report: focused feasibility study, baseline risk

Report: Focused Feasibility Study, Baseline Risk Assessment, GCL Tie & Treating Site, Sidney, New YorkEBASCO SERVICES INCORPORATED
GCL TIE & TREATING SITE SIDNEY, NEW YORK
APRIL 1994
NOTICE
THE INFORMATION PROVIDED IN THIS DOCUMENT HAS BEEN FUNDED BY THE UNITED STATES ENVIRONMENTAL PROTECTION AGENCY (USEPA) UNDER ARCS II CONTRACT NO. 68-W8-0110 TO EBASCO SERVICES INCORPORATED (EBASCO). THIS DOCUMENT HAS BEEN FORMALLY RELEASED BY EBASCO TO USEPA. THIS DOCUMENT DOES NOT REPRESENT, HOWEVER, THE USEPA POSITION OR POLICY, AND HAS NOT BEEN FORMALLY RELEASED BY THE USEPA.
D0065.LYN 400318
EBASCO April 15, 1994 ARCS 11-94-66-090
Mr. Carlos Ramos Work Assignment Manager US Environmental Protection Agency 26 Federal Plaza New York, NY 10278
SUBJECT: ARCS H PROGRAM - EPA CONTRACT NO. 68-W8-0110 WORK ASSIGNMENT NO. 066-2L6X - GCL TIE & TREATING SUBMITTAL OF FINAL FFS RISK ASSESSMENT
Dear Mr. Ramos:
Ebasco is pleased to submit 10 copies of the Final Focused Feasibility Risk Assessment under the subject work assignment. All EPA and New York State comments have been addressed and incorporated into the final documents. A detailed accounting of comments and responses is being provided under a separate cover letter.
Please do not hesitate to contact me at (201) 460-6434 or the Site Manager, Mr. Howard Lazarus at (201) 460-6062 if you have any questions or need additional assistance concerning this matter.
Very truly yours,
Dev R. Sachdev, Ph.D., P.E. ARCS n Program Manager
cc: M S Alvi K Moncino D Butler H Lazarus MKuo L Voyce Project File (1.9.3)
(EPA) (w/o attachment) (EPA) (w/o attachment) (EPA) (w/o attachment)
400319
E B A S C O E N V I R O N M E N T A L A Division ofEbasn services incorporated
160 CHUBB AVENUE • LYNDHURST, N.J. 07071-3586 • (201) 460-6500
USEPA WORK ASSIGNMENT NUMBER: 066-2L6X USEPA CONTRACT NUMBER: 68-W8-0110
EBASCO SERVICES INCORPORATED
GCL TIE & TREATING SITE SIDNEY, NEW YORK
APRIL 1994
Prepare^ By: ^ Approved By:
loward Lazar\s^_yy ^ Dev Sachdev, P.E., Ph.D. Site Manager ARCS n Program Manager Ebasco Services Incorporated Ebasco Services Incorporated
Reviewed By:
Ming Kuo, Ph.D., P.E. Technical Support Manager Ebasco Services Incorporated 400320
D0065iYN
RISK ASSESSMENT SUMMARY
The GCL Tie & Treating site is an inactive wood processing and treating facility located in Sidney, Delaware County, New York. Timber was cut and treated with creosote on site in the fabrication of wood products, predominantly railroad ties. Contaminants have been released to the environment through direct contact with the surface soil as a result of open drip-drying of treated products, disposal of end cuts and scrap as fill material, and at least one documented spill.
This Focused Feasibility Study Baseline Risk Assessment addresses the potential human health impacts associated with contaminated soils on the western portion of the GCL site. At the direction of EPA, the scope is limited to exposure to contaminated soils via dermal contact, ingestion, and inhalation of particulates. Potential receptors are young child and adult off-site residents, older child and adult trespassers, off-site workers, and future on-site workers. Site land use is expected to remain industrial/commercial. Data used to evaluate the risk posed by the site were obtained from sampling performed by EPA as part of a 1990 USEPA Removal Action and a Subsurface Sampling Investigation conducted in 1993 by Ebasco at the direction of EPA.
Results of this Risk Assessment indicate potential carcinogenic risks above EPA target levels to all receptors included in the analysis, through incidental ingestion of contaminated soil, should no cleanup be performed. Inhalation risks are within or below, and dermal contact risks are below target levels. No noncarcinogenic effects are predicted for the site, as all receptor Hazard Index values are below 1.0.
Chemicals contributing to excess risks include carcinogenic polycyclic aromatic hydrocarbons
(PAHs), the main constituents of creosote, as well as arsenic and chromium.
This Risk Assessment was performed in accordance with "Risk Assessment Guidance for
Superfund" (USEPA, 1989b), Guidance for Conducting Remedial Investigation and Feasibility
Studies under CERCLA (USEPA, 1988a) and Superfund Accelerated Cleanup Model (SACM)
Guidance (USEPA, 1992d).
400321 D0065.LYN
i r- m O -n O O z H m z H w
400322
3.0 SELECTION OF CHEMICALS OF CONCERN 9
3.1 Database Selection 9
3.2 Selection Criteria 10
3.4 Chemicals of Concern 10
4.0 IDENTinCATION OF EXPOSURE PATHWAYS 13 4.1 Contaminant Source Analysis 14 4.2 Receptor Analysis 14
4.3 Exposure Pathway Analysis 15
5.0 HAZARD IDENTIFICATION 19
5.2 Health Effects Criteria for Potential Carcinogens 20
5.3 Toxicological Assessment 22
6.1 Health-Based Applicable or Relevant and
Appropriate Requirements (ARARs) and To Be Considered (TBC) Criteria 22
6.2 Estimation of Exposure Point Concentrations 26
D0065I.YN . i - f, . ? 11 400323
TABLE OF CONTENTS (Cont'd)
Section Title Page No.
7.0 QUANTITATIVE RISK CHARACTERIZATION 43
7.1 Quantitative Risk Assessment Methods 43
8.0 RESULTS OF RISK CALCULATIONS 45
8.1 Reasonable Maximum Case 45
8.2 Average Case 46
9.0 COMBINING RISK LEVELS AND HAZARD INDEX VALUES 46 ACROSS PATHWAYS
10.0 QUALITATIVE DISCUSSION OF RISKS NOT QUANTITATIVELY 46
EVALUATED IN THE RISK ASSESSMENT
11.0 POTENTIAL PUBLIC WELFARE IMPACTS 50
12.0 UNCERTAINTY IN THE RISK ASSESSMENT 50
12.1 Uncertainties Associated with Sampling and
Analytical Procedures 50
Intake Assessment Methods 54
12.3 Uncertainties Associated with Toxicologic
Models and Parameter Estimates 55
13.0 RISK ASSESSMENT SUMMARY 56 13.1 Human Health Risk Assessment Summary 56
REFERENCES R-1
Chemical Data Summary Toxicological Profiles Human Health Risk Assessment Spreadsheets
.ih^^m^jt iii 400324
LIST OF TABLES AND FIGURES
1
2
3
5-16
17-19
20
21
TABLES
Chemicals of Potential Concem - ARARs/TBCs and Levels
of Detection in Soil
Risk Levels and Hazard Index Values Summary Across
Exposure Pathways - Present/Future Use Scenarios
Sources of Uncertainty in the Risk Assessment
Summary of Potential Risks
2
3
16
17
1.0 INTRODUCTION
1.1 Scope
This Focused Feasibility Study Baseline Risk Assessment (FFSBRA) addresses the potential human health impacts associated with contaminated soils on the western portion of the GCL Tie & Treating site, located in Delaware County, New York. The GCL property encompasses an area of approximately 26 acres. This FFSBRA is designed to focus on the potential health impacts of surface and subsurface soil contamination only. The procedures used in this risk assessment are consistent with USEPA guidelines for risk assessments in general, and Superfund sites in particular, including the "Risk Assessment Guidance for Superfund" (RAGS, 1989) and Guidance for Conducting Remedial Investigation and Feasibility Studies under CERCLA (USEPA, 1988a).
1.2 Site Location
The GCL Tie & Treating site (GCL) is a 60 acre property located at 42n7'N and 75^25^ in a
commercial/industrial section of the Village of Sidney, Delaware County, New York. The site
is bordered to the north by a rail line owned by the Delaware and Hudson Railroad. A
warehouse and the Sidney Municipal Airport are located to the north of the rail line. Route 8
forms the eastern border of the property, with the principal residential and commercial sections
of the Village of Sidney to the east. Delaware Avenue (also indicated as Gifford Road on the
Site Location Map, see Figure 1) runs along the southern border of the site in a northeast to
southwest direction. A drainage ditch and woodlands area lie on the northern side of the road,
while a large shopping center, containing a supermarket, retail stores, motel, restaurants, and
other small businesses, is located on the southern side of Delaware Avenue. The western edge
of the property abuts an impoundment and wetlands area. Figure 1 provides a Site Location
Map, and Figure 2 presents a Site Plan.
1.3 Site History
The GCL Tie & Treating property was owned by the Delaware and Hudson Railroad Company between 1940 and 1979. Railcon Wood Products/Railcon Materials, Inc. acquired the site in 1979 and sold it to GCL Tie & Treating in 1983. GCL filed for bankruptcy in 1987 and abandoned the property in January 1988. Railcon regained control of the property, sold aU inventory and equipment, and then abandoned the property. Present ownership of the site is unclear and is currently being investigated by USEPA.
