PUBLIC HEALTH ASSESSMENT
JOSLYN MANUFACTURING AND SUPPLY COMPANY
BROOKLYN CENTER, HENNEPIN COUNTY, MINNESOTA
Minnesota Department of Health (MDH) staff were requested by Minnesota Pollution Control Agency (MPCA) Superfund Program staff to prepare this Public Health Assessment for the West Area of the Joslyn Manufacturing and Supply Company site in Brooklyn Center, Minnesota. The Joslyn site is a former wood treating operation.
Investigations conducted in the West Area from as far back as the 1940s have documented releases of contaminants from the former Joslyn wood treating operation and its predecessors. Surface soils, sediments, and surface water in some parts of the West Area are contaminated with pentachlorophenol (PCP), polycyclic aromatic hydrocarbons (PAHs), and dioxins at levels that are well above current applicable state and federal human health and environmental screening criteria. The extent of this contamination has not been fully defined, and off-site contamination is possible. Potential impacts to Twin Lakes cannot be ruled out based on available information and the limitations of existing data.
Based on a review of available information in MPCA and MDH files, various site visits, and meetings with neighborhood groups, it appears that exposure to contaminated soil and sediments in the West Area may have occurred in the past. While the extent of possible exposure is difficult to evaluate, adverse health effects are unlikely. Cancer rates for the community around the site are within normal ranges. The construction of a fence around the majority of the West Area as an interim response measure should limit current and future exposures to contaminated soil and sediment. Some uncertainty exists over the extent of continued human exposure, the lack of definition of the full extent of soil contamination, and the lack of adequate surface water, sediment, and fish data for Twin Lakes.
The Joslyn Manufacturing and Supply Company site, West Area (the site), is located in the City of Brooklyn Center, Minnesota, just northeast of the intersection of France Avenue and State Highway 100. The site is bounded on the west by Twin Lakes, on the north by railroad tracks, open space, and commercial/industrial properties with a residential neighborhood beyond, on the south by a residential neighborhood, and on the east by the development portion of the Joslyn site, France Avenue and Highway 100. The location of the site is shown in Figure 1.
Wood-preserving operations were conducted at the Joslyn site from the 1920's until 1980. Groundwater and soils at the site are contaminated with pentachlorophenol (PCP) and polynuclear aromatic hydrocarbons (PAHs). Polychlorinated dibenzo-p-dioxins and furans, common contaminants of PCP, have also been detected in soils and groundwater at the Joslyn site. The entire site was listed on the National Priorities List (NPL) in 1983, but the majority of the Joslyn site (with the exception of the westernmost portion, known as the "West Area") has been de-listed from the Permanent List of Priorities (PLP), the state Superfund list, following cleanup. The de-listed portion of the site is currently undergoing commercial redevelopment under the oversight of the MPCA Voluntary Investigation and Cleanup (VIC) Unit, while ongoing investigation of the West Area remains under the oversight of the MPCA Superfund Program. The West Area is located on private property owned by Joslyn.
MDH has conducted several evaluations of the Joslyn site, with the most recent being a Public Health Assessment dated February 7, 1994 (MDH 1994). A brief update to that document, in the form of a memorandum, was prepared in 1995. These documents did not specifically address the West Area. MDH staff were requested by MPCA Superfund Program staff to conduct a review of available information regarding the West Area of the site, and to develop conclusions and recommendations regarding potential public health concerns. Information reviewed included historical reports and environmental sampling data, information from community members, and data from the Minnesota Cancer Surveillance System (MCSS). This document will focus primarily on the West Area.
The Naugle Pole and Tie Company was the original owner and operator of the site. Wood treating began at the Joslyn site in the early 1920s, but little is known about these early operations. In the 1940s, the Consolidated Pole Treating Company purchased the property. The Joslyn Manufacturing and Supply Company (now Joslyn Corporation), a partial owner of the Consolidated Pole Treating Company, became sole owner of the facility in the early 1960s. Wood treating operations continued under Joslyn's ownership until site closure in September 1980.
Site operations consisted of treating wood products, primarily utility poles. Lesser quantities of posts, railroad ties, and other timbers were also treated. Thermal treating processes involved dipping wood products into tanks of heated preservative solution and consisted of "butt-dip treatment" (partial emersion of poles - used in the 1920s through 1965) and "thermal treatment" (full immersion of poles - used from about 1945 through 1980). In 1965, a pressure treatment system became the main treatment process. The preservatives used at this site for wood treatment included creosote, PCP, and soluble metal salts (copper-chromium arsenate, or CCA). Creosote was the only fluid used in the butt-dip treatment process, and PCP the only treatment fluid used in the thermal treatment process. All three of the preservatives were used at different times in the pressure treatment system, with PCP being the predominant fluid used.
Wastes from these treatment processes included dirt, "wood sugars," wood debris, waste treating fluids, and water (Barr 2000). Sludge (consisting of dirt, grit, wood scraps, and residual wood treatment fluids) was periodically removed from the butt-dip and thermal treatment vessels and from storage tanks. This sludge was buried on site. Boiler blow-down water, which consists of water from boilers used to heat the treating fluids and power steam engines, was generated until approximately 1965 and contained lubricating oils from the steam engines. Until roughly the 1950s, boiler blowdown water was disposed on site in a pond known as Pond C, which was one of three ponds used for the disposal of wastewater . Pond C is located in the West Area, and was constructed between 1937 and 1945 according to aerial photos (Barr 2000). Pond C also received surface water runoff from the site. From the 1950s on, the boiler blowdown water was sent to Pond B, an infiltration pond not located in the West Area. The pressure treatment system produced wastewater that contained residual wood treating fluids, and a disposal pond (known as Pond A) was constructed following the conversion to a pressure treatment system in 1965 to hold this wastewater. This wastewater was routed through a reclamation system, and preservatives and oil were skimmed from the water before it entered Pond A. Pond A was located east of the West Area.
A Remedial Investigation (RI) was completed at the Joslyn site in 1986. Based upon findings of the RI and a subsequent Feasibility Study, final groundwater and soil remedies were selected. The remedy included four major components: 1) a groundwater pump-out and oil recovery system; 2) removal and off-site disposal of heavily contaminated soils; 3) excavation and on-site land treatment of the remaining contaminated soil; and, 4) long-term monitoring of groundwater. The selection process and a description of the remedy were presented in a Record of Decision (ROD) for the site signed by MPCA in 1989.
Over the years, a series of investigations and cleanup activities have been conducted at the Joslyn site by the Joslyn Corporation under MPCA oversight, and have been described in previous MDH documents. They include excavation and off-site disposal of wood-treating sludge and highly contaminated soil, construction of a land-treatment unit (LTU) for the biological treatment (a process using soil bacteria to break down the contaminants) of the remaining less contaminated soils, installation of a groundwater pump-out, oil recovery, and monitoring systems in both the upper and lower aquifers, and continued monitoring of groundwater quality.
The Superfund Record of Decision (ROD) for the Joslyn site, executed in 1989, required that the remaining contaminated soil be excavated and biologically treated on site (MPCA 1989). Numerical soil treatment goals were set for PCP and PAHs. Excavation was to occur to the limits of no visual contamination or to the water table. Decisions about whether soils had been excavated to the limits of no visual contamination were made on the basis of a field screening procedure known as the oil sheen test, or "spoon" test. The procedure for conducting the spoon test consisted of obtaining approximately 25 grams of soil with a stainless steel spoon, and then directing a stream of water onto the soil until saturation was reached and water collected on the spoon. The amount of oil sheen on the water was then observed, and results recorded as follows: no sheen, or trace, moderate, or heavy sheen. Soils with a moderate or heavy sheen were excavated for biological treatment, while soils with no or slight (trace) oil sheen were considered to have met the cleanup goals.
Biological treatment of contaminated soil in the LTU was completed in 1998. With the completion of soil excavation and biological treatment, the Joslyn site consisted of a vacant property surrounded by a chain link fence, which did not include the West Area. The cleanup was consistent with the ROD and with the industrial/restricted commercial land use of the site. The ROD also required a groundwater pumpout and treatment and oil recovery system, which are operating successfully and will be ongoing for the foreseeable future. As described in the most recent annual monitoring report, the operation of the groundwater pumpout system has contained the plume of contaminated groundwater to approximately within the boundaries of the former Joslyn site (Barr 2001a). Pumped water is pretreated on-site to separate oil and discharged (under a permit from the Metropolitan Council) to the sanitary sewer system. The oil is disposed separately off-site.
