FALLS TOWNSHIP GROUNDWATER CONTAMINATION
(a/k/a CORCO CHEMICAL, PARASCIENTIFIC, MEENAN OIL)
FALLS TOWNSHIP, BUCKS COUNTY, PENNSYLVANIA
ENVIRONMENTAL CONTAMINATION AND OTHER HAZARDS
A. CONTAMINATION
ATSDR must rely upon sampling and analysis data developed by others, such as facility owners and state and federal agencies, for evaluating the nature and extent of contamination. ATSDR has received sample and analyses information primarily for the Corco Chemical and Meenan Oil recovery areas and for private wells in the vicinity. No sampling investigations have been conducted around the building Para Scientific Company leases, except for some soil gas evaluations conducted there by Corco Chemical as a part of its property evaluations. Most of the sampling information received was developed between 1982 and 1985; data for 1986 and later are quite limited. Essentially all sample analyses were for VOCs. The contaminants most frequently detected are several chlorinated compounds: trichloroethane (TCA), TCE, PCE, and chloroform.
Sampling data that provide the framework for the evaluations in this public health assessment are summarized in discussions that follow. Groundwater, soil, and soil gas data also are summarized in Tables 1A, 1B, and 1C in Appendix A, along with comparison values ATSDR considered in its evaluation. Sampling data are presented in greater detail in Tables 2-6 (Appendix A). Comparison values for a public health assessment are contaminant concentrations in specific media that are used solely to select contaminants for exposure and public health evaluation. The presence of a contaminant in a table or its presence at a concentration greater than the comparison values in Tables 1A, 1B, or 1C does not mean that it will cause adverse health effects through exposure. The health significance of contaminants and exposure is described in a later section (Public Health Implications).
Country Lane Trailer Park Water Supply Well
Sample data for groundwater once used for the residents' potable water supply are shown in Table 2 (Appendix A) (9). Chlorinated VOC compounds [TCA, TCE, PCE, dichloroethene (DCE), and chloroform (maximum concentrations from 19-308 ppb)] were detected in the sample as were nonchlorinated VOC compounds like benzenes, xylene, and naphthalenes (25-242 ppb). Five of the compounds were found to be at levels that exceed ATSDR's comparison values for drinking water.
Area-Wide Groundwater
Information about several contaminants found most frequently and at the highest concentrations at selected private wells within the area is summarized here and in Table 3 (Appendix A).
Private wells in the study area north of Penn Valley Road are primarily in the following locations: on Cedar Lane, on Old Tyburn Road, on the northern portion of Corbin Lane, on Bristol Pike, and on the portion of Penn Valley Road east of Bristol Pike. South of Penn Valley Road, groundwater users are along or near Bristol Pike. Immediately south of the study area, private wells are along Wheatsheaf Road and near the intersection of Wheatsheaf Road and Bristol Pike, south of the canal.
ATSDR has examined one or more sets of groundwater data for 29 private wells in the study area that were sampled between 1982 and 1992 (2, 3, 9, 10, 11, 12, 13, 14, 15). During the 1990 site visit, ATSDR was shown approximately 40 residences and 15 businesses in the study area that were not being served with public water. ATSDR assumes most of those locations were being served by private wells. Thus, it appears not all the wells in the study area have been sampled. By 1992, county health personnel report that owners of about 14 of the 29 wells had connected to the public water supply, use bottled water, or installed a filter system. However, the effectiveness of the treatment systems have not been confirmed. The health department reports that the data ATSDR examined were for untreated well water, except for one treated sample taken in 1990 at a residence on Corbin Lane.
Most of the 29 wells sampled are north of Penn Valley Road. A few other private wells were sampled along Bristol Pike between Penn Valley Road and the canal. Sampling data also were obtained south of the study area in 1992 for eight of the private wells that are along Wheatsheaf Road and also along Bristol Pike within about 1,500 feet of its intersection with Wheatsheaf Road (15).
TCA, TCE, PCE, and chloroform were the principal VOC compounds detected in private wells. The concentrations of those chlorinated compounds in several geographic groups of wells are reported in Table 3. ATSDR's evaluation of the data shows that, of the substantively potent chemicals present, PCE and TCE typically occur together, are the most prevalent, and can be considered the primary--but not exclusive--indicators of groundwater quality for this public health assessment.
The available data show that the groundwater contamination varies with respect to concentration, location, and time. Sampling conducted between in 1982, 1985, and 1987 for 26 private wells suggests that well water is least contaminated on Cedar Lane, along Tyburn and Old Tyburn Roads, and at the northern end of Bristol Pike. Data also suggest that private well water is most contaminated near the intersection of Bristol Pike and Penn Valley Road, an area not close to either Corco Chemical or the Meenan Oil release locations. More moderate concentrations were noted in groundwater samples from wells on Corbin Lane and elsewhere on Bristol Pike. Analyses of water from those private wells, excluding the previously described Country Lane Trailer Park well (no longer in use), have shown substantive levels of TCA (625 ppb), TCE (1,400 ppb), and PCE (175 ppb). Eighteen of the 26 wells sampled showed one or more of those compounds at levels greater than ATSDR's comparison values for drinking water. Sixteen of the wells contained two or more of those chemicals above comparison values. Chloroform (8 ppb) was also found in a few wells above its comparison value. Nonchlorinated VOC compounds were detected in only one private well (benzene; 0.9 ppb), except as noted previously for the Country Lane Trailer Park well.
Sampling was conducted in the study area in 1989, 1990, or 1992 for nine wells. Five of the nine wells contained PCE (20 ppb) or TCE (49 ppb). Three of those five had PCE or TCE at levels greater than comparison values. The residence at which the greatest concentrations were found has a filter system; hence, if the filter is performing satisfactorily, the quality of filtered water should be substantially better than reported for the raw water sample. PCE was present in one of the other two wells at a level of 3.4 ppb, and TCE was present in both at levels as much as 13.8 ppb.
