PETTITIONED PUBLIC HEALTH ASSESSMENT
KOPPERS COMPANY, INCORPORATED (OROVILLE PLANT)
[a/k/a KOPPERS (OROVILLE)]
OROVILLE, BUTTE COUNTY, CALIFORNIA
The following sections contain evaluations of the environmental and health outcome data available for the Koppers site. In preparing this evaluation, ATSDR used established methodologies for determining how persons may contact (or be exposed to) potential contamination related to the Koppers site and what harmful effects, if any, may result from such exposure. Chemical exposure pathways (or routes of physical contact with chemicals) that ATSDR evaluated are ingestion, inhalation, and skin (dermal) contact.
If exposure was or is possible, ATSDR then considers whether chemicals were or are present at levels that might be harmful to people. ATSDR does this first by screening the concentrations of contaminants detected in an environmental medium against their health-based comparison values. ATSDR's comparison values are often based on animal studies because relevant human data are lacking. Comparison values are therefore derived using very conservative assumptions and often have large safety factors built into them to be protective of human health. Some comparison values may be hundreds or thousands of times lower than exposure levels shown to produce effects in laboratory animals. Although concentrations at or below the relevant comparison value may be considered safe, any concentration that exceeds a comparison value would not necessarily be expected to produce adverse health effects in the community. Chemicals detected above comparison values simply require a more detailed evaluation of site-specific exposure conditions. ATSDR also emphasizes that regardless of the level of contamination, a public health hazard exists only if people come in contact with, or are otherwise exposed to, harmful levels of contaminated media.
For a more detailed discussion of ATSDR's public health assessment methodology (quality assurance considerations, human exposure pathway analyses, ATSDR's health comparison values, and the methods of selecting contaminants of potential health concern), refer to Appendix C.
ATSDR identified ways in which on-site workers and nearby residents may have come in contact with site-related contamination, including:
- Ingestion and skin contact with PCP in groundwater both on and off property.
- Short-term exposures to possible contaminants in smoke during past fire episodes.
- Ingestion of homegrown farm products (e.g., chickens and eggs) containing dioxins.
- Limited contact with contaminated on-site soils by site employees.
- Limited contact with surface water and sediment in streams along the site periphery.
Site clean-up activities have essentially eliminated these exposures. CDHS continues, however, to research the significance of dioxin levels detected in chickens and eggs in the Oroville area where chickens have access to soils. A summary of exposure pathways that ATSDR evaluated for the Koppers site is presented in Table 1 and described in more detail below. In the subsections below, ATSDR's conclusions for each topic are presented in italicized text followed by more detailed discussion.
- Groundwater beneath the Koppers property is contaminated with chemicals associated with wood treatment processing activities. PCP contamination has been documented in on-property supply wells since 1971. PCP-contaminated groundwater was first reported in off-property domestic wells immediately south of Koppers in 1973 and in domestic wells further south in 1983.
- Alternate drinking water supplies were provided on site starting in the mid-1970s and to potentially affected residences south of Koppers beginning in 1984. Some exposures to varying levels of PCP could have occurred before provision of alternate water supplies.
- ATSDR's review of available groundwater and biologic monitoring data indicate that no long-term (chronic) health effects are expected as a result of past exposures to the PCP levels detected in groundwater.
- In the past, residents reported various symptoms (e.g., headaches, rashes) perceived to be associated with their drinking water supplies. Although no firm evidence exists linking known exposures and reported symptoms, ATSDR does not have sufficient data on past exposures to categorically rule out the association. However, ATSDR understands that the reported symptoms have since subsided and no evidence exists that any chronic effects have occurred as a result of past exposures.
- Most groundwater in off-property areas now meets health-based clean-up goals. Groundwater will continue to be treated, monitored, and have restricted uses, as necessary, to ensure the continued protection of public health.
In drawing these conclusions, ATSDR reviewed the following information:
- Site hydrogeology (What aquifers exist beneath the site and in what direction[s] does groundwater flow?
- Groundwater use (Who may have been exposed and when?)
- Groundwater quality (What contaminants were detected and at what levels? What has been done to restore the groundwater quality?)
- Health implications (Could exposures result in adverse health effects?)
Groundwater is present in three aquifer zones under the Koppers property (B', C', and D'). The most shallow aquifer is the B' aquifer, which consists of 60-80 feet of gravel and sand. Groundwater flow from the B' aquifer to the lower C' and D' aquifers is slowed by "aquitard" formations. Four aquifer zones are present off property (A, B, C, and D). The shallowest A aquifer exists off property only in the southern and western portions of the study area. Groundwater flow in all the aquifers is generally in a southerly direction (Ebasco 1988).
Groundwater is used to supply domestic, irrigation, and industrial wells at and in the vicinity of the Koppers site. On-property groundwater was not used for drinking purposes after the mid-1970s, although the specific date when use ended is not clearly documented (HSI GeoTrans 1997). Since the mid-1970s, groundwater at Koppers has been used only for industrial purposes (spraying of lumber and other process activities) and possibly personal hygiene purposes (e.g., hand washing) (HSI GeoTrans 1997).
