FINAL PATHWAYS ANALYSIS REPORT FOR THE BASELINE RISK ASSESSMENT FOR ANNISTON PCB SITE OPERABLE UNIT 4 ANNISTON, ALABAMA

Transcription

1 FINAL PATHWAYS ANALYSIS REPORT FOR THE BASELINE RISK ASSESSMENT FOR ANNISTON PCB SITE OPERABLE UNIT 4 ANNISTON, ALABAMA Prepared for: U.S. Environmental Protection Agency Region 4 Atlanta, Georgia Contract No. EP-S December 2009 Prepared by: A Service Disabled Veteran Owned Small Business

2 Table of Contents 1. INTRODUCTION Overview of the PAR PAR Contents SITE BACKGROUND AND SETTING Site Location and Description Site History Land and Water Use EXPOSURE UNITS Characterization Areas Determination of Exposure Units HUMAN EXPOSURE PATHWAYS Conceptual Site Model Source of Contamination, Release and Transport Mechanisms, and Receiving Media Primary Exposure Media Identification of Exposure Pathways Direct Contact Exposure Agricultural Products Consumption Fish Consumption Characterization of Potentially Exposed Populations Residents Recreational Users Utility Workers Farmers Recreational Fisherman EXPOSURE ASSESSMENT Agricultural Modeling i

3 Table of Contents, Continued Soil-to-Plant Transfer Mechanisms Prediction of Concentrations in Vegetables Prediction of Concentrations in Animal Feed Prediction of Concentrations in Animal Products Exposure Parameters Constant Exposure Parameters Recreational User Exposure Parameters Utility Worker Exposure Parameters Farmer Exposure Parameters Recreational Fishermen Exposure Parameters REFERENCES ii

4 List of Figures Figure 2-1 Figure 2-2 Figure 3-1 Figure 3-2 Figure 3-3 Figure 3-4 Figure 3-5 Figure 3-6 Figure 3-7 Figure 3-8 Figure 3-9 Figure 3-10 Figure 4-1 Anniston PCB Site Location Map Operable Unit Locations Operable Unit 4 Location Characterization Area C1 Characterization Area C2 Characterization Area C3 Characterization Area C4 Characterization Area C5 Characterization Area C6 Characterization Area C7 Characterization Area C8 Characterization Area C9 Conceptual Site Model iii

5 List of Tables Table 3-1 Table 5-1 Table 5-2 Table 5-3 Table 5-4 Table 5-5 Table 5-6 Table 5-7 Description of Characterization Areas Soil Contact Exposure Parameters Agricultural Product Ingestion Exposure Parameters Vegetables and Beef Agricultural Product Ingestion Exposure Parameters Dairy, Chickens, and Eggs Fish Ingestion Exposure Parameters Summary of Agricultural Product Intake Rates (As Consumed) Fraction of Food Intake that is Home Produced Derivation of Agricultural Product Ingestion Rates iv

6 Acronyms and Abbreviations ABS ADEM AF ALT AT Bs BTF BTF ag BTF beef BTF bg BTF chicken BTF eggs BTF fat BTF milk BW C ag C beef C bg C chicken C eggs C grain C i dermal absorption factor Alabama Department of Environmental Management adherence factor Alabama Land Trust averaging time soil bioavailability factor biotransfer factors biotransfer factor for above ground plants biotransfer factor for fat content in beef biotransfer factor for root vegetables biotransfer factor for fat content in chicken biotransfer factor for fat content in eggs biotransfer factor for animal fat biotransfer factor for fat content in milk of dairy cattle body weight concentration of PCBs in above ground produce due to root uptake concentration of PCBs in beef concentration of PCBs in below ground produce due to root uptake concentration of PCBs in chicken concentration of PCBs in eggs concentration of PCBs in grain concentration of PCBs in plant type i eaten by the animal v

7 Acronyms and Abbreviations C soil CA CERCLA cm 2 cm 3 CSM CTE ED EF EPA EPC EU FI FI grain FI i FI soil HHRA HHRAP IAF IR grain IR i IR soil concentration of PCBs in soil characterization area Comprehensive Environmental Response, Compensation, and Liability Act square centimeter cubic centimeter conceptual site model central tendency exposure exposure duration exposure frequency U.S. Environmental Protection Agency exposure point concentration exposure unit fraction ingested fraction of grain grown on contaminated soil and ingested by the animal fraction of plant type i (forage, silage, and grain) grown on contaminated soil and ingested by the animal fraction of ingested soil from floodplain Human Health Risk Assessment Human Health Risk Assessment Protocol intestinal absorption factor ingestion rate of grain ingestion rate of plant type i eaten by the animal per day ingestion rate of soil eaten by the animal per day vi

8 Acronyms and Abbreviations IRS JMWA Kds kg Kow MF mg/cm 2 mg/kg OU PAR PCB RCF RCRA RFI/CS RME SA UCL soil ingestion rate J.M. Waller and Associates soil-water partitioning coefficient kilogram octanol/water partitioning coefficient metabolism factor milligram per square centimeter milligram per kilogram Operable Unit Pathways Analysis Report polychlorinated biphenyl root concentration factor Resource Conservation and Recovery Act RCRA Facility Investigation/Confirmatory Sampling reasonable maximum exposure exposed skin surface area upper-confidence limit vii

9 1. INTRODUCTION J.M. Waller and Associates, Inc. (JMWA) was tasked by the U.S. Environmental Protection Agency (EPA) to perform a human health risk assessment (HHRA) for Operable Unit 4 (OU-4) of the Anniston Polychlorinated Biphenyl (PCB) Site (the Site). The risk assessment will be performed under Contract No. EP-S , Task Order No. 01. The Site refers to the area where hazardous substances, including PCBs (associated with releases or discharges as a result of the operations and waste disposal from the Anniston Plant by Solutia Inc. (Solutia), Monsanto Chemical Company (Monsanto), and their predecessors), have come to be located. Solutia s Anniston plant encompasses approximately 70 acres of land and is located about 1 mile west of downtown Anniston, Alabama. OU-4 encompasses the length of Choccolocco Creek and its floodplain from the confluence with Snow Creek, including the backwater area and upstream on Snow Creek to Highway 78, to Lake Logan Martin. This Pathways Analysis Report (PAR) was developed to characterize the exposure setting and receptor characteristics for OU-4. It identifies the current and future-use exposure pathways by which populations may be exposed to contaminated media. Exposure pathways were identified based on consideration of the sources and locations of contaminants, the likely environmental fate of the contaminants, and the location and activities of the potentially exposed populations. The PAR identifies the potential exposure points and routes of exposure for each exposure pathway, as well as parameters regarding receptor characteristics and behavior (e.g., body weight, ingestion rate, exposure frequency). It focuses on the potential exposure to PCBs, the primary Site contaminant. Other contaminants such as dioxin-like PCB congeners, dioxins, furans, and metals (i.e., mercury) will also be considered in the HHRA but the extent to which these contaminants will be evaluated is contingent upon the results of the OU-4 sampling efforts. The PAR does not include any site related concentration data, toxicity criteria, or risk estimates. This information will be included in the HHRA. The main purpose of the PAR is to allow the stakeholders to review and comment on the approach to the exposure assessment before the HHRA is completed so that appropriate changes can be made to the assumptions that will be used to estimate exposure and risk. 1-1

10 1.1 OVERVIEW OF THE PAR In preparation of this PAR, the JMWA team reviewed the available information pertaining to the Site from other OUs (i.e., OU-1/2 and OU-3), as well as available information on land and water uses along the Choccolocco Creek. Members of the JMWA team also visited the area on several occasions, floated major portions of the Choccolocco Creek, and researched current and future land use trends in the area. This information was applied to the development of the exposure assessment presented in this document. The PAR was developed in accordance with EPA Guidance set forth in the following documents: Specific risk assessment guidance from EPA Region 4. Risk Assessment Guidance for Superfund: Human Health Evaluation Manual, Part A (EPA, 1989). Human Health Evaluation Manual, Supplemental Guidance: Standard Default Exposure Factors (EPA, 1991). Guidelines for Exposure Assessment (EPA, 1992). Exposure Factors Handbook, Volumes I, II, and III (EPA, 1997). Supplemental Guidance to RAGS: Region 4 Bulletins, Human Health Risk Assessment Bulletins (EPA, 2000). Risk Assessment Guidance for Superfund: Human Health Evaluation Manual, Part D (EPA, 2001). Supplemental Guidance for Developing Soil Screening Levels for Superfund Sites (EPA, 2002a). CSFII Analysis of Food Intake Distributions (EPA, 2003a). Risk Assessment Guidance for Superfund: Human Health Evaluation Manual, Part E, Supplemental Guidance for Dermal Risk Assessment. Final (EPA, 2004). Human Health Risk Assessment Protocol for Hazardous Waste Combustion Facilities (EPA, 2005). Child-Specific Exposure Factors Handbook (EPA, 2008). Risk Assessment Guidance for Superfund, Volume I: Human Health Evaluation Manual, Part F, Supplemental Guidance for Inhalation Risk Assessment. Final (EPA, 2009). 1.2 PAR CONTENTS The PAR is comprised of six sections, as follows: Section 1 Introduction. Section 2 Site Background and Setting Describes the Site location and description, the Site history, and the land use. 1-2