400326 D0065.LYN 1
N
t
juntoxttuTt sau.e ojp 0i/za/9i
EBASCO SERVICES INCORPOP T D
s
MILl]
UOUNO WITH A S S O R U B DEBRIS
SAW
VILLAGE OF SIDNEY. NY
DJP OI /Ba/9 i SHOPPING PLAZA ^
CREOSOTE STORAGE TANKS
U.S. ENVIRONMENTAL PROTECTION AGENCY
GCL TIE & TREATING SITE
««ieooi
Wood preserving operations began on-site as early as the 1940s. The property was used as a
railroad tie manufacturing and treating plant since at least the early 1960s. Logs were brought
on site, cut, and treated with creosote. The practice of drip drying creosote soaked lumber, with
no containment safeguards resulted in numerous areas of stained soil. A sequence of aerial
photographs taken between 1963 and 1983 indicate a gradual expansion of site operations from
the eastern section of the property to the west. The greatest growth occurred between 1977 and
1983, when clearing of vegetation and filling of wetland areas occurred.
In October 1986, the site came to the attention of the New York State Department of
Envkonmental Conservation (NYSDEC) after a spill of 9,000 to 10,000 gallons of creosote was
reported. The spill occurred at night and attempts to remediate the area were made by GCL.
Sidney police noticed the spill the next morning and notified NYSDEC, which instructed GCL
to excavate the contaminated soil, place it on a polyurethane liner, and cover it pending final
disposal. Contaminated soil was to be placed in 55-gallon containers and removed from the site.
As of November 1992, uncovered, unlined contaminated soil remained on site.
Contamination at this site is not limited to the creosote spill. Stained soils are present throughout
the site and are assumed to be a result of creosote dripping from the tie-drying process. Former
GCL employees have stated that creosote contaminated material was routinely disposed of in the
wetland area adjacent to the site. In addition, contaminated material was added to the stockpiled
soil and unspecified amounts of material were transferred from the waste pile and placed in the
wetlands. A large pile of sawdust and wood debris abuts the wetland area.
In December 1989, USEPA Region II conducted a sampling program in conjunction with their criminal investigation. Samples were collected from soils, the waste pile, and materials in the above ground storage tanks. In addition, a sample of creosote was obtained from Allied Chemical, GCL's supplier of this material, and used as a fingerprint to match the chemical analyses of on-site samples. Results of these analyses confirmed the presence of creosote in all samples.
In July 1990, NYSDEC boarded up the buildings and erected a fence along the eastern edge of the GCL property. These actions were taken to limit access and potential exposure to contaminated materials. Before this action, access to the site and all buildings was unrestricted. Public access to the site is still easily obtained as there is no fence around the site perimeter and no gate across the access road.
In addition to the creosote stained soils, seven 55-gallon drums of muriatic (hydrochloric) acid were discovered in the abandoned buildings. This material had a pH of 1. At one time there were at least two underground storage tanks present on-site. One 2,000-gallon storage tank
D0065.LYN 400329
contained fuel oil for the boiler and one 1,000-gallon tank held fuel oil for the furnace. Five
above ground tanks, presumed to have contained creosote, were also on site.
In September 1990, NYSDEC sent USEPA a written request to conduct a removal action at the
site pursuant to Section 104 of CERCLA. As a result of this request, USEPA initiated a removal
action in March 1991 to undertake mitigation including fence installation, site security, and
stabilization and disposal of hazardous substances. Under the removal action, 14,159 gallons of
creosote were removed from the tanks and associated piping. An additional 500 gallons of
creosote were removed from floors, tanks, and sumps. Approximately 4,000 cubic yards of soil
were screened and staged pending final disposal.
1.4 Current Conditions
The site is divided into sections, defined by USEPA as the GCL property and non-GCL property.
The GCL property section has been partially fenced and is comprised of the 26 acre inactive
sawmill and wood pressure treating facility known as GCL Tie & Treating. Present ownership
of this section of the site is under investigation. The current conditions on the GCL property are
discussed in detail below. The non-GCL section of the site has unrestricted access and has two
active businesses present. A small sawmill operation is located adjacent to the restricted GCL
property. The company has stockpiles of cut and uncut wood as well as shavings in open
storage. The sawmill appears to lease the property from the unidentified owners of the GCL
property. On the other side of the sawmill, a different company runs a wood laminating
operation in the buildings located on the eastern end of the property. All manufacturing and
storage is located indoors, and both interior and exterior areas are apparently well maintained.
The GCL property section of the site underwent a removal action by USEPA. The activities
included removal and cleaning of storage tanks and associated piping, stockpiling of contaminated
soils, removal of drums, screening and relocation of the debris pile, and performance of a pilot
scale composting operation. Entry to this section of the site from the access road has been
limited by the construction of a chain link fence along the eastern edge of the GCL property.
Soil sampling was performed along the fence line during construction. Additional soil sampling
was performed in known hot spots, and water samples were taken from the impoundment on the
western edge of the site.
A site visit was conducted by USEPA, NYSDEC, and Ebasco Environmental personnel on November 5, 1992. A tour of the site was provided by the USEPA On Scene Coordinator (OSC). During the site walk, it was noted that the site exhibited a strong creosote aroma and poor Surface drainage. The abandoned saw mill and pressure-treating buildings are in poor condition. The warehouse building that was providing shelter to the composting study and an open shed
D0065.LYN t ^ U m 5 400330
providing cover for the EPA decontamination area had been better maintained. The piles of
contaminated soil were partially covered. The western edge of the property, adjacent to the
impoundment, had been filled using an assortment of end-cuts from wood products and other
assorted materials. A second site visit was conducted by Ebasco Environmental personnel on
July 1-2, 1993. This visit was specifically made to identify potential exposure pathways,
sensitive populations, access points, as well as patterns and likely routes of exposure to
contaminants on-site and potentially migrating off-site, given current conditions at the site.
2.0 RISK ASSESSMENT METHODOLOGY
Methods used in this assessment are in accordance with the RAGS document As outlined below,
this document presents a six-step methodology for conducting a risk assessment:
1. Identify chemicals of potential concern (Section 3).
2. Define human exposure pathways (Section 4).
3. Assess contaminant toxicity (Section 5). 4. Estimate exposure point concentrations (Section 6). 5. Assess human contaminant intakes (Section 7).
6. Characterize the human health risks (Section 8).
A brief summary of this methodology is presented below.
1. Identify Chemicals of Potential Concem - Identification and selection of site-specific "chemicals of potential concem" (site-related contaminants whose data are of sufficient quality to be used in a quantitative risk assessment) for soils at the GCL Tie & Treating property were based on:
frequencies of occurrence
the historic data base of contamination at the site, and
the toxicological, physical, and chemical characteristics of the chemicals detected
The USEPA Region II and RAGS selection criteria for an organic compound to be retained as a chemical of potential concem for quantitative evaluation includes meeting one or more of the following: the compound occurred in at least 5 percent of the samples in a given medium in a single data set; the chemical is either a known human carcinogen (Group A) detected at any level or the chemical is a probable human carcinogen carcinogen (Group B1 and B2) detected above 1 ug/kg in the soil; the compound was detected above the applicable analytical detection limit;
i ^BQOk 6 400391
and toxicological data were available for the compound. Unless specifically stated, these criteria apply to both carcinogenic and noncarcinogenic compounds.
As with organics, an inorganic compound was selected if the toxicological data were available for the compound; the compound was detected above the applicable analytical detection limit; the compound occurred in at least 5 percent of the samples in a given medium in a single data set; the compound is either a known human carcinogen (Group A) or the compound is a probable human carcinogen detected above 1 ug/kg in soil. Those metals which were present at levels not significantly higher than local background, which are naturally occurring trace nutrients, or which were present at levels far below levels of potential concern from a health standpoint were eliminated. Based on these criteria, the inorganics arsenic and chromium, both Class A carcinogens, were chosen as chemicals of concern.
A toxicity/concentration screen was not performed for this Risk Assessment because the contaminants detected in soil which have toxicity criteria were limited. All chemicals detected that have current toxicological data and met the other chemical of potential concem criteria were included in the risk analysis (see Section 3).
2. Define Human Exposure Pathways - Prior to selecting the chemicals of concern, all potential
human exposure pathways for the site (i.e., inhalation, ingestion and dermal contact) were
defined. Each potential pathway was then evaluated considering site-specific conditions to
determine if the pathway could be present at the site. The area demography and land use
characteristics were taken into consideration when the pathways were developed. If a pathway
between the source of contamination and a human receptor could potentially be complete, and
was included in the scope of work for this risk assessment, it was retained for further quantitative
or qualitative evaluation (see Section 4).
3. Assess Contaminant Toxicity - All contaminants detected in site matrices were reviewed for their toxicity to humans. Data on contaminant toxicity was obtained from EPA's Integrated Risk Information System (IRIS), from the Health Effects Assessment Sununary Tables (HEAST) and from the scientific literature. On the basis of these data, contaminants were separated into two groups, those exhibiting carcinogenic effects (carcinogens) and those exhibiting noncarcinogenic effects (noncarcinogens), based largely on EPA classifications (see Section 5). Although all contaminants were reviewed for toxicological effects, only those chemicals with EPA promulgated toxicity criteria were considered for quantitative evaluation. Those not having toxicity criteria were evaluated qualitatively.