The closest domestic wells downgradient (south) of the Koppers plant are in the vicinity of Baggett Marysville Road (also known as Ophir Road). The next closest in-use domestic wells directly downgradient of the site are nearly 2 miles further south. Several additional domestic wells border Lone Tree Road (southwest of site). Figure 3 in Appendix A shows the types and locations of wells in the site area.
All domestic wells in the area were used for drinking water until March 1984 when Koppers provided bottled water to homes that had known or possible PCP contamination. Domestic well water was still used for bathing and irrigation (Ebasco 1988). Koppers installed the "Herfi Test Well" in March 1984, in the deeper aquifer, to provide water to the seven households with the greatest PCP contamination (HSI GeoTrans 1997). In February 1986, approximately 50 households in the area were connected to the Oroville-Wyandotte Irrigation District (OWID) public water supply, of which 34 received subsidy from Koppers (EPA 1997b; CDHS 1997a).
The provision of alternate water supplies (i.e., bottled water, Herfi Test Well water, and ultimately OWID water) beginning in 1984 should have largely eliminated current exposures.
Groundwater quality at and around Koppers has been monitored since the early 1970s. In the course of off-property investigations, 128 domestic wells have been sampled, primarily south of Koppers (HSI GeoTrans 1997). The most extensive sampling occurred during the remedial investigation (RI).
PCP contamination was first reported in on-property supply wells in 1971. The highest detected PCP concentration in the three on-property supply wells was 3,140 ppb (SW-1). Average concentrations in SW-1 and SW-2 ranged from 118 to 364 ppb. The highest PCP concentration detected in SW-3 was 2 ppb (CDHS 1997b).
PCP contamination was first identified in a domestic well immediately adjacent to Koppers (well 31C2) in 1973 (Ebasco 1988; EPA 1998). Well 31C2 was monitored on roughly a monthly basis from 1973 to 1983. PCP concentrations during that period ranged from <1.0 ppb to 2,100 ppb (1974), with reported concentrations between 1973 and 1975 averaging 228 ppb. PCP concentrations did not exceed 60 ppb after 1975.
Domestic wells farther south of Koppers were not monitored until late 1983 when Koppers initiated a well-inventory and sampling program in response to taste and odor complaints (CDHS 1997b; Schmidt 1984). Sampling for PCP and other contaminants of 25 domestic wells south of Ophir Road revealed detectable levels of PCP in five active wells (wells 59, 60, 61, 62, and 81). Investigators concluded that the source of contamination was Koppers. Contamination was believed to have left the plant in the early 1960s and taken approximately 20 years to reach these locations about 2 miles south of the property line (Schmidt 1984).
The highest level of PCP (5,500 ppb) was detected in Well 59 in December 1983, but was measured below 28 ppb in the same well in 1984 (8 sampling events), below 8 ppb in 1985 (5 sampling events), and below detection limits or at trace levels in RI samples collected after 1985. Wells 60, 61, 62, and 81 also had consistent hits of PCP during the 1980s, with concentrations ranging from <1 ppb up to 640 ppb; post-January 1984 samples detected less than 210 ppb PCP (EPA 1998).
EPA and contractors have questioned the quality of some of the early groundwater data (pre-RI). Specifically, some false positives may have been reported (i.e., actual PCP concentrations in groundwater may have been lower than those reported by the laboratory) (EPA 1997b; Ebasco 1988).
In 1986, a more comprehensive groundwater study was initiated as part of the RI. During the RI, PCP was detected both on and off property in a narrow plume extending approximately 2 miles south of the Koppers property along Lone Tree Road up to Palermo Road, with the highest concentrations detected in the B/B' aquifer (see Figure 4 in Appendix A). Samples were collected from 39 on-property monitoring wells and 105 off-property monitoring and domestic wells (Ebasco 1988) and analyzed for volatile and semivolatile organic compounds, metals, and dioxins/furans.
Benzene, dioxins, polycyclic aromatic hydrocarbons (PAHs), and PCP exceeded comparison values in on-property monitoring wells. Arsenic, PAHs, and PCP exceeded comparison values in some off-property monitoring wells, but only PCP was detected above comparison values in domestic wells, and it was detected consistently in only seven domestic wells (see Tables 2 and 3 in Appendix B). PAHs were detected at concentrations slightly above comparison values and dioxins/furans were detected at concentrations below comparison values in isolated domestic wells, but the monitoring data did not indicate a distinct plume of contamination (Ebasco 1988). Investigations have therefore focused primarily on PCP. Wells, both domestic and monitoring, have generally been monitored on at least a quarterly basis.
To clean up the PCP plume, Beazer installed two groundwater pump and treat systems (one on- and one off-property). Monitoring well data indicate that groundwater conditions have continued to improve since treatment began in 1993. Contaminant concentrations and the size of the plume continue to decrease. As can be seen in Figure 3, the "plume" of PCP contamination in groundwater has receded. Residual PCP contamination still exists to the south of the Koppers property boundary, but not near any domestic wells currently in use. EPA requires that Beazer continue to treat groundwater beneath the Koppers' property and the residual contamination in off-property groundwater and to monitor area groundwater to ensure that the quality of the water is maintained (Dames & Moore 1996; EPA 1997a, 1997b, 1998, 1999a, 1999b, 2000; and HSIGeoTrans 1999).
