categorize 1,1-DCE as to its carcinogenicity because of theinconclusive evidence of its cancer causing ability in animals.
1,1-DCE was detected in the ambient air at a maximum annual concentration of approximately22 µg/m3 between the years of 1987-1989. This level would not be expected to cause illness in exposed people. Additionally, the residents would not have experienced a significant increase in their risk of cancer from exposure to these levels. 1,1-DCE levels were not reported for air samples collected before 1987, but it is likely that 1,1-DCE levels were higher before 1987. It is uncertain if prior to 1987, the residents surrounding the site were exposed to levels which would have been likely to cause illness.
Although, 1,1-dichloroethene was tentatively identified in one residential well sample in 1993, it was not detected in subsequent samples.
Methylene chloride, also known as dichloromethane, is a man made chemical that is widely usedas an industrial solvent and as a paint stripper. It is also found in some aerosol and pesticideproducts and has been used to extract caffeine from coffee and tea. It has a sweet odor which isnoticeable only at high concentrations in the air (over 100 ppm). Methylene chloride ismanufactured in the United States, but production has decreased upon the discovery thatlong-term exposure caused cancer in laboratory animals. Methylene chloride can cause adversehealth effects in people if it is breathed at high levels in the air. Exposure to high levels ofmethylene chloride in the air (above about 500 parts methylene chloride per one million parts air,or ppm) can irritate the eyes, nose, and throat (33). Acute (short-term) and chronic (long-term)exposure to methylene chloride in the air has also been shown to adversely affect the nervoussystem. People exposed to 300 ppm of methylene chloride in air for a short period of timedeveloped problems with their vision and hearing, which cleared up following exposure.
Studies in laboratory animals have shown that frequent or lengthy exposures to methylenechloride can affect the liver and kidneys. The results of these studies and those of exposedworkers, suggest that it is unlikely that methylene chloride would cause serious liver or kidneydamage in humans unless exposure is very high (33). Based on the neurological effects onhumans, ATSDR has developed a minimal risk level (MRL) for acute inhalation exposure tomethylene chloride of 0.4 ppm (1,400 µg/m3) (33). ATSDR also developed an intermediateMRL for methylene chloride, for exposure up to 1 year in length, of 0.03 ppm (106 µg/m3).
Epidemiological studies of humans exposed to methylene chloride for up to 30 years in the workplace have not found evidence of increased cancer deaths (33). Conversely, inhalation studieswith laboratory animals have shown a link between methylene chloride exposure and cancer. Exposure of laboratory mice to very high levels of methylene chloride for a lifetime caused anincrease in cancers of the liver and lungs. Based on these studies, EPA, NTP, and IARC haveclassified methylene chloride as a probable human carcinogen (Group B2). A Group B2classification denotes that there is insufficient evidence of carcinogenicity in humans andsufficient evidence of carcinogenicity in animals.
Methylene chloride was detected in the off-site air in 1970 at a maximum concentration of55,680 µg/m3 (16 ppm). Persons exposed to this level for a short period of time could experienceadverse health effects. Due to the lack of air sampling data for methylene chloride from 1970 to1984, it is not possible to estimate levels of methylene chloride in off-site air and the resultingpossibility of health effects from this exposure. If typical off-site concentrations of methylenechloride during the 1970s continued at the levels measured in 1970, area residents exposedduring that time may be at increased lifetime risk of developing cancer. It is likely, however, thatthe discontinuation of open lagoons in 1970 contributed to the dramatic decline in measuredlevels of airborne methylene chloride between 1970 and 1984.
Methylene chloride was detected in 1984 at a concentration of 7,578 µg/m3 (2 ppm), over seventimes lower than the maximum concentration from 1970. The maximum off-site airconcentration measured in 1987 was about 880 µg/m3, with a maximum average annualconcentration (1987-1989) of 190 µg/m3. It is unlikely that people would experience healtheffects from exposure to these levels of methylene chloride. As the concentrations of methylenechloride in the air decreased over the years, the risk of illness from this exposure (both cancerand non-cancer effects) would also decrease.
Methylene chloride was also detected in off-site seeps, off-site surface water, and on- and off-sitegroundwater. People would have the potential to be directly exposed to this contaminant in seepsand surface water; however, this exposure would only be expected to occur on a very limitedbasis, and thus is not considered a serious public health concern. Methylene chloride has notbeen detected in any of the residential wells sampled.