Since 1998, activities at the Joslyn site (with the exception of the West Area) have centered on the redevelopment of the property for industrial and restricted commercial use. The redevelopment, when complete, will consist of several large commercial warehouse buildings with associated access roads, truck loading docks, parking lots, and landscaped areas. The city of Brooklyn Center also plans to extend Azalia Avenue across the site. Prior to the beginning of redevelopment activities, surface soil sampling was conducted across the entire Joslyn site to determine residual concentrations of PCP, PAHs, and dioxins in the uppermost, or accessible, soil layer. The results of this sampling were used to develop appropriate response action and construction contingency plans for the protection of workers, surrounding residents, and the ultimate users of the property. Residual soil contaminants are generally managed on-site by placing the soil under buildings or parking lots where the likelihood of future human exposure is minimal. Other areas will be covered with clean fill and vegetated. Institutional controls (in the form of deed restrictions on the future use of the property) are in place to help ensure that the contaminated soil is not made accessible for human contact. The development is consistent with the ROD and will provide additional protections that were not required by the ROD, but were identified by the MPCA during a mandated five-year review.
The Joslyn site is relatively flat and covered by sandy soil. Sandy fill was placed over much of the site during the early period of facility construction or operation. The site is underlain by an 80 to 140 foot thick sequence of unconsolidated (loosely arranged) sand, peat and sandy fill materials. The uppermost glacial deposit is a stratified fine- to medium-grained sand that ranges from 26 to 50 feet in thickness and is continuous under the site. This sand contains some beds of gravel and becomes progressively finer-grained with depth. The lower portions of the upper sand contains interbeddings of silt and silty clay. Beneath these materials is a layer of low-permeability silt, sand, and clay which ranges from 20 to 60 feet thick. This layer, referred to as the middle confining unit, covers the eastern two-thirds of the site. A buried bedrock valley under the western one-third of the site, including the West Area, is filled with sand and gravel. The valley cuts through the St. Peter and into the Prairie du Chien Group. The middle confining unit is absent in the buried valley.
The hydrogeology of the Joslyn site consists of an upper and lower aquifer. The upper aquifer is located within the unconsolidated materials and extends to depths of approximately 80 feet below grade. The lower aquifer, located below the middle confining unit, consists of the St. Peter Sandstone, the Prairie du Chien formation, and a sand and gravel unit that overlies these bedrock formations.
Groundwater in the upper aquifer moves from the west to east -- from Twin Lakes to the Mississippi River. The bedrock aquifers between the site and the Mississippi River are thought to be used only for industrial and residential, non-potable purposes. The St. Peter and Prairie du Chien-Jordan (deeper) aquifers also discharge to the Mississippi River. The Prairie du Chien-Jordan bedrock aquifers are important sources of drinking water for the Twin Cities metropolitan area. The municipal well nearest to the site draws from the Jordan and is about 1 mile to the northeast of the site.
Investigations in the West Area
The West Area was not directly used for wood treating operations, but rather for support activities. A railroad spur was located on the West Area, as was a small bunkhouse or barracks. The area now consists of a mixture of wooded uplands and defined wetlands, as shown in Figure 2. Contamination may have occurred in the West Area through a number of activities, including (Barr 2000):
- Sludge disposal or burial;
- Placement of contaminated fill;
- Use of Pond C for disposal of boiler blowdown wastewater; and,
- Overland flow of contaminated stormwater runoff.
The first soil samples collected in the West Area were from the former Pond C area in 1981. The soil samples were collected by Barr Engineering Company (Barr), an environmental consultant working on behalf of the Joslyn Corporation. Oil contamination was found at depths of 7 to 10 feet below ground level. No detailed chemical analyses of the soil samples were conducted, other than for the presence of arsenic and chromium; several aquatic toxicity tests were also conducted. The soil was determined to not meet the definition of a hazardous waste under Minnesota rules. In 1997, a soil boring (C-1A) was advanced near the location of one of the 1981 soil samples. Laboratory analysis of soil samples from the boring found PCP and PAH contamination extending to at least 8 feet below ground. The locations of historical soil samples, soil borings, and excavation areas are shown in Figure 3.
One soil boring (PB-1) was drilled to a depth of 141 feet in 1985 at the northern end of the West Area. A creosote odor was noted from a depth of 0 - 4.5 feet; below that level no evidence of contamination was noted. In 1989, 2,500 cubic yards of soil were excavated from this area. The soil was moderately contaminated (PCP concentration of 41 mg/kg, carcinogenic PAHs (cPAHs) concentration of 35 mg/kg), and was treated biologically in the LTU.
In 1997, additional soil borings were drilled in the area of Pond C to depths of up to 20 feet below ground. Analysis of soil samples from the borings showed low levels of PCP. Elevated levels of PAHs were found at depths between four and eight feet in one of the borings (PSS1). An oily sheen was also observed in this boring between four and 14 feet below the surface.
A series of hand-drilled borings was conducted around previous borings C-1A and PSS1 in 1997 to visually define the area of contamination in the surficial soils around these borings. No soil samples were collected for laboratory analysis from these borings. However, a composite soil sample from the identified contaminated area was collected and analyzed for PCP and PAHs. PCP was detected at a concentration of 1,300 milligrams per kilogram (mg/kg), and cPAHs were detected at a total concentration of 638 mg/kg. Five additional hand auger borings were then drilled, and soil samples collected and analyzed for PCP and PAHs to confirm the results of the previous visual borings. PCP ranged from 2.8 to 160 mg/kg and total cPAHs ranged from 62 to 886 mg/kg. This area (identified as the 1997 Excavation Area in Figure 2) was then excavated to a depth of 3 feet (at or below the level of the water table) and 650 cubic yards of soil were removed for treatment in the LTU. The area was then backfilled with clean soil from off site. Soil samples collected from the base of the excavation showed contamination remains below the clean fill.
Three samples of soil/sediments (DSS1, DSS2, and DSS3) were collected from the former ice chute and adjacent wetland area in 1997 by Barr and analyzed for PCP and PAHs. The ice chute was a ditch that ran across part of the site, and was used for transporting blocks of ice cut from Twin Lakes. It was suspected that the ice chute could have served as a channel for contaminated water from the site to reach Twin Lakes. The ice chute is still visible today, but has been partially filled by vegetation. Low levels of total PAHs (1.91 mg/kg) were found in sample DSS1, which was located at the mouth of the former ice chute where it enters Twin Lakes. PCP was below the laboratory detection limit of 5.7 mg/kg. PCP and PAHs were not detected above the laboratory detection limits in DSS2 and DSS3, which were collected in the ice chute further in from Twin Lakes and from the former outlet of Pond C respectively.
In 1998, a broad sampling program was implemented by Earth Tech, Inc., a consultant for the developer of the Joslyn site, in an attempt to characterize the extent of remaining shallow soil contamination across the entire Joslyn site, including the West Area. For the West Area, this sampling was designed to provide a final data set prior to delisting of the site from the state Superfund list. For the purposes of this sampling, the West Area was divided into seven sub-areas based generally on geographic features (an eighth was later added). The sampling sub-areas are shown in Figure 4, and the sample locations are shown in Figure 3. From each sub-area, up to five surficial (0-18 inches in depth) soil samples were collected and combined (or composited) to form one sample designed to be representative of the entire sub-area. Samples from area WA-7, which contains an open water marsh, were collected from sediments at a depth of 3 feet below the water level. The composite samples were then analyzed for PCP and PAHs, and some samples were also analyzed for polychlorinated dioxins and furans.