11 Section 3 Exposure Units Describes the manner in which the OU-4 area will be divided based on land use and contamination patterns for quantitative evaluation in the HHRA. Section 4 Human Exposure Pathways Presents a conceptual site model and identifies the exposure pathways and the potentially exposed receptors. Section 5 Exposure Assessment Describes the agricultural modeling approach and presents the receptor-specific exposure parameters. Section 6 References Contains the report references. 1-3

12 2. SITE BACKGROUND AND SETTING 2.1 SITE LOCATION AND DESCRIPTION The Anniston PCB Site is located in Calhoun County in the north-central part of Alabama (Figure 2-1). The Anniston PCB Site consists of the entire geographic area in Anniston and its environs where PCBs have come to be located. EPA believes that the vast majority of the PCBs in the Anniston area were released from the operations of the former Monsanto Company's Anniston PCB manufacturing plant. Today the former PCB plant property is owned by Solutia, Inc. Solutia s Anniston plant encompasses approximately 70 acres of land and is located about 1 mile west of downtown Anniston. Solutia currently produces para-nitrophenol and polyphenyl compounds at the Anniston plant. EPA has been performing investigations in Anniston under the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA) to evaluate the threat to public health, welfare, or the environment posed by hazardous substances, including PCBs. The Anniston PCB Site has been divided into OUs to facilitate the investigation and cleanup. OU-1/OU-2 is generally the potentially impacted residential and non-residential areas from the Solutia facility to Highway 78, along Snow Creek. OU-3 is the Solutia facility. OU-4, the focus of this report, is defined as Snow Creek and its floodplain from Highway 78 to the confluence with Choccolocco Creek, and Choccolocco Creek and its floodplain from the confluence of Snow Creek to Lake Logan Martin in Talladega County. OU-4 is the largest of the Anniston PCB Site areas. It encompasses over 35 miles of the Choccolocco Creek and the associated floodplain area, and is used for a variety of purposes. Figure 2-2 presents the locations of the Anniston PCB Site Operable Units. 2.2 SITE HISTORY A thorough discussion of the manufacturing history at the Solutia facility was included in the Resource Conservation and Recovery Act (RCRA) Facility Investigation/Confirmatory Sampling (RFI/CS) Work Plan for the Anniston, Alabama, Facility (Golder, 1997). As reported therein, manufacturing operations began in 1917 with the production of ferro-manganese, ferro-silicon, ferro-phosphorous compounds, and phosphoric acid (added later) by the Southern Manganese 2-1

13 Corporation. In 1927, the production of organic chemicals began with the introduction of biphenyl, which remains a major product today. In 1930, Southern Manganese Corporation became Swann Chemical Company (Swann); in May 1935, Monsanto Chemical Company purchased Swann. PCBs were produced at the plant from 1929 until In 1997, Monsanto Company formed Solutia Inc. and transferred ownership for certain chemical divisions. Solutia Inc. currently produces polyphenyl compounds at the Anniston plant. During its operational history, the plant disposed of hazardous and nonhazardous waste at various areas, including the west end landfill and the south landfill, which are located adjacent to the plant. The west end landfill encompasses six acres of land, located on the southwestern side of the plant. The west end landfill was used for disposal of the plant s wastes from the mid- 1930s until approximately In 1960, Monsanto Company began disposing of wastes at the south landfill. Disposal of wastes at the south landfill ceased in approximately During the time that the west end landfill and the south landfill were used to dispose of wastes, there was a potential for hazardous substances, including PCBs, to be released from the landfills via soils and sediments being transported in surface water leaving the property. In addition, during the time that PCBs were manufactured by Monsanto Company at its Anniston plant, an aqueous stream flowing to a discharge point (currently identified as DSN0001) on Monsanto Company s Anniston plant Site contained PCBs. Discharge from that discharge point flowed to a ditch, the waters of which flowed toward Snow Creek. Sampling by EPA, Solutia Inc., Alabama Department of Environmental Management (ADEM), and other parties has indicated that sediments in drainage ditches leading away from the plant, Snow Creek, and Choccolocco Creek, as well as sedimentary material in the floodplains of these waterways, contain varying levels of PCBs and other contaminants. The Site has been evaluated extensively since Environmental work has included a combination of investigative and remedial efforts conducted pursuant to a variety of environmental permits. The environmental response efforts under RCRA included the general areas of the Solutia manufacturing plant, which were termed the "On-Site" area, and areas downstream of the Solutia manufacturing plant, termed the "Off-Site" area. 2-2

14 2.3 LAND AND WATER USE The HHRA will evaluate the potential exposure and risks associated with the current and reasonably anticipated future uses of the Choccolocco Creek and its floodplain that includes area in both Calhoun and Talladega counties. The current uses of the floodplain and Creek will form the basis for the evaluation of baseline (existing) conditions. Potential future land and Creek uses will form the basis of the evaluation of the future uses within OU-4. Information about land use trends is critical in developing realistic assumptions regarding future use. A number of information sources will be used to identify the land and Creek uses, including: Aerial photographs and maps. Field notes and observations of EPA and contractor personnel. Studies of fishing and consumption habits along the Creek and other waterbodies within Alabama. Discussions with agricultural extension and farm service agents from Calhoun and Talladega Counties regarding the typical farming practices. Alabama Land Trust acquisitions and easements. The OU-4 area includes numerous properties owned by private and public entities that are used for residential, recreational, agricultural, and commercial/industrial purposes with a significant portion currently forested. The floodplain area is approximately 6,000 acres. The percentage of each land use in the floodplain is as follows (Arcadis, 2008): Agriculture 40 % Forest 38 % Scrub 10 % Commercial/Industrial 7 % Residential 3 % Park 1 % Waste-water treatment plant 1 % Agriculture and forest land are the predominant land uses in OU-4. The primary agricultural use of the floodplain along the Creek is open pasture land where beef cattle are grazed. According to local agricultural and farm service agents, at the current time there are no dairy cattle and only limited row crop production in Calhoun County in the floodplain other than crops such as corn and soybeans that can be used as silage for cattle (Butler, 2009 and West, 2009). Further 2-3

15 downstream in Talladega County, row crops are more common (wheat, cotton, corn and soybeans) and acreage in row crops exceeds acreage used to raise beef cattle (Browning, 2009 and Jurriaans, 2009). As with Calhoun County, there are no current dairy farms with grazing cows in the floodplain in Talladega County. Agricultural Extension agents and Farm Service agents for both counties indicated that locally raised beef consumption is not typical and that the common practice is to sell livestock to local and/or regional buyers (Butler, 2009, Browning, 2009, Jurriaans, 2009, and West, 2009). While dairy operations are not currently occurring in the floodplain, it is possible that such practices could occur at some point in the future. Small backyard gardens and chicken raising operations are present at many locations in both counties, although it is unclear whether that practice occurs in the floodplain areas. It is possible, however, that backyard gardens and chicken raising operations could occur in the floodplain at some point in the future. Fishing is possible anywhere along the Choccolocco Creek but it is likely that the majority of the fishing occurs at and around bridge crossings where access to the bank fishing area is easy, or is engaged in by local landowners with private access. In addition, given the nature, size and accessibility of the Creek, it is likely that fishing is more common at locations further downstream than at locations closer to the confluence with Snow Creek. Solutia, Inc. is currently developing a fish consumption study that will provide some useful information on current fishing habits along the Creek, although the results of the survey are not likely to reflect baseline conditions (i.e., fishing conditions likely to occur without the contamination present and a fish consumption advisory in place). There are a number of recreational areas located throughout OU-4. The forested areas provide attractive habitat for various recreational activities including hiking, fishing, canoeing, wading, etc. It is also likely that local adolescents frequent specific areas along the creek. Hunting is common at many areas as demonstrated by the deer hunting blinds (shooting houses) interspersed throughout the floodplain. There are a limited number of residential areas within the floodplain. However, there are a few residences and residential developments located adjacent to the floodplain area. The 2-4

16 commercial/industrial areas consist of the airport property and two waste-water treatment plants. Natural gas pipelines, a railroad, and aboveground utility lines transect the floodplain at various locations. Future plans regarding land use in OU-4 are important to the determination of reasonably anticipated future uses of OU-4 and the potential for future contact with contaminated media. The Alabama Land Trust (ALT) is currently in the process of attempting to develop a Protection Corridor for Choccolocco Creek. The Protection Corridor is a conservation easement that limits the development and use of the floodplain within certain distances from the Creek bank. This is important information for the HHRA because the land use and potential exposure to contaminants within the easement may be different from exposure outside of the easement. The properties that have been either purchased by the ALT or have entered the Protection Corridor will be identified and evaluated accordingly in the HHRA. 2-5

17 3. EXPOSURE UNITS Because of the large area associated with OU-4, it will be necessary to identify the Exposure Units (EUs) that will be the focus of the HHRA. The EUs will initially be based on the Characterization Areas (CAs) developed by Solutia. Additional information obtained after the development of the CAs also will be used to determine the EUs, including floodplain contamination levels and site-specific land uses. Because the analytical data results from the floodplain sampling program are not part of the PAR process, final EUs will not be identified in the PAR. 3.1 CHARACTERIZATION AREAS The OU year floodplain area encompasses over 35 miles of the Choccolocco Creek and approximately 6,000 acres. To adequately characterize the nature and extent of the contamination, sub-divisions of the area were developed by Solutia. Existing PCB data indicate that PCB concentrations tend to decrease with both distance downstream and distance out into the floodplain. This resulted in Solutia segmenting the OU-4 area in lateral and longitudinal directions. Based on topographical and hydraulic features, OU-4 was segmented into nine sections along its length and up to four sections laterally for characterization purposes. The main lateral split was along Choccolocco Creek itself (i.e., north and south side of the creek). An additional lateral segmentation was based on the distance of 500 feet from the creek bank or the edge of the 100-year floodplain, whichever is closer. If the floodplain in the CA extended beyond 500 feet, a final lateral segmentation was established as 500 feet from the creek bank to the edge of the 100-year floodplain. Table 3-1 summarizes the CAs. Figure 3-1 through 3-10 present the CAs that were developed by Solutia. For characterization purposes, floodplain soil sample numbers were determined by Solutia based on the size of the CA and its location within OU-4. Higher numbers of samples per acre were collected at CAs closest to the creek bank and farthest upstream. A maximum sample density of one sample per five acres or a minimum of 20 samples per CA was the goal for the 2009 floodplain sampling effort (Arcadis, 2008). Table 3-2 summarizes the intended number of samples per CA. 3-1