4. Estimate Exposure Point Concentrations - Estimation of exposure point chemical
concentrations for the reasonable maximum case were based on the upper 95% confidence limit
D0065.LYN t u r n 7 4003. 2
(95% UCL) of the arithmetic mean of the log-transformed data, or the maximum detected
concentration if it was less than the 95% UCL, in accordance with EPA (USEPA, 1992)
guidance. For the average case scenarios, the 95% UCL (or the maximum concentration, if it
was less than the 95% UCL) was also used as the exposure point concentration, in combination
with exposure parameters reflecting average case scenarios. Particulate emissions used to
estimate soil dust contaminant concentrations were calculated using the Cowherd Model (USEPA,
1985b).
5. Assess Human Contaminant Intakes - A quantitative assessment of human contaminant
intakes associated with each potential exposure pathway was then developed. Human exposure
levels and chronic and subchronic contaminant intakes were estimated for each contaminant and
matrix through the use of exposure scenarios. Exposure scenarios are reasonable sets of human
exposure pathways that help to define the intake levels of contaminants in site media. The
"reasonable maximum exposure" (RME) scenario employed the exposure point concentration and
reasonable maximum exposure circumstances. Average case exposure scenarios for those RME
scenarios exceeding a 1 x 10" carcinogenic risk, and/or a Hazard Index of 1.0 were also
developed in this Risk Assessment Adult and young child off-site residents, adult and older
child site trespassers, and adult on and off-site worker potential exposure parameters were
included in this evaluation.
6. Characterize Human Health Risks - The final step in this risk assessment was the health risk characterization. For noncarcinogens, exposure pathways were evaluated by comparing site-specific Chronic Daily Intake (CDI) rates to acceptable Reference Doses (RfDs) for each soil exposure pathway and each chemical of potential concem. The RfD values used were obtained from IRIS (February 1994 sessions) and HEAST (1993).
Potential noncarcinogenic effects are evaluated as the ratio of the Chronic Daily Intake (CDI) to the Reference Dose (RfD). The sum of all of the CDLRfD ratios for the selected chemicals of concem is called the Hazard Index (HI) and is calculated as shown below:
n
(1) HI = E ^ ^ i=I RfDi
Where: HI = Hazard index CDIj = Chronic daily intake for chemical i (mg/kg/day), RfDj = Reference dose for chemical i (mg/kg/day), and n = Number of indicator compounds (i.e. compounds of concem) in the medium
under consideration.
D0065. ^.gGOl 8 400333
A hazard index less than 1.0 is unlikely to be associated with health risks and is therefore less
likely to be of concem than a hazard index greater than 1.0. However, a conclusion should not
be categorically drawn that all hazard indices less than one are "acceptable" and all His greater
than 1.0 indicate that health risks will occur. This is a consequence of the uncertainties inherent
in the derivation of the RfD in the exposure assessment and the uncertainties associated with
adding the individual terms in the Hazard Index calculation.
Potential excess lifetime cancer risk due to exposure to a specific carcinogenic compound is
calculated by multiplying the compound specific CDI by its slope factor (SF) as follows:
(2) Excess lifetime cancer risk = CDI x SF
Where:
CDI = Chronic daily intake of the chemical (mg/kg/day), and SF = Slope factor for the chemical (mg/kg/day)"'
This linear equation is valid for excess lifetime cancer risks less than 10" (one in one hundred).
Above this level, individual excess lifetime cancer risks should be calculated using the equation:
(3) Excess lifetime cancer risk = 1 - exp(-CDI x SF)
Slope Factors are defined by USEPA's Carcinogen Risk Assessment Verification Endeavor
(CRAVE) and obtained from IRIS and HEAST. For the purposes of this assessment, cancer risks
for exposure to multiple carcinogenic contaminants were assumed to be additive.
A discussion of the soil pathways identified for evaluation as a potential health risk and the specific chemical constituents of concem within each soil exposure pathway is provided in Section 7. Potential sources and areas of uncertainty in the risk assessment are discussed in Section 12.
3.0 SELECTION OF CHEMICALS OF CONCERN
3.1 Database Selection
The soil sample results of the EPA Removal Action and Focused Feasibility Study were used to
create the data base for this Risk Assessment The analytical results of the samples taken from
soil were used to evaluate the potential risks from exposure to chemicals detected at the site.
D0065.LYN i ^ ( m ' ^''0334
3.2 Selection Criteria
The primary considerations for selection of chemicals of concern were: 1) frequency of
detection; 2) detections of probable carcinogens above 1 ug/kg and known human carcinogens
regardless of the concentration; and 3) the availability of toxicological data. All chemicals
occurring at a frequency of detection of greater than or equal to 5 percent which have
toxicological criteria were carried through the evaluation, with the exception of inorganics which
are required trace nutrients or which were found at levels not significantly higher than local
background. Regardless of concentrations and/or frequencies of detection, site contaminants not
having current toxicological data were not included for quantitative analysis in the Risk
Assessment.
3.3 Chemicals of Potential Concem Selection Summary
On the basis of the factors described above, chemicals of potential concem were selected for the
GCL Tie & Treating site. Chemicals of concem for this Risk Assessment are summarized in
Table 1. Data for these compounds, presenting the frequency of detection, concentration ranges,
arithmetic and geometric means, and the 95% Upper Confidence Limit (UCL), are included in
Appendix A.
3.4 Chemicals of Concem
The compounds which have been chosen as the chemicals of concem (COCs) in soil meet the
foregoing criteria for carcinogenic and noncarcinogenic indicators and are discussed below.
Table 1 is the hst of COCs for soil at the site. Table 2 provides the typical composition of
creosote and an analysis of the creosote used at GCL Tie & Treating.
Carcinogenic COCs selected for incidental soil ingestion and/or inhalation pathways include
methylene chloride, chloroform, tetrachloroethene, carcinogenic polycyclic aromatic hydrocarbons
(PAHs); including benzo(a)anthracene, benzo(b)fluoranthene, benzo(k)fluoranthene,
benzo(a)pyrene, chrysene, indeno(l,2,3-cd)pyrene, fluorene, and dibenz(a,h)anthracene. The
PAHs are the major contributors to the carcinogenic risks at this site.
Other carcinogenic COCs include bis(2-ethylhexyl phthalate), heptachlor, heptachlor epoxide, DDE, DDT, alpha chlordane, Aroclor 1248, arsenic, and hexavalent chromium.
Do^^OOi 10 400395
Methylene Chloride
DDT Heptachlor
Heptachlor Expoxide
COMPOUND OR COMPONENTS
Naphthalene Methyl Naphthalene
TOTAL 100.0 100.0
* Lorenz and Gjovik, 1972. ** Analysis of standard supplied to GCL by Allied Chemical.
9m}in 12 400337
Noncarcmogenic COCs (some of which also have carcinogenic properties and are included previously) for ingestion and/or inhalation of on-site soil include methylene chloride, chloroform, tetrachloroethene, toluene, ethylbenzene, xylenes, benzo(a)pyrene, fluorene, naphthalene, bis(2- ethylhexyl)phthalate, di-n-octyl-phthalate, acenaphthylene, anthracene, fluoranthene, fluorene, pyrene, phenol, 2,4-dimethylphenol, 4-chloroaniline, aniline, hepthachlor, heptachlor epoxide, DDT and alpha-chlordane.
The only COC assessed for exposure via dermal contact for this site is Aroclor 1248 (PCBs), as this is the only COC for which EPA guidance allows quantitative evaluation. The potential effects of dermal exposure to other COCs, including PAHs, are described in toxicological profiles in Appendix B.
4.0 IDENTIFICATION OF EXPOSURE PATHWAYS
The purpose of this section is to identify the most significant potential pathways through which
individuals may be exposed to the contaminants of concem in various media at and in the
vicinity of the GCL Tie & Treating site. It is based on the Exposure Pathways Analysis Report
submitted to USEPA in July 1993. The analyzed exposure pathways were developed based upon
data gathered during the Removal Action, Subsurface Soil Investigation (SSI) and from a site
visit by Ebasco Environmental personnel in July 1993. Hydrology, geology and current land use
information from the Removal Action and SSI was used. In identifying these pathways both
current and potential future land use of the site and surrounding area will be considered.
As defined in the Risk Assessment Guidance for Superfund (RAGS, 1989), an exposiu-e pathway is composed of the following elements:
• A source and mechanism of chemical release to the environment
• An environmental transport medium (e.g., soil) for the released chemical
• A point of potential contact by humans or animals with the contaminated medium
and
• A route of exposure (e.g., ingestion, inhalation, and dermal contact)
In this RA pathways are identified for the No Action alternative, assuming no site remediation has occurred. The calculated risks include contaminant concentrations found in site soil prior to the EPA Removal Action, as well as later SSI soil sampling results. This RA also assumes that no additional restrictions to site access or use exist The goal is to determine whether it .is
D0065I,' vfi^'OOl 13 400338
feasible for individuals to enter the site areas or engage in activities resulting in exposure to site-
related contaminants.
There are three general routes through which individuals could potentially be exposed to chemical
contamination in soil at the GCL Tie & Treating site; ingestion, inhalation and dermal contact.
