Attach 2.1 SE Lighthouses Alternate Cleanup Levels.pdf

TECH MEMO_ACL_Final_update.doc 1

T E C H N I C A L M E M O R A N D U M

Alternate Cleanup Levels for Soil at USCG Light Stations in Southeast Alaska PREPARED FOR: Mark Ridgway/USCG CEU Juneau PREPARED BY: Dennis Shelton/CH2M HILL

Donna Morgans/CH2M HILL COPIES: Jane Gendron/CH2M HILL

Win Westervelt/CH2M HILL DATE: February 20, 2003— Updated March 19, 20041

Introduction This document presents al ternate cleanup levels (ACLs) developed for soil at selected U.S.

Coast Guard (USCG) light stations in Southeast Alaska. As authorized under 18 AAC 75.340, ACLs can be developed using Alaska Department of Environmental Conservation approved methods in order to provide more realistic site specific cleanup levels.

ACLs are risk-based target concentrations derived for each chemical constituent that was detected at least once at any light station. ACLs are developed for the most plausible potential future land uses at these light stations. The remote nature and logistical challenges of accessing the light stations is considered in developing ACLs which are protective of human health and the environment while minimizing the cost of remediation.

The results of the ACL comparison will support remedial decisions for these light stations. Site data from each individual light station will be evaluated to determine whether cleanup is needed, and if so, which scenario-specific ACLs are most attainable, given the reasonably anticipated future land uses. Additionally, the results could support a decision for restricting future land uses. Cleanup decisions for specific lighthouses will be evaluated as part of a separate deliverable.

Eight of the light stations were investigated by CH2M HILL between May 5 and May 17, 2002 to collect soil data, which are presented in the Phase III Site Investigations Draft Report (CH2M HILL 2002). Two light stations underwent remedial action efforts in the summer of 2002 based on prior studies. Cleanup concentrations remaining in soil at these sites have been incorporated into this analysis for a comprehensive evaluation of all light stations.

Generic Conceptual Site Model for USCG Light Stations This section provides a generic description of the conceptual site model (CSM) for the USCG Light Stations in Southeast Alaska. This CSM provides a current understanding of the physical setting, sources of contamination, potential future land uses, potential groundwater use, and identifies potentially complete exposure pathways for the light stations.

1 Material in this updated Technical Memorandum is shown within a box

Attachment 2

ALTERNATE CLEANUP LEVELS FOR SOIL AT USCG LIGHT STATIONS IN SOUTHEAST ALASKA

TECH MEMO_ACL_FINAL_UPDATE.DOC 2

The USCG historically operated light stations at the following locations in Southeast Alaska:

• Cape Decision, on the southwestern end of Kuiu Island, 63 miles south of Sitka

• Cape Spencer*, an island at the entrance to Cross Sound, 70 miles west of Juneau

• Eldred Rock*, on the southeasterly part of Eldred Rock in Lynn Canal, about 50 miles north of Juneau

• Fairway Island*, an island in Peril Straits, 28 miles northeast of Sitka

• Five Finger, an island in Stephens Passage, 70 miles east of Sitka

• Guard Island*, the largest of two islands at the easterly entrance to Tongass Narrows, 9 miles north of Ketchikan

• Lincoln Rock*, an island at the westerly end of Clarence Strait, 54 miles northwest of

Ketchikan

• Mary Island*, an island in Revillagigedo Channel, 10 miles southeast of Ketchikan

• Point Retreat, on the tip of Admiralty Island about 20 miles northwest of Juneau

• Point Sherman, on the mainland 38 miles north of Juneau

• Sentinel Island*, an island in Lynn Canal, 23 miles northwest of Juneau

• Tree Point*, a peninsula on the mainland, 40 miles southeast of Ketchikan

The sites are all remote and accessible only by boat or helicopter. Figure 1 shows the relative location of these light stations. Of these twelve sites, all but Point Retreat will be evaluated for potential human health risks using the ACLs presented in this Technical Memorandum. Eight of these stations (marked with an “*” above) were included in a Phase III site investigation conducted in the spring of 2002. The site descriptions, photos and maps for these stations are provided in the Phase III Site Investigations Draft Report (CH2M HILL 2002). Two sites are futher described in the Five Finger and Cape Decision Remedial Action Report (Jacobs, 2002)

Contaminant Sources In the early 1900's, the lighthouses were painted with paints designed to chalk in order to maintain a freshly painted appearance on the building exteriors. The paints contained lead, which was then deposited in the soils surrounding the buildings. In addition, maintenance activities on the exterior paint surfaces released paint chips into the surrounding soils, a practice that continued until the 1990's. As a result of the type of paint and maintenance activities of the past hundred years, many lighthouses now have or are suspected to have significant lead contamination in the soil.