Groundwater monitoring requirements vary depending on the contaminant concentrations detected in individual wells. Quarterly monitoring is required for wells in which contamination continues to be detected at levels exceeding the cleanup goals. If contamination is not detected in a given well for four consecutive sampling periods (quarters), then sampling of that well will be required only every 6 months. After 4 years of non-detects, sampling of that well will be required every 2 years. If at any time contamination is detected, quarterly sampling will be resumed in that well (EPA 2000).
In April 1998, EPA informed 26 of the 34 well owners that their well water or the nearest monitoring well indicated PCP concentrations less than 1 ppb for 12 consecutive months and that the restriction on well use for domestic drinking water for these properties had been removed. Owners can exercise their prerogative about returning to use of well water or continuing to use OWID water. The remaining eight domestic wells will undergo further review and evaluation. Well owners will continue to be paid an OWID allotment until it is demonstrated that their wells meet the clean-up criteria for 12 consecutive months. Five of these deactivated wells are in the area of the residual plume (EPA 1999a, 1999b).
The municipal water currently being used by area residents meets drinking water standards; therefore, no current or potential future exposures to harmful levels of groundwater contaminants are expected. The OWID water supply, which currently serves many area residents, is fed by water from surface reservoirs in foothills east and upgradient of the site. OWID is required under EPA's Safe Drinking Water Act and state and local regulations to regularly test the public water supply and maintain safe water.
Some Koppers workers and residents whose domestic wells are in the path of the PCP plume may have been exposed to PCP-contaminated groundwater in the past through drinking or by skin contact, although the time over which people may have been exposed is uncertain. The 1970s is the most likely exposure period for on-property groundwater exposures. For affected domestic wells located immediately south of Koppers, some exposures may have occurred as early as 1973, but persons living further south are not thought to have been affected before 1983. Some uncertainty exists regarding actual exposure levels because sampling data are not available before 1983 for most domestic wells.
To evaluate potential past exposures, ATSDR looked at PCP levels to which people may have been exposed. Because PCP levels fluctuated over time, using the average PCP levels measured in individual well during the multiple sampling rounds better reflects the amount of PCP that people could have been exposed to over time. The highest average PCP concentration in any of the off-property domestic wells was approximately 200 ppb, based on data available from 1973 to 1986. Average concentrations in the most contaminated on-property supply wells (SW-1 and SW-2) ranged from 118 to 364 ppb.
These estimated exposure concentrations are lower than the health-based comparison value of 1,000 ppb for noncancer, long-term health effects. That is, no chronic health problems would be expected if 1,000 ppb PCP in water were ingested on a daily basis over the course of a lifetime. Exposures south of Koppers were for shorter time periods and at lower concentrations.
ATSDR also evaluated potential health implications of shorter-term (acute) exposures to the highest PCP levels measured in certain wells. Before using alternate water supplies, some residents may have been exposed to PCP in groundwater at levels as high as 5,500 ppb. Some higher PCP detections exceeded comparison values for acute (1-14 days) and intermediate (14-365 days) exposures. ATSDR reviewed the health literature to better understand whether health problems might be associated with these highest detected levels of PCP.
While numerous studies exist that have identified adverse health effects associated with PCP exposures in the work setting or in homes with PCP-treated wood, little human data exist describing the effects of ingesting PCP in drinking water. Therefore, in evaluating potential health effects associated with drinking water exposures, ATSDR relied on data from experimental animal studies to provide additional perspective.
Available animal studies show effects at doses above 1 milligram per kilogram of body weight per day (mg/kg/day) (largely liver and kidney effects) (ATSDR, 1994). Exposures to 5,500 ppb PCP would result in a dose lower than that. If we assume a child weighing 16 kg (35 pounds) drank one liter of water containing 5,500 ppb PCP, exposure doses would be approximately 0.3 mg/kg/day. A 70 kg adult would receive a slightly lower dose of 0.2 mg/kg/day. These calculations do not account for children growing over time, in which case estimated doses would be less; assume all fluids come from well water; and assume no fluctuation in the PCP concentration in well water.
Determining whether symptoms that some residents reported in the 1980s (e.g., rashes, headaches) were linked with past exposures to domestic well water is difficult. Studies are not available that evaluate the health effects associated with contacting PCP at the levels detected in the vicinity of the Koppers site. Most available studies looked at exposures to PCP solutions versus diluted levels in potable water. One study indicates that contact with a 0.4% (or 4,000,000 ppb) PCP solution resulted in localized redness and pain after a 10-minute exposure (Bevenue et al. 1967), but this is approximately 700 times higher than the highest PCP level detected in domestic wells. Other studies have shown various effects in people exposed to PCP via treated wood, but no specific exposure data are available (ATSDR 1994).
ATSDR does not have sufficient data on past exposures to rule out the possibility that they were associated with the symptoms reported but believes that if any outcomes were triggered by exposure to domestic water supplies, the outcomes were quickly reversible once exposure ended. To the best of ATSDR's knowledge, no recent health complaints have been received from people living/working in the vicinity of the Koppers site. No specific studies or surveys, however, have been conducted to collect health status information from area residents.