Contamination variation over time was reviewed using information for six wells on Corbin Lane and East Penn Valley Road that were sampled one or more times during 1982-1987 and also 1989-1992. Those data show that maximum total TCE and PCE concentrations present in each of those wells during the 1982-1987 period were substantially lower during the 1989-1992 period. Examples of total TCE and PCE decrease over time include; 29 ppb decreased to not detected, 32 ppb decreased to 0.7 ppb, and 1,575 ppb decreased to 69 ppb. Many of the other 26 wells sampled during 1982-1987 were not resampled later and, hence, provide no insight to contamination variation with time. The 1982-1987 data showed some wells that were not resampled contained no TCE or PCE, some contained TCE or PCE below ATSDR's comparison values, and some contained TCE or PCE at levels greater than comparison values.
ATSDR also reviewed the latest available sampling data for each of the 29 wells tested to identify current water quality at wells that continue to be used for potable water. Results of this evaluation suggest 4 wells in use contained contaminants, when last tested, at levels that exceed ATSDR's comparison values; 1 well in use contains contaminants that would exceed ATSDR's comparison values if the filter system does not function properly, and 1 well contains contaminants that exceed comparison values but current use of the well is not known.
The areal distribution of contaminants shown in well data was evaluated to assess whether the existing sampled well locations might adequately circumscribe and characterize groundwater contamination within the study area and provide confidence about water quality at locations that have not been sampled. This evaluation provided two examples from the 1985 and 1987 sampling data that show that evidence of acceptable water quality at one private well location does not necessarily mean that water quality at an adjacent well can be assumed to be acceptable; 1) One deep well near East Penn Valley Road showed no contamination, while two nearby shallow wells showed substantially elevated levels, and 2) two deep wells on Bristol Pike showed no contamination, while a nearby well (depth unknown) between the two showed unacceptable levels of contamination. From this information, ATSDR concludes that the character of groundwater identified at the specific wells sampled may not be indicative of groundwater quality nearby or elsewhere in the study area.
Groundwater data are available for the Corco Chemical monitoring wells (Tables 3 and 4) and for the Meenan Oil recovery well system (Tables 3 and 5) (2, 3, 4, 12, 14, 16, 17, 18, 19, 24, 26). Several contaminants were found at levels that exceed ATSDR's comparison values for drinking water. Elevated concentrations of the principal chlorinated VOCs [TCA (350 ppb), TCE (3,700 ppb), PCE (1,200 ppb), and chloroform (1,880 ppb)] were detected in groundwater at the Meenan Oil fuel recovery well system near Cedar Lane. The compounds also were detected at Corco Chemical monitoring wells. Levels were low at the Corco wells in earlier years, but, according to the most recent data, concentrations have increased [TCA (25 ppb), TCE (214 ppb), PCE (431 ppb), and chloroform (77 ppb)]. Groundwater at the Meenan Oil recovery well system also was found to contain elevated, nonchlorinated VOCs [benzene (778 ppb) and toluene (356 ppb)]. Nonchlorinated VOC levels at the Corco Chemical monitoring wells are much lower [benzene (3 ppb), toluene (6 ppb), xylene (17 ppb)]. Lead (50 ppb) also was reported in groundwater at one Corco monitoring well during one sampling. Low levels of carbon tetrachloride (10 ppb) and dichloroethane (DCA) (5 ppb) were detected at both locations.
Corco Chemical Samples
Detailed sampling data for Corco Chemical are shown in Table 4 (Appendix A). Groundwater information was summarized previously in the Area-Wide Groundwater section of this public health assessment. Analyses of samples from the company's potable water supply well through 1985 showed low levels of VOCs (3, 14). TCE (5 ppb in 1982; <1 ppb in 1985) had the highest concentration; the TCE level reported in 1982 slightly exceeds ATSDR's comparison value for drinking water; the concentration in the 1985 sample is below its comparison value. The county health department reports that well is still in use.
Analyses of lake water west of the plant property showed low levels of two chlorinated VOCs (TCA and TCE); TCA was of greater magnitude (2 ppb) (3, 14, 19). TCE (1.6 ppb) was present at values slightly below ATSDR's comparison value for drinking water.
The cooling water effluent that discharges into the lake east of Cedar Lane was analyzed frequently through early 1990. Many contaminants have been detected in the effluent. Some range greatly in magnitude over time; the highest levels generally occurred in the earlier years (3, 20). The highest concentrations were noted for chlorinated VOCs, such as TCE (>1000 ppb), PCE (>500 ppb), and chloroform (>500 ppb); those levels are well above ATSDR's comparison values for drinking water. Nonchlorinated compounds included xylene (120 ppb). Now that drains have been disconnected from the cooling water system, contaminant levels in its effluent probably are minimal. However, sampling data are not available to confirm that.
Water samples taken at the lake east of Cedar Lane showed low levels of three chlorinated VOCs [TCE (2.9 ppb), PCE (3.4 ppb), and chloroform (1.1 ppb)] and a nonchlorinated VOC, toluene (4 ppb) (4, 14, 21). Lead also was detected (3.0 ppb). Of these, the PCE level exceeds its comparison value for drinking water, and TCE was about equal to its comparison value. Analyses of sediments from the lake showed low levels of four VOCs. DCE was at the highest level (10 ppb), and TCE and PCE also were present (21). Now that the cooling water effluent no longer contains drain discharge, lake contaminant levels might be lower than shown by the available data.
Soil gas analyses were conducted in 1989 at shallow depth below ground at the plant area, near the building leased by Para Scientific Company, on the east edge of Cedar Lane, and east of Cedar Lane near the landfill (22). At the plant area, the principal chemicals detected included TCE, PCE, benzene, and toluene at levels ranging from 130,000-340,000 ppb, by volume. At the east edge of Cedar Lane and by the Para Scientific building, TCE and PCE were the primary soil gas constituents at levels of 50,000 ppb, by volume. By the landfill, PCE was detected at a level of 300 ppb, by volume. The concentrations of benzene, PCE, and TCE exceed ATSDR's comparison values for chronic exposure in ambient air. Benzene, PCE, TCA, and toluene levels exceed comparison values for acute exposure in ambient air. Sampling data are not available for outdoor ambient air or for interior air (businesses and residences).