The analytical results of this sampling are shown in Table 1. The results showed levels of PCP below the MPCA Soil Reference Value (SRV) for residential soil of 71 mg/kg, and below the recreational soil SRV of 67 mg/kg with the exception of one area (WA-3). Visual evidence of contamination was observed in sub-sample location 5-WA-3. Total benzo(a)pyrene equivalents (a measure of the relative toxicity of mixtures of cPAHs) exceeded the residential SRV in four of the areas. The reported PAH data did not include the full list of cPAHs now considered of concern to MDH; benzo(a)pyrene equivalents were calculated using relative potency factors recently communicated by MDH (MDH 2001). Conservatively, one-half the detection limit was used for those samples where levels of PAHs were below the detection limit; detection limits were quite high in some samples. The assumption of one-half the detection limit is commonly made when evaluating such data. Dioxins and furans exceeded SRVs in some samples. Recreational SRVs are numerically the same as residential SRVs for cPAHs and for dioxins and furans. The SRVs are human health-based values and represent the concentration of a contaminant in soil at or below which normal dermal contact, inhalation, and/or ingestion is unlikely to result in adverse human health effects. The SRVs are generic criteria, and exceedances of SRVs only indicate the potential that the contamination could pose an unacceptable long-term human health risk.
To further characterize the contamination observed in area WA-3, five additional soil samples were collected (6WA-3 through 10WA-3). Analysis of these samples showed PCP concentrations of between 20 and 130 mg/kg. PAH levels were below the SRVs; dioxins and furans were not analyzed for. Subsequently, hand augers borings and test pits were used to further define the area of visibly contaminated soil around 5-WA-3. In 1999, approximately 1,000 cubic yards of visibly contaminated soil was excavated from this area and disposed of off-site in an approved landfill (Barr 1999a). The results of the analyses of the composite soil samples from the base of the excavation show that contaminated soil remains beneath this area, at a depth of at least 3 feet below ground. This depth is below the water table, which is the depth at which the groundwater surface is found.
Surficial soil, sediment, and surface water samples were collected by Barr in 2000 to fill in gaps in the data collected by Earth Tech, Inc. in 1998 (Barr 2000, Barr 2001b). The goal of this sampling was to confirm and further supplement previous data for PCP, PAHs, and dioxins and furans that were available for each of the identified sub-areas in the West Area. A similar sampling approach was used, i.e. composite samples were collected from up to seven sub-sample locations in each area. The sample locations used by Earth Tech, Inc. were duplicated in some instances. The area containing the former Pond C (WA-6) was divided further into three smaller areas (WA-6S, WA-6MID, and WA-6N) to better characterize the variability of residual contamination in different parts of the former Pond C. Sediment samples were again collected from the open water portion of WA-7, and one surface water sample was also collected from this area. A sample (WA-7M) was collected from the marsh (cattail area) of WA-7. Sample locations are shown in Figure 5.
The soil, sediment, and water samples were analyzed for pH, organic carbon, PCP, PAHs, and dioxins and furans according to the data gaps that were identified. Two soil samples and the sediment sample were also subjected to a water leach test. Due to perceived "significant" differences in the data between individual sub-areas, the reserved samples from WA-6MID and WA-6S were later submitted to two separate labs for duplicate analyses for dioxins and furans. In general, the results of these duplicate analyses correlated well. The analytical results of the soil sampling are presented in Table 1, and the results of the sediment samples are presented in Table 2. Surface water results are presented in Table 3. The results of the water leach test can be found in Table 4.
The results of the analysis of the surficial soil samples collected in the West Area show that soils exceed the residential and recreational SRVs for PCP, PAHs, and/or dioxins and furans in five of the eight sub-areas (WA-1, WA-2, WA-3, WA-6, and WA-8). In these areas, PCP and cPAHs are present at levels only slightly in excess of their respective SRVs (with one exception - PCP in sample area WA-3). The highest concentrations of dioxins and furans, as expressed in 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) toxic equivalency factors (TEFs), exceeded the SRV by a factor of several hundred. The TEF concept is a way of expressing the toxicity of mixtures of dioxins and furans, and is described later in this document.
The results of the analysis of sediment samples collected in area WA-7 show that concentrations of dioxins and furans exceed available MPCA SRVs for residential soil, and concentrations of PAHs approached the SRV in one sample. No applicable sediment screening criteria have been developed by the MPCA. Analysis of the surface water sample collected in area WA-7 showed low levels of PAHs and dioxins and furans; PCP was below the laboratory detection limit. Only one PAH (benzo(a)pyrene), and TCDD were present at a level in excess of MPCA surface water standards or criteria. The results of the soil water leach test showed that PAHs, and very low levels of dioxins and furans were leached from sediments from site soils.
Investigations of Twin Lakes
Oil in the wetlands abutting the West Area and Twin Lakes was reported as early as the mid-1940s, when representatives of the Minnesota Department of Conservation and MDH found oil from the site on the surface of Twin Lake (MDH 1944). The oil was believed to have reached the lake from a drain pipe which emptied into a system of ditches, ultimately reaching the lake through a channel formerly used as an ice chute. A ditch connecting Pond C to the ice chute is clearly visible on aerial photographs from 1945 (Barr 1996). Oil was also noted by MDH staff in the swamp that lies along the east shore of the lake. Analysis of water samples (it is not clear how many) collected by MDH from the outlet of the drainpipe and from the lake itself showed low levels of phenol and creosote (4.17 and 0.0107 parts per million phenols, respectively). The MDH report went on to discuss three factors involved in the problem of waste disposal from the site: 1) pollution of the lake with dissolved phenols, creosols, and possible other chemicals; 2) unsightly pollution of the lake surface by oil films which may develop from continual discharge of oily waste; and, 3) fire hazards (MDH 1944). Along with a recommendation for enclosing part of the marsh with an earthen dike and use of an oil filter, the report also recommended that phenol concentrations in the lake be monitored. An MDH investigation of Twin Lakes, conducted in response to a fish kill in April of 1950, did not find evidence of phenolic contamination from the site in water and bottom fauna samples collected from Twin Lakes, but again noted heavy oil contamination from the site in the swamps along the lake. The 1950 report noted that the company had reportedly stopped the use of the marsh adjacent to the lake for wastewater disposal, and concluded that the source of the fish kill could not be identified.
Because of concerns over the potential impact of the Joslyn site on Twin Lakes, in 1982 the U.S. Environmental Protection Agency (EPA) collected three surface water and three sediment samples from Twin Lakes and analyzed them for PCP, PAHs, metals, phenolic compounds, volatile organic compounds (VOCs), pesticides, and TCDD (EPA 1983). The samples were collected from the north end of middle Twin Lakes (near the railroad bridge), from the bank of the lake on the edge of the West Area, and from the outlet to lower Twin Lakes near the Highway 100 crossing. The results of analysis of surface water samples showed low levels of one PAH, pyrene, in the sample collected near the lake outlet. PAHs were also found in the sediment sample collected from this location, but were not detected in the two other samples. One pesticide compound, alpha-BHC (also known as alpha hexachlorocyclohexane) was found at low levels in the sediment sample collected at the north end of middle Twin Lakes. This pesticide is not believed to be related to the Joslyn site. Arsenic found in some of the surface water and sediment samples (from the inlet area, and near the site) was attributed by EPA to natural background sources. Arsenic containing compounds have reportedly also been used for weed control in Twin Lakes. Detection limits were high for some analytes, such as PCP (25 mg/kg) and TCDD (100 micrograms per kilogram (µg/kg)).
Surface water samples were collected from six locations on Twin Lakes in January and May of 1985 (Barr 1986). The sampling locations are shown in Figure 6. Two samples were collected from Upper Twin Lakes, two were collected adjacent to the West Area, one was collected adjacent to the swimming beach located in the park south of the site, and one from the outlet at the Highway 100 bridge. The samples were analyzed for PAHs, heterocyclics, and phenolic compounds. Results are presented in Table 5. Levels of PCP and other phenolics were not detected above laboratory detection limits (5 µg/L for both), while detectable levels of PAHs were found at all six sampling locations in both sampling events. Two PAHs (benzo(a)anthracene and benzo(a)pyrene) exceeded MPCA surface water screening criteria in two and four of the six locations respectively. The highest levels were found in Upper Twin Lake, in a location that was some distance from the site and was intended to serve as a background sample location. PAH levels were generally higher in the January 1985 samples than in the May 1985 samples.