The following subsections describe the possible sources, receptors and exposure pathways
relevant to soil considering both current and potential future land use. An identified pathway
does not imply that exposm-es are actually occurring, only that the potential exists for the
pathway to be complete.
4.1 Contaminant Source Analysis
Based upon observations and limited data, the major source area for the principal contaminant, creosote, is the westem portion, or GCL property section, of the site. Due to past practices of open drying, the documented spill, and nature of the wetlands fill material, the entire area of approximately 26 acres must be considered to have some degree of contamination. The area of greatest contamination, with soil concentrations over 50,000 mg/kg of PAHs, is located in the former tank area, north of the large primary building containing the pressure tanks. Contamination may be released to the air, wetlands, surface water sediments or groundwater pathways from the contaminated soils through solubilization of creosote components, volatilization of some components, migration of nonaqueous phase liquids, and physical transport of contaminated soil particles. Only soil pathways are to be considered in this FFS; other media- specific pathways will be evaluated in the RI/FS for the site.
GCL Tie & Treating and other industries used a variety of chemicals over a period of many
years, for various purposes. Accidental spills, discharges, storage tanks, drip areas, dump areas
and other processes are all considered potential contaminant sources. The resulting soil
contamination is considered a source of contamination to groundwater, which will be addressed
in a subsequent investigation.
4.2 Receptor Analysis
Six potential receptor populations were identified for exposure to GCL Tie & Treating site-related
contaminants in soil in present/future use scenarios. Off-site adult and young child residents;
adult and older child trespassers; and on and off-site worker potential exposure pathways were
included in this risk evaluation.
eSBM^ 14 400339
4.3 Exposure Pathway Analysis
There are three general routes through which individuals could potentially be exposed to chemical contamination in soil and other environmental media at the GCL Tie & Treating site; inhalation, ingestion, and dermal contact. An identified pathway does not imply that exposures are actually occurring, only that the potential exists for the pathway to be complete. Several media-specific pathways for contaminant migration and exposure exist. These include inhalation and ingestion of contaminants in airborne soil, surface water mnoff and groundwater, and direct dermal contact with water, soil or sediment. Figures 3 and 4 illustrate the potential current and future exposure pathways for this site.
Pathways Assessed in the FFS Risk Assessment
There is potential exposure resulting from windborne contamination. The soil mounds on site are largely covered. However, additional creosote contaminated soil on-site is still exposed to the environment, potentially venting the volatile components of creosote to the atmosphere. According to a NYSDEC official, there have been complaints from the surrounding businesses of the odors emanating from the facility during the summer months (Site Assessment Sampling Report, D. Perera, undated). The exposed soil is also subject to windborne movement of contaminated soil particles. The proximity of both working and residential communities represent significant receptors for the airborne volatile and soil contaminants via inhalation and ingestion. Site trespassers may also be exposed to site windborne contaminants. As directed by USEPA, the inhalation pathway risk analysis is limited to inhalation of particulates.
Exposure through direct contact with creosote contaminated soils and materials is also a potential
pathway of concem for site trespassers and future on-site workers. Employees on adjacent sites,
as well as other trespassers, may be exposed in areas where wood treating operations previously
occurred. In addition, employees from surrounding industries use the numerous footpaths through
and around the site to gain access to the shopping center across Delaware Avenue from the site.
This potential exposure route has been partially mitigated by erection of additional fencing during
the EPA removal action.
Pathways to be addressed in the RIIFS Risk Assessment
The following pathways will not be addressed in the FFS Risk Assessment, but during the RI/FS
and are included only to provide a full characterization of potential exposure pathways for this
site:
FIGURE 3
Current Use Receptors Current Use Receptors
Primary Source Secondary Source
5*5 nGURE4
GCL TIE & TREATING SITE RISK ASSESSMENT FUTURE USE SOIL EXPOSURE PATHWAYS
Future Use Receptors
Site Trespassers
Child Adult
Industrial and ->
Commercial Activities
Spill/Discharge -> Soil
O CD
D0065.LYN 17
There is potential contaminant migration via surface water runoff. The GCL site is located in
a low-lying area that receives a large amount of water during storms. The ground quickly
becomes saturated, generating surface runoff. In areas where the rainwater forms pools, the water
has been observed as dark brown in color. In addition, a dark colored leachate with a noticeable
oily sheen has been observed emanating from the base of the soil mounds. The runoff from the
western portion of the site drains into the adjacent wetlands area, posing a threat to the biotic
communities. The ranoff on the eastern portion of the site is channeled through a drainage
network toward the main section of the Village of Sidney, potentially exposing the population
to dermal contact and possible inhalation hazards. Sediment in the wetlands and drainage
network may also be impacted. Both the drainage network and the effluent from the wetlands
impoundment empty into the Susquehanna River. This poses a potential threat to the river's use
for recreation and as a fishery.
There is also potential migration through groundwater contamination. Leaching from the
contaminated soils through the overburden poses a threat to existing drinking water wells. There
are also several springs in the area where ingestion, inhalation, and dermal contact represent
hazards. The presence of free creosote product at the surface and potentially in the subsurface
increases the potential for contamination of the bedrock aquifer with dense nonaqueous phase
liquids (DNAPLs).
Present Use Exposure Pathways
At the GCL Tie & Treating site chemical contaminants exist in surface, and potentially in
subsurface soils. Ingestion and inhalation of particulates by off-site workers and local residents
(adult/young child) and ingestion, inhalation and dermal contact by older child and adult
trespassers will be considered completed present use pathways in the FFS Risk Assessment.
Reasonable Maximum Exposure (RME) scenarios will be analyzed, and Average Case (central
tendency) scenarios will be considered should the calculated potential carcinogenic risks exceed
1x10" , or if the Hazard Index is greater than 1.0.
Future Use Exposure Pathways
The present use exposure scenarios via all pathways previously identified will be assumed to extend indefinitely into the future in the exposure analysis. It is expected that site use will remain the same or be used for similar industrial activities for an indefinite period of time. Local land use pattems (mixed residential/commercial/industrial) are not expected to change.
Current conditions at the site indicate that soil ingestion, inhalation, and dermal contact for
industrial/commercial receptors should be considered completed future use pathways. Future on-
Doo65iyN 18 400^43
site workers are assumed to be exposed to contaminants via dermal contact, in addition to the
other present use pathways. RME, and, if calculated risks exceed 1x10" or the HI exceeds 1.0,
Average Case scenarios will be analyzed.
Figures 3 and 4 summarize the current and future soil exposure pathways, illustrating potential contaminant pathways from their sources to these six receptors.
5.0 HAZARD IDENTIHCATION
For chemicals that exhibit noncarcinogenic (e.g., systemic) effects, many authorities consider
organisms to have repair and detoxification capabilities that must be exceeded by some critical
concentration (threshold) before the health effect is manifested. For example, an organ can have
a large number of cells performing the same or similar functions that must be significantly
depleted before an effect on the organ is seen. This threshold view holds that a range of
exposures from just above zero to some finite value can be tolerated by the organism without an
appreciable risk of adverse effects.
Health criteria for chemicals exhibiting noncarcinogenic effects for use in risk assessment are generally USEPA reference doses (RfDs) or reference concentrations (RfCs) developed by the RfD Work Group. The RfD is an estimate of a daily exposure level for humans that is likely to be without an appreciable risk of deleterious effects during a lifetime. For those chemicals for which the USEPA has not derived verified RfDs, health criteria used in a risk assessment may be derived from information provided in the USEPA Health Effects Assessment Summary Tables (HEAST), IRIS, Office of Drinking Water Health Advisories (HAs), Office of Drinking Water Maximum Contaminant Level Goals (MCLGs), or National Ambient Air Quality Standards (NAAQS). The RfD is expressed in units of mg chemical/kg body weight-day. In general, the RfD is an estimate of an average daily exposure to an individual (including sensitive individuals) below which there will not be an appreciable risk of adverse health effects during a lifetime. The RfD is derived using conservative safety factors (e.g., to adjust from animals to humans and to protect sensitive subpopulations) to ensure that it is unlikely to underestimate the potential for adverse noncarcinogenic effects to occur. A RfC is an estimate of the daily exposure that is likely to be without deleterious effects via inhalation. The purpose of the RfD/RfC is to provide a benchmark against which estimated doses (e.g., those projected from human exposure to various environmental conditions) might be compared. Doses that are significantly higher than the RfD/RfC may indicate that an inadequate margin of safety could exist for exposure to that substance and that an adverse health effect could occur.
400344
5.2 Health Effects Criteria for Potential Carcinogens
For chemicals that exhibit carcinogenic effects, USEPA as well as other scientific authorities recognize that one or more molecular events can evoke changes in a single cell or a small number of cells that can lead to malignancy. This is the non-threshold theory of carcinogenesis, which purports that any level of exposure to a carcinogen can result in some finite possibility of causing cancer. Generally, regulatory agencies assume the non-threshold hypothesis for carcinogens in the absence of information concerning the mechanisms of carcinogenic action for the chemical. USEPA's Carcinogen Risk Assessment Verification Endeavor (CRAVE) has developed slope factors (i.e., dose-response values) for estimating excess lifetime cancer risks associated with various levels of lifetime exposure to potential human carcinogens. The slope factor estimate [in units of (mg/kg body weight-day)"*] is a number which, when multiplied by the lifetime average daily dose of a potential carcinogen (in mg/kg body weight-day), yields the upper-bound lifetime excess cancer risk associated with exposure at that dose. Upper-bound is a term used by USEPA to reflect the conservative nature of the slope factors; risks estimated using slope factors are considered unlikely to underestimate actual risks but they may overestimate actual risks for a given exposure. Excess lifetime cancer risks are generally expressed in scientific notation and are probabilities. An excess lifetime cancer risk of 1x10" (one in one million), for example, represents the incremental probability that an individual will develop cancer as a result of exposure to a carcinogenic chemical over a 70-year lifetime under specified exposure conditions.