ATSDR also looked at potential cancer threats posed by PCP at measured levels. ATSDR does not believe that PCP levels were high enough or exposure was long enough to cause cancer. PCP has been classified as a "probable human carcinogen." Although inadequate data exist to show a link between PCP and cancer in people, PCP has been shown to cause cancer in laboratory animals under experimental conditions where high doses were administered. Because possible effects resulting from lower level exposures are difficult to study, scientists are uncertain about the effects of such exposures, especially in humans. That is why screening values (e.g., ATSDR's Cancer Risk Evaluation Guide) are set very low to help ensure that people are not exposed to contaminant levels even remotely close to effects levels seen in experimental studies. It is important to note, however, that even if a 150 pound (70 kg) person were exposed to the average PCP concentration of 200 ppb detected in area groundwater (drinking 2 liters per day, every day, over the course of a lifetime), estimated doses would be 0.006 mg/kg/day, which are 3,000 times lower than the lowest levels resulting in cancer in laboratory studies (18 mg/kg/day) (NTP 1989). Lastly, cancer statistics compiled for the census tracts in the vicinity of Koppers through 1989 have not shown increased cancer rates (see Health Outcome section below for further discussion on available cancer statistics, including their limitations).
- Air samples collected during the RI revealed no harmful levels of contaminants.
- During site fires, site workers and neighbors reported adverse health effects, but no air data are available from the time of the fires to enable an evaluation of possible causative factors.
- Blood and urine samples collected following the 1987 fire showed background PCP levels. Health effects appear to have been short-term and reversible with no lasting adverse health effects.
The only site-related air data were collected from June 1986 through January 1987 as part of the RI. The contaminants detected and their concentrations are listed in Table 4 in Appendix B. No contaminants exceeded available comparison values, indicating that no health threat was posed by air releases at that time. No air data are available to evaluate conditions at other times. This represents a data gap. A brief overview of air issues is presented below.
On-site Air: The operation of the aeration lagoon and two wastewater spray fields and production-related spills and drippings of contaminants over past years may have been a source for air emissions, but, again, because of limited air data, full evaluation is not possible. Site cleanup efforts are assumed to have eliminated the primary source(s) of air releases.
Off-site Air: Past fires periodically contaminated ambient air surrounding the site. Although no historical off-site air monitoring data are available, residents report that past fires reduced ambient air quality. Using observations and analysis of meteorologic data at the time of the 1987 fire, CDHS concluded that wind and atmospheric conditions likely resulted in dispersion and dilution of air contaminants in a southerly direction the morning of the fire and then in a northern direction late in the fire. Following the fire, CDHS conducted a health screening, documenting health complaints (including headaches, nausea, and skin irritations) of residents exposed to contaminated ambient air (CDHS 1987). Although temporary health effects were reported, blood and urine sampling did not verify exposures to PCP, which was a chemical of potential concern. These findings are discussed in more detail in the Health Outcome section below.
Because air and exposure data immediately following the fire are lacking, no firm conclusions can be made regarding the cause of reported health effects. Exposures and effects were of short duration, however, and no long-term adverse health effects resulting from exposures during the fire are expected.
- Limited sampling of home-produced eggs and poultry revealed elevated levels of dioxins/furans. Detected levels were high given the fact that 1) dioxins/furans were present in only trace amounts in soils where chickens forage, and 2) chicken feed, water, and bedding were not contaminated.
- A review of the available sampling data and toxicologic literature shows that consumption of products with the detected levels of dioxins/furans is not likely to result in adverse health effects. However, given some of the uncertainties regarding the potential harmful effects of low-level dioxin exposures, residents are encouraged to adhere to the 1989 CDHS health advisory for the Oroville area that warns of potential chicken and egg contamination and lists precautions to be taken.
- CDHS continues to evaluate additional area soil and egg data collected more recently. If necessary, ATSDR will re-evaluate its current conclusions based on the findings of CDHS's latest study.
Summary of Sampling Results
CDHS conducted a study after the 1987 Koppers fire in which it examined dioxin/furan concentrations in backyard-raised eggs in the greater Oroville area. In 1987 and 1988, CDHS collected egg samples from 25 locations, including two "index homes" (those in close proximity to Koppers and to detected trace levels of dioxins/furans in soils). CDHS also collected commercial eggs and a limited number of chicken, beef, and milk samples.
- The dioxin/furan concentrations in chicken eggs from Index Homes #1 and #2 ranged from 5.6 to 18.26 parts per trillion (ppt) measured in International Toxicity Equivalents (ITEQ) on a whole egg basis. The average of those values was approximately 10 ppt.(1)
- Eggs collected from 23 locations within an 11 kilometer radius of Koppers contained between 0.08 and 13.16 ppt ITEQ.
- The average concentrations in eggs from the 25 locations was 2.8 ppt ITEQ, including the two index homes.
- Commercial eggs from five stores in the Oroville area contained less than 1 ppt ITEQ.
- The fat of chickens from the two index homes measured 177-228 ppt ITEQ.
- Single beef and milk samples provide too limited a data set for evaluating public health significance. Dioxins were detected in a pre- and post-fire cow liver sample at similar levels, suggesting a pre-fire source of dioxins.