Methyl Ethyl Ketone (MEK) is an organic solvent which is also referred to as 2-butanone. It is acolorless liquid with a noticeable odor that can be smelled at about 5 ppm (parts per million) inair. MEK occurs in very small amounts in nature; for example, it may be found at lowconcentrations in fruits and vegetables. MEK is also manufactured and used as an industrialsolvent in a variety of common items such as paints and glues. Relatively high levels of MEKmust be present in the air before adverse health effects begin to occur. When workers breathed100 ppm (parts per million, 293 mg/m3) of MEK, they complained of irritation of their nose andthroat. Exposure of humans to MEK levels at or below 200 ppm, has not caused depression ofthe central nervous system (CNS) (34). The level at which MEK will cause depression of theCNS in humans is not known. Some symptoms which characterize CNS depression includeheadaches, nausea, incoordination, vomiting, motor function difficulty, and general malaise. Animal studies have shown depression of the central nervous system when animals were exposedto levels above 10,000 ppm. In addition to CNS depression, studies performed with rats andmice have shown that MEK caused developmental effects in offspring of animals exposed tohigh levels during pregnancy. Developmental effects include delays or disruptions of normalgrowth and development through adolescence. Although no human studies were available, oneanimal study using pregnant mice inhaling high levels of MEK (3,020 ppm, 8849 mg/m3) for 7days resulted in a majority of the offspring born with skeletal abnormalities and decreased birthweights (34). Using this exposure level which caused decreased birth weights in mice, EPA hascalculated a RfC (Reference Concentration) for MEK of 1000 µg/m3. A RfC is an estimate ofthe concentration of a substance in air which would be unlikely to cause adverse health effectsfollowing long term (lifetime) exposure.
MEK has not been conclusively shown to cause cancer in humans. Epidemiological studies ofworkers exposed to airborne MEK continuously for long periods of time showed an increase innasal and cheek cancer. The link between an increased risk of these cancers and MEK exposureis not conclusive, however, because other factors, such as lifestyle habits of the workers (e.g.smoking) which have been linked to cancer were not evaluated. There have been few animalstudies examining the cancer potency of MEK; thus, its ability to cause cancer in animals is stilluncertain. EPA has classified MEK as a Group D carcinogen. A Group D classification denotesthat a chemical is not classifiable as to its human carcinogenicity.
People living and working near the Spectron site would have been exposed to MEK in the air. MEK has been detected in past off-site air samples near the Spectron site at levels greatlyexceeding the RfC of 1000 µg/m3. The maximum concentration of MEK detected in off-site airin 1970 was 94,000 µg/m3 (32 ppm), and in 1976 and 1977 (Table 2B) at mean concentrations of5,070 and 900 µg/m3, respectively, in the samples in which it was detected. The decrease inairborne concentrations of MEK is probably due to the discontinuation of open lagoons in 1970and other process changes which occurred after that time. Air samples collected in 1984 werenot analyzed for MEK. Air sampling in 1987-1989 detected MEK in the air at a maximumconcentration of 22.5 µg/m3, which is well below the RfC.
People living near the site chronically exposed to the level of MEK detected in 1970, 1976, and1977 may have an increased risk of adverse health effects such as having babies with low birthweights. Additionally, people around the site may have experienced CNS depression duringperiods of high concentrations of MEK. MEK was also detected in off-site surface water andseeps; however, because direct contact with these sources is expected to be infrequent, this is notconsidered a serious public health concern. Residential well sampling in 1992 did not detect anyMEK contamination.
Tetrachloroethylene (PCE) is a common chemical that is widely used for dry cleaning and as anindustrial solvent. It is also found in some consumer products such as silicone lubricants, spotremovers, and adhesives. PCE has a sweet odor in air and taste in water. PCE can be smelled atabout 5 ppm and tasted in water at about 0.3 ppm. PCE can cause toxic effects if it is breathed oringested. Short-term exposure to high levels of PCE in the air has been shown to cause eye andnasal irritation. PCE has also been shown to cause central nervous system (CNS) depression(i.e., dizziness, headache, confusion) in people. When human volunteers were exposed torelatively high levels of PCE for 5 days, they experienced a decrease in physical coordination. Based on this study, ATSDR developed an acute (short-term) MRL of 0.6 ppm of PCE in air. PCE has also been shown in animal studies to damage the liver and kidneys when inhaled oringested. When mice breathe PCE in air at relatively high concentrations for 30 days, a largepercentage experienced liver damage (35). Based on liver damage in this study, ATSDRdeveloped an intermediate (exposure up to 1 year) MRL of 0.009 ppm (62 µg/m3) of PCE in air.