In 1997, MPCA staff collected a series of five sediment core samples from Twin Lakes for visual inspection (MPCA 1997). The sediment samples were collected parallel to the shore of Twin Lakes, at various distances from shore. Sample locations were from approximately the end of the former rail spur to the channel leading to upper Twin Lakes. The samples ranged from six inches to two feet in length, and were examined visually for signs of oil contamination. No signs of oil contamination, or odors, were found in any of the five sediment samples. No chemical analyses of the sediment samples were conducted.
On July 6, 2000, and on several occasions since that date, Jim Kelly of MDH has conducted site visits to the Joslyn Site, West Area. The West Area is located along Twin Lakes, and was reportedly not used for wood treating operations by Joslyn except for possible drainage or discharge of process water, as described above. A railroad spur, bunkhouse, and ice chute were the only physical features observed on historical aerial photos. The area now consists of mostly wooded upland areas, wetlands, and the shoreline of Twin Lakes.
Prior to the installation of fences around the West Area, visible trails could be observed. The most heavily used trail began in a yard at the end of Twin Lake Avenue North, on the southern end of the West Area. This trail was obviously used fairly frequently judging by the lack of vegetation on it, and the presence of footprints (both human and canine), bike tire tracks, and crude "bridges" over low spots. There were also beer and soda cans as well as other litter scattered along the trail. The trail split at several locations, with one side-path leading to a small sandy area or beach on Twin Lakes. There were several empty cans at this location, as well as an old tire in the water that was surmised to have been used as a swing based on the presence of sections of rope on the tire and on a nearby tree. The remains of a campfire were also visible nearby. Wild raspberries were observed growing in this area as well. Another side-path led to the wetland on the north end of the West Area; it appeared that this path would not be traversable except in very dry or frozen conditions.
The main trail continued through the West Area along the base of the berm on the west side of the Wickes warehouse. A silt fence at the base of this berm was in poor condition. Along the path, near a small stormwater pond outlet from the Wickes building, is an area of exposed objects including small pieces of burned tire, scrap metal, and small miscellaneous auto parts. Vegetation is sparse in this area. It is not clear how these wastes came to be here. The trail ultimately exited the West Area at the railroad tracks to the north. It may have continued on the other side of the railroad tracks. The railroad line appears to be quite active, with many trains passing by each day from a rail yard located east of the site.
It was obvious that people used the West Area for some recreational purposes, such as walking, biking, dog walking, "partying,"and perhaps wading or swimming. The shoreline of Twin Lakes (with the exception of the small sandy area) and the wetlands on the site were relatively difficult to access from the trails due to the amount of vegetation.
Subsequent site visits conducted during the winter of 2000-2001 showed that the West Area was being accessed from the lake by snowmobiles and perhaps ice fishermen. In the spring of 2001 a large portion of the West Area flooded due to the rapid snow melt and heavy spring rains. Much of the West Area is below the 100-year flood elevation, and runoff from the Joslyn site development appears to have at least been partially directed onto the West Area.
During the fall of 2000, a fence was placed along the north and south property lines of the West Area to restrict access. A third fence connecting the two previous fences was erected along the east border of the West Area in the spring of 2001. A fourth section of fence was completed along the west side in the fall of 2001. At this time, the West Area is completely fenced except for a small section through the wetlands along Twin Lakes where installation of a fence would be difficult. No trespassing signs stating that the West Area is private property have been posted by the site owner. The Joslyn Manufacturing Company has also reportedly hired a security guard to regularly police the property (Joslyn 2001). The fence restricts access to the site and should limit the types of activities observed on the West area that are described above to all but the most determined individual. As a further precaution, the MPCA has posted signs along the fence at various locations stating the following:
Warning - contaminated soil, sediments and water. Walking or playing within the fenced area should be avoided. Contaminants may be released from soil or sediments if disturbed.
The site is located in a densely populated, mixed use area of Brooklyn Center. Single-family and multi-family residential properties are located in proximity to the site. A park, including a swimming beach, is located on Twin Lakes approximately ¼ mile south of the site. Twin Lakes, as an urban lake, is used for recreational activities such as wading, swimming, boating, water skiing, and fishing. According to 1990 census data (the most recent census data is not yet available) the population within a one-mile radius of the site was estimated to be 15,687 people (ATSDR 2001). Of this number, there were an estimated 1,512 children under the age of six.
A search of the MDH County Well Index identified 33 water wells within a one-mile radius of the site (MGS 1997). These wells are a mixture of private wells, public supply wells, and commercial/industrial wells. This does not include the numerous monitoring and groundwater pumpout wells located on the remainder of the Joslyn site. None of the wells appear to be in a directly down-gradient location from the site, and are unlikely to have been affected by contaminants from the site.
There are numerous leaking underground storage tanks (LUST) sites, hazardous waste generators, and VIC Program sites located within a one-mile radius of the Joslyn site. None are located in an upgradient location in terms of groundwater flow, and none are known to have impacted the site. General regional issues, as with many urban areas, would include overall groundwater quality, surface water runoff, air pollution, etc. Surface water quality has been identified by area residents as a specific area of concern with regards to Twin Lakes.
A local community group, the Brooklyn Center Community Association (BCCA), as well as individual citizens living near the site, have expressed concern over contamination remaining in the West Area and the possible impacts on public health, including a possible increase in the incidence of cancer in humans and animals near the site (human cancer rates in the area of the site are discussed below). Specific concerns cited by community members (and MDH's response) during a meeting held on May 15, 2001 with MDH and MPCA staff included:
- Prior to the fence being erected, the area was used quite regularly by many residents in the neighborhood for dog walking, hiking, biking, and as a shortcut to the beach area. Kids frequented the area, using it for play, and older kids or young adults used it for parties. It was also reported that vagrants had been living in the area, though not necessarily in the West Area itself.
- MDH has received numerous reports over the years of past uses of the West Area as described by the residents.
- One resident noted that children were still using the West Area for play by simply going around the partial fence that was constructed in the fall of 2000.
- This type of activity has been made more difficult, but not impossible, by the extension of the fence.
- Also described was the continued use of the West Area by boaters, who had easy access from the lake, and by snowmobilers and ice fishermen in the winter.
- Again, the extension of the fence should limit the occurrence of trespassing. Winter activities would likely result in limited exposure due to frozen soil conditions.
- Concern was expressed about runoff from the rest of the Joslyn Site through the West Area and associated wetlands and into Twin Lakes. Holding ponds were described as ineffective, allowing water to pass through them.
- MDH shares the concern over stormwater runoff.
- Odors have been reported in the West Area, and off of the site to the north. The odors were described as "creosote-like," or "rotten." They have been reported on the north side of the Wickes building, and along the railroad tracks.
- The source of the odors is unknown. The still active Canadian Pacific Railway line is a possible source, as are wetlands.
- Fishing is popular in Twin Lakes, and is more concentrated near the park and Highway 100 bridge on Lower Twin. Ice fishing is also common in the winter. It is unclear, and perhaps unlikely that the fishing could be considered "subsistence" fishing. Fishing piers were reportedly being considered by the cities of Brooklyn Center and Crystal that could increase fishing opportunities in the lake.
- MDH is aware that Twin Lakes is used for recreational fishing.
- A proposed extension of a road across the development site could increase the potential for people to be exposed to contaminants in the West Area.
- This document is focused on the West Area. The possible effect of construction activities in the developed portion of the Joslyn site on potential exposure in the West Area cannot be predicted.
A draft version of this Public Health Assessment was made available for public comment from April 8 to June 7, 2002. A press release announcing the availability of the document was issued. A story about the site also appeared on a local television news program, and included an interview with MDH staff. Comments were received from several community members, as well as from the responsible party. These comments, along with MDH's response, are summarized in Appendix V.