18 3.2 DETERMINATION OF EXPOSURE UNITS Although the CAs are designed to characterize the nature and extent of the contamination in the floodplain, a variety of land uses and exposure scenarios can exist within a single CA. This makes it necessary to consider subdividing the CAs into smaller areas (i.e., Exposure Units) to facilitate a meaningful exposure assessment. Thus, EUs, as a subset of CAs, may need to be identified for evaluation in the HHRA in certain CAs. The results of the 2009 floodplain soil sampling program will be critical in the determination of the EUs. The intent of the EU determination process is to focus on CAs that have elevated levels of PCBs. For example, if the CA-specific exposure point concentration (EPC), which can be defined as the 95% upper confidence level of the mean (95% UCL), is less than 1 mg/kg, the entire CA may be eliminated from further consideration and no subdivision into EUs would be necessary, assuming that the data did not show the presence of elevated concentrations in sensitive locations such as residences or agricultural areas. Alternatively, a CA may have a 95% UCL PCB concentration greater than 1 mg/kg with significant contamination in one area with a specific land use and non-detected results at other areas with different land uses. In this case, an EU would be identified for the area with the land use with elevated concentrations and other parts of the CA without contamination may be eliminated from further evaluation. The basis for any subdivision of EUs will be land use and/or property ownership and may cross the borders of the CAs if necessary. 3-2

19 4. HUMAN EXPOSURE PATHWAYS Potential human exposure pathways for the Site will be defined based on the current and reasonably anticipated future land and water uses as well as the results of the ongoing sampling program. The objective of this section is to present the conceptual site model for human exposure and the potential exposure pathways that will be used to characterize the potentially exposed populations. 4.1 CONCEPTUAL SITE MODEL A conceptual site model (CSM) describes the contaminant sources, the release and transport mechanisms, the receiving media, the exposure media, the exposure routes, and the potentially exposed populations. The primary objective of the conceptual site model is to identify complete and incomplete exposure pathways. A complete exposure pathway has all of the above-listed components, whereas an incomplete pathway is missing one or more. Figure 4-1 illustrates the CSM that was developed for OU-4. Each component of the conceptual site model is examined in detail in the following sections Source of Contamination, Release and Transport Mechanisms, and Receiving Media PCBs released in the past from the current Solutia facility have been transported in storm water in Snow Creek and ultimately discharged into the Choccolocco Creek. The release and transport processes affecting the fate and effect of PCBs within the Choccolocco Creek and its floodplain are interrelated and complex. The following potential contaminant transport pathways have been identified: Surface runoff and drainage from the Solutia facility in Anniston. Erosion and downstream transport of contaminated bank soil. Sediment contamination via runoff carrying suspended soil particles contaminated with PCBs. Floodplain soil contamination via deposition of suspended river sediment during flood events. Erosion of contaminated floodplain soil (surface and subsurface) during flood events, and subsequent deposition as contaminated river sediment. 4-1

20 Bioaccumulation and cycling of PCBs within the terrestrial and aquatic food chains exposed to contaminated soil, surface water, and sediment Primary Exposure Media Based on the review of the current and potential land and water uses, the following primary exposure media are of potential concern to humans in OU-4: Soil (floodplain). Sediment. Surface water. Agricultural products. Fish. 4.2 IDENTIFICATION OF EXPOSURE PATHWAYS The length of the Choccolocco Creek within OU-4, and the size and multiple uses of the floodplain pose a significant challenge to effectively assessing human health risk from direct and indirect exposures for both current and potential future uses. Children and/or adults may be exposed to soil while engaging in a variety of activities around their homes or recreational activities at other locations. Adults may be exposed to soil while working in agricultural, landscaping, utility maintenance, and other occupations. Sediment and surface water exposure may occur along the riverbanks or in shallow areas of the Creek during recreational activities such as fishing, canoeing, swimming, or wading. Anglers, farmers, and hunters and their families may be exposed to Site contaminants from consumption of fish caught from the Creek, crops and other agricultural products raised in the floodplain, and/or consumption of game. The potential exposure associated with consuming game (e.g., deer and turkey) taken from the floodplain is expected to be negligible given the home ranges of the game and the limited time that contact with the affected media in OU-4 is likely to occur. Therefore, consumption of game will not be quantitatively evaluated in the HHRA unless floodplain soil data indicate that a potential exposure over an applicable range is possible. For OU-4, three main points of potential contact between contaminated media and humans are likely: Direct contact with contaminated media (soil, sediment, and surface water). 4-2

21 Consumption of agricultural products (e.g., vegetables, beef) from the floodplain. Consumption of fish. The following sections describe the possible receptors and exposure pathways considering both current and potential future land and water uses. An identified pathway does not imply that exposures are actually occurring, only that the potential exists for the pathway to be complete Direct Contact Exposure The direct contact portion of the HHRA will evaluate the potential exposure to floodplain soil, sediment, and surface water Floodplain Soil Exposure For direct floodplain soil contact, the following exposure pathways will be evaluated: incidental soil ingestion, dermal contact and absorption, and inhalation of particulate. Typically, the inhalation of particulate exposure pathway results in exposure and risks that are minimal compared to the exposure and risks associated with the incidental ingestion and dermal contact and absorption exposure pathways. An analysis will be performed in the HHRA assuming the worst-case PCB concentrations in the soil and the most conservative inhalation exposure parameters to determine if the inhalation of particulate pathway warrants further evaluation in the HHRA. If the conservative analysis results in negligible exposure and risks, the inhalation of particulate pathway will be eliminated from further consideration in the HHRA Sediment and Surface Water Exposure Consistent with EPA Region 4 guidance, direct contact with sediment in underwater areas will not be quantitatively evaluated in the HHRA because of infrequent contact by human receptors. Based on the low levels observed in the available surface water data, it is assumed that the surface water contact exposure scenarios will be eliminated from consideration in the HHRA. A risk-based surface water screening evaluation will be performed in the HHRA Agricultural Products Consumption The focus of this portion of the HHRA will be the current and potential future food production activities within the floodplain. Consideration will be given to the homeowner (farmer) who 4-3

22 grows vegetables and crops in the floodplain and raises livestock in the floodplain. The ingestion of agricultural products will take into account the current agricultural practices in OU-4. It will also consider the reasonably anticipated future agricultural practices. In contrast to the direct contact and fish consumption portions of the HHRA, PCBs in the agricultural products consumed by humans will be estimated using models. The models will predict the degree to which PCBs measured in the floodplain soil are transferred to plants (root uptake) and animals (incidental soil ingestion and ingesting feed grown in the floodplain). Model input values will be based on site-specific information (when available) that is relevant to the site-specific conditions, including regional farm management practices Fish Consumption The potential exposure and risks attributable to consuming fish caught in the Choccolocco Creek will be evaluated. Recreational fish consumption will be evaluated in the HHRA. The analytical data used to determine the fish exposure point concentrations will be derived from samples that represent fish species, fish length, and fish tissue (fillet) that are most typically caught and consumed by the local population. 4.3 CHARACTERIZATION OF POTENTIALLY EXPOSED POPULATIONS Residents Areas used for residential purposes are located within the floodplain area. It is assumed that young children and adults contact the surface soil in the areas immediately around the location of the residence. The residentially used areas will be delineated and evaluated in the HHRA by comparing the PCB concentrations (95% UCL) to 1 mg/kg, the residential remedial level used in OU-1/2. If any of the 95% UCL values (or maximum detected values if 95% UCL cannot be calculated) are greater than 1 mg/kg, the area will be evaluated further. These properties will not be evaluated in the HHRA Recreational Users Exposure to individuals who recreate along the Choccolocco Creek and the floodplain will be examined. This population will be assumed to contact the surface soil in the floodplain through 4-4