In addition to systemic (non-cancer) effects, PCE has been shown to cause cancer (leukemia,liver, and kidney cancer) in laboratory animals following long-term oral or inhalation exposures. EPA has classified PCE as a Group B2 (probable human) carcinogen (35).
PCE was initially detected in off-site air in 1984 at a maximum concentration of 135 g/m3, andan average concentration of 35 µg/m3). The maximum annual average level of PCE measuredduring 1987-1989 was approximately 30 µg/m3. These levels of PCE in air would not beexpected to cause illness in exposed people. People nearby the site would not have experienced asignificant increase in cancer risk as a result of breathing PCE at the maximum level recorded in1984. Air samples taken prior to 1984 were not analyzed for PCE. Due to the lack of airsampling data, it is uncertain whether levels of PCE prior to 1984 posed a significant health riskto nearby residents, although airborne PCE levels were probably considerably higher during siteoperation.
PCE was also tentatively identified in eleven residential wells in 1992. The maximumconcentration of PCE detected in the wells was 14 ppb (14 parts PCE in one billion parts ofwater). Exposure to this level of PCE in drinking water would not be expected to cause illness inthe exposed residents. No PCE was detected in the 21 residential wells sampled in September1995.
PCE was detected in off-site leachate seep samples, off-site creek samples, on-site monitoringwells, and off-site groundwater. Because of its likely infrequent occurrence, direct contact withPCE in seeps and creek samples is not be considered a serious health hazard.
Toluene is a solvent that is widely used in industry and as a component of many consumerproducts such as gasoline, paints, fingernail polish, and adhesives. People can be exposed totoluene during the use of these consumer products in the home or workplace.
The major effect of toluene in humans is on the central nervous system. Studies of peopleexposed to moderate levels of toluene in workplace air have reported effects which includefatigue, dizziness, confusion, and memory loss (17). Long-term exposure to toluene in theworkplace has also been associated with hearing loss, with some evidence that combinedexposure to certain common medications, such as aspirin, might increase this effect. There is no evidence of an association between toluene exposure and cancer in either humans or laboratory animals.
Levels of toluene detected in the air off of the Spectron site in 1970 and in 1976 could putchronically exposed people at increased risk of experiencing effects on the central nervoussystem, described above. Average daily levels detected in the air in the summer of 1984 wouldnot be expected to cause health effects in exposed people. Toluene has also been detected inseeps entering Little Elk Creek, in groundwater beneath the creek, and in surface water samplestaken directly from the creek. Toluene concentrations detected in Little Elk Creek would not be expected to endanger the health of people who contacted the water.
1,1,1-Trichloroethane (1,1,1-TCA) is a man made chemical that is commonly used as acomponent of industrial solvents and is also found in consumer products such as glues, cleaningproducts, and aerosol sprays. It has a noticeable odor and can be smelled at about 120 ppm (partsper million) in the air. Humans exposed to high vapor concentrations of 1,1,1-TCA haveexperienced central nervous system depression characterized by dizziness, lethargy, andincoordination. 1,1,1-TCA has also been shown to cause permanent neurological damage inlonger term studies with laboratory animals (36). When mice were exposed to 210 ppm of1,1,1-TCA continuously for a relatively long period of time, it caused damage to the brain andspinal cord nerves, resulting in permanent damage to the nervous system. Based on this study,ATSDR has developed an intermediate MRL (Minimal Risk Level) for an exposure period of upto a year of 0.7 ppm (3,885 µg/m3) for 1,1,1-TCA in air. ATSDR also developed an acute MRL(exposure periods up to 15 days) for 1,1,1-TCA of 2 ppm (11,100 µg/m3).
Studies on the carcinogenic (cancer causing) potential of 1,1,1-TCA exposure are inconclusive. Because no studies have shown conclusive evidence of cancer as a result of long-term exposure,EPA has classified 1,1,1-TCA as a Group D carcinogen. A Group D classification means that1,1,1-TCA is not classifiable as to its carcinogenicity (36).