In addition, all of the sites had one or more fuel storage tanks. The tanks were used to store heating oil or gasoline for the light towers, keeper’s dwellings and shop buildings. At some locations, spills or leaks occurred resulting in contaminated soil around the tanks or along fuel pipelines. Refuse at the light stations was disposed of either in burn pits, carried back to stations on supply vessels, or thrown off docks. Burn pits are still in evidence at several light stations.

TECH MEMO_ACL_FINAL_UPDATE.DOC 4

Potential Future Land Uses at Light Stations Most of the lighthouses were established between 1902 and 1925 and until the 1970s operated as fully staffed facilities. These facilities are currently automated and unmanned. Limited USCG funding prohibits the upkeep of unused lighthouse buildings and so most of these historic facilities are deteriorating. The primary objective of the USCG is to convey lighthouse properties to other federal agencies and to nonprofit corporations under the authority of the National Historic Lighthouse Preservation Act of 2000.

The Act amends the National Historic Preservation Act of 1966 and establishes a National Lighthouse Preservation Program. This program allows for historic preservation and public access. The USCG will continue to have the right of access to maintain and service the active lights.

Following transfer to other owners, potential future site users could include seasonal visitors (possibly including both adults and children), or onsite worker/residents (e.g., bed and breakfast owner/keeper). The remote locations of the lighthouses naturally limit the future potential use of the sites.

Potential for Groundwater Use In general, the soil around the facilities is relatively shallow (a few inches to a few feet). Bedrock is either exposed or is very near the land surface. The soils are typically poorly drained, fine grained soils with high organic content.

Groundwater that may exist at the sites is generally not considered a drinking water source by application of 18 AAC 75.350. In addition, groundwater is generally not within a zone of contribution of an active private or public drinking water system, is not within a recharge area for a public or private drinking water well, or a wellhead protection area, or a sole source aquifer.

When these facilities were manned, rainwater was typically collected from the roof of the lighthouse for storage in water cisterns onsite. This suggests that no dependable source of groundwater, or that non-potable groundwater, is found at these locations. The USCG has no plans to try to locate groundwater at the sites or provide onsite groundwater or surface water drinking water supplies at these locations. It is unlikely that future owners or leaseholders of the sites would consider groundwater, if present, for use as potable water. Future use of the sites would rely on outside sources of drinking water. Migration to groundwater ACLs were not developed for the light stations because groundwater is not used for drinking water purposes.

Potential Human Exposure Pathways This section describes the potential exposure pathways resulting from site contaminants, based on currently available site information. These potential pathways were identified in accordance with ADEC guidance, with the use of professional judgement and information on contaminant sources, release mechanisms, routes of migration, potential exposure points, potential routes of exposure, and potential receptor groups expected to be associated with these sites.

An exposure pathway can be described as the physical course that a chemical of potential concern (COPC) takes from the point of release to a receptor. Chemical intake or route of

TECH MEMO_ACL_FINAL_UPDATE.DOC 5

exposure is the means by which a COPC enters a receptor. For an exposure pathway to be complete, all of the following components must be present:

• A source (e.g., lead-based exterior paint)

• A mechanism of chemical release and transport (e.g., wind erosion)

• An environmental transport medium (e.g., soil)

• An exposure point (e.g., onsite near light station)

• An exposure route (e.g., incidental ingestion)

• A receptor or exposed population (e.g., seasonal visitors)

In the absence of any one of these components, an exposure pathway is considered incomplete and by definition, there is no risk or hazard. Figure 1 presents the CSM schematic for the site.