Surface soil (depth to 12 inches) was sampled at the plant area (21, 26). VOCs were essentially absent in the soils, but several other contaminants were detected, including polychlorinated biphenyls (PCBs) (5,300 ppb), bis(2-ethylhexyl)phthalate (1,200 ppb), and low- and high-molecular-weight polynuclear aromatic hydrocarbons (PAHs) (1,154-1,674 ppb). The concentration of PCBs exceeds ATSDR's comparison value for incidental soil ingestion. Analysis of a background soil sample showed PAHs at slightly lower concentrations than those of samples taken on the property.
A solvent trap and junction box were sampled several times before the system was disconnected (22, 23). The trap was removed, and the box was cleaned. Sampling in 1989 showed many contaminants, including TCA, TCE, PCE, and chloroform (8,647-12,334 ppb) and acetone (107,801 ppb). The level of PCE exceeds ATSDR's comparison value for incidental ingestion.
Meenan Oil Recovery System Samples
Detailed sample data for the Meenan Oil recovery system are shown in Table 5 (Appendix A). Groundwater analyses associated with this system are summarized previously in the Area-Wide Groundwater section of this public health assessment.
Effluent from the recovery system that discharged groundwater to Warner Lake, south of Penn Valley Road, also showed elevated levels of chlorinated compounds (TCE, PCE, chloroform, DCE, and DCA [1,600 to 4,500 ppb]) (4, 19, 24). Sampling showed that the effluent also contained benzene (778 ppb). Several compounds exceed ATSDR's comparison values for drinking water.
Analysis of a water sample from Warner Lake taken in 1986 while effluent was being received showed chlorinated compounds were present, but at much lower concentrations (16-153 ppb) than were found in the recovery well effluent (4). The greatest concentrations were of DCE, DCA, and PCE. Nonchlorinated compounds, benzene (191 ppb) and toluene (2.4 ppb), also were present. Several compounds were at levels that exceed ATSDR's comparison values for drinking water. Water quality in the lake is likely to be substantially better now that the recovery activities have terminated. However, recent sampling data are not available to confirm that.
Samples of surface soils from the vicinity of the recovery well and from elsewhere within Pennwood Crossing were analyzed; no organic compounds were detected (4).
K-Mart Samples
Data for several samples of water that pooled on the K-Mart property and in ditches in 1982 during attempts to put out the K-Mart fire are shown in Table 6 (Appendix A) (10). VOCs were detected in only one sample. Primarily chlorinated compounds, including TCA (330 ppb), were present; benzene (30 ppb) also was detected. The concentrations of both compounds exceed ATSDR's comparison values for drinking water.
The results of groundwater monitoring at K-Mart in 1983 also are shown in Table 6 (25). Analyses were not performed for VOCs. Pesticides and PCBs, the only organic compounds analyzed for, were not detected.
B. QUALITY ASSURANCE AND QUALITY CONTROL
ATSDR has received little quality assurance information for laboratory analyses and none for field sampling. In preparing this assessment, ATSDR has presumed that appropriate protocols were followed and that analytic results are valid. The completeness and reliability of available information could affect the validity of ATSDR's conclusions.
C. PHYSICAL AND OTHER HAZARDS
ATSDR observed nothing in the area of concern that should pose a usual type of physical hazard to the public. Soil gases could be hazardous if VOCs are released below ground from contaminated groundwater or subsurface chemical wastes, migrate into buildings, and accumulate at explosive levels. Ambient air monitoring data are not available, and information about soil gases is not sufficient to fully evaluate whether a hazard exists.
A. ENVIRONMENTAL PATHWAYS (FATE AND TRANSPORT)
Chemicals can enter the environment from a source as a result of a planned or permitted release or an accidental release to the ground, air, surface water or sediment, groundwater, and, sometimes, plants and animals. In an active workplace, contaminants can be transported by production and maintenance activities, vehicle traffic, or construction. Workers may transport contaminants on their clothing and skin. Once in the environment, contaminants may begin to degrade or migrate depending on climatic factors, physical factors associated with the release setting, chemical factors associated with the contaminant, and factors associated with the receiving environmental media. VOCs, which were detected primarily in groundwater, surface water and soil gas at the Falls Township site, are highly volatile, and most are relatively mobile in soil and water. The phthalate, PAHs, and PCBs found in soils at Corco Chemical, are only slightly volatile and relatively immobile in soil and water. Substantial sampling and analytic data usually are required to confirm the actual source(s) of contaminants and the environmental media that are important transport pathways.
Information about several potential sources of the observed groundwater contamination have been considered in the course of this public health assessment: Corco Chemical, Para Scientific Company, the Meenan Oil fuel leak, the landfill, the K-Mart fire, septic system cleaners, and trucking companies. The absence of comprehensive sampling over time and uncertainty about groundwater flow patterns probably preclude identifying any primary, or dominant, source of either the chlorinated or nonchlorinated VOCs. For example, available data show the highest levels of chlorinated VOCs have been found in private wells near the junction of Penn Valley Road and Bristol Pike. Those findings suggest that either a separate unknown source in the vicinity released contaminants to groundwater, or a mass of more highly contaminated groundwater from another source migrated there.
Groundwater is relatively shallow (10-20 feet), and water levels in the lakes probably reflect the approximate groundwater table elevation throughout the area of concern. Soils beneath the area are primarily sand and gravel and extend well below the groundwater table. Because soils are sandy, the permeability and groundwater flow rate in some soil zones could be relatively high. The general direction of groundwater flow in the region is likely to be toward the Delaware River, which is within about a mile to the east of the area and somewhat farther away to the southeast and south. Some investigators suggest that flow direction in the immediate area is east and southeast, toward Van Sciver Lake (2). However, the specific effects of local topography, geology, and the temporary fuel oil recovery activities on groundwater flow across the area are not defined.