MDH, under a cooperative agreement with the Agency for Toxic Substances and Disease Registry (ATSDR), evaluated the public health significance of contamination associated with the Joslyn, West Area site. More specifically, MDH and ATSDR cooperated to determine whether health effects are possible and to make recommendations to reduce or prevent possible health effects. ATSDR, located in Atlanta, Georgia, is a federal agency within the U.S. Department of Health and Human Services. ATSDR is mandated by the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA) of 1980, as amended by Superfund Amendments and Reauthorization Act (SARA 1986), to conduct public health assessments at contaminated sites. In cooperation with ATSDR, the MDH has evaluated the public health significance of the entire Joslyn site in the past, issuing a Public Health Assessment on February 7, 1994. The 1994 Public Health Assessment concluded that the Joslyn site represented an indeterminate health risk, stating that "available data do not indicate that people are being or have been exposed to levels of site contaminants that would be expected to cause adverse health effects, but data or information are not available for all environmental media to which people could be exposed."
MDH staff evaluated data compiled over the last 10 years from the Minnesota Cancer Surveillance System (MCSS) for the zip code 55429, which includes the cities of Crystal and Brooklyn Center. While this area is much larger that the immediate area around the former Joslyn site, it is the smallest area available for analysis in the MCSS. The number of cancer cases reported in this zip code over the last 10 years is within the normal expected range for both men and women. In addition, no individual cancer sites showed a statistically significant difference from the expected rates for the population of Minnesota.
Soils and Sediment Contamination in the West Area
Surface soils on the West Area are contaminated with PAHs, PCP, and dioxins at levels that exceed current MPCA Soil Reference Values (SRVs) for soil in residential and recreational areas, as well as current ATSDR soil screening criteria. Soil screening criteria are discussed in greater detail in Appendix III. Based on composite sample data, and as indicated in Tables 1 and 2, PAHs exceed the SRVs in subareas WA-1, WA-3, and WA-6, while PCP exceeded the SRVs in subareas WA-3 and a portion of WA-6 (WA-6MID). Concentrations of dioxins and furans (as expressed in TCDD TEFs) exceed the SRVs and the ATSDR and EPA residential soil screening criteria in soils (or sediment) in six of the eight subareas (WA-4 and WA-5, along the shore of Twin Lakes, were the exceptions). Four of the eight areas exceeded the EPA action criteria of 1 ppb for residential soils. There is relatively little information about levels of contaminants in the sub-soil; qualitative evidence (from past soil borings) suggests that contaminants are present at depth in some areas.
The highest concentrations of dioxins in surface soils were found in areas WA-6MID and WA-6S, in the area of the former Pond C. The highest dioxin concentrations are associated with high organic carbon concentrations in the soil, which in the case of area WA-6 may be from sediment deposits in the former Pond C. In terms of the specific dioxins and furans found, the hepta- and octa-chloro dioxins and furans predominate. This is usually the case with wood treating sites, where the more highly chlorinated dioxin and furan compounds are often found in high concentrations relative to the other compounds (Copeland et al 1993; Laine et al 1997). This may be due to these dioxins and furans being present in higher concentrations in the PCP originally used in the wood treatment process, or as a result of the relatively faster breakdown of the less chlorinated compounds in the environment.
The use of composite sampling is best suited when the goal is to determine if contamination is present or not, as was the case with the initial sampling conducted for delisting of the site from the state Superfund list. It is also useful for determining average exposure concentrations across an area. While composite sampling is consistent with EPA guidelines, the use of composite sampling methods in this case made it difficult to determine the actual extent and the maximum level of the contamination in surface soil in the impacted areas. Without analysis of the individual point samples, it cannot be determined if one or several of the sub-samples used to make up the composite are contributing the majority of the dioxins, or if the contamination is present at that level in all of the sub-samples. Based on the existing composite data, the southern extent of contamination has not been defined. It is important to define the extent of contamination and to determine if it extends off of the site closer or into residential areas where exposure could be more frequent through routes other than directly contacting contaminated soil. It has been estimated that as much as 32% of indoor dust could originate from outdoor soil through foot tracking or other transport mechanisms, thus providing a secondary exposure route (Calabrese and Stanek 1992). This also contributes to the fact that residential screening criteria are typically lower than industrial or recreational criteria.
The results of the soil-water leach tests (Table 4) show that very low levels of dioxins can be leached from the surface soil and sediment, either in dissolved phase or bound to fine particulates that may have passed through the filter used in the test. Dioxins that are bound to soil particles may also therefore be carried by surface water runoff. This indicates that contamination may be carried towards Twin Lakes during the spring thaw or periods of heavy rain. At least a portion of the stormwater runoff from the adjacent development appears to have been routed through the West Area. In addition, contamination may have been carried from the West Area to the south, into the adjacent residential area, through the past use (and overflow) of Pond C, or more recently due to flooding of the site. The 100-year floodplain elevation in the area is 855.5 feet above sea level. This elevation encompasses almost the entire West Area, and extends off of the site to the south. During at least one site visit, standing water was observed on the site (in the area of Pond C) and extending off the site to the south in one continuous pond. These factors indicate that soil sampling off of the West Area to the south in low elevation areas is warranted.
Sediments in the open water area (WA-7) are contaminated with dioxins at levels in excess of MPCA SRVs for residential soil (see Table 2). No applicable ecological-based sediment screening criteria are available for comparison at this time, although they may be developed on site-specific basis by the MPCA. This indicates that sediments may pose a potential threat to human health if people come into regular contact with the sediments, which is unlikely. The surface water samples collected in area WA-7 also showed detectable levels of PAHs and dioxins, with levels of the PAHs benzo(a)anthracene and benzo(a)pyrene and TCDD exceeding the MPCA's surface water criteria (Table 3). In both sediments and surface water, once again the hepta- and octa-chloro dioxins and furans predominate.
The MPCA is currently working with Joslyn to finalize a plan for additional sampling in the West Area to better define the extent and magnitude of the contamination. This sampling effort will include discreet samples of soil, sediment, and surface water at various locations and depths. Implementation of this sampling plan will help address some of the uncertainties with the existing data.
In the past, oil from the Joslyn wood treating operation discharged into Twin Lakes. Two investigations were conducted by MDH in the 1940s and 1950s as a result of complaints of oil on Twin Lakes and a reported fish kill. Phenolic compounds were detected in lake water samples collected by MDH in 1944. Figure 7, which is a map produced by the predecessor to the Minnesota Department of Natural Resources (DNR) in the 1950s, shows a ditch running from a pond (most likely Pond C) on the site to Twin Lakes. Thus, there is a possibility that wastewaters from Pond C may have directly reached the lake. In the years since, contamination may also have reached the lake through the former ice chute that connects area WA-7 and Twin Lakes, or through stormwater runoff from the site. Based on the results of the most recent site investigations, small amounts of contaminants are capable of being leached from surface soils. Contaminants bound to soil organic particles (which is where PAHs and dioxins would concentrate) could also have been carried by storm or floodwaters into the lake.
Twin Lakes have been part of past investigations. In 1982, three surface water and three sediment samples were collected by EPA. While PCP and PAHs were not detected in the water samples (with one exception), laboratory detection limits for the surface water samples were not reported. Laboratory detection limits for the sediment samples were well above the current relevant screening criteria for PCP and dioxins, so although they were not detected in the sediment samples, the presence of these contaminants above levels of concern cannot be ruled out based on these analyses alone.
Six surface water samples collected in 1985 showed PAHs were present in many of the samples, with some levels exceeding MPCA surface water criteria (see Table 5). PCP was not detected in any of the samples; analyses for dioxins were not conducted. The PAHs were detected in samples collected both near the site and further away from the site. It is certain that there are other sources of PAH contamination to the lake, such as surface runoff from highways and streets, the railroad tracks and bridge, and boat traffic. In their report on the surface water sampling, Barr indicated that the PAH contamination was likely the result of outboard motors, motor fuels, or lubricants based in part on the similarity of the data (Barr 1986). However, for five of the six samples, which were collected at the same locations in January and May of 1985, the PAH levels were higher for the January sample when the lake was presumably frozen than in the May sample, when boat traffic would be expected to be at or near its peak. PAHs, which would likely be adsorbed to particulate matter, would tend to settle out of the water column during the winter months. Volatilization could account for some of the concentration differences, especially for the non-carcinogenic PAHs. The unknown effect of the spring and fall turnover, other sources of PAHs, and large runoff events make overall interpretation of the PAH results difficult.