Based on current understanding of reasonably anticipated future uses of the USCG light stations, the most plausible exposure scenarios that are considered these stations include the following:

• Incidental ingestion, inhalation, and dermal contact with onsite soil by future seasonal visitors (possibly including both adults and children).

• Incidental ingestion, inhalation, and dermal contact with onsite soil by future lighthouse worker/residents (children and adults, e.g., bed and breakfast owner/keeper).

Derivation of Soil ACLs ACLs are risk-based target concentrations derived for each chemical constituent that was detected at least once at any light station, and are developed for the most plausible exposure scenarios identified for the USCG light stations. Table 1 summarizes the chemical constituents detected at each of the ten light stations.

Due to the implicit differences in methodologies for evaluating exposure and risk for lead and non-lead COPCs, the approaches and derivation of ACLs are presented separately for non-lead COPCs and lead in the following sections.

TECH MEMO_ACL_FINAL_UPDATE.DOC 6

TABLE 1

Chemical Constituents Detected at U.S. Coast Guard Light Houses in Southeast Alaska US Coast Guard Light Stations

Chemical Group

Analyte Cape Decision

Cape Spencer

Eldred Rock

Fairway Island

Five Finger

Guard Island

Lincoln Rock

Mary Island

Sentinel Island

Tree Point

BTEX Ethylbenzene X BTEX Toluene X X X BTEX Xylenes, Total X X METAL Arsenic (RCRA) X X METAL Barium (RCRA) X X X X X METAL Cadmium (RCRA) X X X X METAL Chromium (RCRA) X X X X X METAL Lead (RCRA) X X X X X METAL Mercury (RCRA) X X X X X METAL Lead (Total X X X X X X X X X X PAH Acenaphthylene X PAH Anthracene X PAH Benzo(a)anthracene X X PAH Benzo(a)pyrene X X X PAH Benzo(b)fluoranthene X X PAH Benzo(g,h,i)perylene X X PAH Benzo(k)fluoranthene X PAH Chrysene X X X X PAH Dibenzo(a,h)anthracene X PAH Fluoranthene X X X X PAH Fluorene X PAH Indeno(1,2,3-cd)pyrene X X PAH Naphthalene X X X PAH Phenanthrene X X X X PAH Pyrene X X X X TPH Diesel Range Organics X X X X X X X X TPH Gasoline Range Organics X X TPH Residual Range Organics X X X X VOC 1,2,3-Trichlorobenzene X VOC 1,2,4-Trichlorobenzene X VOC 1,2,4-Trimethylbenzene X VOC 1,3,5-Trimethylbenzene X VOC 1,4-Difluorobenzene X X VOC 2-Butanone X X VOC 4-Isopropyltoluene X VOC Acetone X X VOC Benzene X X VOC Ethylbenzene X VOC isopropyl alcohol X VOC Isopropylbenzene X VOC m&p Xylenes X VOC Methylene Chloride X X VOC n-Propylbenzene X VOC Naphthalene X VOC O-Xylene X VOC tert-Butyl Methyl Ether X VOC Toluene X VOC Trichlorotrifluoroethane X X

TECH MEMO_ACL_FINAL_UPDATE.DOC 7

Soil ACLs for Non-Lead Constituents This section provides the equations used for computing ACLs for carcinogens and noncarcinogens other than lead. The symbol definitions are provided in the footnote2.

Methods for Carcinogens

The following equation was used to calculate the remote lighthouse worker/resident soil ingestion cleanup levels for carcinogenic constituents:

adjsoil

IFEF

mg kg o

SF

yeardaysATCTR kgmg ingestion

ACL

/365

Where:

a aa c cc adjsoil BW

EDIR

BW

EDIR

IF

The following equation was used to calculate the seasonal visitor soil ingestion ACL concentrations for carcinogenic constituents:

a

IREF

a

ED

mg kg o

SF

yeardaysATC a

BWTR

kgmg ingestion

ACL

/365

The following equation was used to calculate the remote lighthouse worker/resident, and adult seasonal visitor inhalation ACL concentrations for carcinogenic constituents:

2 ACLingestion = alternate concentration limit for the ingestion route (mg/kg) ATN = averaging time for noncarcinogens ATC = averaging time for carcinogens BWa = adult body weight BWc = child body weight EDa = adult exposure duration EDc = child exposure duration EF = exposure frequency IFsoil/adj = age-adjusted soil ingestion factor Inh = inhalation rate IRa = adult soil ingestion rate IRc = child soil ingestion rate RfC = reference concentration RfDi = inhalation reference dose Sfi = inhalation slope factor Sfo = oral slope factor THQ = target hazard quotient TR = target risk URF = unit risk factor VF = volatilization factor

TECH MEMO_ACL_FINAL_UPDATE.DOC 8

VF

EDEF

mg g

URF

yeardaysATTR kgmg inhalation

ACL

/365 μ

The following equation was used to calculate the child seasonal visitor inhalation ACL concentrations for carcinogenic constituents:

iSF

VF

InhEFED yeardaysAT c

BWTR

kgmg inhalation

ACL

Methods for Noncarcinogens

The following equation was used to calculate the remote lighthouse worker/resident, and adult and child seasonal visitor soil ingestion ACL concentrations for noncarcinogenic constituents:

o RfDmg kg

IREDEF

yeardaysATNBWTHQ kgmg ingestion

ACL

/365

The following equation was used to calculate the remote lighthouse worker/resident, and adult seasonal visitor inhalation ingestion ACL concentrations for noncarcinogenic constituents:

VFRfC

EDEF

yeardaysATNTHQ kgmg inhalation

ACL

/365

The following equation was used to calculate the child seasonal visitor inhalation ACL concentrations for noncarcinogenic constituents:

VF

i

RfD InhEDEF yeardaysATNBWTHQ kgmg inhalation

ACL

Per ADEC guidance, the most stringent of the ingestion or inhalation ACLs calculated is selected as the ACL for each detected constituent. Some constituents are known to cause both carcinogenic and noncarcinogenic health effects. When this occurs, a carcinogenic and noncarcinogenic ACL is calculated, and the most stringent value is selected.

Exposure Assumptions

Exposure assumptions for development of soil ACL concentrations for each of the exposure scenarios evaluated are summarized in Table 2. Where possible, exposure assumptions were obtained from the ADEC Guidance on Cleanup Levels Equations and Input Parameters (ADEC 2001).

TECH MEMO_ACL_FINAL_UPDATE.DOC 9

For site-specific assumptions, best judgement was used. For each of the exposure scenarios, the only exposure routes considered include incidental ingestion and inhalation of volatiles from ambient air. In accordance with ADEC guidance, the ingestion of soil pathway is considered protective of exposure from inhalation of fugitive dust and the dermal contact pathway, and therefore were not considered in this evaluation. The exposure assumptions for each potential pathway are summarized as follows:

• For future remote lighthouse worker/residents, a soil ingestion rate of 100 mg/day for adults and 200 mg/day for children, an exposure frequency of 210 days per year, over a 25 year duration (6 years for a child plus 19 years for an adult) is assumed. The exposure frequency selected is based on an average rainfall of greater than 40 inches per year.

• For seasonal visitors, a soil ingestion rate of 100 mg/day for adults and 200 mg/day for children, an exposure frequency of 90 days per year, over a six year exposure duration is assumed.

Risk-based concentrations for each exposure scenario are based on a target excess lifetime cancer risk of 1x10-5 for carcinogens and a hazard quotient of 1, per ADEC guidance.

Toxicity Values

When available, the toxicity values for constituents detected at the light stations were obtained from the ADEC Guidance on Cleanup Levels Equations and Input Parameters (2001). If a toxicity value was not available through ADEC, then it was obtained from one of the following sources listed below:

• The Integrated Risk Information System (IRIS), a database available through the EPA National Center for Environmental Assessment (NCEA). IRIS, prepared and maintained by EPA, is an electronic database containing health risk and EPA regulatory information on specific chemicals (EPA, 2002).

• The Health Effects Assessment Summary Tables (HEAST), provided by the EPA Office of Solid Waste and Emergency Response (EPA, 1997) is a compilation of toxicity values published in various health effects documents issued by EPA.