23 the incidental ingestion and dermal contact and absorption exposure routes. The potential exposure associated with the recreational user population will be based on a number of recreational activities that can occur within the floodplain (e.g., bank fishing, hunting, hiking, walking, riding all-terrain vehicles, etc). Young child, adolescent, and adult receptors will be evaluated depending on the EU. Adolescents (7 through 16 years) and adults will be the most frequently evaluated recreational receptors based on the nature of the area and the difficulty a young child would likely experience attempting to recreate in the floodplain area. The young child (1 through 6 years) will be considered at areas with easy access to the floodplain area (near a residence) Utility Workers Utility workers may be exposed to contaminants in surface and subsurface soil via incidental ingestion and dermal contact during activities such as easement or equipment maintenance, and/or the installation of new equipment such as utility poles or piping. This potential exposure is assumed to be intensive for a short duration. The exposure will be assumed to occur along utility lines within the floodplain. A construction worker scenario is not considered to be a complete exposure scenario because flooding events will preclude major construction in the floodplain Farmers Similar to the residents, contact with soil in the delineated residential areas associated with the farm property will be evaluated by comparing the PCB concentrations (95% UCLs) to 1 mg/kg. If any of the 95% UCL values (or maximum detected values if 95% UCL cannot be calculated) are greater than 1 mg/kg, the area will be evaluated further. In the fields, the farmer (adult) will be assumed to intensively contact the floodplain surface soil (incidental ingestion and dermal contact and absorption) while tilling the soil and planting and harvesting the crops. Depending on the EU being evaluated, the farmer, including a young child, may be assumed to consume agricultural products (e.g., vegetables and beef) raised in the floodplain. The degree to which agricultural products are consumed will be determined on an EU-specific basis and will incorporate site-specific information. 4-5

24 4.3.5 Recreational Fisherman This population, including a young child and an adult, will be assumed to ingest fish caught in the Choccolocco Creek. The fish data will be reviewed to determine if the data from adjacent fish collection locations can be combined in the HHRA. Subsistence level fish ingestion from fish caught in the Choccolocco Creek was considered for evaluation but it was determined to be unreasonable based on the local demographics, a lack of any evidence supporting this practice, the likely inability of the Creek to support subsistence level ingestion, and more attractive fishable waterbodies nearby. 4-6

25 5. EXPOSURE ASSESSMENT The objective of the exposure assessment is to estimate the nature, extent, and magnitude of potential exposure of humans to PCBs and potentially other Site contaminants that were measured or modeled in the affected media, considering both current and future land and water uses. To provide a range of exposure and risks, the reasonable maximum exposure (RME) and central tendency exposure (CTE) scenarios will be evaluated (EPA, 1992). The RME, an estimate of the high-end exposure in a population, is based on a combination of average and high-end estimates of exposure parameters typically representing the 90 th percentile or greater of actual expected exposure. The CTE represents an estimate of the average exposure in a population and is based on central estimates of exposure parameters. This section presents the approach for modeling PCBs in agricultural products, and the exposure parameters for the receptors that will be evaluated in the HHRA. 5.1 AGRICULTURAL MODELING An investigation of current agricultural practices indicates that the primary uses of the floodplain in OU-4 are cattle grazing (for beef production) and crops (for direct sale and to a lesser extent, cattle feed) (Butler, 2009, Browning, 2009, Jurriaans, 2009, and West, 2009). Dairy production is no longer practiced in the floodplain areas of OU-4, according to farm service agents in Calhoun and Talledega Counties, and it is not clear that chickens, eggs, or garden vegetables are commonly raised in floodplain soil, although it is possible that this could change in the future (Butler, 2009, Browning, 2009, Jurriaans, 2009, and West, 2009). To account for current practices in floodplain areas, namely beef cattle and crop production, measured soil concentrations in specific CAs and/or EUs will be used to estimate exposure to local land owners (farmers) from consumption of those products to the degree that PCB concentrations are measured in areas that are currently used for those purposes. To account for potential future uses of the floodplain that could result in exposures, such as dairy operations, chicken and egg production, and home garden vegetable production, estimated exposure will be modeled based on a range of floodplain soil concentrations developed from the 5-1

26 results of the current sampling being performed by Solutia. This will allow for all interested parties to determine whether current soil concentrations are likely to result in unacceptable risks if farming practices change in the future. If additional investigation of floodplain sampling results and local uses indicates that these practices (i.e., dairy, chickens/eggs, vegetables) currently occur at specific locations in the floodplain in OU-4, modeling will be performed on a CA or EU-specific basis. Models will be used to predict the concentrations of PCBs in plants (i.e., vegetables and animal feed) and animal products. The models described below will be based on total PCBs, but if other site-related contaminants (e.g., dioxin-like PCB congeners, dioxins, and furans) are determined to be of concern, the models will also be used for those compounds. The approach and models presented in EPA s Human Health Risk Assessment Protocol for Hazardous Waste Combustion Facilities (HHRAP) (EPA, 2005) will be used. The types of plants that will be evaluated include above ground vegetables, below ground (root) vegetables, and animal feed (e.g., pasture grass and silage). The predicted concentrations of PCBs in vegetables will be used to estimate exposure from human consumption of home grown garden vegetables. The predicted concentrations in animal feed will be used to model uptake into animals grazing in the floodplain and consuming feed raised in the floodplain. The models in the EPA HHRAP that will be used in the HHRA are designed to be conservative and may result in an over-estimate of the concentrations of PCBs in the agricultural products of interest and a potential overestimate of risk. However, given the lack of site-specific concentration data for these products, the generic models recommended by EPA in the HHRAP (EPA, 2005) will be used as an initial screening approach and the conservatism will be addressed in the HHRA. This could be done by simply describing the likely conservative nature of the models, refining EPA s models, reviewing the current literature on the PCB uptake into agricultural products and developing alternative models, or collecting site-specific data. The decision on exactly how to deal with these issues will be determined based on the results of the screening approach. 5-2

27 5.1.1 Soil-to-Plant Transfer Mechanisms This section describes the mechanisms by which PCBs can migrate from the soil to plant tissue. Contaminants such as PCBs are transferred from soil to plant tissue by: 1) Root uptake from soil and transfer into above ground vegetation. 2) Partitioning from soil to root vegetables. The biotransfer factors (BTF) for above ground plants (BTF ag ), including vegetables and animal feed, will be calculated on a dry weight basis using the correlation equation from Travis and Arms (1988) as presented in Equation 5-1. As previously described by EPA (1995), the BTF values for most compounds are a function of water solubility, which is inversely proportional to octanol/water partitioning coefficient (Kow). Thus, for compounds with a high Kow value (e.g., PCBs), which indicates very low water solubility, the potential transfer is expected to be minimal. The correlation equation developed by Travis and Arms does not distinguish between above ground produce, forage, silage, or grain. The equation below was derived from experiments performed on compound classes such as DDT, pesticides, dioxins, furans, and PCBs. Therefore, because of the similarities between the test compound classes and the OU-4 contaminants, it is considered by EPA to be a valid modeling approach. log BTFag = (log Kow) Equation 5-1 The log Kow value used in the modeling analyses will be 6.5. This value is for Aroclor 1254 and was obtained from EPA s HHRAP (EPA, 2005). The BTF for root vegetables (BTF bg ) will be based on a root concentration factor (RCF). The RCF value will be calculated on a wet weight basis based on experiments by Briggs et al. (1982) using Equation 5-2. The equation below is specific to compounds with a log Kow value of greater than

28 Equation 5-2 log RCFwet weight = 0.77 x log Kow The log RCF and a soil-water partitioning coefficient (Kds) value will be used to calculate the BTF bg on a wet weight basis (Equation 5-3). A Kds value of 24,535 (cm 3 /gram) based on Aroclor 1254 will be used (EPA, 2005). An empirical correction factor of 0.01 will be applied to the calculated BTF bg value to reduce the PCB uptake to root vegetables. Because of the protective outer skin, size, and shape of below ground produce, transfer of PCBs to the center of the produce is unlikely (EPA, 2005). Equation 5-3 BTF bg = log RCF Kds wet weight x Prediction of Concentrations in Vegetables Home grown produce will be evaluated in two categories: above ground vegetables and below ground (root) vegetables. The soil-to-plant BTFs described in the previous section will be applied to the exposure point concentration in the EU of interest to yield an area-specific estimate of the concentration of PCBs in home grown produce (see Equations 5-4 and 5-5). In the case of future exposure, uptake of PCBs into vegetables will be estimated based on a range of floodplain soil concentrations developed from the results of the current sampling being performed by Solutia. The modeled above ground produce concentrations will be in dry weight. For consistency with the vegetable ingestion rates discussed in Section 5.2, it will be necessary to convert the produce concentrations to wet weight. A moisture content of 94% will be used for above ground vegetables. This value represents the average moisture content of cucumbers, peppers, and tomatoes (EPA, 1997). 5-4

29 Equation 5-4 Above ground produce C = C ag soil x BTF ag Where: C ag = Concentration of PCBs in above ground produce due to root uptake (mg/kg dry weight). C soil = Concentration of PCBs in soil (mg/kg dry weight). BTF ag = Soil-to-plant biotransfer factor for above ground produce (unitless). Equation 5-5 Below ground produce Where: C = C bg soil x BTF C bg = Concentration of PCBs in below ground produce due to root uptake (mg/kg wet weight). C soil = Concentration of PCBs in soil (mg/kg dry weight). BTF bg = Soil-to-plant biotransfer factor for below ground produce (unitless). bg Prediction of Concentrations in Animal Feed PCBs in pasture grass, silage, and grain will be predicted to determine the potential intake of livestock. The BTF ag value derived using Equation 5-1 will be applied to the exposure point concentration observed in the EU of interest where the animals graze or where the feed is grown to derive the levels of PCBs in the feed of animals in the floodplain area (Equation 5-4). Because the animal feed consumption rates are on a dry weight basis, there is no need to convert the silage and pasture grass to wet weight. 5-5