VOCs possibly migrated to groundwater by percolating downward through soils following spills, leaks, septic tank discharge, and pooling of water used to put out the K-Mart fire. Surface water in the lake east of Corco, which receives the chemical company's effluent, probably is a source of groundwater recharge; therefore, some contaminants released by Corco to the lake could enter the groundwater system. Also, VOCs or other contaminants that might be in landfilled materials or oil reclamation residues next to the lake might enter area groundwater by percolation through soil or by runoff into lake water. Surface runoff from Corco Chemical into the lake to its west could also transport contaminants that would subsequently enter the groundwater through recharge by lake waters. The layer of fuel oil that floated on groundwater at the Meenan Oil fuel leak prior to cleanup undoubtedly resulted in direct migration of some nonchlorinated VOCs, and probably soluble fractions of non-VOCs, to groundwater. Information is insufficient to show whether any VOCs contributed by the fuel oil release to groundwater migrated to any private wells. Clean-up operations included discharge of contaminated groundwater to Warner Lake, where some portion of the diluted discharge might enter the groundwater system through recharge by lake water. Runoff from the K-Mart fire might have transported chemicals from there to the Delaware River via a creek that passes well west of the area of concern.
VOCs in contact with ambient air tend to volatilize and migrate as a gas to surrounding areas. The principal sources of airborne VOCs are likely to be from workplaces and from spills or disposal on the ground. Volatilization could also occur readily from contaminated water in area lakes. If chemical releases lessen or stop, continued volatilization usually decreases the concentrations of source chemicals on the ground and in lakes. VOCs present below the ground surface because of waste burial or percolation and VOCs transported by groundwater also will volatilize. The resulting soil gas occupies voids in the soil matrix and may move along vertical or horizontal paths toward areas of lower pressure. Some soil gas may exit the ground surface and become diluted in ambient air; it also may exit to buildings where gases could accumulate in enclosed space.
Non-VOCs (for example, phthalate and biphenyls) detected in soil at Corco Chemical tend to sorb strongly to the soil. Therefore, if contaminated soil erodes and runs off to the adjacent lake, the compounds could become part of the sediment. Another migration mechanism for the compounds may be suspension of contaminated particles by wind, vehicle, or worker activity and subsequent wind transport and deposition elsewhere to the groundwater or lake waters.
If non-VOC compounds detected in soils at Corco Chemical are in water or sediments in the adjacent lake, they may bioaccumulate in fish. The VOCs known to have been in area lake waters do not bioaccumulate substantively in fish. Likewise, VOCs found in groundwater would not tend to be taken up in garden produce irrigated with contaminated water.
B. HUMAN EXPOSURE PATHWAYS
ATSDR identifies completed and potential human exposure pathways by examining environmental and human components that lead to contact with contaminants. A pathway analysis considers five elements: a source of contamination, transport through an environmental medium, a point of exposure, a route of human exposure, and an exposed population. A completed exposure pathway exists when all five elements are evident and indicates that exposure to a contaminant has occurred, is occurring, or will occur. A potential exposure pathway exists when one or more of the five elements is not clearly defined but could be present. Potential pathways indicate that exposure to a contaminant is plausible and could have occurred, could be occurring, or could occur.
ATSDR's pathway analyses indicate that residents' and workers' use of contaminated groundwater is a completed exposure pathway in portions of the study area. Analyses also show that residents' or workers' activities associated with surface water, sediment, air, soil gas, and soil and their workplace are plausible potential exposure pathways.
Completed Exposure Pathway
Available information about groundwater sampling and well use indicates that during the early 1980s some of the persons in the study area that have relied on private wells for their potable water supply were exposed to substantive concentrations of VOCs in the groundwater. In 1990, ATSDR was shown the locations of approximately 40 residences and 15 businesses in the study area that relied on groundwater. The sampling data for 29 wells shows that most, but not all of the wells tested had contaminated water. Review of the latest available test data for each of the 29 wells that are still used for potable water suggests that 4 contain VOCs that exceed ATSDR's comparison values for selecting contaminants for exposure and public health evaluation and 1 contains VOCs that would exceed comparison values if the filter system malfunctions. An additional well contains VOCs that exceed comparison values, but current use of the well is uncertain. ATSDR believes the available data cannot be extrapolated with confidence to characterize water quality at unsampled well locations; hence groundwater for some other past and currently used wells in the study area that have not been sampled might also be contaminated at levels that exceed ATSDR's comparison values.
The water authority reports that by the end of 1992 more private well users in the study area had connected to the public water supply.
The primary exposure routes of concern for well users are ingestion of contaminated groundwater (drinking water and food) and inhalation (showering, bathing, and cooking). In residences where contaminated water is used for domestic purposes, such as showering and cooking, as well as drinking, the amount of exposure to VOCs through inhalation can be about as great as the exposure that occurs from drinking. Dermal contact is a less important exposure route.
Potential Exposure Pathways
Workers at the Meenan oil recovery well might have been exposed to VOCs while installing or removing equipment or from recovered groundwater and oil, primarily through inhalation and dermal contact.
People who swim, boat, or fish in nearby lakes might be exposed intermittently to VOCS, probably at low concentrations, primarily through incidental ingestion and inhalation if the water has been contaminated by chemicals in groundwater influent or by chemicals in industrial-type discharges. Potential exposure issues cannot be fully evaluated because sampling data either are not available, are too old, or are of insufficient quantity to determine whether water at any of the lakes presently contains contaminants at levels of concern. PADER reported to ATSDR they contacted various agencies in 1992 and could not determine whether nearby lakes are used for recreation--except for Van Sciver Lake which is used for swimming, boating and fishing. An area resident reported to ATSDR in 1990 that swimming and fishing had been observed in the past in the lake east of Corco Chemical and swimming had been observed in Warner Lake. Warner Lake is well posted in 1992 and appears to be quite shallow; thus, that lake may be unattractive for recreation in the future.