In practice, slope factor estimates are derived from the results of human epidemiology studies or chronic animal bioassays. The animal studies must usually be conducted using relatively high doses in order to detect possible adverse effects. Since humans are expected to be exposed at lower doses than those used in the animal studies, the data are adjusted by using mathematical models. The data from animal studies are typically fitted to the linearized multistage model to obtain a dose-response curve.
The 95th percentile upper confidence limit slope of the dose-response curve, subject to various adjustments and an inter-species scaling factor is applied to conservatively derive the slope factor estimate for humans. Dose-response data derived from human epidemiological studies are fitted to dose-time-response curves on an ad-hoc basis. These models provide rough, but reasonable, estimates of the upper limits on lifetime risk. Slope factor estimates based on human epidemiological data are also derived using very conservative assumptions and, as such, they too are considered unlikely to underestimate risks.
^ . i - 20 ^^^3^5
Therefore, while the actual risks associated with exposures to potential carcinogens are unlikely to be higher than the risks calculated using a slope factor estimate, they could be considerably lower (if exposure estimates are conservative).
In addition, there are varying degrees of confidence in the weight of evidence for carcinogenicity
of a given chemical. USEPA (1989) has proposed a system for characterizing the overall weight
of evidence for a chemical's carcinogenicity based on the availability of animal, human and other
supportive data. The weight-of-evidence classification is an attempt to determine the likelihood
that an agent is a human carcinogen and thus qualitatively affects the estimation of potential
health risks. Three major factors are considered in characterizing the overall weight of evidence
for carcinogenicity: 1) the quality of evidence from human studies, 2) the quality of evidence
from animal studies which are combined into a characterization of the overall weight of evidence
for human carcinogenicity, and 3) other supportive information which is assessed to determine
whether the overall weight of evidence should be modified. USEPA's final classification of the
overall weight of evidence has the following five categories:
Group A - Human Carcinogen
This category indicates that there is sufficient evidence from human epidemiological
studies to support a causal association between an agent and cancer.
Group B - Probable Human Carcinogen
This category generally indicates that there is at least limited evidence from epidemiological studies of carcinogenicity to humans (Group Bl) or that, in the absence of adequate data in humans, there is sufficient evidence of carcinogenicity in animals (Group B2).
Group C - Possible Human Carcinogen
This category indicates that there is limited evidence of carcinogenicity in animals in the
absence of data on humans.
Group D - Not Classified
This category indicates that the evidence of carcinogenicity in animals is inadequate.
400346 D0065.LYN ,, , 2 1
Group E - No Evidence of Carcinogenicity in Humans
This category indicates that there is no evidence for carcinogenicity in at least two adequate animal tests in different species or in both epidemiological and animal studies.
Slope factors are developed based on epidemiological or animal bioassay data for a specific route
of exposure, either oral or inhalation. For some chemicals, such as chloroform, sufficient data
are available to develop route-specific slope factors for inhalation and ingestion exposure routes.
5.3 Toxicological Assessment
Table 3 summarizes the chronic oral and inhalation reference doses and reference concentrations
(RfDs/RfCs) and slope factors (SFs) used to analyze noncarcinogenic effects and carcinogenic
risks of the COCs. These criteria were the most current data, obtained from February 1994
sessions of the Integrated Risk Information System (IRIS) and 1993 Health Effects Assessment
Summary Tables (HEAST). For carcinogenic PAHs, toxic equivalency factors (TEFs), as related
to benzo(a)pyrene are also included. These TEFs were obtained through communications with
EPA Region II Risk Assessment staff. Chronic toxicity indices were used for residential
exposure. Toxicological profiles are included in Appendix B for those chemicals detected on-
site, for which information is available.
6.0 HEALTH RISK CHARACTERIZATION
According to guidelines for preparing risk assessments for RI/FS purposes, the potential adverse
effects on human health should be assessed where possible by comparing chemical concentrations
found in environmental media at or near the site and at receptor locations with numerical
Applicable or Relevant and Appropriate requirements (ARARs) or other guidance that has been
developed for the protection of human health or the envu-onment. Thus this section presents a
comparison to available ARARs and "To Be Considered" criteria (TBCs). Exposure point
concentration estimates, modeling, chemical-specific methods of evaluation and exposure
frequency parameters are also presented.
6.1 Health-Based Applicable or Relevant and Appropriate Requirements (ARARs) and To
Be Considered (TBC) Criteria
ARARs or other guidance are first identified for the chemicals of potential concem. Where chemical-specific ARARs are available for an environmental medium, they are compared with average and maximum concentrations observed in that medium at points of potential exposure. USEPA interim guidance on ARARs (USEPA, 1987) defines them as follows:
D0065.LYN 2 2
TOXICITY DATA FOR NONCARCINOGENIC AND CARCINOGENIC RISK EVALUATION
Sheet 1 of 2
(mg/Kg-day) (mg/Cu.m) (mg/Kg-day)
Carcinogen Slope Factor
SF Weight Unit Risk SF Weight (Oral) of (Inhalation) inhalation) of
(mg/Kd-day)-l Evidence (ug/Cu.m)-l (mg/Kg-day)-I Evidence
7.5E-03
6.1E-03
5.2E-02
TOXICITY DATA FOR NONCARCINOGENIC AND CARCINOGENIC RISK EVALUATION
Sheet 2 of 2
(mg/Kg/day) mg/Cu.m) (mg/Kg-day)
Carcinogen Slor>e Factor
SF Weight Unit Risk SF Weight (Oral) of (Inhalation) (Inhalation) of
(mg/Kg-day)-l Evidence (ug/Cu.m)-l (mg/Kg-day)- Evidence
1 1 7.3E+00
Group A: Human Carcinogen. Sufficient evidence from epidemiologic studies to support a casual association between exposure and cancer. Group Bl: Probable Human Carcinogen. Limited evidence of carcinogenicity in human from epidemiological studies. Group B2: Probable Human Carcinogen. Sufficient evidence of carcinogenicity in animals. Inadequate evidence of carcinogenicity in humans. Group C: Possible Human Carcinogen. Limited evidence of carcinogenicity in animals. Group D: Not classified. Inadequate evidence of carcinogenicity in animals. Note: - No data/Not available.
Dq YN
"Applicable requirements" means those cleanup standards, standards of control, and other substantive environmental protection requirements, criteria, or limitations promulgated under Federal or State law that specifically address a hazardous substance, pollutant, contaminant, remedial action, location, or other circumstance at a CERCLA site. "Applicability" implies that the remedial action or the circumstances at the site satisfy all of the jurisdictional prerequisites of a requirement.
"Relevant and appropriate requirements" mean those cleanup standards, standards of control, and
other substantive environmental protection requirements, criteria, or limitations promulgated under
Federal or State law that, while not "applicable" to a hazardous substance, pollutant, contaminant,
remedial action, location, or other circumstance at a CERCLA site, address problems or situations
sufficiently similar to those encountered at the CERCLA site that their use is well suited to a
particular site.
The relevance and appropriateness of a requirement can be judged by comparing a number of
factors, including the characteristics of the remedial action, the hazardous substances in question,
or the physical circumstances of the site, with those addressed in the requirement. It is also
helpful to look at the objective and origin of the requirement For example, while RCRA
regulations are not applicable to closing undisturbed hazardous waste in place, the RCRA
regulation for closure by capping may be deemed relevant and appropriate.
A requirement that is judged to be relevant and appropriate must be complied with to the same
degree as if it were appUcable. However, there is more discretion in this determination: it is
possible for only part of a requirement to be considered relevant and appropriate, the rest being
dismissed if judged not to be relevant and appropriate in a given case.
Non-promulgated advisories or guidance documents issued by federal or state governments do not have the status of potential ARARs. However, these "To Be Considered" (TBCs) criteria may be examined in determining the necessary level of cleanup for protection of health or the environment.
Only those ARARs, advisories, guidance or TBCs that are chemical-specific requirements [i.e., those requirements which "set health or risk-based concentration limits or ranges in various environmental media for specific hazardous substances, pollutants, or contaminants" (USEPA, 1987)], as opposed to ARARs/TBCs which are classified as action-specific or locational requirements, are used in this Risk Assessment. For soils, only TBCs from the NYSDEC Technical and Administrative Guidance Memorandum (TAGM) HWR-94-4046: Determination of Soil Cleanup Objectives and Cleanup Levels are available.
v^ 9. 400350
The classes of ambient or chemical-specific health-based ARARs/TBCs that are considered
pertinent to the Risk Assessment for the site are discussed below.