30 5.1.4 Prediction of Concentrations in Animal Products Absent site-specific contaminant data in animal products, the potential transfer of contaminants from soil and food into animal tissue will be predicted using regression models. Equations developed by Travis and Arms (1988) have been commonly used to predict contaminant transfer from affected media and food into beef and milk. However, there is a significant amount of uncertainty surrounding the Travis and Arms approach based on the limited log Kow range upon which the regression equation is based and questions surrounding the validity of the underlying biotransfer data set (EPA, 2005). As a result, EPA developed a new methodology for predicting transfer into beef and milk (RTI, 2005). Basically, the updated methodology predicts transfer into animal fat (BTF fat ) where lipophilic compounds such as PCBs tend to sequester (see Equation 5-6), The BTF fat values are then adjusted to account for the assumed fat content in animal products. Equation log BTFfat = x (log Kow) x log Kow Beef As previously discussed, beef cattle are common in the OU-4 area. Therefore, uptake of PCBs into beef will be modeled if the soil sampling results in areas used for grazing indicate a need to evaluate this exposure. PCBs may accumulate in the tissue of beef cattle that graze in the floodplain as a result of ingesting pasture grass and soil or feed grown in the floodplain. The degree of livestock exposure to contaminated soil or food will be EU-specific and dependent on the amount of area in the EU that is located within the floodplain. The BTF fat value calculated in Equation 5-6 will be adjusted to account for the assumed fat content in beef on a wet weight basis as shown in the Equation 5-7. Equation 5-7 BTF beef = 10 log BTFfat x

31 The beef cattle ingestion rates of food items (forage, silage, and grain) and soil will be obtained from the HHRAP (EPA, 2005). The degree to which cattle ingest forage, silage, and grain will be considered on an EU-specific basis. Further, given the limited transfer of PCBs from soil to animal feed plants, it is likely that incidental soil ingestion by the cattle will be the primary contributor to the overall PCB intake. Equation 5-8 presents the general equation for calculating the concentration of PCBs in beef tissue on a wet weight basis. Where: ( ( FI x IR x C ) FI x IR x C x Bs) x BTF x MF C beef = i i i + C beef = Concentration of PCBs in beef (mg/kg wet weight). soil FI i = Fraction of plant type i (forage, silage, and grain) grown on contaminated soil and ingested by the animal (unitless). This factor will be specific to the EU of interest and will be dependent on the amount of pasture within the floodplain. IR i = Ingestion rate of plant type i eaten by the animal per day (kg dry weight plant/day). Forage 8.8; Silage 2.5; and Grain C i = Concentration of PCBs in plant type i eaten by the animal (mg/kg dry weight). FI soil = Fraction of ingested soil from the floodplain. This factor will be specific to the EU of interest and will be dependent on the amount of pasture within the floodplain, IR soil = Ingestion rate of soil eaten by the animal per day (0.5 kg/day dry weight). C soil = Concentrations of PCBs in soil (mg/kg dry weight). Bs = Soil bioavailability factor (unitless). A value of 1.0 will be used. BTF beef = Beef biotransfer factor (day/kg wet weight tissue) Dairy Products MF = Metabolism factor (unitless). A value of 1.0 will be used. soil soil beef Equation 5-8 As previously discussed, dairy cattle are not common in the OU-4 area. However, it is possible that dairy operations could occur in the future. Therefore, uptake of PCBs into milk will be estimated based on a range of floodplain soil concentrations developed from the results of the current sampling being performed by Solutia. 5-7

32 PCBs may accumulate in the milk of dairy cattle that graze in the floodplain as a result of ingesting pasture grass and soil or feed grown in the floodplain. The BTF fat value calculated in Equation 5-6 will be adjusted to account for the assumed fat content in milk on a wet weight basis as shown in the Equation 5-9. Equation 5-9 BTF milk = 10 log BTFfat x 0.04 The dairy cattle ingestion rates of food items (forage, silage, and grain) and soil will be obtained from the HHRAP (EPA, 2005). Given the limited transfer of PCBs from soil to animal feed plants, it is likely that incidental soil ingestion by the cattle will be the primary contributor to the overall PCB intake. Equation 5-10 presents the general equation for calculating the concentration of PCBs in dairy milk on a wet weight basis. Where: ( ( FI x IR x C ) FI x IR x C x Bs) x BTF x MF C milk = i i i + C milk = Concentration of PCBs in milk (mg/kg wet weight). soil FI i = Fraction of plant type i (forage, silage, and grain) grown on contaminated soil and ingested by the animal (unitless). This factor will be adjusted to reflect a number of different cattle ingestion scenarios. IR i = Ingestion rate of plant type i eaten by the animal per day (kg dry weight plant/day). Forage 13.2; Silage 4.1; and Grain 3.0. C i = Concentration of PCBs in plant type i eaten by the animal (mg/kg dry weight). FI soil = Fraction of ingested soil from the floodplain. This factor will be adjusted to reflect a number of different cattle ingestion scenarios. IR soil = Ingestion rate of soil eaten by the animal per day (0.4 kg/day dry weight). C soil = Concentrations of PCBs in soil (mg/kg dry weight). Bs = Soil bioavailability factor (unitless). A value of 1.0 will be used. BTF milk = Milk biotransfer factor (day/kg wet weight tissue). MF = Metabolism factor (unitless). A value of 1.0 will be used. soil soil milk Equation

33 Chickens and Eggs Chicken and egg production has not been observed in the floodplain in the OU-4 area. However, it is possible that these operations could be occurring or could occur in the future. Therefore, uptake of PCBs into chicken and eggs will be estimated based on a range of floodplain soil concentrations developed from the results of the current sampling being performed by Solutia. PCBs may accumulate in chicken and subsequently eggs as a result of incidentally ingesting floodplain soil or feed (grain) grown in the floodplain. The BTF fat value calculated in Equation 5-6 will be adjusted to account for the assumed fat content in chicken and eggs on a wet weight basis as shown in the Equation Equation 5-11 BTF chicken = 10 log BTFfat x 0.14 BTF eggs = 10 log BTFfat x 0.08 The chicken ingestion rates of grain and soil will be obtained from the HHRAP (EPA, 2005). Equation 5-12 presents the general equation for calculating the concentration of PCBs in chickens and eggs on a wet weight basis. Equation 5-12 C Where: chicken = C eggs = ( FIgrain x IR grainx Cgrain ) x FIsoil x IR soil x Csoil x BTFchicken x MF ( FI x IR x C ) x FI x IR x C x BTF x MF grain grain grain C chicken = Concentration of PCBs in chicken (mg/kg wet weight). C eggs = Concentration of PCBs in eggs (mg/kg wet weight). soil FI grain = Fraction of grain grown on contaminated soil and ingested by the animal (unitless). This factor will be adjusted to reflect a number of different chicken ingestion scenarios. IR grain = Ingestion rate of grain (0.2 kg dry weight plant/day). C grain = Concentration of PCBs in grain (mg/kg dry weight). soil soil eggs 5-9

34 FI soil = Fraction of ingested soil from the floodplain. This factor will be adjusted to reflect a number of different ingestion scenarios. IR soil = Ingestion rate of soil (0.022 kg/day dry weight). C soil = Concentrations of PCBs in soil (mg/kg dry weight). Bs = Soil bioavailability factor (unitless). A value of 1.0 will be used. BTF chicken = Chicken biotransfer factor (day/kg wet weight tissue). BTF eggs = Eggs biotransfer factor (day/kg wet weight tissue). MF = Metabolism factor (unitless). A value of 1.0 will be used. 5.2 EXPOSURE PARAMETERS This section presents the exposure parameters that will be used to quantify exposure in terms of contaminant intake (exposure dose). Tables 5-1 through 5-4 present the exposure parameters for each receptor by media. The formulas used in estimating exposure intakes are also shown on these tables. When applicable, exposure parameters used in the HHRAs for OU-1/2 and OU-3 will be used in OU-4. Site- and regional-specific information will be used when available as it relates to fish consumption and ingestion of home grown agricultural products. Absent specific information, professional judgment will be used. Exposure parameters will be separated into two categories. The first category is the constant exposure parameters that are similar for all exposure scenarios. These parameters are described in Section These parameters are not repeated in each scenario-specific discussion. The second category of exposure parameters is the variable exposure parameters. These parameters are usually different for each exposure scenario and are presented in the exposure scenariospecific discussions in Section through Constant Exposure Parameters The exposure parameters values that will be constant for all of the exposure scenarios are listed below: Body weight (BW). Averaging time (AT) cancer and noncancer. Dermal absorption factor (ABS). Intestinal absorption factor (IAF). 5-10

35 Body Weight The average BW values for the young child (1 through 6 years) and the adult will be 15 kg and 70 kg, respectively (EPA, 1989, 2008). For the adolescent (7 through 16 years), the BW will be 45 kg (EPA, 1997, 2000). These values will be used in the RME and CTE evaluations and are constant across all scenarios Averaging Time The cancer-based AT will be based on a 70-year lifetime for all age groups and equates to 25,550 days (70 years x 365 days/year) (EPA, 1989). The noncancer AT for each of the scenarios will be based on the receptor- and scenario-specific exposure duration (ED) in years multiplied by 365 days/year. The noncancer-based AT is constant across all of the scenarios in that it is always the ED multiplied by 365 days/year Dermal Absorption Factor The ABS term (unitless) represents the fraction of PCBs that are assumed to penetrate the skin following dermal contact with contaminated soil. Similar to the HHRAs performed for OU-1/2 and OU-3, an ABS value of 0.06 will be used for PCBs (Solutia, 2002). This value will be used in the RME and CTE evaluations for all direct contact scenarios Intestinal Absorption Factor The IAF term (unitless) represents the fraction of PCBs that are assumed to be absorbed through the gastrointestinal tract following the ingestion of the contaminated item (e.g., soil, agricultural products, fish). Similar to the HHRAs performed for OU-1/2 and OU-3, an IAF value 0.3 will be used for PCBs (Solutia, 2002). This value will be used in the RME and CTE evaluations for all of the scenarios involving the ingestion route of exposure Recreational User Exposure Parameters Recreational users are potentially exposed to PCBs in surface soil through incidental ingestion and dermal contact and absorption. The recreational receptors will include the young child, the adolescent, and the adult that use the Site for various recreational activities, including walking, hiking, picnicking, riding all-terrain vehicles, hunting, fishing, and related activities. The 5-11