• The U.S. EPA Region IX Preliminary Remediation Goal Table (October, 2002) at www.epa.gov/docs/region09/waste/sfund/prg/index.html

Toxicity values (i.e., cancer slope factors, unit risk factor, oral reference doses, and reference concentrations) used to calculate the soil ACLs are presented in Tables 3 and 4.

TECH MEMO_ACL_Final_update.doc 10

Table 2

Summary of Exposure Assumptions for Select Exposure Scenarios

US Coast Guard Light Stations

Parameter Symbol Units Remote Lighthouse Worker/Resident Scenario

Source Child Seasonal User *

Source Adult Seasonal User

Source

Target Risk TR unitless 1 x 10-5 -- 1 x 10-5 -- 1 x 10-5 -- Target Hazard Quotient THQ unitless 1 -- 1 -- 1 -- Oral Reference Dose RfDo mg/kg-day chemical specific a chemical specific a chemical specific a Oral Slope Factor SFo kg-day/mg chemical specific a chemical specific a chemical specific a Reference Concentration RfC mg/m3 chemical specific a -- -- -- -- Unit Risk Factor URF ug/m3 chemical specific a -- -- -- -- Inhalation Reference Dose RfDi mg/kg-day -- -- chemical specific a chemical specific a Inhalation Slope Factor SFi kg-day/mg -- -- chemical specific a chemical specific a Body Weight - Adult Bwa kg 70 b -- -- -- -- Body Weight - Child BWc kg 15 b 15 b 70 b Carcinogenic Averaging Time ATC yrs 70 b 70 b 70 b Noncarcinogenic Averaging Time ATN yrs 6 b 6 b 6 b Exposure Frequency EF day/yr 210 c 90 c 90 c Exposure Duration - Adult EDa yrs 19 d -- b 6 b Exposure Duration - Child EDc yrs 6 b 6 Incidental Soil Ingestion Rate - Adult IRa mg/day 100 b -- b 100 b Incidental Soil Ingestion Rate - Child IRc mg/day 200 b 200 -- Age-adjusted soil ingestion factor Iradj mg/year/kg-day 107 -- -- -- Inhalation Rate Inh m3/day 20 e 10 e 20 e Volatilization Factor VF m3/kg chemical-specific a chemical-specific a chemical-specific a

* Note: A child is defined as a person under the age of 7 when the risks from exposure to lead are the greatest.

a. See Tables A-2 and A-3

b. ADEC. Cleanup Level Guidance, August 17, 2001.

c. Professional judgement; based on seasonal accessability of the light station.

d. Based on a total work tenure of 25 years for a lighthouse worker.

e. USEPA Exposure Factors Handbook Volume I Gneral Factors. EPA/600/P-95/002Fa. EPA, 1997.

TECH MEMO_ACL_FINAL_UPDATE.DOC 11

Table 3 Summary of Toxicity Values for Non-Lead Constituents US Coast Guard Light Stations Constituent Name SFo

(mg/kg-day)-1

Source URF (ug/m3)

Source Chronic RfDo (mg/kg-day)

Source RfC (mg/m3)

Source SFi (mg/kg-day)-1

Source Chronic RfDi (mg/kg-day)

Source Subchronic RfDo (mg/kg-day)

Source Subchronic RfDi (mg/kg-day)

Source Volatilization Factor (m3/kg)