6.1.1 Soil Contamination Compared with ARARs and TBCs
Chemical-specific TBCs for chemicals of concem and the contaminant concentrations detected
are presented in Table 4. Both health-based cleanup levels and the recommended soil cleanup
objective, which may or may not be health-based are included. The range of detected values,
arithmetic means and frequency of detection are summarized in Appendix A. A comparison of
these values reveals soil contamination at levels exceeding TBC criteria.
6.2 Estimation of Exposure Point Concentrations
Estimates of exposure point concentrations are needed as part of the quantitative risk evaluations since these estimates are used along with the exposure scenarios to estimate chronic daily intake and subsequent human health risks.
Estimation of exposure point concentrations for all ingestion, dermal contact and soil inhalation
pathways are based on measured concentrations of the COCs. The representative exposure point
concentration was taken as the 95% UCL of the arithmetic mean of the log transformed data (or
the maximum measured concentration if the 95% UCL exceeds the maximum) of the compound
for the matrix in question using the reasonable maximum exposure case. The 95% UCL was calculated using current USEPA criteria (USEPA, 1992) from the average, variance and standard error of the natural log transformed data using the following equation:
95% UCL = EXP [x + 0.5S^ + HSJ
where x is the mean of the natviral log transformed data, S is the variance on the transformed data, S(, is the standard error on the transformed data and H is the t-value for the transformed data. (The latter value differs from the tabulated t-values because of the natural log transformation of the data.) This calculation includes all analyses for a given compound in a given matrix with the non-detect analyses evaluated at one-half of the sample specific quantitation limit. This is considered a reasonable and conservative, representative concentration for non-detect results.
For the average case scenarios, the 95% UCL (or the maximum concentration, if it was less than the 95% UCL) was used as the exposure point concentration.
''W)wiM~w» 26
• O
GCL TIE AND I'REATING SITE CHEMICALS OF POIENTIAL CONCERN
ARARSATBCS AND LEVELS OF DETECTION IN SOIL
Maximum Concentration Detected (mg/kg)
2.00E-01 4.10E+00 4.80E-01 1.20E-01 6.80E-K)0 3.50E+00 4.10E+02 1.67E+04 1.36E+04 4.40E+03 7.06E+03 2.11E+03 1.40E+04 4.74E+02 5.54E+04 3.68E+04 4.54E+04
NY State Health-Based Criteria** (mg/kg)
9.30E+01 8.00E+03 1.14E+02 1.40E+01 2.00E+05 2.00E+04
NA 2.00E+04 2.24E-01 6.10E-02
NA NA NA
NY State Soil Cleanup Objective**
(mg/kg)
l.OOE-01 5.50E-H00 3.00E-01 1.40E+00 1.20E+00 1.50E+00 4.10E+01 5.00E+01 2.24E-01 6.10E-02 l.lOE+00 l.lOE+00 4.00E-01 1.43E-02 5.00E+01 5.00E+01 5.00E+01
to 1 ^
D0065.LYN 27
TABLE 4
ARARSA'BCS AND LEVELS OF DETECTION IN SOIL
Sheet 2 of 2
Maximum Concentration Detected (mg/kg)
5.50E-01 6.79E+04 2.64E+03 1.67E+03 2.80E+01 2.30E-01 1.66E+02 2.20E-02 2.40E-02 5.00E-03 4.80E-02 2.10E-04 1.60E-01 9.70E+00 1.15E-K)2* 1.15E+02*
NY State Health-Based Criteria** (mg/kg)
5.00E+01 3.00E+02 5.00E+04
NA 2.00+02
2.00E+03 1.23E+02 1.60E-01 7.70E-02 2.10E+00 2.10E+00 5.40E-01 l.OOE+00
NA NA NA
(mg/kg)
5.00E+01 1.30E+01 3.00E-02
NA 2.20E-01 5.00E+01 l.OOE-01 l.OOE-01 2.00E-02 2.10E+00 2.10E+00 5.40E-01 l.OOE+00 10 or SB 10 or SB 10 or SB
o * - Chromium sampling was unspeciated, maximum provided is total chromium. ** - NYSDEC Technical and Administrative Guidance Memorandum:
Determination of Soil Cleanup Objectives and Cleanup Levels, Revised 1/24/94. NA - Not Available SB - Site Background
D0065.LYN 28
6.3 Exposure Point Concentration Modeling
Estimation of exposure point concentrations for all ingestion and dermal contact pathways are based on measured concentrations of the contaminants of concern. In accordance with RAGS (1989), the exposure point concentration was taken as the 95% UCL of the arithmetic mean of the log-transformed data of all the analyses for the matrix in question in a given area, for the reasonable maximum case. For those compounds where the maximum detected value was less than the 95% UCL, the maximum detected value was used. As required by EPA guidance, the average case exposure point concentrations are the same as the RME exposure point concentrations.
Exposure point concentrations for soil particulate inhalation scenarios were estimated using
chemical concentrations in soil and a standard default measurement of suspended soil particles
in air. The models and other methods which were used to estimate the exposure point
concentration are discussed in detail below.
Cowherd Model
A predictive emission factor model was used to estimate respirable particulate emissions. The
Cowherd model uses surface and meteorological data to determine the concentration of
particulates suspended by wind erosion. These calculated emissions are then used to develop
ambient soil bome contaminant concentrations and the resulting potential exposure via inhalation.
This model is referenced in "Rapid Assessment of Exposure to Particulates Emissions from
Surface Contamination Sites" (USEPA, 1985b).
A combination of default factors and site-related values were used in applying this model to the
GCL property risk assessment.
6.4 Estimation of Pathway-Specific Parameters
In this section, parameters are defined for each of the exposure pathways presented in Section 3.0 for both the reasonable maximum and average case scenarios. Average case scenarios were developed for those RME scenarios that exceeded a 1 x 10" carcinogenic risk or Hazard Index of 1.0. These parameters are specific to each exposure scenario and will be used along with the exposure point concentrations previously defined to determine daily chronic or subchronic intake rates for each of the indicator contaminants. One set of parameters is defined for each exposure pathway for the reasonable maximum and average exposure case. As discussed in Section 4, exposure pathways were defined for soil for several populations under present use and potential future use scenarios. Parameters specific to each exposure pathway were developed for each
^ * . « ,Q 4003?4
population as appropriate for present use and futirre use scenarios under reasonable maximum, and when necessary, average case exposure conditions. The present and futvu-e use scenario parameters used in this risk assessment are defined in detail in Tables 5 through 16.
For the evaluation of present-use risks to human health, present-use conditions are assumed to
continue indefinitely into the future. The future-use conditions are evaluated assuming that the
site wUl be used for commercial or industrial purposes.
General Exposure Parameters
Age-specific exposure parameter distributions were developed for each exposed population to account for variation in exposure over an individual's lifetime for each exposure scenario. The parameters used in this FFS Risk Assessment were included in the July 1993 GCL Tie & Treating Exposure Pathways Analysis Report and approved by USEPA. These distributions were largely based on data contained in the Risk Assessment Guidance for Superfund (RAGS, 1989). Three age groups were specified ages 0-6 (young children), ages 6-12 (older children) and ages 18-70 years (adults). For carcinogens only, a six-year exposure rate was assumed for children (USEPA Region II criteria).
Residential exposure durations were considered to be the years spent at one residence based on
maximum and average case estimates (USEPA, 1989). Noncarcinogenic hazard index values
were calculated for each group, but were not considered to have additive effects over a lifetime.
Frequencies of exposure to contaminants for the various populations were developed based on
RAGS (1989) and site-specific information.
The number of days per year that an individual might be exposed to site contamination varied
depending upon age, vocation and exposure route. Resident exposures are assumed to occur on
a daily basis, with 15 days per year spent away from home.
Three body weight groups were specified for the receptors, adults (18-70 years) at 70 kg, older children (6-12 years) at 35 kg, and young children (0-6 years) at 15 kg, in accordance with RAGS and other risk assessment guidance.
Soil Exposure Pathway Parameters
In this baseline Risk Assessment only soil contamination was evaluated. In the current and
future-use scenarios, residents, trespassers and workers were assumed to be exposed to site soils.