Two lakes that received industrial-type discharges are of primary interest. Relatively high concentrations of VOCs had been discharged over time from Corco Chemical to the lake to its east and from the Meenan Oil recovery well to Warner Lake. The reported lake water concentrations are much lower than the concentrations in the discharged materials because of the diluting effect of the large volume of receiving water and because of their continuing volatilization from the receiving water. Also, the major contaminant sources have been eliminated; Corco Chemical eliminated key sources of VOCs from its cooling water discharge, and Meenan Oil stopped its oil recovery operations. Very low levels of VOCs also have been recorded in the lake water west of Corco Chemical, the company's source of cooling water. Thus, the cooling water discharge is likely to transfer those VOCs unabated to the lake east of Corco.
Investigators suggest that locally, groundwater flows east and southeast toward Van Sciver Lake. Sampling information is not available for Van Sciver or lakes other than those previously described. However, contaminant levels would likely be low in Van Sciver Lake, and at any lake into which contaminated groundwater discharges, because of dilution and through volatilization of VOCs from the water to the atmosphere. The same would hold for any lake into which contaminated groundwater might discharge. However, actual water quality in the lakes has not been confirmed. The large lake that is west and southwest of the lake from which Corco draws its cooling water is posted along Penn Valley Road, which may inhibit its use for recreation.
Swimmers or waders may be exposed intermittently to contaminants in sediments in some lakes during such activities, primarily through incidental ingestion. Also, if sand and gravel removal is re-initiated at an affected lake, workers could be exposed intermittently to contaminants primarily through incidental ingestion and inhalation of contaminated particles. Sediment samples, obtained only at the lake east of Corco Chemical, contained low levels of VOCs. Sediments at Warner Lake could contain some VOCs because of the Meenan Oil discharge. Because the major contaminant sources at those two lakes have been eliminated, the current concentrations of VOCs in sediments may not be consequential. Sediments at Warner Lake also may be contaminated with non-VOC compounds from the Meenan Oil recovery operation, and the lake sediments west of Corco Chemical may be contaminated with PAHs or PCBs from runoff at that property. Runoff from the landfill to the lake east of Corco also may have released non-VOC contaminants to the sediments. Sampling data are not available to confirm whether such contamination exists.
Water used to put out the K-Mart fire produced water shown by analysis to contain low levels of a few VOCs. No pesticides or herbicides were detected. Because of the diversity of products stored, however, some non-VOCs probably also were contained in and transported by runoff, principally as residue of the burned materials or bound to soil particles. Runoff waters were carried from the area by a drainage system that ultimately discharges into the Delaware River. Runoff may have deposited chemicals as sediment along the drainage route and discharged others into the river.
The brief exposures that might have occurred to the low levels of VOCs in runoff water are unlikely to have been a health concern. Data are not available to evaluate what types of non-VOCs may have been in the runoff water; hence, exposure effects of non-VOCs, if any, cannot be evaluated. Therefore, the K-Mart runoff water is not addressed further in this public health assessment.
People could be exposed by inhalation to VOCs in ambient air in the workplace, and, possibly, in residences, but such exposure is less likely outdoors, where substances are rapidly diluted. In the workplace and at the landfill, people also might be exposed by inhalation to airborne contaminated particulates. Issues potentially associated with ambient air and soil gas cannot be fully evaluated because air monitoring data are not available, and soil gas and groundwater data are too limited.
VOCs in contaminated groundwater will tend to volatilize when well water is used in residences and businesses and will accumulate in indoor ambient air. Occupants at such locations would be exposed through inhalation. ATSDR's experience also shows that soil gas released from VOCs in contaminated groundwater and soil can enter residential and business structures and accumulate. Sampling near Corco Chemical has shown high concentrations of soil gas below ground on their property and diminished levels at the landfill, the farthest location evaluated. If soil gas does enter buildings, occupants would be exposed to contaminants through inhalation. At high concentrations, some types of gases conceivably might explode and cause physical harm.
Workers outdoors at Corco Chemical, potentially have been exposed intermittently to contaminants in soil, primarily through incidental ingestion or inhalation of contaminated particles. If soil contamination found around the Corco buildings extends as far as the Para Scientific building, any workers outside that building potentially could be exposed intermittently to contaminants in that medium. Likewise, if soils around the landfill contain contaminants, former workers there potentially were exposed intermittently through incidental ingestion and inhalation. Results of sampling conducted throughout the Pennwood Crossing area show that the Meenan Oil recovery operations did not contaminate surface soils. Thus, residents there should not be exposed to the chemicals of concern through the soil medium.
The K-Mart fire could have resulted in substantial residual soil contamination. If so, construction workers for the new building and grounds and maintenance workers potentially have been exposed intermittently, primarily by incidental ingestion or inhalation. No soil sampling data are available to evaluate those issues. Therefore, the K-Mart soil is not addressed further in this public health assessment.
Residents who consume garden produce irrigated with contaminated groundwater are not likely to be substantively exposed to the chemicals of concern because plants generally do not bioaccumulate VOCs. Although fish generally do not bioaccumulate VOCs in surface water either, they can bioaccumulate some non-VOCs through water and the foodchain. Thus, compounds such as the PCBs found at low levels in surface soils at Corco Chemical might have reached lake water and sediments west of the plant and been taken up by fish. Therefore, persons who consume fish from that lake might be exposed intermittently to selected non-VOCs that have bioaccumulated in edible tissue. Information about non-VOCs in that lake (and other lakes) and about fish consumption are not available to fully evaluate potential exposures.