Routes of Exposure
There are several routes through which people may be exposed to contaminants from the site. The routes of exposure from contaminated soil at a former wood treatment facility that was similar to the Joslyn operation were discussed in a study conducted by Copeland et al (1993). The study primarily focused on dioxins and furans, as concentrations of PCP and PAHs were generally below levels of concern. The potential routes of exposure they evaluated that may be applicable to this site include:
- Ingestion of contaminated soil / sediments;
- Inhalation of airborne particulates; and
- Dermal (skin) exposure to contaminated soil / sediments.
This discussion will focus on ingestion of and dermal contact with contaminated soil or sediments, which are the most likely routes of human exposure at the site. Inhalation of airborne dusts is not believed to be a major exposure route for contaminants in the West Area due to the fact that the majority of the contamination is in low-lying soils or sediments that are vegetated and often wet (or frozen). In addition, studies of inhalation hazards posed by dioxin contaminated soils have shown that inhalation will rarely be a significant route of exposure based on cancer risks (Paustenbach et al 1991). The source of odors reported by some local residents is unclear, but could be related to the railroad tracks and bridge. Creosote is still used for the treatment of wooden railroad ties and bridge timbers. Other potential routes of exposure may include consumption of contaminated fish from Twin Lakes, incidental ingestion of and dermal contact with contaminated surface water in Twin Lakes, and ingestion of mother's milk. Due to the lack of data, it is not possible to evaluate fish consumption, exposure to surface water, or the ingestion of dioxins in mothers milk at this time. As with any lake, MDH's general fish consumption advisory should be followed (see Appendix IV).
The ingestion of contaminated soil (which can include sediments) is usually viewed as the primary means of exposure to non-volatile contaminants in soil, including dioxins. Such ingestion is usually incidental, and occurs from hand-to-mouth contact as a result of such activities as gardening or work activities (in the case of adults) or outdoor play activities (in the case of children). An exception to this is pica behavior in children, in which the intentional ingestion of sometimes large amounts of soil is a defining characteristic.
The amount of contaminant due to incidental ingestion that is absorbed into the body and available to cause an adverse effect is affected by a number of factors, including (Copeland et al 1993):
- The chemical concentration in surface soil;
- The bioavailability of the contaminant in the soil;
- The half-life of the contaminant in soil;
- The soil ingestion rate; and,
- The frequency of soil ingestion.
The oral bioavailability of dioxins in soil may be dependent on the soil organic content, and for TCDD has been found to range from 0.5% to 50% in animals (Copeland et al 1993). The bioavailability of other dioxin compounds, such as the octa-CDDs may be less, perhaps 10% the absorption rate of TCDD. In a study of digestive absorption of dioxins and furans in humans (from food) using a mass-balance approach, the maximum absorption of TCDD was 63%; again the absorption of the more highly chlorinated congeners was reportedly much less (Schlummer et al 1998). The same study also found considerable variability in absorption rates among the test subjects, with age being a key factor. Absorption rates in older test subjects were much less than in younger subjects.
The frequency and amount of soil ingestion are the most difficult parameters to estimate. Anecdotal reports by local residents near the Joslyn site have indicated that adults and children could have entered the site as frequently as several times per week as a result of hiking, dog walking, and other activities. Most screening exposure scenarios utilize a residential setting, where exposure to soil could be expected to occur on a regular basis. In a survey study of soil contact behavior by adults in a similar climate to Minnesota, the adults surveyed reported contact (at the 95th-percentile) with soil at their residence through such activities as home repairs or digging a little more than one time per week (Garlock et al 1999). The median soil exposure rate was less, on the order of once per month for home repairs or digging. Exposure frequency from activities such as gardening, other yard work, and team sports appeared to be much more frequent, although the survey units for the two groups of activities made direct comparison difficult. Some adults may also have higher soil ingestion rates. People who have frequent contact with soil, such as gardeners, also tend to ingest more soil. Behaviors that involve frequent hand to mouth contact, such as smoking, can also lead to higher soil ingestion rates.
The intake of contaminants and the potential effects of dermal (skin) exposure to contaminants in soil are influenced by several factors, including (Copeland et al 1993):
- The chemical concentration in surface soil;
- Skin surface area available for contact;
- Skin adherence properties of soil; and,
- Dermal bioavailability.
The area of skin available for contact with soil can vary. Typically, it is assumed that skin contact involves the hands and lower arms, but can include the legs, feet, or other body parts. In the soil contact study described above (Garlock et al 1999), median skin area potentially exposed during various outdoor activities during warm months was approximately one-third of the skin area of the body (32%) as reported by adults responding to the survey. The researchers assigned the estimated skin area exposed based on the respondents replies using values for various body parts that were similar, but not identical, to values used by EPA. The skin area potentially exposed does not necessarily equal the area of skin that actually comes into contact with soil.
When skin comes into contact with soil, only a small amount is usually left on the skin surface once the contact has ceased. Contaminants that remain on the skin may be absorbed through it, at a rate that is based on the properties of the contaminant. EPA summarized several studies of dermal soil loading in children and adults and cited values of between 0.5 and 1.5 milligrams of soil per square centimeter of skin (mg/cm2) (EPA 1999). It should be noted that these values were derived mainly from studies using the hands, which typically have a higher soil adherence factor than other body parts. Theoretically, there is a point at which an increase in soil loading does not result in further absorption of a chemical due to the establishment of a uniform layer of soil on the skin - any additional soil is not in contact with the skin. In a study of pesticide absorption from soil using cadaver skin, this value was estimated to be between 1 and 5 mg/cm2 (Duff and Kissel, 1996). The type of soil (clay, sand, etc) may influence the value.
The amount of a chemical contaminant that can be absorbed through the skin from soil is dependent on the condition of the skin, the amount of contaminated soil applied, the soil characteristics, and the physical properties of the chemical. Typical values cited in the literature for the dermal bioavailability of dioxin based on animal studies range from 0.1 to 2% (Copeland et al 1993). In a study using pharmacokinetic modeling instead of animal testing, higher absorption rates were predicted, typically between 14 and 26% (Kissel and McAvoy 1989). These results have not been validated through animal testing, however, and may represent a theoretical rather than a practical maximum. An important factor affecting absorption is the speed at which it is absorbed. Dioxins appear to be absorbed slowly through the skin, indicating that if the exposed area is adequately washed within a reasonably short time after exposure, much of the absorption can be prevented (Banks and Birnbaum 1991).
Properties of the Contaminants of Concern
Pentachlorophenol (PCP) has traditionally been one of the most widely used chemicals for the preservation of wood products. It was recognized and used as an insecticide, fungicide, herbicide, molluscicide, and algicide in a wide variety of applications (ATSDR 1994). The most common use for the treated wood was for utility poles; other uses have included fence posts, railroad ties, cross arms, and other common industrial wood products. Because of its widespread use, PCP is nearly ubiquitous in the environment, having been found across the United States in surface waters, sediments, rainwater, groundwater, soils, food, and living organisms, including humans. It has historically been estimated that volatilization from the surface of PCP-treated wood products releases as much as 760,000 pounds of PCP to the air per year in the U.S. (ATSDR 1994).
In the environment, PCP may adsorb to soils depending on the pH of the soil and its organic matter content. The amount of PCP adsorbed at a given pH increases with increasing organic content of the soil (ATSDR 1994). PCP is more mobile in soil under neutral or alkaline conditions, and adsorption is minimal at pH values above 6.8. PCP may also have the ability to bioaccumulate, or build up, in the tissues of animals (such as fish) exposed to it. It has not been shown to become concentrated in animal tissues as it moves up the foodchain, however. PCP is able to be broken down by microorganisms in the soil, and biodegradation is thought to be the major pathway of PCP degradation in the environment. Biodegradation was successfully used to treat contaminated soils from the site in the LTU area.