36 exposure parameters for the recreational user scenario will be developed to cover the potential exposure associated with the most soil intensive recreational activity. The age groups of the recreational user receptors evaluated at an EU will be determined based on the EU s access characteristics. The young child receptor will be evaluated at EUs located close to residences. The adolescent and adult will be evaluated at every recreational EU. RME The incidental soil ingestion rates (IRS) for residential exposure in the list below will be used in the RME evaluation for the recreational users. Young child 200 mg/day (EPA, 1991, 1997). Adolescent 100 mg/day (EPA, 1991, 1997). Adult 100 mg/day (EPA, 1991, 1997). The following exposed skin surface area (SA) values will be used in the RME evaluation: Young child exposed skin surface includes head, hands, forearms, lower legs, and feet. This equates to a SA value of 2,800 cm 2 (EPA, 2004). Adolescent exposed skin surface includes head, hands, forearms, and lower legs. This equates to a SA value of 5,300 cm 2 (EPA, 2004). Adult exposed skin surface includes head, hands, and forearms. This equates to a SA value of 3,300 cm 2 (EPA, 2004). The following soil-to-skin adherence factor (AF) values will be used in the RME evaluation: Young child a value of 0.3 mg/cm 2 will be used, which is the 95 th percentile value for the daycare children activity (EPA, 2004). Adolescent a value of 0.4 mg/cm 2 will be used, which is the 95 th percentile value for children playing in dry soil activity (EPA, 2004). Adult a value of 0.1 mg/cm 2 will be used, which is the 95 th percentile value for the commercial/industrial groundskeeper activity (EPA, 2004). The following ED values will be used in the RME evaluation: Young child a value of 6 years will be used, based on the age range of 1 through 6 years. Adolescent a value of 10 years will be used, based on the age range of 7 through 16 years. Adult a value of 30 years will be used. This value is consistent with EPA s default residential ED (EPA, 1997). The duration of 30 years is supported by 2006 Census data for Calhoun and Talladega Counties related to the year an individual moved in their current 5-12

37 residence. The data indicate that approximately 10% of the respondents have been in their current dwelling since 1969 or earlier (U.S. Census Bureau, 2007a, 2007b). For soil ingestion, a fraction ingested (FI) value of 1.0 will be used. This assumes that the exposed individual receives 100% of their daily soil intake while engaging in recreational activities at the EU. The frequency of soil contact can vary at different EUs as a function of the accessibility of the EUs. At the majority of the EUs, the recreational users will be assumed to be exposed to soil 52 days/year which assumes exposure one day per week over the course of a year (52 weeks). This exposure frequency (EF) is half of the recreational user EF value used in the OU-1/2 HHRA (CDM, 2008b). Many of the floodplain areas are not readily accessible as a result of vegetation and property ownership issues. Thus, a reduced recreational user EF will be used. At recreational EUs located near residential properties where access is not restricted by vegetation (e.g., along maintained pathways), a greater EF value will be used (104 days/year). CTE The RME parameters for SA will also be used for the CTE analysis and will not be repeated in this subsection. The young child and adolescent RME ED values will be used for the CTE. The IRS values in the list below will be used in the CTE evaluation. Young child 100 mg/day (EPA, 1991, 1997). Adolescent 50 mg/day (EPA, 1991, 1997). Adult 50 mg/day (EPA, 1991, 1997). The following AF values will be used in the CTE evaluation: Young child a value of 0.04 mg/cm 2 will be used, which is the geometric mean value for the daycare children activity (EPA, 2004). Adolescent a value of 0.04 mg/cm 2 will be used, which is the geometric mean value for the children playing in dry soil activity (EPA, 2004). Adult a value of 0.02 mg/cm 2 will be used, which is the geometric mean value for the commercial/industrial groundskeeper activity (EPA, 2004). An ED value of 15 years will be used for the adult recreational user. This value is half of the RME value. A soil FI value of 0.5 will be used. This assumes that the exposed individual receives 50% of their daily soil intake from within the EU. At the majority of the recreational 5-13

38 EUs, the recreational users will be assumed to be exposed to soil 26 days/year which assumes exposure one day every two weeks over the course of a year (52 weeks). An EF value of 52 days/year (once a week) will be used at recreational EUs located near residential properties Utility Worker Exposure Parameters Utility workers may be exposed to PCBs in surface and subsurface soil along utility lines within OU-4 via the incidental soil ingestion and dermal contact and absorption routes of exposure. The utility worker will be an adult. The exposure will be based on intense soil contact activities that last for a short duration. RME The IRS will be 330 mg/day (EPA, 2002). The SA value will be 3,300 cm 2 (EPA, 2004) and assumes that the head, hands, and forearms are exposed. The AF value will be 0.3 mg/cm 2, which corresponds to the 95 th percentile value for the construction workers activity (EPA, 2004). The utility worker ED will be 1 year. The EF will be 10 days/year which assumes the utility worker maintains easements, and inspects, repairs and replaces equipment. The FI will be 1.0. CTE The RME parameters SA and ED will also be used for the CTE analysis and will not be repeated in this subsection. An IRS value of 100 mg/day will be used (EPA, 2003b). The AF value will be 0.1 mg/cm 2, which corresponds to the geometric mean value for the construction workers activity (EPA, 2004). The EF will be 5 days/years, which is half of the RME value. The FI will be Farmer Exposure Parameters The farmer exposure scenario will consist of an adult who might contact floodplain soil during typical farming activities such as planting and harvesting. It will be applied to locations that are currently used for agricultural purposes. Consumption of home grown vegetables and beef will be evaluated for the adult and young child at areas where these activities likely occur or are assumed to occur at some point in the future. 5-14

39 RME Higher soil ingestion rates than those used for residential exposure are recommended for contactintensive activities such as farming. EPA recommends a soil ingestion rate of 330 mg/day for construction work activities (EPA, 2002a). This value represents the 95 th percentile rate based on a study by Stanek at al. (1997). The 90 th percentile ingestion rate from the Stanek study was 200 mg/day. The IRS of 200 mg/day will be use in the RME for the adult farmer. This rate applies to the planting and harvesting activities in which heavy equipment can be used and fugitive dust generated. The RME EF for the adult farmer contact with floodplain soil will be 10 days/year. This value is based on a 200-day growing season and assumes that a farmer spends 5 days/year planting and 5 days/year harvesting in the floodplain. A SA value of 3,300 cm 2 will be used. An AF value of 0.4 mg/cm 2, which is the 95 th percentile value for the farmer activity, will be used (EPA, 2004). The farmer based ED value of 40 years will be used in the RME evaluation (EPA, 2005). A FI value of one will be used. For the potential food consumption pathways, an EF of 350 days/year will be used for the child and adult. The child will be 1 through 6 years old. Information presented in EPA s CSFII Analysis of Food Intake Distributions (EPA, 2003a) will be used to estimate the potential exposure resulting from the consumption of food products grown or raised in the floodplain. Per capita food intake estimates on an as consumed basis will be used. As consumed intake rates are based on the weight of the food in the form that it is consumed. As a result, preparation and cooking losses will not be applied to the intake rates. For the RME analysis, the average of the 95 th percentile intake values across the appropriate age categories will be used. The per capita intake rates will be multiplied by the fraction of the intake that is home produced to arrive at the estimate of the as consumed home grown intake rate that will be used in the HHRA. The fraction of intake that is home produced will be obtained from the Exposure Factors Handbook (EPA, 1997). Tables 5-5 through 5-7 present the derivation of the RME ingestion rates (IRs) for above ground vegetables, root (below ground) vegetables, beef, dairy products, chickens, and eggs. 5-15

40 The fraction of produce (above ground and below ground vegetable) that is ingested from the floodplain (the FI term) will be based on the fraction of the planted area within the floodplain. This will be determined on a case-by-case basis. In the event of future vegetable gardens, a range of FI terms will be used. CTE The RME parameters for SA and ED will also be used for the CTE analysis and will not be repeated in this subsection. The IRS will be 100 mg/day (EPA, 2003b). The CTE EF for the adult farmer contact with floodplain soil will be 5 days/year. An AF value of 0.1 mg/cm 2, which is the geometric mean value for the farmer activity, will be used (EPA, 2004). A soil FI value of 0.5 will be used. For the CTE analysis, the average of the median (50 th percentile) intake values will be used to estimate exposure from the ingestion of agricultural products. Tables 5-5 through 5-7 present the derivation of the CTE IRs. The fraction of produce that is ingested from the floodplain will be based on the fraction of area of planted area within the floodplain Recreational Fishermen Exposure Parameters The recreational fisherman scenario will consist of an adult or child who may be exposed through ingestion of fish from the Choccolocco Creek. A range of fish ingestion rates will be evaluated including rates estimated by Solutia s ongoing creel/angler survey as well as regionalspecific fish consumption rates applicable to rivers without fish consumption advisories, consistent with EPA policy. Given the likelihood that the current fish consumption advisory posted on this portion of the Creek would reduce the local population s frequency of fishing and the amount of fish consumed, it is anticipated that the creel/angler survey will identify a current fish consumption rate that is lower than it would likely be for similar rivers and streams without an advisory. At the time of the release of this document, Solutia s creel survey has not been fully reviewed by EPA. When completed, it is anticipated that the results of the creel survey will be used to 5-16