1,2,3-Trichlorobenzene -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 1,2,4-Trichlorobenzene -- -- -- -- 0.010 a 0.20 a -- -- 0.057 d 0.010 0.57 d 21,421 1,2,4-Trimethylbenzene -- -- -- -- 0.050 c 0.0060 c -- -- 0.0017 c -- -- -- -- 1,3,5-Trimethylbenzene -- -- -- -- 0.050 c 0.0060 c -- -- 0.0017 c -- -- -- -- 2-Butanone -- -- -- -- 0.60 b 1.0 b -- -- 0.29 b 2.0 d 0.29 d 10,020 4-Isopropyltoluene -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Acenaphthylene -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 65,791 Acetone -- -- -- -- 0.10 a -- -- -- -- -- -- 1.0 d -- -- 14,677 Anthracene -- -- -- -- 0.30 a -- -- -- -- -- -- 3.0 d -- -- 378,834 Arsenic 1.5 a 0.0043 a 3.00E-04 a -- -- 15 b -- -- 3.00E-04 d -- -- -- Barium -- -- -- -- 0.070 a 5.00E-04 a -- -- 1.43E-04 d 0.070 d -- -- -- Benzene 0.055 a 8.30E-06 a 0.0030 c -- -- 0.029 b 0.0017 c -- -- -- -- 2,066 Benzo(a)anthracene 0.73 a -- -- -- -- -- -- -- -- -- -- -- -- -- -- 4.58E+06 Benzo(a)pyrene 7.3 a -- -- -- -- -- -- -- -- -- -- -- -- -- -- 1.19E+07 Benzo(b)fluoranthene 0.73 a -- -- -- -- -- -- -- -- -- -- -- -- 2.24E+06 Benzo(g,h,i)perylene -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Benzo(k)fluoranthene 0.073 a -- -- -- -- -- -- -- -- -- -- -- -- -- -- 1.91E+07 Cadmium -- -- 0.0018 a 5.00E-04 b -- -- 6.3 b -- -- -- -- -- -- -- Chromium -- -- 0.012 a 0.0030 a 1.00E-04 a 42 b 2.20E-06 b 0.020 d -- Chrysene 0.0073 a -- -- -- -- -- -- -- -- -- -- -- -- -- -- 1.32E+06 Dibenzo(a,h)anthracene 7.3 a -- -- -- -- -- -- -- -- -- -- -- -- -- -- 5.05E+07 Diesel Range Organics -- -- -- -- -- Ethylbenzene -- -- 1.10E-06 b 0.10 a 1.0 a 0.0039 b 0.29 b -- -- -- -- 3,028 Fluoranthene -- -- -- -- 0.040 a -- -- -- -- -- -- 0.40 d -- -- 1.48E+06 Fluorene -- -- -- -- 0.040 a -- -- -- -- -- -- 0.40 d -- -- 248,159 Gasoline Range Organics -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Indeno(1,2,3-cd)pyrene 0.73 a -- -- -- -- -- -- -- -- -- -- -- 2.77E+07 isopropyl alcohol -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 26,384 Isopropylbenzene -- -- -- 0.10 b 0.39 b -- 0.11 b 0.0090 0.0026 d 5,501 m&p Xylenes -- -- -- -- 0.70 b 0.10 b -- -- 0.029 b -- -- -- -- 3,783 Mercury -- -- -- -- 3.00E-04 b -- -- -- -- -- -- 0.0030 d -- -- -- Methylene Chloride 0.0075 a 4.70E-07 a 0.060 a 3.0 a 0.0016 b 0.86 b 0.060 d 0.86 2,452 Naphthalene -- -- -- -- 0.020 a 0.0030 a -- -- 8.57E-04 b -- -- -- -- 27,619 n-Propylbenzene -- -- -- -- 0.040 c -- -- -- -- -- -- -- -- -- -- -- O-Xylene -- -- -- -- 0.70 b 0.10 b -- -- 0.029 b -- -- -- -- 3,783 Phenanthrene -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Pyrene -- -- -- -- 0.030 a -- -- -- -- -- -- 0.30 d -- -- 1.84E+06 Residual Range Organics -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --tert-Butyl Methyl Ether -- -- -- -- -- -- 3.0 b -- -- 0.86 b -- -- 4,582 Toluene -- -- -- -- 0.20 a 0.40 a -- -- 0.11 d 2.0 d 2,444 Total Lead -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Trichlorotrifluoroethane -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Xylenes, Total -- -- -- -- 0.70 b 0.10 b -- -- 0.029 b -- -- -- -- 3,783

a. ADEC. Cleanup Levels Guidance. Appendix C - Chemical Specific Parameters - Carcinogens and Noncarcinogens. August 17, 2001 b EPA. The Integrated Risk Information System (IRIS, 2002), a database available through the EPA National Center for Environmental Assessment (NCEA). http://www.epa.gov/iris/ c EPA Region IX Preliminary Remediation Goal Toxicity Factor Table (October, 2002).