Sampling and analysis of chicken feed, bedding, vegetation, and water revealed no dioxin contamination. CDHS concluded that the low concentrations of dioxins and furans in soil likely led to bioaccumulation in animals in contact with the soil (Stephens et al. 1990; CDHS 1991). A controlled follow-up study conducted by CDHS in an Oroville backyard supported this premise; specifically, eggs from chickens raised in a small cage with soil contact had higher concentrations of dioxins in their eggs than did eggs from chickens with no soil contact (Petreas et al. 1996). CDHS continues to evaluate the nature and extent and possible sources of dioxin/furan contamination in area soils and eggs. CDHS is compiling the results and analyzing the data for geographic and temporal trends and will provide recommendations for public health action. If requested, ATSDR will review any new data as it becomes available.
In 1989, CDHS issued a Health Advisory which recommends that people reduce their consumption of home-produced eggs, chickens, and beef. It also recommends changing practices that may contribute to dioxin exposure, including raising chickens away from soil and discontinuing placement of PCP-treated wood in areas where animals forage. Because contaminated eggs also have been found outside of South Oroville and the full extent of contamination is unknown, the Health Advisory applies to all of Oroville and the surrounding area, including South Oroville, Palermo, Thermalito, and East Biggs.
According to CDHS, the extent to which area residents may have consumed home-raised eggs varies. Before the advisory, some residents reported consuming no home-produced eggs, and some households reported consuming 3-35 eggs per week (CDHS 1997c; Goldman et al. 2000).
To evaluate possible health implications, ATSDR studied the levels of dioxins/furans detected in eggs in the advisory area. The toxicology of dioxins/furans is somewhat complicated (see toxicologic profile in Appendix E). Dioxins/furans exist in a variety of forms (or congeners), with some forms shown to be more harmful than others. Using rough estimates of the relative toxicity of selected congeners of chlorinated dioxins and furans based on very limited animal and in vitro studies, regulators have devised a way of weighting the concentrations of different congeners against the most harmful form (TCDD) and expressing the total concentration of selected CDDs and CDFs in a mixture in terms of toxicity "equivalents" (i.e., the ITEQ concentration mentioned above).
The TEQ approach provides a convenient tool for screening and regulation, but uncertainties regarding differences in congener toxicity and the mechanism of toxicity in humans make it less helpful in predicting human health effects, which is the purpose of this public health assessment. Therefore, ATSDR also reviewed the available congener-specific egg data to provide additional perspective.
The highest ITEQ concentration detected in tested eggs was 18.26 ppt, or 18.26 picograms/gram (pg/g). Consumption of eggs with this level of dioxins is not expected to result in adverse noncancer health effects. An adult (weighing 70 kg) who eats one egg per day (weighing 70 grams) would receive a dose of 18.26 pg per kilogram of body weight per day (pg/kg/day), which is below ATSDR's conservatively derived screening values or Minimal Risk Levels (MRLs) for acute and intermediate effects. This dose does exceed ATSDR's chronic MRL for dioxins of 1 pg/kg/day. However, ATSDR's chronic MRL is approximately two orders of magnitude (i.e., 100 times) lower than the effects levels in experimental and epidemiologic studies. The MRL is set low to account for recognized areas of uncertainty in dioxin toxicity (see Appendix E). The chronic MRL is essentially equal to doses associated with average background exposures. In addition, MRLs are even more conservative when, as in this case, they are applied to exposures of ITEQ concentrations rather than to the congener (i.e., 2,3,7,8-TCDD) on which they are specifically based. Furthermore, available testing data show that egg consumers were likely exposed on average to lower levels of dioxins in eggs.
In looking at the data collected in 1987 and 1988, ATSDR found that approximately 98% of the total TEQ in eggs consists of the hepta-, hexa-, and penta- forms of dioxins/furans. All known health effects observed in rodents for these congeners are greater than 100,000 pg/kg/day, or at least 7-200 thousand times higher than the estimated average daily exposure from eating one egg per day in the vicinity of Koppers (ATSDR 1998) (See Table 6.) Humans are thought to be at least 10 times less dioxin-sensitive than most laboratory animals (Aylward et al. 1996; Kimbrough 1992; Roberts et al. 1985). See Appendix E for a further discussion of the limitations of using animal data to assess the human health impacts of dioxins.
Lastly, ATSDR considered the cancer-causing potential of dioxins. 2,3,7,8-TCDD is classified by EPA as a probable human carcinogen. This means that sufficient animal data exist linking cancer effects with dioxins. However, human data are not sufficient to conclude that dioxins produce cancer effects in people. The lowest cancer effect level recorded to date in experimental animals is at doses greater than 100,000 pg/kg/day (See Appendix E concerning IARC's recent reclassification of TCDD). The carcinogenicity of the other dioxin/furan congeners is less well studied, but thought to be less potent than 2,3,7,8-TCDD. As mentioned above, daily doses from consuming one egg per day with the detected levels of dioxins (at a maximum TEQ concentration of 18.26 ppt) would be much lower than doses associated with cancer effects in laboratory animals, and 2,3,7,8-TCDD is only a very small fraction of the dioxin mixture measured in tested eggs.