^'S!^M!^^' 30 4003^5
TABLE 5
GCL TIE & TREATING SITE PARAMETERS AND ASSUMPTIONS TO CALCULATE RME
SOIL EXPOSURE PATHWAYS CURRENT/FUTURE USE SCENARIO
Receptor: Oif-Site Residents - Young Children
Exposure Route
12<^^
0.83®
(1) Risk Assessment Guidance for Superfund (EPA, 1989) (2) Human Health Evaluation Manual, Supplemental Guidance, "Standard Default Exposure
Factors" (EPA, 1991) (3) Dermal Exposure Assessment: Principles and Applications (EPA, 1992)
400356
Receptor: Off-Site Residents - Adults
8(3)
0.83®
32 400357 D0065I.YN
GCL TIE & TREATING SITE PARAMETERS AND ASSUMPTIONS TO CALCULATE AVERAGE
Receptor: Off-Site Residents - Young Children
Exposure Route Ingestion
Ingestion Rate (mg/day) 200®
D006:
400358
Receptor: Off-Site Residents - Adults
Ingestion
43 ^
9(2)
70®
1
100^^>
*.#p M ituui ' I ^ S ^ V N ' 34 40035^
TABLE 9
Receptor: Site Trespassers - Older Children
Exposure Route
5®
0.83®
DOO65/YN" t s k v n 1)1 35 4 U U a b
TABLE 10
Receptor: Site Trespassers - Adults
Exposure Time (hours/day)
8(3)
0.83®
0<Jfc i- {•*' >i A n of t I D0065iYN 3 6 4UUwVX
TABLE 11
Receptor: Site Trespassers - Older Children
D0065
4003P2
Receptor: Site Trespassers - Adults
Ingestion
43®
g(i)
70®
1
100®
s;%oui D0065I.YN 38 400363
TABLE 13
GCL TIE 8c TREATING SITE PARAMETERS AND ASSUMPTIONS TO CALCULATE RME
SOIL EXPOSURE PATHWAYS FUTURE USE SCENARIO
Receptor: On-Site Workers
gO)
0.83®
Doo65.i-'^||^y y ^ 39 4003B4
TABLE 14
SOIL EXPOSURE PATHWAYS FUTURE USE SCENARIO
Receptor: On-Site Workers
Factors" (EPA, 1991)
TABLE 15
Receptor: Off-Site Workers
Inhalation
250®
25®
70®
1
g(3)
0.83®
D0065:
Receptor: Off-Site Workers
Ingestion
250®
25®
70®
1
50®
Factors" (EPA, 1991)
400367 D0065.LYN 4 2
Residents were assumed to be exposed to soil via ingestion, inhalation and dermal contact. The
adults were assumed to be exposed for 30 years, 350 days per year, and children were assumed
to be exposed for 6 years (carcinogenic scenario only), 350 days per year, in the reasonable
maximum case. In the average case, adults were assumed to be exposed for 9 years (RAGS,
1989).
For soil ingestion exposure parameters, adult residents were assumed to consume an average of
100 mg/day. Other potential receptors' exposure parameters vary, as shown in Tables 5 through
16.
7.0 QUANTITATIVE RISK CHARACTERIZATION
The results of the quantitative assessment of human health risks associated with exposure
pathways and scenarios described previously, and the methods used to perform the quantitative
analysis are presented below. Potential health risks were calculated for baseline conditions, i.e.,
with no treatment of soil prior to contact with the receptor. The approach taken in this section
is to combine the concentration data, exposure scenarios and chemical intake models, and critical
toxicity values, to generate quantitative estimates of carcinogenic and noncarcinogenic health
risks for the present and future use reasonable maximum case and average case (when necessary)
exposure scenarios. These factors are combined using methods defined by the EPA for exposure
and risk assessment for Superfund and other hazardous waste sites. These methods are discussed
in detail in Section 7.1.
Section 8 provides a discussion of the risk results for the various exposure pathways and
scenarios.
7.1 Ouantitative Risk Assessment Methods
To quantitatively assess the potential risks to human health associated with present and future use
scenarios, chronic daily intakes (CDIs) were calculated for each ingestion and inhalation exposure
pathway, and Dermally Absorbed Doses (DADs) were calculated for the dermal exposure
pathway using the estimated exposure point concentrations. Formulae for each soil exposure
pathway are summarized in each spreadsheet in Appendix C. CDIs and DADs are expressed as
the amount of a chemical an individual would be exposed to per unit body weight per day (e.g.,
mg/kg/day). The CDI and DAD are averaged over a lifetime for carcinogens for adults (RAGS,
1989) and over a six year period for children. The CDI and DAD are averaged over the annual
exposure period for noncarcinogens (RAGS, 1989).
DooeS^COUt 43
The estimated CDIs and DADs are then combined with health effects criteria (reference doses and slope factors) to quantitatively estimate potential human health risks. For potential carcinogens, excess lifetime cancer risks are obtained by multiplying the CDI/DAD for the contaminant under consideration by its slope factor (SF). Cancer risks are assumed to be additive. Therefore, the sum of these values is then evaluated as the potential for excess cancer risks. The goal of developing a quantitative risk assessment is to estimate the potential upper-bound risk at the site in the absence of remediation.
Potential risks for noncarcinogens are presented as the ratio of the CDI/DAD to the reference dose (RfD) (i.e., CDI/RfD). The sum of all the ratios of chemicals under consideration for a given pathway is called the hazard index. The hazard index is useful as a reference point for gauging the potential noncarcinogenic effects of environmental exposures to complex mixtures. In general, hazard index values which are less than one are not likely to be associated with any health risks. If the hazard index is greater than one for the total of all chemical HI values, the compounds can be segregated according to their critical effects (target organs) and separate hazard index values can be derived for each effect (USEPA 1986a). A conclusion should not be categorically drawn, however, that all hazard index values less than one are "acceptable" or that hazard index values greater than one are "unacceptable". This is perhaps a consequence of the one order of magnitude or greater uncertainty inherent in estimates of the RfD, as discussed further in Section 11.
In accordance with USEPA's guidelines for evaluating the potential toxicity of complex mixtures
(USEPA, 1986b), it is assumed that the toxic effects of the site-related chemicals would be
additive. Lifetime excess cancer risks and the CDI/RfD ratios are summed to indicate the
potential risks and effects associated with mixtures of potential carcinogens and noncarcinogens,
respectively. In the absence of specific information on the toxicity of the mixture to be assessed
or on similar mixtures, USEPA guidelines generally recommended assuming that the effects of
different components in the mixtures are additive when affecting a particular organ or system.
Synergistic or antagonistic interactions may be taken into account if there is specific information
on particular combinations of chemicals. In this risk assessment, it is assumed that the potential
effects of the site-related chemicals are additive, and no synergistic or antagonistic effects exist.
The CDIs/DADs of chemicals of potential concem for potentially exposed individuals are first calculated. To determine these CDIs/DADs, the assumptions concerning chemical concentrations (exposure point concentrations) and exposure conditions such as fi-equency, duration, and time of exposure, are used together with media intake parameters. For each exposure scenario, a reasonable maximum case is considered. The reasonable maximum case scenario is intended to place an upper-bound limit on the potential risks by combining plausible maximum exposure estimates with upper-bound health criteria. For the reasonable maximum case, the exposure point
^ ^ £ ^ 44 4003^9
and duration of exposure.
8.0 RESULTS OF RISK CALCULATIONS
The following is a receptor-specific summary of the results of the potential carcinogenic and
noncarcinogenic risk calculations for the reasonable maximum and average case soil exposure
scenarios for the GCL Tie & Treating site. Reasonable maximum exposure (RME) case potential
carcinogenic risks and Hazard Indices were calculated for all exposure pathways approved by
EPA following review of the July 1993 GCL Tie & Treating Exposure Pathways Analysis
Report. Average case potential risk was calculated for those RME exposure pathways exceeding
either a Hazard Index of 1.0 or carcinogenic risk of 10" . This includes potential carcinogenic
risk from soil ingestion for all receptors. The risk calculation spreadsheets are included as
Appendix C.
Residents, trespassers and workers were assumed to be exposed to soil via several pathways in
the cmrent/future use scenarios.
8.1 Reasonable Maximum Case
The carcinogenic risk of incidental ingestion of soil contaminants for adult and young child
residents is 2.67E-04 and 6.23E-04, respectively. Adult and child trespasser risks are similar, at
2.67E-04 for both potential receptors, adults and older children. On-site and off-site worker
potential risks are higher, at 9.54E-04. The majority of the risk is due to the presence of
carcinogenic PAHs, particularly benzo(b)fluoranthene, benzo(a)pyrene, dibenz(a,h)anthracene,
benzo(a)anthracene, benzo(k)fluoranthene and indeno(l,2,3-cd)pyrene. Arsenic also significantiy
contributes to excess risk. The HI for incidental ingestion for adult and child residents is below
the threshold of 1.0, at 4.92E-02 and 4.59E-01, respectively. Adult and older child trespassers
have an HI of 4.94E-02 and 1.98E-01, respectively. Worker noncarcinogenic potential impacts
were calculated to be 1.69E-01, also below the level of concern.
Potential carcinogenic risk from the inhalation of airborne particulate soil contaminants for adult and child residents is 3.49E-06 and 6.11E-06, respectively. Potential risks are 3.49E-06 and 1.09E-06 for adult and older child trespassers and 2.60E-06 for both off-site and on-site workers. The risk is largely due to the presence of carcinogenic PAHs, particularly benzo(b)fluoranthene. Benzo(a)pyrene, dibenz(a,h)anthracene, arsenic and chromium also contribute to risk. The calculated HI for inhalation is 6.19E-04 for adult, and 6.07E-03 for young child residents. Trespassers His are 8.67E-04 and 1.08E-03 for adults and older children, respectively. On-site
J ^ - O e i 45 460170
and off-site worker potential noncarcinogenic risks were calculated to be 6.19E-04. Like ingestion, all inhalation His are below levels of concem.
Dermal contact risk was calculated as 1.98E-07 for adult trespassers and 5.51E-08 for older child
trespassers. On-site worker risks from dermal contact are at 5.09E-08.
8.2 Average Case
Ingestion average case potential carcinogenic risk is 1.23E-05 for adult, and 2.32E-04 for young
child residents. Adult and older child trespassers face an average case potential risk of 1.23E-05
and 9.92E-05, respectively. On-site worker risks were calculated at 1.99E-04, and off-site worker
risks at 9.94E-05.