The maximum off-site concentration of 1,1,1-TCA detected during the 1984 air sampling was2,158 µg/m3 (0.6 ppm) (Table 2C). The maximum annual average concentration of 1,1,1-TCAdetected in the air in 1987 was 180 µg/m3. People exposed to the levels of 1,1,1-TCA in the airat the 3 locations monitored from 1987-1989 was 111 µg/m3 (Table 2D). Due to a lack ofsampling data, levels of 1,1,1-TCA in the air prior to 1984 are unknown. The levels of1,1,1-TCA measured in the air during the 1980s would not be expected to cause illness inexposed people. As with the other contaminants found near the site, it is likely that past levels of 1,1,1-TCA in the air were considerably higher than those measured in the 1980s, with a resulting increase in the risk of illness.
1,1,1-TCA has also been detected in on- and off-site leachate seeps, surface water, monitoringwells, and groundwater samples. Because regular exposure to contaminants in seeps and surfacewater is unlikely (and concentrations in the creek are relatively low), this exposure pathway is not currently considered a serious health hazard.
Trichloroethylene (TCE) is a man made chemical which is used as an industrial solvent and as acomponent of a number of different consumer products, including paint removers, glues, andcleaning fluids. Historically, TCE was used as an anesthetic before its toxicity was fullyunderstood. TCE has a sweet odor and can be smelled at about 100 ppm (parts per million) inthe air. Short-term exposures to much higher concentrations of TCE can produce central nervoussystem depression characterized by dizziness, incoordination, sleepiness, and slowed reactiontimes (37). The central nervous system can be permanently damaged through long- termexposure to TCE. When rats breathed 1,000 ppm of TCE continuously for 18 weeks, theyexperienced a loss of vision, incoordination, and a reduction in reaction time. Based on thisstudy, ATSDR developed an intermediate (for exposure up to 1 year) inhalation MRL (MinimalRisk Level) of 2 ppm (10,920 µg/m3) of TCE in air. Long-term exposures with laboratory animals suggest that TCE may cause the liver to become enlarged, and the kidney not to function properly.
Additionally, TCE exposure has also been shown to cause cancer in laboratory animals. Animalstudies have shown an increased incidence of cancer in a number of different organs followinglong-term oral or inhalation exposures. Studies of humans exposed to TCE in drinking waterhave shown a possible, although inconclusive, association with leukemia (37). EPA is currentlyevaluating the data on the carcinogenicity of TCE.
TCE was detected at a concentration of 655 µg/m3 in one of 138 off-site air samples collectedduring one day in October 1976. A maximum average annual concentration of approximately 25µg/m3 TCE was reported from air sampling conducted from 1987-1989. It is unlikely that theTCE levels measured in 1976 or 1987-1989 would cause illness or symptoms in exposed people. Air samples taken before 1977 were not analyzed for TCE. Due to lack of air sampling data, itis uncertain if residents were exposed to levels of TCE before 1977 which were sufficiently highto cause illness. It is likely that airborne levels of TCE were at times considerably higher than thelevel measured in 1977. Highest levels were likely present in the air before the waste lagoon wasremoved in 1970.
TCE has also been detected in off-site leachate seeps, surface water, groundwater, and on- andoff-site monitoring well samples. To date, TCE has not been detected in any residential wells.
Ten inorganic contaminants (metals) were detected in on-site monitoring well samples inconcentrations exceeding comparison values (Table 1 in Appendix B). No exposures to thesecontaminants are known to be occurring at this time. During the most recent residential wellsampling round, the samples were not analyzed for metals. The movement of metals throughgroundwater to off-site residential wells would not be expected, however, because metals do nottend to migrate readily in groundwater. Also, on-site metal concentrations were lower than VOCconcentrations, and metal contamination of residential wells would not be expected in theabsence of VOC contamination. Because there is no known human exposure to the inorganiccontaminants found in the on-site monitoring wells, they will not be further discussed.
In order to examine the possibility that past exposure to contaminants from the Galaxy/Spectronsite may have resulted in a greater-than-expected rate of disease in the exposed population, it isnecessary to: 1) define the population of concern; 2) define the diseases of concern; and 3) obtainrecords of disease outcome in the study population for the time period of interest. Unfortunately,health outcome data (for example, vital statistics) are not compiled for population groups smallerthan counties. Thus, this evaluation cannot be performed with existing resources, but wouldrequire a specially designed study.