Scientists continue to study the potential harmful effects of low-level dioxin exposures and to evaluate the similarities of responses in animals and humans. In light of uncertainties, it is prudent public health practice to limit exposures to the extent possible. Therefore, ATSDR recommends that residents continue to adhere to the CDHS advisory.
- Past processing activities resulted in varying amounts of contamination of site soils.
- Employees may have come in contact with contaminated surface soils in the past, but contact is believed to have been infrequent, especially with the most contaminated soils. Therefore, long-term adverse health effects are not likely.
- Cleanup measures have reduced soil contamination to safe levels for industrial land use. The cleanup included the removal of debris resulting from past fires.
- The low levels of PCP and dioxins/furans detected in a limited number of off-site soils were well below levels known to result in health problems resulting from soil contact.
PCP (up to 33,000 parts per million [ppm]), PAHs (up to 600 ppm), dioxins (up to 0.118 ppm), and arsenic (up to 681 ppm) were detected in on-site soil (see Table 7 in Appendix B). Detected levels exceeded comparison values for residential and industrial soils.
Soil samples were collected at increments of 6 inches to a depth of 40 feet below the ground surface. Sampling occurred in the area of the processing facilities where the fires occurred and where the treated lumber was stored. Figure 5 (Appendix A) shows soil sampling locations. The highest soil levels of dioxins/furans were observed in samples from the process area. The highest concentrations of most chemicals were detected in the top 2 feet of soil (Ebasco 1988).
Contaminated soils have been excavated and safely placed in an on-site landfill. After the removal of contaminated soils, clean soils were brought in and affected areas were regraded (TRC 1998). On-site soil contamination no longer serves as a potential source of off-site contamination, and on-site soils present no public health hazards assuming continued industrial/commercial use of the site.
In the past, Koppers workers may have been exposed to contaminated soils in areas used for waste chemical storage, Cellon process treatment, and areas of processed lumber storage, as well as in areas subject to fires and used for burial of fire debris, but direct contact is believed to have been infrequent (Ebasco 1988). Some individuals working in these areas may have been exposed to contaminants through skin contact, ingestion, or inhalation of soil particles. Most worker exposures were likely sporadic and for short periods and therefore not likely linked with any long-term adverse health effects. Although trespassers could have accessed the site, they would not likely have come into contact with soils in these areas on any regular basis.
During the RI, a limited number of off-site soil samples were collected north of the facility to establish background levels of metals. Following the 1987 fire at Koppers, EPA and CDHS sampled off-site soils for PCP and dioxins/furans (CDHS 1987). Available soil data are summarized in Table 7 of Appendix B.
PCP was detected in only a few samples and at levels below the comparison value for PCP of 6 parts per million (ppm) (CDHS 1987). CDHS sampling of soils in South Oroville for dioxins/furans following the fire revealed concentrations ranging up to 46 parts per trillion (ppt) (or 0.000046 ppm TEQ) (CDHS 1987). Because contaminants in off-site soils were detected below the ATSDR comparison value of 0.00005 ppm for ingestion of soil, no adverse health effects would be expected from contact with these soils.
- Workers and site trespassers were unlikely to have come in contact with levels of contaminants in site surface waters or sediment that would result in adverse health effects. Exposures, if any, were likely of short duration and frequency.
Surface water runoff and suspended sediment from the Koppers facility flow primarily southwest and leave the site via a single drainage ditch. The ditch leads westward toward the neighboring L-P facility and on toward the Feather River. During the RI, surface water and sediment samples were collected from more than 20 locations, many in areas thought to be most affected by on-site contaminant sources (Dames & Moore 1988).
Some contaminants were detected in the surface water of the ditches and seasonal ponds and streams on the western periphery of the site. The surface water just southwest of the Koppers Fire Pond and L-P Ditch on the western site boundary contained the highest concentration of PCP in surface water. The maximum concentrations of dioxins/furans in surface water were approximately the same in the various ponds and ditches, as were the inorganic substances (e.g., arsenic). The concentrations of contaminants detected in surface water are listed in Table 8 in Appendix B.
As shown in Table 7 in Appendix B, PAHs, dioxins, and arsenic exceeded comparison values in sediments. The highest concentration of PCP was detected in the on-site Koppers ditch while the greatest concentrations of dioxins/furans and PAHs were found in the Koppers fire pond and L-P Ditch (Ebasco 1988).
Past, present, or future exposure to surface water and sediment is possible, but unlikely, because no indication exists that employees (Koppers or L-P) work in the areas where surface water collects, or that residents use the area for recreation, although occasional trespassing has been reported. Any contact with sediments is likely to be infrequent and of short duration. Crops are not grown in the immediate vicinity of the site. No data were received indicating that foraging animals or birds in the area are or were used as food sources for humans (Dames & Moore 1988a).
Children are at greater risk than are adults from certain kinds of exposure to hazardous substances emitted from waste sites and emergency events. They are more likely to be exposed for several reasons. Children play outside more often than do adults, increasing the likelihood that they will come into contact with chemicals in the environment. Because they are shorter than adults, they breathe more dust, soil, and heavy vapors close to the ground. Children are also smaller, resulting in higher doses of chemical exposure per body weight. The developing body systems of children can sustain damage if toxic exposures occur during certain growth stages. Therefore, to determine whether detected levels might be associated with any reproductive or developmental effects, ATSDR evaluated the types and quantities of chemicals detected in air, water, and soil at and around the site and how children might be exposed.