Short-term exposure to high concentrations of PCP is associated with adverse effects to the kidneys, blood, lungs, nervous system, immune system, and gastrointestinal tract (ATSDR 1994). It can also cause a potentially serious increase in body temperature as the body attempts to metabolize it. Dermal contact can irritate the skin, eyes, and mouth. These types of exposures and concentrations are usually only seen in the workplace. Long-term exposure to lower levels of PCP can cause damage to the liver, kidneys, blood, and nervous system. PCP is considered a probable human carcinogen. Some of the adverse effects associated with exposure to PCP may be caused by impurities present in commercially produced PCP, such as dioxins and furans.
Polynuclear Aromatic Hydrocarbons
Polynuclear aromatic hydrocarbons (PAHs) are produced by the incomplete combustion of organic materials such as coal, oil, wood, tobacco, and even food products (ATSDR 1995). They are also found in such products as asphalt, coal tar, creosote, and roofing tar. As a result, they are very common in the environment from such processes as volcanic eruptions, forest fires, home wood burning, and vehicle exhaust. Over 100 PAHs have been identified, and they are usually found in the environment as mixtures. PAHs generally fall into two groups based on their potential health effects: those that are carcinogenic (cancer causing, known as cPAHs), and those that are not (non-carcinogenic PAHs, or nPAHs). The PAHs found on the site (a mixture of cPAHs and nPAHs) are likely present as a result of the use of creosote in wood treatment. Creosote itself is usually derived from coal tar, and is described as a thick, oily liquid that is amber or black in color, and contains hundreds or even thousands of different chemicals including PAHs and phenols (ATSDR 1996). It has been in use as a wood preservative and waterproofing agent for over 100 years.
PAHs tend to bind to soil particles, especially organic matter, and therefore tend to remain in soils and sediments. Because of their affinity for organic matter, PAHs can accumulate in aquatic and terrestrial organisms, but unlike PCP, can become concentrated as they move up the foodchain (ATSDR 1995). This effect is somewhat balanced by the ability of many organisms, such as fish, to metabolize PAHs. In soil, PAHs can be broken down by microorganisms. The rate and extent of biodegradation can be influenced by environmental factors, the composition of the soil, the type of microbes present, the presence of other toxic compounds, and the properties and concentrations of the mixture of PAHs present (ATSDR 1995).
Individual cPAHs are classified as probable or possible human carcinogens by the International Agency for Research on Cancer (IARC) (ATSDR 1995). MDH uses information developed by the California Environmental Protection Agency to evaluate the carcinogenicity of cPAHs, and the list of cPAHs of concern has been expanded from prior lists typically reported by EPA (MDH 2001b). Exposure to high levels of PAHs in general has also been associated in animals with reproductive difficulties and adverse effects on the skin and immune system. Adverse effects on the liver and gastro-intestinal tract have also been noted.
Polychlorinated dibenzo-p-dioxins were the most elevated of the contaminants found at the site. They consist of a family of 75 different chemical compounds, commonly referred to as polychlorinated dioxins or just dioxins (ATSDR 1998). This family is further divided into eight groups, based on the number of chlorine atoms in the particular dioxin compound. There are a number of different dioxin compounds in each group (also known as congeners), based on the position of the chlorine atoms on the dioxin molecule. The names of individual dioxin compounds denote both the number and position of the chlorine (Cl) atoms. Furans are a similar family of compounds, differing from dioxins only by the lack of one of the two oxygen (O) atoms between the benzene (six-carbon atom, circle-shaped) ring structures. The chemical structures of two of the most studied (and toxic) compounds, 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and 2,3,7,8-tetrachlorodibenzo-p-furan (TCDF) are shown below:
Dioxins and furans have never been manufactured, but are formed as a result of the incomplete combustion of fossil fuels, organic matter, and waste materials, during the bleaching of paper in pulp and paper mills, and as a by-product in the production of other chemicals such as PCP and the herbicide 2,4,5-T (ATSDR 1998). In the environment, dioxins and furans always occur as mixtures. Once in the environment, dioxins tend to bind to small particles or organic matter. They do not dissolve easily in water or air. As a result, they tend to settle out of the air or water and end up in soils or sediments. In sediments, dioxins are taken up by microscopic plants or animals through feeding or direct contact. Dioxins can then pass through the foodchain and become concentrated in the tissues of larger animals, especially in the fatty tissue. The bioconcentration effect is more pronounced than in the case of PCP because dioxins are much more stable in the environment, and are much harder for most organisms to break down. Dioxins in soil can be transported to surface water bodies via runoff, and animals may be exposed to them through indirect ingestion or dermal contact. Plants do not efficiently take up dioxins through their roots, but may have dioxins on their surfaces as a result of particle deposition (ATSDR 1998). Animals that eat the plants may then ingest the dioxins.
On the surface of the soil, dioxins may be broken down by sunlight, a process known as photodegradation. The half-life of TCDD on soil may be on the order of 15 years at the soil surface (Paustenbach et al 1992). This process is only effective in the top few millimeters of soil where ultraviolet light can penetrate (EPA 2000). Burial in place (by the constant accumulation of airborne dust and dirt, erosion, and the buildup of organic matter) or erosion to surface water bodies are likely the main environmental fate of dioxins in soil. Once buried (i.e. in the sub-soil), TCDD has been shown to have a half-life of up to 100 years, and likely becomes tightly bound to soil organic matter (EPA 2000).
As a result of natural and man-made processes, dioxins are found nearly everywhere in the environment. Dioxins have been found in the fat tissue of humans across the U.S., even in those who have no known exposure to dioxins. This indicates that exposure is widespread, and is likely occurring through the food supply. Foods containing animal fat, such as meat, fish, and dairy products are the most common dietary sources. Dioxins may also be passed from mother to infant through breast milk, which is high in fat.
According to an EPA summary of available studies, background levels of dioxins in soils in rural areas in North America average 2.5 parts per trillion (ppt, or 0.0025 ppb) as expressed using TEFs, with a range of between 0.1 to 6 ppt (EPA 2000). In urban areas, the average cited by EPA is 9.4 ppt (0.0094 ppb), with a range of between 2 and 21 ppt. Background levels in sediments average 5.31 ppt (0.00531 ppb) with a range of from less than 1 ppt to 20 ppt.
Exposure to high levels of dioxins is associated with chloracne, a severe skin disease, as well as other skin disorders. Such skin diseases usually result from exposure to high concentrations for an extended time period, such as in the workplace or from an industrial accident. Studies in animals have shown that long-term exposure to lower levels of dioxins can affect the liver, and may cause reproductive or developmental effects. Dioxin exposure may also be associated with changes in the immune system (Stehr-Green et al 1987). The dioxin 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) was recently listed as a "known human carcinogen" by the National Institutes of Health (NIH) based on studies in humans and animals (NIH 2001). EPA characterizes TCDD as "carcinogenic to humans," and considers mixtures of dioxins to be highly potent "likely" carcinogens (EPA 2000). Exposure to TCDD is thought to be associated with an increased risk of all cancers, rather than a specific type of cancer. TCDD is believed to be a cancer promoter, rather than an initiator (Aylward et al 1996). Cancer initiators cause direct genetic damage that can also lead to mutations. The initial mechanism by which dioxins are thought to induce adverse health effects, including cancer promotion, is by binding with a cellular protein known as the aryl hydrocarbon receptor (AhR). The chain of events that may lead to an adverse health effect following this action is poorly understood (EPA 2000). The AhR protein is part of a family of cellular proteins that play an important role in normal physiological function.
Not all dioxins and furans are as toxic as TCDD, but all are thought to cause adverse effects through the same mechanisms. Penta- and hexachloro-dioxins with chlorine atoms in the 2,3,7 and 8 positions appear to have similar toxicities, while other dioxins that do not have chlorine atoms in those positions are relatively less toxic (ASTDR 1998). To assess the toxicity of mixtures of dioxins and furans, a series of toxicity equivalency factors (TEFs) have been developed that compare the toxicity of other dioxins and furans to TCDD. The overall toxicity of a mixture can then be calculated in terms of total TCDD equivalents. The TEFs currently proposed for use by EPA were published by the World Health Organization (WHO) in 1998 (EPA 2000). The TEFs are based on existing toxicological data on individual dioxins and furans, or are estimated using a number of different methodologies. They are designed to be an interim approach pending additional research on specific dioxin and furan compounds. The current WHO TEFs are as follows (EPA 2000):
Dioxin/Furan TEFs, WHO 1998
|Dioxin (D) Congener||TEF||Furan (F) Congener||TEF|
Current estimates of the mean daily exposure in the general U.S. population to dioxins and furans is one picogram per kilogram of body weight per day (1 pg/kg/day) of TCDD equivalents (EPA 2000). A picogram is one-trillionth of a gram (0.000000001 gram). Estimates of the 95th and 99th percentile intake rates are two times the mean and three times the mean, respectively. Intake rates may be as much as three times the mean for children. The vast majority of this exposure is through the diet. Certain sub-populations, such as those who eat a particularly fatty diet, subsistence fishermen, and nursing infants may have a higher daily intake. Studies have shown that levels of dioxins and furans measured in human body fat samples have declined from the early 1980s to the present as a result of the increased regulation of emission sources and the subsequent decrease in levels measured in the environment (EPA 2000).