d. HEAST. Health Effects Assessment Summary Tables, U.S. EPA, 1997

TECH MEMO_ACL_Final_update.doc 12

Table 4 Summary of Toxicity Values for TPH Fractions US Coast Guard Light Stations

Chemical Name RfDo (mg/kg-day) Source RfC (mg/m3) Source

C6-C10 Aliphatics 5.00E+00 a 1.84E+01 a C6-C10 Aromatics 2.00E-01 a 4.00E-01 a C10 - C25 Aliphatics 1.00E-01 a 1.00E+00 a C10 - C25 Aromatics 4.00E-02 a 2.00E-01 a C25 - C36 Aliphatics 2.00E+00 a -- -- C25 - C36 Aromatics 3.00E-02 a -- -- Notes:

-- = not applicable or not available

a. ADEC. Guidance on Cleanup Levels Equations and Input Parmeters, July 28, 1999.

Results for Non-Lead Constituents

ACLs were developed for a total of 35 chemical constituents that have been detected in soil at least once at a USCG light station. Table 5 provides a summary of the ACLs for each of these constituents, for each of the exposure scenarios evaluated. These ACL values will be compared with station-specific data in a separate deliverable, to support risk management decisions for each individual light station.3

Soil ACLs for Lead Lead has been detected in soils at every light station at concentrations above background levels, due to the prevalent use of lead-based exterior paint. The ACLs for lead were derived for each of the three exposure scenarios (residential, industrial/commercial, and seasonal visitor). Due to the different methodologies associated with evaluation of children and adults, each is assess separately, as described in the following subsections.

ACL Method for Remote Lighthouse Worker/Resident Scenario

For addressing exposures to potential future remote lighthouse worker/resident, the most sensitive individual would be expected to be a child included as part of the worker/resident family. ACLs for child receptors were derived using the U. S. Environmental Protection Agency (USEPA) integrated exposure uptake biokinetic (IEUBK) model (Version 1.0, build 252). The IEUBK model predicts blood lead levels in children up to 7 years of age and the percentage of children with blood lead levels above 10 micrograms per deciliter (μg/dL), resulting from cumulative exposure to lead from air, diet, water, and dust/soil. The soil ACL for lead was calculated as the concentration resulting in a predicted blood-lead level of 10 μg/dL in greater than 5 percent of the potentially exposed population. All of the default model assumptions and parameters were used, with the exception of soil/dust bioavailability. A physiologically-relevant extraction was used to estimate the site-specific bioavailability of lead in surface soil.

The results of the bioavailability evaluation are provided as a separate technical memorandum in Appendix A.

3 In March 2004, the ACLs for DRO, GRO, and RRO were updated. During derivation of the original ACLs, in lieu of actual information on TPH composition, it was conservatively assumed that all petroleum fractions were aromatics. We have now adjusted the ACLs to match the ADEC default percent fraction of aromatic/aliphatic for each petroleum group.

The adjusted ACLs for these compounds are shown in Table 5.

TECH MEMO_ACL_FINAL_UPDATE.DOC 13

Table 5 Summary of Alternate Cleanup Levels for Non-Lead Constituents US Coast Guard Light Stations Constituent Name Units Remote Lighthouse

Worker/Resident Child Seasonal Visitor Adult Seasonal Visitor

Arsenic mg/kg 7.6 24 221 Barium mg/kg 9,125 21,292 198,722 Cadmium mg/kg 130 152 2,839 Chromium mg/kg 391 6,083 8,517 Mercury mg/kg 39 913 852 Benzene mg/kg 12 54 127 Ethylbenzene mg/kg 5,264 5,343 12,466 Toluene mg/kg 1,699 1,636 3,816 Xylenes, Total mg/kg 260,714 212,917 >10+6

Gasoline Range Organics mg/kg 9,044 24,057 56,133 Diesel Range Organics mg/kg 7,242 16,898 157,716 Residual Range Organics mg/kg 12,474 29,107 271,664