A special study of Providence Valley residents was carried out by a task force organized in 1974by Dr. Neil Solomon, director of Maryland's Department of Health and Mental Hygiene(DHMH). Because of concerns that were expressed at the time, the group examined the numberof cancer deaths that had occurred in the study population during the period of concern. Indefining the study area, the task force decided to include the entire area from which odorcomplaints had been received (an area of about 2.75 miles north-to-south and 0.75 mileseast-to-west). Using death certificates, the investigators tried to identify all cancer deaths thathad occurred in the study population from January 1, 1967, to June 30, 1976. They thencompared the number of observed deaths with the number that would be expected based oncancer death rates for Cecil County as a whole during this period (3).
The group reported that during this period, there were 1.8 times more cancer deaths in the studypopulation (80% greater than expected) as compared to the number expected based on the ratesfor Cecil County. During the study period, a total of thirteen cancer deaths were identified in thestudy population. This number was considered significant based on statistical analysis using a95% level of significance. The study also observed a significantly greater-than-expected numberof cancer deaths from specific categories of cancer, including: 1)lymphosarcoma and reticulumcell sarcoma, and 2)other lymphoma. The results of this study are summarized in Appendix C.
Authors of the task force report concluded that at the time, no conclusions could be made aboutthe cause, or causes, of the greater-than-expected number of cancer deaths. Included in thethirteen cancer death cases identified in the valley were cases which were very unlikely to becaused by solvents in the valley air. For example, there were two cases of men who lived morethan a mile from the site and who died of prostate cancer at advance ages (82 and 83). Therewere other cases, however, in which there is a greater possibility that the cancers were related toexposure from the site. Specifically, three cases of lymphoma (cancer of lymphatic tissue) wereidentified in people who lived distances of 0.06 to 0.3 miles from the site. At the time of deaththese individuals had lived in their homes for periods ranging from 7 to 72 years. The task forcestudy did identify one category of lymphoma (there are different categories of this based on thespecific tissues/cell types involved) as being significantly elevated in the study population. Studies of workers exposed to solvents, including some of those identified at the Spectron site,have shown a significant association with lymphoma (38). Chemicals identified at Spectronwhich have been linked by some studies to an increased risk of lymphoma include benzene,trichloroethylene, and tetrachloroethylene (39). This observation does not prove that thelymphoma cases identified near Spectron were caused by chemicals from the site, but it doesraise this question.
Dr. Pietro Capurro was the author of a paper published in 1979 which identified additional casesof individuals who were likely exposed to airborne contaminants from the site, and who had diedfrom or developed cancers of lymphatic or hematopoietic (blood forming) tissues (7). Inconversations with people living near the site during the preparation of this health assessment,MDE staff were also informed of two additional cases of individuals who had lived near the siteand who had developed either leukemia or lymphoma. One of these cases may have beenidentified in Dr. Capurro's paper.
Also included in the Task Force report was a review of death certificates from former Galaxyemployees. From 1961 to 1977, 166 people are reported to have worked at the site since theinception of activities. It was known that of the 166 employees who had ever worked at Galaxy,131 or 78.9% had been employed for a period of one year or less. As of 1977, four of the formeremployees had died; two had died in car accidents and two had died of cancer. The first cancercase was a 50 year old man, who died in 1974 of cancer of the larynx. This man had been aGalaxy employee for seven years and a Valley resident for 11 years. This case is complicated bythe fact that he was also a heavy smoker. The second cancer death was a 22 year old male whohad died of cancer (adenocarcinoma) of the gastrointestinal tract or the pancreas. This man was aGalaxy employee for 21 months (1969-1971) and had never been a Providence Valley resident. This case was complicated by the fact that the patient had been diagnosed with ulcerative colitisat age 15, a condition which is known to predispose to carcinoma of the colon(3). Figures on thenumber of employees after 1977 are not available and there have not been any follow-up healthstudies conducted on this group of workers since the task force investigation in 1977.
Using readily available data, it is possible to compare recent health outcome data (for example,cancer deaths and birth defects) for residents of Cecil County and the State of Maryland as awhole. This, however, cannot be used as an indication of disease rates in a small populationwithin Cecil County (for example, potentially affected Providence Valley residents). Thisinformation is presented in this document under "cancer mortality" because it may be of interest to the reader to examine recent rates in the population of Cecil County as compared to the State of Maryland.