Past and current workers at Corco Chemical, if they have not been adequately protected, may have been exposed intermittently through inhalation, dermal contact, or incidental ingestion to contaminants while processing or re-packaging chemicals. Para Scientific workers are not likely to be exposed in the workplace unless a substantive breakage occurred. Worker's safety- and health-related activities at active facilities like Corco Chemical and Para Scientific Company are subject to regulation by the U.S. Occupational Safety and Health Administration. Workers at the Meenan Oil recovery well and the closed landfill and possibly its oil reclamation unit also might have been exposed intermittently to a variety of chemical compounds in oil or in wastes. Truck maintenance workers also may have been exposed intermittently to contaminants from oil and various chemicals. Unless workers have exercised exceptional care, some chemicals could have been transported to their homes via clothing, and, in turn, family members may have been exposed intermittently by dermal contact, inhalation, or incidental ingestion. Information about past or present work practices and related sampling data for the locations identified are not adequate for determining whether potential exposures are of concern.
PUBLIC HEALTH IMPLICATIONS
Some citizens may be or may have been exposed to hazardous substances in groundwater at the Falls Township study area. Exposures may result from ingestion of contaminated water used for drinking or cooking; from inhalation of and dermal contact with contaminated water used for showering or bathing; or a combination of these routes. Contaminants of concern in the groundwater include PCE, TCE, and chloroform. As is further discussed in the Health Effects By Contaminant section of the Toxicologic Evaluation section below, chronic (long-term) exposures to those contaminants through drinking water may be a public health hazard; acute (short-term) exposures via drinking water pose no apparent public health hazard. Exposure to 1,1,1-trichloroethane through drinking water is at a level below public health concern. Exposure to benzene through drinking water is at a level below public health concern, but exposure to any human carcinogen such as benzene should be limited.
In addition, persons who use lakes (for swimming, boating, wading and fishing) to which groundwater has transported contaminants might also intermittently have been exposed, by incidental ingestion, inhalation, and dermal contact to 1,2-dichloroethane (1,2-DCA). Individuals who have swum regularly (more than once per week) in the lakes may have been exposed to 1,2-DCA at a level of public health concern. Without additional information about fish consumption habits and levels of contaminants in fish, ATSDR cannot evaluate the health risk associated with fish consumption, although VOCs such as 1,2-DCA are not expected to bioaccumulate in fish. Using the reference dose developed by EPA for contaminants, ATSDR staff members have determined that there is no apparent public health hazard to people using the lakes for other activities.
Workers at Corco Chemical may have been exposed to polycyclic aromatic hydrocarbons (PAHs) and polychlorinated biphenyls (PCBs) in soil around the facility. Further characterization of contaminant location in relation to occupational activities is necessary to identify whether an exposure of health significance is occurring or may have occurred.
Because Corco Chemical property is an occupational rather than a residential setting, the public's access to the soil may be limited. Based on the information provided, any current exposures associated with the soil do not appear to represent a health threat based on short-term exposure. The level and extent of soil contamination, as well as worker activity patterns, must be evaluated to determine whether there is a health hazard based on long-term exposure. Therefore, workers potentially exposed to contaminants in soil around Corco Chemical are not discussed further in this public health assessment.
A. TOXICOLOGIC EVALUATION
ATSDR staff have examined the available scientific literature on various adverse health effects associated with exposure to site-related contaminants. The adverse health effects are broadly grouped as cancer and non-cancer. Following is a brief discussion of adverse health effects in general followed by discussions of health effects associated with exposures to specific contaminants of concern by way of potential and completed pathways.
Cancerous and Non-cancerous Health Effects
Cancer is a group of diseases characterized by uncontained growth and spread of abnormal cells. Cancer cells multiply uncontrollably, destroying normal cells, and can spread from their site of origin to other parts of the body (27). A chemical capable of causing damage leading to cancer is called a carcinogen. The latency period, or amount of time between exposure and development of disease, ranges from a few years to decades (28). One contaminant of concern at the Falls Township study area, benzene, has been classified as a human carcinogen, a designation given to chemicals for which sufficient experimental evidence indicates its ability to cause cancer in humans. Several contaminants of concern, including PCE, chloroform, and 1,2-DCE, have been classified as probable human carcinogens--chemicals shown experimentally to cause cancer in animals. There is inadequate evidence on their human carcinogenicity; but, based on a lifetime exposure, they are suspected of causing cancer in people. One other contaminant of concern, TCE, has inconclusive evidence of causing cancer in animals or people. ATSDR estimates cancer risks by using the EPA's Cancer Potency Factor. That method assumes that animal data gathered under high-dose exposure conditions can be used to estimate the risk of low-dose exposures in humans (29). The method also assumes that there is no safe level of exposure (30). There is little experimental evidence to confirm or refute those two assumptions. Lastly, the method computes the 95% upper bound for the risk, rather than the average risk. In other words, there is a 95% chance that the risk of cancer is actually lower, perhaps by several orders of magnitude (31).
Chemical exposure may result in adverse health effects other than cancer. Those effects may be acute (resulting from a short-term exposure of less than 14 days), intermediate (resulting from an exposure of more than 14 but less than 365 days), or chronic (resulting from an exposure of at least a year's duration). ATSDR has developed Minimal Risk Levels (MRLs) for use as health comparison guidelines of contaminant exposure. An MRL is an estimate of daily human exposure to a chemical likely to be without an appreciable risk of deleterious non-cancerous effects over a specified duration (acute, intermediate, or chronic) of exposure. The mere presence of a contaminant does not imply that harm will result from exposure. A contaminant at a concentration lower than that chemical's MRL should pose no appreciable public health hazard with respect to non-cancerous adverse health effects.