41 establish a current use fish ingestion rate. Therefore, it is assumed that the creel survey will be used as the basis for the CTE. The RME fish consumption rate will be based on the Alabama Department of Environmental Management s (ADEM) 1993 Estimation of Daily Per Capita Freshwater Fish Consumption of Alabama Anglers. The mean consumption rate calculated by the serving size method for all respondents was 30 g/day. This consumption rate was calculated based on data applicable to the interview site (i.e., not all lakes and rivers in Alabama). Note that the fish consumption rate suggested herein is equal to the 30 g/day that ADEM uses to establish water quality criteria for the protection of human health associated with the consumption of fish and shellfish. Also note that there were advisories on some Alabama waterbodies when the ADEM study was conducted (although it is not known if the advisories were emplaced on the waters on which the interviews were conducted) and the 30 g/day value may be biased low in this study. Child consumption rates for recreationally caught freshwater fish were not available from the ADEM (1993) study. The child consumption rates were assumed to be a fraction of the adult rate. Using fish consumption data available from EPA s Estimated per Capita Fish Consumption in the United States (2002b), ratios of child to adult consumption rates range from 0.48 to 0.49 depending on the consumption rate statistic (i.e., mean, median, 90 th percentile) considered. Based on these ratios, one-half of the adult consumption rate of 30 g/day, that is 15 g/day, was selected as a reasonable estimate of the consumption rate for the dependent child of a recreational angler. Fraction ingested can be particularly important when assessing fish consumption. Given that the fish consumption rates were based on site-only values instead of consumption from all Alabama waters, the starting point for an FI is assumed to be 1.0 for the recreational angler scenario. That is, it is assumed that the recreational angler catches and consumes all of their fish from one waterbody, which may not be a reasonable assumption for portions of the Choccolocco Creek. Based on professional judgment regarding the areas most likely to be fished, stream characteristics, species of fish in the Creek, and the average ingestion rate, the portion of 5-17

42 Choccolocco Creek downstream of Jackson Shoals, i.e., river mile 0 to 10, will be assigned an FI of 1. However, the portion of the creek between river mile 10 and 37 is unlikely to consistently provide catch amounts high enough to support a 30 g/day adult ingestion rate for the avid recreational angler. For one adult to ingest an annual average of 30 g skin-off fillet/day, approximately 50 lbs. of fish would need to be caught (assuming a conservative dress-out ratio of 0.5). The average number of days Alabama anglers fish rivers and streams is 21 (DOI, 2006; 90 percent confidence interval = 15 to 27 days); therefore, on average, approximately 2.2 lbs of fish would need to be caught at each outing to obtain the necessary mass. This is more fish than can reasonably be expected to be caught in this portion of the Creek; therefore, the FI proposed for river miles 10 to 37 will be Note that cooking loss will not be considered because the fish tissue concentrations are based on skin-off fillet samples. PCBs tend to sequester in the fat and skinning the fillets effectively removes the majority of the fat deposits, resulting in what are likely similar concentrations to cooked skin-on fillets. 5-18

43 6. REFERENCES ADEM (Alabama Department of Environmental Management) Estimation of Daily Per Capita Freshwater Fish Consumption of Alabama Anglers. Arcadis Anniston PCB Site, Phase 2 Field Sampling Plan for Operable Unit 4. Prepared for Pharmacia Corporation and Solutia, Inc. September Briggs, G.G., R.H. Bromilow, and A.A. Evans Relationships between Lipophilicity and Root Uptake and Translocation of Nonionized Chemicals by Barley. Pesticide Science. Volume 13. Pages Browning, B Personal communication with Bill Browning, Talladega County Extension Agent, regarding farming and livestock practices in Talladega County along the Choccolocco Creek. Butler, S Personal communication with Scott Butler, Calhoun County Farm Service, regarding farming and livestock practices in Calhoun County along the Choccolocco Creek. CDM. 2008a. Anniston PCB Site Operable Unit 3 Human Health Baseline Risk Assessment b. Anniston PCB Site Operable Units 1 and 2 Human Health Baseline Risk Assessment. EPA (U.S. Environmental Protection Agency) Risk Assessment Guidance for Superfund (RAGS), Volume I, Human Health Evaluation Manual (Part A) Interim Final. Office of Emergency and Remedial Response, Washington, DC. EPA/540/1-89/002. December RAGS, Volume I: Human Health Evaluation Manual, Supplemental Guidance, Standard Default Exposure Assumptions. Office of Emergency and Remedial Response, Toxics Integration Branch. Interim Final. Publication March Guidelines for Exposure Assessment. National Center for Environmental Assessment. EPA/600Z-92/001. May Review Draft Development of Human Health-Based and Ecologically-Based Exit Criteria for the Hazardous Waste Identification Project. Volumes I and II. Office of Solid Waste. March Exposure Factors Handbook, Volumes I, II, and III. Office of Research and Development, EPA/600/P-95/002F. Washington, DC Supplemental Guidance to RAGS: Region 4 Bulletins, Human Health Risk Assessment Bulletins. EPA Region 4. Originally published November 1995, Website version last updated May 2000: 6-1

44 RAGS, Volume I, Human Health Evaluation Manual (Part D, Standardizing, Planning, Reporting and Review of Superfund Risk Assessment). Final. EPA 540-R December a. Supplemental Guidance for Developing Soil Screening Levels for Superfund Sites. OSWER Office of Solid Waste and Emergency Response, Washington, DC. December b. Estimated per Capita Fish Consumption in the United States. August a. CSFII Analysis of Food Intake Distributions. National Center for Environmental Assessment. EPA/600/R-03/029. March b. Recommendations of the Technical Review Workgroup for Lead for an Approach to Assessing Risks Associated with Adult Exposures to Lead in Soil. EPA-540-R OSWER Dir # January RAGS, Volume I, Human Health Evaluation Manual (Part E, Supplemental Guidance for Dermal Risk Assessment). Final. EPA/540/R/99/005. NTIS No. PB Office of Emergency and Remedial Response, Washington, DC. December Human Health Risk Assessment Protocol for Hazardous Waste Combustion Facilities. Final. EPA-530-R Office of Solid Waste and Emergency Response. September Child-Specific Exposure Factors Handbook. National Center for Environmental Assessment, Office of Research and Development. EPA/600/R-06/096F. September RAGS, Volume I, Human Health Evaluation Manual (Part E, Supplemental Guidance for Inhalation Risk Assessment). Final. EPA-540-R January Golder Associates, Inc RCRA Facility Investigation/Confirmatory Sampling (RFI/CS) Work Plan. Jurriaans, W Personal communication with Wanda Jurriaans, Talladega Calhoun County Extension Agent, regarding farming and livestock practices in Talladega County along the Choccolocco Creek. RTI (Research Triangle Institute) Methodology for Predicting Cattle Biotransfer Factors. Prepared for U.S. Environmental Protection Agency (EPA) Office of Solid Waste. EPA Contract No. 68-W August. Solutia RCRA Facility Investigation/Confirmatory Sampling Report for the Anniston, Alabama Facility. October. 6-2

45 Stanek, Edward J. III, Edward J. Calabrese, Ramon Barnes, Penelope Pekow Soil Ingestion in Adults Results of a Second Pilot Study. Ecotoxicology and Environmental Safety, 36, Travis, C.C. and A.D. Arms Bioconcentration of organics in beef, milk, and vegetation. Environ. Sci. Technol. 22: U.S. Census Bureau. 2007a. Calhoun County, Alabama Selected Housing Characteristic Accessed April 1, Available at: qr_name=acs_2007_3yr_g00_dp3yr4&-ds_name=acs_2007_3yr_g00_&- tree_id=3307&-redolog=true&-_caller=geoselect&-geo_id=05000us01015&-format=&- _lang=en. 2007b. Talladega County, Alabama Selected Housing Characteristic Accessed April 1, Available at: context=adp&-qr_name=acs_2007_3yr_g00_dp3yr4&- ds_name=acs_2007_3yr_g00_&-tree_id=3307&-redolog=true&-_caller=geoselect&- geo_id=05000us01121&-format=&-_lang=en West, D Personal communication with David West, Calhoun County Extension Agent, regarding farming and livestock practices in Calhoun County along the Choccolocco Creek. 6-3

46 TN Coldwater Spring Branch UV 202 kj Anniston PCB Site 431 ANNISTON MS 20 Legend Anniston, AL FL GA Interstate Hwy. State Hwy Local Roads Railroad Snow Creek River/Stream Lake/Pond 78 County Boundary C a l h o u n C o u n t y Oxford Lake Robertsons Lake UV21 Wolfskull Creek Coldwater Spring Branch µ Scale in Miles 20 T a l l a d e g a C o u n t y Dry Branch Choccolocco Creek Anniston PCB Site - Operable Unit 4 Pathways Analysis Report FIGURE 2-1 ANNISTON PCB SITE LOCATION MAP