Data for the occurrence of twelve separate birth defects (referred to as sentinel defects) areavailable for Maryland at the county level for the years 1984 to 1991 (Maryland Department ofHealth and Mental Hygiene, Provisional summary statistics, 1984-1991). During this period, theincidence rate for infants born with one or more of the 12 defects was 91.2 per 10,000 births forCecil County (52 infants out of 5,704 total births) as compared to 57.8 per 10,000 (2,939 infantsout of 508,179 births) for the State of Maryland as a whole. Birth defects which occurred at leasttwice as frequently among Cecil County live births as compared to the state as a whole include:
|anencephaly||8.8 per 10,000 Cecil County births|
vs. 3.1 per 10,000 Maryland births
(w. or w/o
|10.5 per 10,000 Cecil County births|
vs. 4.6 per 10,000 Maryland births
|rectal/anal atresia||7.0 per 10,000 Cecil County births|
vs. 2.3 per 10,000 Maryland births
|10.5 per 10,000 Cecil County births|
vs. 3.8 per 10,000 Maryland births
|Downs syndrome||12.3 per 10,000 Cecil County births|
vs. 7.6 per 10,000 Maryland births
Age-adjusted (to 1970 U.S. population), cancer death rates per 100,000 population are availableat the county level for the period of 1983 to 1987. Cancer death rates are adjusted for agebecause age is an important risk factor for cancer, with people in the oldest age groups having thegreatest risk for developing cancer. County cancer death rates are age-adjusted to account forpossible differences in the age distribution of residents living in the different Maryland counties. To account for the possible effect of age on cancer death rates, the county rates are adjusted usingthe age-distribution of the 1970 U.S. population. Average annual cancer death rates were notsignificantly different (based on statistical analysis at a 95% confidence level) for Cecil Countyas compared to the State of Maryland as a whole for the most common forms of cancer (see databelow).
|(cancer death rates per 100,000)|
It can be concluded that from 1983 to 1987, residents of Cecil County did not die from thecancers listed above at rates that differed significantly from the residents of the State of Marylandas a whole. As comparable data are not available specifically for the Providence Valley area,conclusions regarding the possibility of increased cancer death rates for Valley residents cannotbe made.
In the past, especially during the late 1960s and early 1970s there was concern expressed overhealth problems resulting from exposure to airborne emissions from the Galaxy/Spectron facility. Dr. Pietro Capurro, a former pathologist at Union Memorial Hospital in Elkton, Maryland,reported observing patients admitted to the hospital whom he believed were suffering effectsfrom exposure to airborne chemicals from the site. Common symptoms reported to Dr. Capurroby hospitalized patients included vomiting, abdominal pain, dizziness, and headaches. Dr.Capurro also expressed concern over what he believed to be an unusually high rate of canceramong residents of Providence Valley (7).
Some area residents have recently expressed concern over possible long-term health effectsresulting from past exposures to site-related contaminants. Specific concerns that wereexpressed focused on cancer and heart disease. One local resident was also concerned about thepotential association between past exposure (during the 1970s) to airborne contaminants andadverse reproductive outcomes (miscarriage). The following is a brief discussion of concernsthat have been expressed to date:
I. Concern regarding exposure to site-related chemicals and development of cancer.
It is possible that extended exposure to contaminants from the site could increase a person's riskof developing cancer during their lifetime. People exposed in the past to the highest levels ofcontaminants (those who worked on and live closest to the site) would likely be at greatest risk. At least one known human carcinogen (benzene) has been measured at elevated concentrations inoff-site air. Another known human carcinogen (vinyl chloride) has been found in on-sitegroundwater and off-site surface water (seeps). A number of other contaminants which havebeen detected in off-site air have been classified as possible or probable human carcinogens byEPA. The cancer mortality study, conducted by the Maryland DHMH (Department of Health andMental Hygiene) and completed in 1977, reported a greater number of valley residents dying ofcancer than would be expected based on the experience of the Cecil County population as awhole. No conclusions regarding a possible association between cancer and exposure tosite-related contaminants could be made by the authors of the study report. Since completion ofthe DHMH study, some studies of human populations (epidemiological studies) have reportedfinding significant associations between lymphoma (cancer of the lymph tissue) and exposure tosome of the chemical solvents which were detected in the air near the Galaxy/Spectron site,including benzene, trichloroethylene, and tetrachloroethylene. The original task force studyidentified three people who had died of some form of lymphoma and who had lived near the site. An author of this health assessment has also recently been informed of cases of two individualswho had lived near the site during the period of high emissions and who subsequently developedsome form of lymphoma, although these have not been confirmed with medical records.