Health Effects by Contaminant
Tetrachloroethylene (PCE)
Contaminated well water users may ingest tetrachloroethylene (PCE) and inhale PCE which has volatilized. The maximum level detected at this study area was 175 ppb (in a private well). PCE is a probable human carcinogen because animal studies indicate it is an animal carcinogen; however, available information is not adequate to determine whether exposure to PCE may be associated with cancer in people (32). ATSDR estimated the cancer risk associated with exposure to tetrachloroethylene at the Falls Township study area. We have determined that there is a low to moderate increased cancer risk, depending on the level of contamination in the well and on the length of time residents drank the contaminated water. Recent information indicates that PCE levels may have declined significantly in some private wells and is no longer adding to the cancer risk. However, other wells in the area have not been sampled and may contain levels of PCE that could be increasing the risk of cancer.
The maximum PCE concentration detected in the well water is below the ATSDR MRL for intermediate exposure via ingestion. Therefore, noncancerous adverse health effects would not be expected. ATSDR has not developed an MRL for chronic ingestion of PCE.
Limited information about the effects of acute exposure to PCE in people indicates that the central nervous system, liver, and kidney may be affected (32). Information on the exposure levels associated with those effects is inadequate, however, those levels are several orders of magnitude higher than those associated with the study area.
Indoor air concentrations of PCE were not measured. They are likely to be significantly lower than levels experimentally shown to be associated with adverse health effects in humans and animals (32). Inhalation exposure could contribute to an individual's total exposure to PCE.
Trichloroethene (TCE)
Contaminated well water users may ingest trichloroethene (TCE) and inhale TCE which has volatilized. The maximum level detected at this study area is 1,400 ppb (in a private well).
TCE is considered a possible to probable human carcinogen; EPA has not conclusively determined TCE's cancer classification.
Increased incidences of tumors have been observed in some animals experimentally exposed to TCE by ingestion or inhalation. Some laboratory studies indicate that some mice exposed to TCE by ingestion developed liver cancer. Additional studies in mice suggest that inhalation may result in liver and lung cancer. Some of the TCE studies use questionable methods and have inconclusive results, making it difficult to determine TCE's "true" carcinogenicity (33). ATSDR used available data to estimate the increased cancer risk associated with exposure to TCE at the Falls Township study area. We have determined that there is a low to moderate increased cancer risk, depending on the level of contamination in the well and on the length of time residents drank the contaminated water. Recent information indicates that TCE levels may have declined significantly in some private wells and is no longer adding to the cancer risk. However, other wells in the area have not been sampled and may contain levels of TCE that could be increasing the risk of cancer.
The maximum TCE concentration detected in well water at the study area is below the ATSDR MRL for intermediate exposure by ingestion. Therefore, noncarcinogenic adverse health effects would not be expected. ATSDR has not developed an MRL for chronic ingestion of TCE.
Indoor air concentrations of TCE were not measured. They are likely to be significantly lower than levels experimentally shown to be associated with adverse health effects in humans and animals (33). Inhalation exposure could contribute to an individual's total exposure to TCE.
Chloroform
Contaminated well water users may ingest and inhale chloroform. The maximum level detected at the study area was 142 ppb (in a private well).
Epidemiologic studies indicate a possible relationship between exposure to chlorinated drinking water and cancer of the bladder and large intestine and rectum in people (34). Chloroform is one of several VOCs frequently found in chlorinated drinking water that is considered to have carcinogenic potential, but it has not been identified as the sole or primary cause of excess cancer rates associated with chlorinated drinking water. Although available information is not adequate to determine whether exposure to chloroform may be associated with cancer in people, chloroform is a probable human carcinogen because animal studies indicate it is an animal carcinogen (34). Mouse studies indicate that ingestion of chloroform may result in liver cancer. Additional studies in rats suggest that ingestion may result in cancer of the kidney. ATSDR estimated the cancer risk of exposure to chloroform at the Falls Township study area. The maximum levels of chloroform found here may represent a slightly increased risk of cancer for anyone ingesting the water regularly for many years. Long-term oral exposure to chloroform may cause damage other than cancer to the liver and kidney.
The maximum chloroform concentration detected in the well water is below the ATSDR MRL for chronic exposure by ingestion. Therefore, noncancerous adverse health effects would not be expected.
Indoor air concentrations of chloroform were not measured. They are likely to be significantly lower than levels experimentally shown to be associated with adverse health effects in humans and animals (34). Inhalation exposure could contribute to an individual's total exposure to chloroform.
Benzene
Contaminated well water users may ingest benzene and may inhale benzene which has volatilized. The maximum level detected at the study area was 18 ppb (in a private well). Because numerous studies indicate that occupational exposure to benzene may be associated with an elevated incidence of leukemia, benzene has been classified as a human carcinogen (35). Those studies involved inhalation exposures; no studies have been found of cancer effects in people after oral exposure to benzene. Animal studies indicate oral exposure to benzene may cause cancer (35). ATSDR estimated the risk of cancer from benzene ingestion at this study area and concluded that the maximum levels of benzene found may represent a slightly increased risk of cancer for anyone ingesting the water contaminated at this level over a lifetime. Since recovery wells closed approximately 10 years after the spill was detected, ATSDR estimated the risk of cancer from benzene ingestion over a 10 year period and found no increased risk of cancer.
Studies addressing adverse health effects in people who have been chronically exposed to the levels of benzene associated with the study area were not found. Clinical effects associated with people who ingested benzene once have been reported (35, 36). In those reports, the concentrations of benzene were not noted, but estimated exposure doses were several orders of magnitude (an order of magnitude is 10 times) higher than exposure doses estimated to result from ingestion of benzene-contaminated water at the Falls Township study area. Effects following ingestion included coughing, stomach pain, nausea, vomiting, and adverse central nervous system effects. Additionally, studies with rats and mice indicate that chronic exposure to low levels of benzene is associated with adverse immunologic effects (35). Although the benzene levels in those experiments were low, they were still several orders of magnitude higher than levels associated with this study area. ATSDR does not expect adverse health effects related to benzene exposure at this study area.
1,2-Dichloroethane
Frequent swimmers in contaminated lakes may intermittently ingest and inhale 1,2-dichloroethane. The maximum level detected at the study area was 109 ppb (in lake water).