Children live in the vicinity of the Koppers site, but, in general, children do not have access to the site. While evaluating this site (e.g., air exposures, trespassing scenarios, use of domestic wells), ATSDR closely reviewed possible exposure situations for children. In its evaluation, ATSDR also used the Environmental Media Evaluation Guidelines (EMEGs) for children, who are considered the most sensitive segment of the population. No special hazards to children were identified on the basis of the available data.
ATSDR reviewed three health studies conducted by CDHS that examined possible health effects associated with contaminant releases from the Koppers plant. ATSDR's conclusions regarding the significance of these studies are summarized below. A more detailed account of the findings of these studies can be found in Appendix D. In addition, ATSDR reviewed and evaluated available cancer statistics for neighborhoods in the vicinity of the Koppers site.
Biologic Monitoring and Health Interview Study among Residents of an Area with Pentachlorophenol Contaminated Water
This study revealed that residents living in the vicinity of the Koppers plant reported having more health problems in the mid-1980s than did a control group. Testing of tap water and urine in both groups revealed no clear evidence of PCP exposures. The study, therefore, provides no definite link between reported symptoms and cause.
In late 1985 and early 1986, CDHS interviewed people living in households south of Koppers, measured PCP concentrations in tap water, and tested urine for PCP. CDHS questioned study participants whether they had ever experienced a variety of symptoms (from a comprehensive list ranging from headaches, eye and nose irritation, sore throats, respiratory problems, etc. to more chronic conditions, including cancer). Compared to a control group not known to have the potential for exposure to contaminated water supplies, residents living near Koppers generally reported poorer health. Many residents living near Koppers also reported odors in their water in the past.
PCP levels measured in study participant's urine were below levels known to be associated with health problems. PCP levels in tap water (0.16 ppb) were comparable to those typically found in US drinking water. From the statistical analyses they performed, CDHS concluded that living in the area south of Koppers and independently reporting water odor were strongly associated with self-reported symptoms (CDHS 1997a).
Because complete exposure data (i.e., PCP levels in water and urine) for the time period when symptoms were experienced were not available, a definite link cannot be made between exposures and reported symptoms. The study simply revealed that self-reported symptoms were higher in the area south of Koppers. PCP exposure is one possible explanation.
Koppers Fire Report
- Area residents and workers showed signs of exposure to irritants from smoke following the April 1987 fire, including eye, nose, throat, and skin irritation. Because no exposure data are available during or immediately after the fire (air or biologic samples), no conclusions can be drawn regarding possible causative agents.
- No one tested had toxic levels of PCP in their blood or urine.
In 1987, CDHS conducted a health screening of area residents within a week of the April 6 fire at Koppers, including telephone interviews and medical screening (CDHS 1987). The most prevalent symptoms reported by residents were headache (40.5%), nausea 28.6%), and skin rash (19%). Other symptoms reported included skin irritation, diarrhea, dizziness, and intense thirst. Symptoms were confirmed during clinical screening.
PCP levels in blood and urine samples collected 5-7 days after the fire were consistent with those typically found in the general public and below levels shown to be associated with adverse health effects. The low levels of PCP in tested blood and urine also provide evidence that no significant chronic exposures to PCP (non-fire related) had occurred during that time period.
Although all reported symptoms could plausibly be associated with acute exposure to PCP or to smoke in general, specific causative factors are speculative in the absence of air data.
Koppers' workers also participated in the medical screening and reported symptoms similar to those reported by residents, including sore throat (26%), headache (26%), cough (18%), nausea (7%), intense thirst (7%), and eye irritation (7 %) in the week following the fire. Only 4% reported skin irritation (CDHS 1987).
PCP levels in blood and urine were higher in some workers than those tested in nearby residents. This suggests that some workers were chronically exposed to PCP, but workers with the highest levels also held jobs that were most likely to bring them into direct contact with PCP. Therefore, these measurements do not necessarily reflect PCP exposure resulting from the on-site fire.
Serum Levels of Dioxins/Furans
- In a small study conducted by CDHS, dioxin levels in the blood of persons known to eat eggs and beef that had elevated dioxin levels were higher than in those who consume unaffected eggs and beef.
- Reported dioxin blood levels show that people have been exposed, but levels are lower than those shown to result in adverse health outcomes.
- Nonetheless, as stated earlier, uncertainties in the scientific literature regarding the effects of low-level dioxin exposures in humans support prudent public health practice (that is, taking measures to prevent contamination of home-raised animals and limiting the consumption of potentially affected food products).
In 1988, CDHS collected blood (serum) samples from nine individuals (from the two Index Homes, #1 and #2) who consumed eggs and meat that had detected levels of dioxin. A control group of nine persons also had their blood tested for dioxin. The control group consumed locally produced eggs and meat but resided in a rural area with no known dioxin sources. Results indicated that residents from Index Homes #1 and #2 who reportedly ate dioxin-contaminated eggs and meat had serum levels of tetra-, penta-, and hexa- dioxin that were elevated (two- to six-fold increases) compared with the control group (Goldman et al. 2000). Although the results demonstrate greater exposure to dioxins in the study group, measured serum levels are lower than those shown in available studies to be associated with adverse health effects.