2. Concern regarding exposure to contaminants and heart disease.
Long-term exposure to some of the chemicals found in on- and off-site samples might contributeto certain types of heart disease (25). Exposure to sufficiently elevated levels of methylenechloride, which has been consistently found in the air near the site, could pose a greater healthrisk to people with heart disease. When exposed to high concentrations, metabolism (breakdown) of methylene chloride by the body produces carbon monoxide, which is normally in thebody at very low concentrations. Carbon monoxide takes away the ability of the blood to carryoxygen to vital tissues and organs. This may result in depression of the central nervous system,and decreased ability of the vital organs, especially the heart, to function properly. This can bedebilitating to a person with a weak heart and may increase the risk of more critical heartcomplications. Methylene chloride has not been measured in the air near the Spectron site inconcentrations that would result in increased carbon monoxide levels in exposed residents;however, higher concentrations may have been present before removal of the evaporation lagoonin 1970. Additionally, because cigarette smoke also contains carbon monoxide, smoking, whichis known to cause heart disease, in conjunction with methylene chloride exposure can increasethe amount of carbon monoxide in the body, which may result in an additive toxic effect on theheart.
3. Concern about exposure to site contaminants and miscarriage.
A number of the contaminants that have been identified in on- and off-site environmental mediahave been found to cause adverse reproductive effects in exposed laboratory animals, and somehave been associated with adverse reproductive outcome in women (40). Three site-relatedcontaminants have been shown, through laboratory animal studies, to cause fetal death and/orcongenital malformation (birth defects) in the offspring. These three site-related contaminantsinclude 1,2-dichloroethane, benzene, and methyl ethyl ketone. The laboratory animals wereexposed to much higher levels of these chemicals than would have been experienced by peopleliving near the Spectron site; however, the possible effect of exposure to a mixture of thesechemicals at lower levels (as would have been experienced by area residents) is not known. Exposure to sufficiently high levels of a fourth chemical, methylene chloride, can reduce theblood's ability to carry oxygen, which could adversely affect the developing fetus of an exposedwoman, especially if the woman also smoked cigarettes. Such exposure levels may have beenexperienced by some area residents before the on-site evaporation lagoon was eliminated in1970. Therefore, it is possible that a woman's exposure to sufficient quantities of thesecontaminants from the site prior to 1970 might have increased the risk of miscarriage and/or birth defects in her children.
4. Concern about potential effects of site contaminants on the health of children.
Although children can be more sensitive to the effects of chemicals than adults, what we estimateto be current levels of airborne contaminants are not considered a health threat to children. Thisis based on the levels of contaminants measured in the air in 1987-1989 and the fact thatchemicals were removed from the site in 1990, which would have further reduced the levels ofairborne chemicals. Levels of chemicals in the creek downstream of the site would not be highenough to cause illness in children who occasionally contacted the water. It is important,however, that children's contact with site contaminants be limited by keeping them away fromthe site, and the seeps and that portion of the stream immediately adjacent to the site.
5. Concern about bioaccumulation of contaminants in fish.
Bioaccumulation refers to chemicals that enter the body, are not easily metabolized (brokendown), and tend to store in specific tissues (i.e. fat deposits) or organs (i.e. the liver) for longperiods of time. The type of contaminants found at the Spectron site, volatile organic chemicals,do not bioaccumulate to a significant degree, although some tissue accumulation will occur. TheMaryland DNR and the U.S. EPA sampled fish downstream of Spectron in 1981. Up to elevendifferent VOC's were detected in some of the 1981 fish samples, but the concentrations weresimilar to the levels that have been measured in the creek samples. Detectable levels ofchemicals would only be expected in fish sampled immediately downstream from the site. This isnot a route of human exposure because there are not edible game fish in this part of Little ElkCreek. These chemicals tend to be metabolized and excreted by fish relatively quickly. Because of this, the fish in Little Elk Creek would not be expected to accumulate the VOC contaminants in water to a significant degree.