1,2-Dichloroethane (1,2-DCA) is a probable human carcinogen. Available information is inadequate to determine whether long-term exposure to 1,2-DCA in people may cause cancer (37). Mouse and rat studies indicate several tumor types were associated with exposure to 1,2-DCA (38). Cancers in experimental animals were noted at dosages several orders of magnitude higher than most persons using contaminated water in Falls Township would be expected to receive or have received. Swimmers may have been exposed during contact with the water of Warner Lake; however, ATSDR is not aware of any individuals who swim in Warner Lake regularly. In addition, not enough samples have been analyzed to determine the spatial and temporal patterns of lake water contamination. Because of the lack of information, health effects associated with swimming exposures cannot be evaluated, but ATSDR expects those exposures to be below levels of health concern. Individuals using the lake for other purposes (e.g., boating) would not have received a significant exposure.
B. HEALTH OUTCOME DATA EVALUATION
ATSDR has identified health outcome data (cancer incidence and mortality rates) for Bucks County and the state of Pennsylvania; ATSDR also has identified an investigation of a reported cancer cluster in Falls Township being conducted by the State Cancer Control Program. Finally, ATSDR has received and reviewed medical records from individuals and/or their legal representatives. Those data are discussed in the following paragraphs.
Cancer Incidence and Mortality Rates, Bucks County
The community has expressed concern about elevated incidence of cancer, particularly cancers of the breast, lung, and stomach. Cancer mortality data are available from 1970 through 1986. The only years for which final county-specific cancer incidence data in Pennsylvania are available are 1985 and 1986. Neither cancer incidence in nor mortality of Bucks County residents in 1985 were significantly higher in any categories of community concern than for all residents of Pennsylvania and, in some cases, were significantly lower. Rates reported to ATSDR by the Pennsylvania Cancer Control Program are in Table 7.
Although evaluating county cancer levels may constitute a first look at the community's health, because the exposed population consists of only a fraction of the county, county rates are an inaccurate proxy for community rates. In addition, it is possible that cancers that may result from exposure to site-related contaminants might not be detectable as of 1985 because of the latency of cancer development.
Review of Residence Information from Cancer Incidence Data for Falls Township in Bucks County, Pennsylvania
The Cancer Control Program at the Bureau of Epidemiology and Disease Prevention, Pennsylvania Department of Health, has investigated a possible cancer cluster in Falls Township as identified by members of two citizen groups, Bucks People United to Restore the Environment (B-PURE) and the Delaware Valley Toxics Coalition (DVTC), with the assistance of the Pennsylvania Department of Health. The groups analyzed available cancer incidence data for 1985 and 1986. An incidence rate measures the number of people who develop cancer as compared to the number of people in the community. The Cancer Control Program staff members reviewed the data, paying special attention to possible inaccuracies in residence information. They found no significant increase in cancer incidence rates in Falls Township. The review concluded that the apparent increase in cancer rates in Falls Township originally found by the citizen groups may be attributed to significant misassignments of residence, and that the overall risk of developing cancer in Falls Township is not significantly different from the County (39).
Medical Records
ATSDR received medical records from individuals and/or their legal representatives. ATSDR reviewed medical records or medical record summaries of five children who live, or have lived, in the study area. Physical exams indicated enlarged and tender livers, rashes, and, in one case, failure to thrive; bloodwork indicated elevated liver enzyme levels; and complaints included abdominal pain. ATSDR also reviewed medical records or medical record summaries of 15 adults who live, or have lived, in the Falls Township study area. General complaints included stomach problems, diarrhea, poor liver function, rashes, headaches, and joint stiffness. Two patients suffered from heart disease, two from hypertension, and one had a confirmed diagnosis of breast cancer. As was discussed in the Community Health Concerns section, residents reported concern about several additional types of cancer.
Individuals' adverse health conditions are of concern, and many of the symptoms noted in the medical records and summaries or brought to ATSDR's attention by individuals are known effects of exposure to chemicals contaminating the groundwater at the Falls Township study area. It cannot be established that those noncancer health effects could have resulted from estimated exposures associated with the study area because current findings in the literature indicate that the levels are too low to cause health effects. In addition, because of low contaminant levels and the long latency period of cancer, it cannot be established that the reported cancers are a result of exposures to site-related chemicals.
Further activities have been recommended to investigate adverse health effects in the community. Refer to the Public Health Actions section for more information.
C. COMMUNITY HEALTH CONCERNS EVALUATION
Some community health concerns were evaluated in the Health Outcome Data Evaluation section. Remaining concerns are about birth defects and miscarriages.
The causes of individual birth defects are often difficult to identify. ATSDR staff are not aware of any studies indicating that the contaminants of concern identified in this PHA cause birth defects. In addition, if exposure to contaminants ceased before conception, unless exposure caused damage to the egg, the defects cannot be attributed to exposure. ATSDR medical staff are not aware of any studies which associate exposure to the contaminants of concern with egg damage. This does not mean that the birth defects are certainly unrelated to exposure; it means only that ATSDR cannot determine whether they are associated.
The birth defects described could all possibly have genetic components. Genetic testing of the
sisters by their physicians could confirm or rule out that component. In addition, exposure of the
mother to other insulting chemicals, such as alcohol or cigarette smoke, can affect the
development of the fetus. Likewise, exposure of the father to chemicals affecting sperm can
affect fetal development. Major birth defects are not uncommon; they occur in 6%-7% of live
births (40). Similarly, miscarriages are common in the population--approximately 15% in the
population of otherwise healthy mothers (40, 41). There are a variety of reasons for miscarriages,
including aspects of the mother's health status which are unrelated to exposure (e.g.
endometriosis). In conclusion: ATSDR cannot determine whether the birth defects and
miscarriages are related to exposure because other common causes of birth defects and
miscarriages have not been ruled out.