9.0 COMBINING RISK LEVELS AND HAZARD INDEX VALUES ACROSS PATHWAYS
Tables 17 through 19 present the combined potential carcinogenic risk and Hazard Index for each
receptor considered to have multiple pathways of exposure under reasonable maximum exposure
conditions, for present and future use scenarios at the GCL Tie & Treating site. Adult and young
child residents, adult and older child tiespassers, and on and off-site workers are all considered
to have multiple potential pathways of exposiu-e.
In the current and future use exposure scenario, off-site residents were assumed to be exposed to soil via ingestion and inhalation. As shown in Table 17, the combined carcinogenic potential risk is 2.70E-04 and 6.29E-04 for adult and child residents, respectively; and the HI is 4.98E-02 for adults and 4.65E-01 for children. Table 18 illustrates the combined risk and HI for site trespassers and Table 19 is a summary of worker potential risks and His.
10.0 QUALITATIVE DISCUSSION OF RISKS NOT QUANTITATIVELY EVALUATED IN
THE RISK ASSESSMENT
The quantitative risk assessment of site matrices does not include several compounds detected in the EPA Removal Action and Subsurface Soil Investigation for several reasons. Some compounds did not meet the frequency of detection criterion, or were not probable human carcinogens detected above 1 ug/kg, while others lacked sufficient toxicological data.
Contaminants are present in soil in a highly varied pattern. Some potential source areas have minimal levels of a few contaminants in soil.
t iV^ i^ 46 400371
TABLE 17
O
GCL TIE & TREATING SITE OFF-SITE RESIDENT RISK LEVELS AND HAZARD INDEX VALUES
SUMMARY ACROSS EXPOSURE PATHWAYS PRESENT/FUTURE USE SCENARIOS
Present/Future Use Scenarios:
Exposure to Soil
Off-Site Resident Adults
3.49E-06 2.67E-04
6.11E-06 6.23E-04
6.19E-04 4.92E-02
6.07E-03 4.59E-01
Summation Results - Off-Site Resident Adults:
Carcinogenic Health Effects = 2.70E-04 Noncarcinogenic Health Effects = 4.98E-02
Summation Results - Off-Site Resident Children:
Carcinogenic Health Effects = 6.29E-04 Noncarcinogenic Health Effects = 4.65E-01
o o CO
1 TABLE 18
GCL TIE & TREATING SITE SITE TRESPASSER RISK LEVELS AND HAZARD INDEX VALUES
Present/Future Use Scenarios:
Exposure to Soil
Carcinogenic Risk Levels Reasonable Maximum Exposure
3.49E-06 2.67E-04 1.98E-07
1.09E-06 2.67E-04 5.51E-08
8.67E-04 4.94E-02
1.08E-03 1.98E-01
Total Health Risk = Soil Inhalation + Soil Ingestion + Soil Dermal Contact
Summation Results - Adult Trespassers:
CO SI V9
Carcinogenic Healtii Effects = 2.68E-04
Noncarcinogenic Health Effects = 1.99E-01
1) Inhalation 2) Ingestion
GCL TIE & TREATING SITE WORKER RISK LEVELS AND HAZARD INDEX VALUES
Carcinogenic Risk Levels Reasonable Maximum Exposure
2.60E-06 9.54E-04 5.09E-08
6.19E-04 1.69E-01
6.19E-04 1.69E-01
Total Health Risk = Soil Inhalation + Soil Ingestion + Soil Dermal Contact
Summation Results - On-Site Worker:
Summation Results - Off-Site Worker:
• M l
o Cdf * ^ J D0065 LYN
lib.
49
11.0 POTENTIAL PUBLIC WELFARE IMPACTS
Contamination at levels exceeding regulatory limits, and posing potential carcinogenic and noncarcinogenic health impacts has been detected in site soil. Additionally, the soil contamination may preclude its future use as a commercial site. This may limit redevelopment and decrease the value of the site and smrounding properties.
12.0 UNCERTAINTY IN THE RISK ASSESSMENT
The quantitative assessment of health effects at hazardous waste sites is inherently uncertain. This uncertainty arises from the need to predict potential future health impacts (often quite subtie or infrequent events) in the absence of observed health effects, and on the basis of limited data concerning contaminant levels, transport mechanisms, receptor behavior and the toxicologic behavior of the chemicals present. In the face of this unavoidable uncertainty, the general approach that has been employed in this assessment is to first develop conservative estimates of contaminant concentrations. Doses and health risks for each medium were calculated for specified exposure routes and exposed populations. For reasonable maximum case exposure, scenarios are defined to provide additional information as to whether the specific contaminant/pathway/receptor scenario is, in fact, likely to be associated with adverse health effects. These reasonable maximum case exposure scenarios, while providing a likely assessment of contaminant exposiue and health risks, are still uncertain, and contain a number of inherentiy conservative assumptions which provide an additional "margin of ertor" for interpreting the quantitative risk estimates. Sources of uncertainty in this risk assessment are summarized in Table 20.
12.1 Uncertainties Associated with Sampling and Analytical Procedures
A major group of factors contributing to uncertainty in the risk analysis are the uncertainties in
exposiue point concentration estimates associated with the sampling, analytical and modeling
procedures used to define contaminant levels in contaminated media. In this assessment,
available data from the USEPA Removal Action and Subsiuface Soil Investigation were used to
assess contaminant levels. This is in accordance with Superfund Accelerated Cleanup Model
(SACM) guidelines.
An analysis of these data sets was performed, using the EPA "Guidance for Data Useabihty in Risk Assessment," Part A. This analysis shows that the data from the Removal Action increases the uncertainty inherent in the calculated potential risks.
D0065.LYN 50 4 0 0 3 7 5
TABLE 20 Sheet 1 of 2
f t '
LIKELY MAGNITUDE OF UNCERTAINTY LEVEL OF BL\S INTRODUCED
Reasonable maximum case exposure
point concentrations calculated using
Highest contaminant levels used to
develop reasonable maximum case
exposure estimates when exceeded
calculation of reasonable maximum risk
Use of historical, removal action
data in risk assessment
meet risk assessment useability criteria
Likely upward bias.
Trench sampling design Moderate to high - data may not be
representative of site conditions
Particulate generation and transport
parameters instead of measured values.
Residential land use unlikely
Exposure estimates assume contaminants
are conservative over time
COCs are persistent
SUght upward bias
GCL TIE & TREATING SITE
CP - J
-a
SOURCE OF UNCERTAINTY LIKELY MAGNITUDE OF UNCERTAINTY LEVEL OF BL .S INTRODUCED
Estimates of physiological, behavioral
residences and use of conservative parameters
Moderate upward bias
Estimates of exposure frequency and duration Moderate Moderate upward bias
Estimates of contaminant contact rates,
intake factors
3. Toxicological Risk Characterization Methods
RfD/CDI ratios to characterize non-cancer
health effects
variable; uncertainty factors vary by orders of magnitude
RfDs are defined conservatively for most pollutants
Lack of toxicity criteria for certain
chemical detected on-site
for dermal contact
Calculated risks may be understated
SFs, Unear low-dose model to assess cancer risks Moderate to high - most SFs are derived from
animal bioassay data
risk slopes
exposures are additive
D0065.LYN
I
52
Use of data from the Removal Action may result in elevated contaminant concentrations being input to the risk assessment by including pre-removal samphng results. There are also several data quality concerns for Removal Action data. These concerns include high detection limits, lack of a specific map for sampling locations, uncharacterized areas of the site, and SOPs and field records not being provided to the risk assessor.
Additional uncertainty comes from the limited analysis done on some samples. Method Detection
Limits (MDLs) which were higher than concenfrations of concem, and unverified data review
procedmes. Several data quaUty indicators, including data completeness, comparability,
representativeness, precision and sampling accuracy are also questionable for the Removal Action
data. All these factors add uncertainty to the resulting risk estimates.
However, the Subsurface Soil Investigation (SSI) data provides an appropriate data quality level
that mitigates some of this uncertainty. It is apparent from the SSI data that while the Removal
Action data may increase overall concentrations, it does not intioduce inappropriate chemicals
of concem. The SSI provides the chemical-specific data necessary to meet the needs of the risk
assessment process.
The SSI sampling does present some additional uncertainty in terms of site coverage and sample representativeness. The tiench locations do not include several areas of the site. Therefore, these areas, and the site are not fully characterized.
The determination of the contaminant concentrations which were used in the exposure models also provides some uncertainty. Pursuant to EPA directives, [Ebasco Environmental's March 7, 1990 meeting regarding guidance issues from the Risk Assessment Guidance for Superfund (RAGS, 1989)] it was determined that the contaminant concentration used for the quantification of risk should be the 95% UCL on the arithmetic mean of the log-transformed data for the RME case. This value provides a conservative estimate of the reasonable maximum risks when used with conservative exposure parameters. This method, however, assumes that the data have a geometiic distribution, which may contribute further to uncertainty, ff the data for a given compound does not have a geometric distiibution, the method used to calculate the UCL results is an inflated value relative to the maximum value in the untransformed data.
Incorporation of these UCLs into the risk calculations would result in an overly conservative estimate of risk which would not be representative of the actual site conditions. Therefore, when the 95 percent UCLs on the arithmetic mean of the log-transformed data exceeded the actual maximum contaminant concentrations detected, the maximum contaminant concentration detected was used for the risk calculations. Similar uncertainty exists fo

report: focused feasibility study, baseline risk

Documents