47 411 Coosa River UV21 UV9 UV UV Lake Logan Martin S t. C l a i r µ Scale in Miles UV 34 UV 77 Acker C reek UV77 Blue Eye Creek UV77 Eastaboga Creek C a l h o u n UV UV202 Cold Water Creek Choccolocco Creek T a l l a d e g a OU-4 Choccolocco Creek OU-3 Facility kj Snow Creek Dry Branch UV281 UV 49 ANNISTON Hillabee Creek UV 49 UV281 C l a y Shoal Creek OU-1 and OU-2 Anniston Residential and Non-Residential 431 C l e b u r n e Anniston PCB Site - Operable Unit 4 Pathways Analysis Report UV 9 FIGURE 2-2 OPERABLE UNIT LOCATIONS

48 Legend OU-4 Characterization Areas C1 C2N, C2S C3NF, C3N, C3S, C3SF C4NF, C4N, C4S, C4SF C5NF, C5N, C5S, C5SF C6N, C6S C7NF, C7N, C7S, C7SF C8N, C8S C9N, C9S Blue Eye Creek C a l h o u n UV 202 Anniston PCB Site kj 431 S now Creek C Eastaboga Creek C7NF C9N C7N C8N C9S C7S C7SF C8S UV 77 C5N C5S C6N C6S Choccolocco Creek T a l l a d e g a UV 21 Coldwater Spring Branch C5NF C5SF Salt Creek C4S C4SF C4N C4NF Dry Branch C2N C2S C3NF C3N C3SF C3S Cheaha Creek µ Scale in Miles Kelly Creek Fayne Creek Anniston PCB Site - Operable Unit 4 Pathways Analysis Report FIGURE 3-1 OPERABLE UNIT4 LOCATION UV 49 UV 281

49 Hale St Legend Interstate Hwy. Highway Major Road Local Roads Railroad Streams County Boundary CA Boundary Snow Creek Hamric Dr Patrick St Roxie Ave Bailey St Corbin St Lakeside Dr Bruce St Lester Ave Central Ave Mccullars Ln Bruce St Recreational Dr «21 Recreational Dr Characterization Area C1 µ Scale in Feet Note: Aerial photography Anniston PCB Site - Operable Unit 4 Pathways Analysis Report FIGURE 3-2 CHARACTERIZATION AREA C1

50 Friendship Rd «21 Grace St 20 C2N Snow Creek C1 C2N Choccolocco Creek Legend Interstate Hwy. Highway Major Road Local Roads Railroad County Boundary CA Boundary Water Body Elm St Characterization Area C2 Plaza Ln C3N Choccolocco Creek Friendship Rd C2N Wolfskull Creek C2S C2N C2S Glover Dr Owens Dr µ C3S Scale in Feet Note: Aerial photography dated Lakeview Dr C3SF Cheaha Dr C a l h o u n T a l l a d e g a Friendship Rd Douglas Dr Clark Ave Clark Ave Douglas Dr Anniston PCB Site - Operable Unit 4 Pathways Analysis Report FIGURE 3-3 CHARACTERIZATION AREA C2 Hinton Dr

51 Glen Davis Rd Legend Tanglewood Dr 78 Dogwood Dr Woodland Dr Interstate Hwy. Highway Major Road Local Roads Railroad Catalpa Ln Laurel Ln Streams Water Body Rose Ln Sherwood Dr County Boundary CA Boundary Pinecrest Ln Meadowood Ln 20 Merimac Dr Airport Rd Old Talladega Hwy Smith St Barry St Miller St Meadow Ave «21 Barry St Rose Ave Howle Rd Maple St Hickory St Lake St Meadowlake Dr C a l h o u n T a l l a d e g a C3S Friendship Rd Lakeview Dr Edgewood Dr C3SF C2N Brownlee Rd Oakdale St Friendship Rd µ ,200 Scale in Feet Cliff Garrett Dr C3N Characterization Area C3 C3NF Pettus Ave Boozer Dr Foster Dr Park Ave Clubhouse Dr Cheaha Dr Note: Aerial photography dated C3S Walker St C3NF C3N C3S «21 Canadian Pass Marsh Ln Canvasback Dr Sleepy Hollow Rd County Line Rd Lataste Dr Gunnels Ln C4N C4S C4SF C3SF Cooper Ln Marsh Ln Pintail Pt Michael Ln Anniston PCB Site - Operable Unit 4 Patricia Ln Vickie Ln Jay St Fitzgerald Rd Pathways Analysis Beck Rd Report Douglas St FIGURE 3-4 CHARACTERIZATION AREA C3 Diana St

52 Robertson Ln Gann Rd Ada Dr Woodridge Trce Legend Interstate Hwy. Highway Major Road Local Roads Railroad Streams Oakland Dr Water Body County Boundary Airport Rd CA Boundary Gatlin Rd Lakewood Dr Dessie Cir Carrie Lee Cir Ina Ln Airport Rd Gidley Rd 78 Gwendale Ln C a l h o u n Watson Dr Rocky Mt. Rd T a l l a d e g a Access Rd Siver Run Rd Brentwood Dr Del Ray Cr Del Ray Dr Nancy Dr 20 C4NF C4N Silver Run Rd C4NF C3NF C3N C3S C4S Choccolocco Creek Ritch Dr Betta View Dr April Ln Alan Dr Hickory Dr Watson Dr Jason Dr Lynn Dr Dorris Dr Reaves Dr Sweetheart Ln Caffey Dr Jackson Ave Creek Dr Earley St Lee Ave Airport Rd Flat Bridge Rd H J Bentley Jr Pky C4NF C4N C3SF Bobwhite Dr C4SF C4S C5N C5S C4S Characterization Area C4 C4N C4SF C5SF Silver Run Rd C4SF Kirby Rd µ Anniston ,600 2,400 Scale in Feet Jennifer Rd Patton Rd Mae Ln Martin Ln Cobb Rd Jenkins Cir Cobb Rd Huntcliff Dr Denney Rd Dry Branch PCB Site - Operable Unit 4 Pathways Analysis Report FIGURE 3-5 CHARACTERIZATION AREA C4 «21

53 Legend Interstate Hwy. Highway Major Road Local Roads Railroad Streams Water Body County Boundary CA Boundary Sears Ln Country Ln Mahaffey Rd 20 County Line Rd T a l l a d e g a Racers Rd Yellow Jacket Ln Bobwhite Dr C4NF C4S C4N Cemetary Ln Freemans Ln C5NF C5NF C5SF Parker Ln Family Ln C5NF Richeytown Rd C5N C5SF Characterization Area C5 Ford Ln Priebes Mill Rd C5NF C5N C5SF C5SF C5S C6N C6S Note: Aerial photography dated Judy Ln µ Anniston ,600 2,400 Scale in Feet Matson Rd Jennifer Rd Jennifer Rd PCB Site - Operable Unit 4 Pathways Analysis Report FIGURE 3-6 CHARACTERIZATION AREA C5

54 Legend C5NF Interstate Hwy. Highway C7NF Richeytown Rd Major Road Local Roads Railroad Streams Water Body County Boundary CA Boundary Cunningham Ln Choccolocco Creek C5N C5S Matson Rd C5SF C6N Judy Ln Creek Rd C7N Champion Rd C6S C7S Characterization Area C6 Rainwater Ln Ledbetter Rd Howard Dr T a l l a d e g a C6S Rose Bud Ln Cole Rd Curry Station Rd C6N C7SF µ Anniston ,200 1,800 Scale in Feet Note: Aerial photography dated Ledbetter Rd Henry Ln PCB Site - Operable Unit 4 Pathways Analysis Report Wildwood St FIGURE 3-7 CHARACTERIZATION AREA C6

55 Legend Interstate Hwy. Highway Major Road Local Roads Railroad Streams Water Body County Boundary CA Boundary Eastaboga Creek Settlement Ln Wallace Rd Eastaboga Rd Brickstore Rd Keswick Dr C7NF C7NF Richeytown Rd T a l l a d e g a C7N Characterization Area C7 Curry Station Rd C7N Choccolocco Creek C7S Creek Rd Champion Rd Chris Leigh Ln Stanley Ln C7NF C7SF Failth Manor C7SF C6S C8N C8S Cheaha Creek Eastaboga Rd Abbie Ln µ Anniston Note: Aerial photography dated Turners Mill Rd ,600 2,400 Scale in Feet Curry Station Rd C6N PCB Site - Operable Unit 4 Pathways Analysis Report FIGURE 3-8 CHARACTERIZATION AREA C7

56 Legend Interstate Hwy. Highway Major Road Local Roads Railroad Streams Water Body County Boundary CA Boundary Chris Leigh Ln C9N C9S C9N C8S T a l l a d e g a C8S C8N Stanley Ln C7NF C7NF Characterization Area C8 C7N C7S C7SF µ Anniston ,000 1,500 Scale in Feet Note: Aerial photography dated Curry Ln Big Oak Rd PCB Site - Operable Unit 4 Pathways Analysis Report FIGURE 3-9 CHARACTERIZATION AREA C8

57 C9S C9N Tipton Rd Lower Mitchell Rd C9N Jackson Trace Rd Jackson Shoals Cir «77 µ Anniston ,000 1,500 Scale in Feet Note: Aerial photography dated T a l l a d e g a Characterization Area C9 C9S Choccolocco Creek PCB Site - Operable Unit 4 Pathways Analysis Report FIGURE 3-10 CHARACTERIZATION AREA C9 Legend Interstate Hwy. Highway Major Road Local Roads Railroad Streams Water Body County Boundary CA Boundary C8S C8N