Cancer Risk Assessment
The potency of a carcinogen is typically estimated using mathematical models. In general, cancer potency is estimated from the linear term in the equation used to describe the observed data. The resulting number is known as a cancer slope factor, and describes the cancer risk per unit dose. For ingestion, it is expressed in terms of the risk per milligrams of contaminant ingested per kilogram of body weight per day (mg/kg/day).
In the evaluation of potential carcinogens, or cancer-causing chemicals, MDH uses a negligible excess lifetime cancer risk of 1 in 100,000, or 1 x 10-5. This means that a person exposed to a concentration of a carcinogen equal to the lifetime risk level of 1 x 10-5 for a lifetime would have up to a 1 in 100,000 chance of developing cancer from this exposure. MDH regards an incremental risk from a single source as negligible at this level, and it is a very small risk compared to the overall existing lifetime cancer rate in Minnesota of approximately 40%.
The cancer slope factor, the MDH negligible lifetime excess cancer risk number, and standard default exposure parameters are used to generate environmental screening criteria such as HRLs, HRVs, and SRVs. Site-specific information may be used where appropriate to develop more refined criteria. The common use of conservative exposure assumptions means that the actual risk from exposure to levels of contaminants at the various screening levels lies somewhere between zero and 1 in 100,000.
A possible shortcoming in this approach is the typical use of a 70-year lifetime exposure model. Chemical exposures are often unequally distributed over a lifetime, and there are critical periods of susceptibility at varying times, especially during pregnancy and childhood. Children may be especially susceptible to during periods of rapid tissue growth and development, and have a longer time in which to develop adverse health effects. A significant portion of lifetime risk may therefore be accumulated in a relatively short time. Traditional risk assessment methods do not adequately address the issue of the proportion of cancer risk accrued during different time periods when exposures are for less than a lifetime. Children also typically receive higher doses per body weight than adults (as in the case of dioxin), and may be able to absorb higher doses of some contaminants than adults, increasing their dose relative to adults for a given level of environmental exposure.
Risk Assessment and Dioxin
While levels of PCP and PAHs do exceed MPCA screening criteria in several of the sub-areas, they are not expected to significantly contribute to the cancer risk compared to dioxins and furans. This section will focus therefore on the health risks from exposure to dioxins and furans in soil and sediment in the West Area.
There has been considerable scientific debate over the potential health risks posed by exposure to dioxins, furans, and dioxin-like compounds. This may be in part due to the wide variety in responses seen in some species of test animals exposed to dioxins. The difference in the amount of dioxin needed to produce death in fifty percent of test animals exposed (a dose referred to as the LD50) differs by more than 1,000 times between some species of rodents (Aylward et al 1996). The recent listing of TCDD as a known human carcinogen by NIH was challenged in court prior to its release, and there has traditionally been debate as to the carcinogenicity of dioxins in humans. Some studies indicate that the cancer hazard to humans posed by exposure to background or environmental levels of dioxins in the diet are not significant (Aylward et al 1996). The U.S. EPA has undertaken considerable effort to conduct a reassessment of the human health risks from exposure to dioxin (EPA 2000). The EPA report concludes that dioxins are highly potent animal carcinogens and likely human carcinogens, and recommends a cancer slope factor (a measure of the relative potency of a carcinogen) of 1 x 106 (mg/kg-d)-1, a value approximately six times higher than the previous EPA cancer slope factor of 1.56 x 105 (mg/kg-d)-1 (EPA 2000). EPA also concludes that non-cancer effects from exposure to dioxin may occur at levels within 10-100 times of average intake and body-burden levels in the U.S. population, based mainly on animal studies. EPA has not established a "reference dose" for assessing exposures for non-cancer risk, but recommends a "margin of exposure" approach where estimated intakes are compared to background exposure levels. A judgment decision must then be made as to whether the incremental exposure represents an unacceptable risk.
The EPA's conclusions regarding the health risks, including cancer risks, from dioxin exposure have been questioned by both EPA's Science Advisory Board (SAB) and industry groups (Pianin 2001). The SAB recently stated, however, that the EPA reassessment document likely represents the best possible review of the current literature, that key issued raised by the SAB have been addressed, and that any shortcomings are likely a result of gaps in the available data that limit EPA's ability to develop a more refined quantitative risk assessment (EPA 2001).
While it is unlikely that exposure has resulted in any adverse health effects in residents near the site, concentrations of dioxins (as expressed in TCDD TEQs) in surficial soil and sediments exceed the current ATSDR residential screening criterion of 50 ppt in six of the eight sub-areas in the West Area. It is important to note that the less toxic hepta- and octa- dioxin congeners predominate at the site, and levels of TCDD are thousands of times less than the concentrations of these congeners.
The ATSDR screening, evaluation, and action criteria (and MPCA SRVs) were developed using a cancer slope factor of 1.56 x 105 (mg/kg-d)-1. If the EPA recommended cancer slope factor of 1 x 106 (mg/kg-d)-1 was used to develop new criteria and SRVs, they would be approximately six times lower if the same methodology were used. While the current ATSDR criterion is designed for screening use only, MDH recommends the use of the ATSDR screening criterion as a site cleanup goal for dioxin contaminated soil in the West Area in this case. This recommendation is due to the West Areas location being immediately adjacent to residential areas, its use for a variety of recreational activities, and its potential contribution of dioxin contaminated soils and sediments to Twin Lakes. A value of 50 ppt or less of dioxins and furans in soil would fall within the midrange of criteria calculated using the proposed EPA cancer slope factor, and is therefore recommended by MDH as an appropriate cleanup goal and is consistent with Minnesota's policy of recommending remediation goals which are based on an excess lifetime cancer risk of 1x10-5 . This is a state goal, not an ATSDR goal. Further discussion of other screening and action levels for dioxin is presented in Appendix 4.
The presence of dioxins in the West Area remains a significant concern for MDH. Due to the potential toxicity of these compounds and the fact that people are already exposed to them through their diet, MDH feels that potential exposure to dioxins from man-made sources such as the West Area should be minimized as much as possible.
ATSDR's Child Health Initiative recognizes that the unique vulnerabilities of infants and children make them of special concern to communities faced with contamination of their water, soil, air, or food. Children are at greater risk than adults from certain kinds of exposures to hazardous substances at waste disposal sites. They are more likely to be exposed because they play outdoors, ingest more soil, and often bring food into contaminated areas. They are smaller than adults, which means they breathe dust, soil, and heavy vapors close to the ground. Children also weigh less, resulting in higher doses of chemical exposure per body weight. The developing body systems of children can sustain permanent damage if toxic exposures occur during critical growth stages. Most importantly, children depend completely on adults for risk identification and management decisions, housing decisions, and access to medical care.
Given the sites wooded nature and its location near a residential area and parks, it has served as an attraction for local youth over the years. While the apparent foot trails were generally not in areas of high contamination, it is possible that children have been exposed to contamination in surface soil or sediments in the West Area during activities such as playing or hiking if these activities occurred in contaminated areas. If lake sediments or surface water are contaminated, exposure also could have occurred to recreational users of Twin Lakes, or from the consumption of fish. The extent of these potential exposures is difficult to evaluate. The construction of a fence around the majority of West Area and the use of warning signs as an interim response measure should limit access and minimize future exposures.