Average serum levels were measured on a lipid or fat basis in females (29 ppt ITEQ) and males (39 ppt ITEQ) living in the index homes. Investigators reported an observed dose-response relationship based on the type of diet or length of time of exposure (i.e., persons who ate more eggs and/or beef and have done so over a longer period of time had higher blood levels of dioxin). Serum dioxin levels in individuals who consumed eggs averaged 26.7 ppt ITEQ and for individuals consuming eggs and beef 63.7 ppt ITEQ . The older members of the study group had the highest average levels of dioxins (65 ppt ITEQ).
Average blood levels detected in the study population are generally comparable to those reported in the general population for persons having no work exposures to dioxins (approximately 13-58 ppt ITEQ) (Needham et al. 1996; Tepper et al. 1997). Blood levels in the highest consumers (beef and egg) were less than 9% higher than the high end of this range. To the best of ATSDR's knowledge, however, no adverse health effects have been reported in the study group to date. No formal followup has been undertaken, however, to assess the health condition of this small population.
Most information regarding possible adverse health effects associated with various dioxin blood levels is for 2,3,7,8-TCDD (as mentioned previously, the most well-studied dioxin and believed to be the most toxic). The hepta-, hexa-, and penta- forms were the most prevalent in the blood of the study group. Measured levels of 2,3,7,8-TCDD in the blood of the study group were only 3.2 ppt in the egg consumers and 5.5 ppt in the egg and beef consumers. These levels are comparable to serum levels reported in the general population (3-7 ppt) (ATSDR 1998). For additional comparison purposes, the serum levels of dioxins measured in this relatively small study group were well below those found in highly exposed populations (e.g., chemical workers). Reported serum lipid concentrations of 2,3,7,8-TCDD in chemical workers range from 61 to more than 2,000 ppt (Patterson et al. 1989; Fingerhut et al. 1989). Studies in persons exposed to 2,3,7,8-TCDD during the Seveso, Italy, industrial accident were as high as 56,000 ppt. To date, chloracne (a severe skin disease) is the only unequivocally TCDD-related adverse effect that has been reported in this population. Not all individuals with elevated 2,3,7,8-TCDD serum lipid levels developed chloracne, although it did occur in all who had levels above 12,000 ppt and in children with blood levels between 830 and 56,000 ppt (Mocarelli et al. 1991). Index home resident blood levels were significantly lower than these.
As mentioned in earlier discussions, the potential for various other chronic effects, including cancer, related to dioxin exposure in humans is not fully understood. However, as noted before, estimated doses of hexa-, and penta- dioxins from eating one egg/day are thousands of times lower than doses showing effects in rodent studies (100,000 pg/kg/day). Furthermore, humans are thought to be at least 10 to 100 times less sensitive to dioxins than are rodents (Kimbrough 1992; Aylward et al.1996). Nonetheless, because uncertainties regarding the true disease potential in humans exist, prudent public health practice calls for limiting exposures involving products potentially contaminated with elevated levels of dioxins.
Cancer Registry Data
No increased rates of cancer have been identified in the areas surrounding the Koppers plant.
The state of California maintains statistics on the number of reported cancer cases on a local (census tract), county, and state level. Many types of cancers are recorded, including lung, skin, and soft tissue. Cancer registry data (state and regional) are available through 1994. Early in the health assessment process, ATSDR reviewed available state and regional cancer data and census block data for the years 1988 and 1989.
Lung, soft tissue, melanoma, other skin, and all other cancer incidence rates were similar to or lower than those for the region and the state. Using census and cancer registry data, however, to evaluate whether a relationship could exist between environmental contamination at Koppers and health effects in the neighboring communities is difficult for the following reasons:
First, the exposed population is probably only a small percentage of the two census tracts, so cancer rates for the exposed population are diluted in the rate calculated for the entire census tract. In other words, an elevated cancer rate for the exposed population could exist but not affect the overall cancer rate for the census tract. Second, in compiling the data, CDHS did not control for variables such as age, sex, and race in the populations. Therefore, differences in rates could be present, but are not evident because these variables were not controlled. In addition, the exposure (whether people actually drank contaminated water) and latency (when people were exposed versus when an effect is seen) are not considered in cancer rates. Cancers, if any, may not be observed for several years following the time of exposure.
The California Cancer Registry continues to maintain records of cancer incidence and mortality throughout the state and compiles statewide, regional, and local statistics. If made aware of unusual cancer trends in the South Oroville area, ATSDR will review and evaluate available statistics as appropriate.
1. All concentrations of dioxin-like compounds reported in this document were converted to toxicity equivalents (TEQs) using 1989 I-TEFs. However, use of a different set of TEFs, including the recently revised TEFs proposed by the World Health Organization, would not have altered any of the health-based conclusions presented in this health assessment. See discussion in the health implications section regarding the significance and interpretation of ITEQ concentrations. Also see Appendix E for a more in depth explanation of ITEQs.