1,2,3-Trichlorobenzene mg/kg -- -- -- 1,2,4-Trichlorobenzene mg/kg 1,304 3,042 17,331 1,2,4-Trimethylbenzene mg/kg 6,518 15,208 141,944 1,3,5-Trimethylbenzene mg/kg 6,518 15,208 141,944 2-Butanone mg/kg 17,415 17,415 40,635 4-Isopropyltoluene mg/kg -- -- -- Acenaphthylene mg/kg -- -- -- Acetone mg/kg 13,036 304,167 283,889 Anthracene mg/kg 39,107 912,500 851,667 Benzo(a)anthracene mg/kg 16 49 454 Benzo(a)pyrene mg/kg 1.6 4.9 45 Benzo(b)fluoranthene mg/kg 16 49 454 Benzo(g,h,i)perylene mg/kg -- -- -- Benzo(k)fluoranthene mg/kg 156 486 4,537 Chrysene mg/kg 1,556 4,861 45,370 Dibenzo(a,h)anthracene mg/kg 1.6 4.9 45 Fluoranthene mg/kg 5,214 121,667 113,556 Fluorene mg/kg 5,214 121,667 113,556 Indeno(1,2,3-cd)pyrene mg/kg 16 49 454 isopropyl alcohol mg/kg -- -- -- Isopropylbenzene mg/kg 3,681 86 8,590 m&p Xylenes mg/kg 260,714 212,917 >10+6

Methylene Chloride mg/kg 254 1,058 2,469 Naphthalene mg/kg 144 144 336 n-Propylbenzene mg/kg 1,304 12,167 28,389 O-Xylene mg/kg 260,714 212,917 >10+6

Phenanthrene mg/kg -- -- -- Pyrene mg/kg 3,911 91,250 85,167 tert-Butyl Methyl Ether mg/kg 23,890 23,890 55,743 Trichlorotrifluoroethane mg/kg -- -- -- -- No toxicity factor exists for these detected constituents.

TECH MEMO_ACL_FINAL_UPDATE.DOC 14

Potential future remote lighthouse worker/residents are assumed to be present at the light station for 7 months (210 days, see Table 2) per year. However, the IEUBK model does not accommodate adjustment of exposure frequency, but assumes continuous long-term exposure to lead from both onsite and offsite sources. To overcome this limitation in the model, a time-weighted average risk-based soil lead concentration was calculated using a three-step process.

• First, a target risk-based soil lead concentration (TCoffsite = 341 mg/kg) was computed assuming that an individual was exposed only to offsite conditions. Under this step, all default model assumptions were used.

• Second, a target risk-based soil lead concentration (TConsite = 822 mg/kg) was computed assuming that an individual was exposed only to onsite conditions. For this computation, all of the default model assumptions and parameters were used, with the exception of soil/dust bioavailability, which was replaced with the estimated site-specific value of 12.6 percent.

• Third, using the two target risk-based soil lead concentrations previously calculated, a time-weighted average target soil lead concentration (TWC = 618 mg/kg) was computed (including both onsite and offsite exposure) using the following equation:

/365

/155()* /365

/210( OffsiteOnsite TC yrdays offsiteyrdaysTC yrdays onsiteyrdaysTWC +=

This concentration represents the continuous arithmetic average soil lead concentration resulting in a predicted blood-lead level of 10 μg/dL in greater than 5 percent of the potentially exposed population. Since offsite background soil is significantly lower than 618 mg/kg, the average onsite concentration can be significantly greater than 618 mg/kg in order to retain a long-term average (both onsite and offsite) of 618 mg/kg. Therefore, the final ACL for onsite lead in soil is calculated using the following equation:

/365

/210(

/365

/155([ yrdays onsiteyrdays

C yrdays offsiteyrdaysTWC

ACL

Bkgd−

The background soil lead concentration (Cbkgd) was assumed to equal naturally-occurring levels identified for twelve regions in southeastern Alaska (USGS, 1984). The USGS reported soil concentrations from 21 samples collected from these regions averaged 13.2 mg/kg.

ACL Method for Seasonal Visitor Scenario

Seasonal visitors include both adults and children that may visit a light station during the Summer months, a period when these light stations are most accessible by boat. For addressing exposures to potential future site visitors, ACLs for seasonal visitor children were developed using the IEUBK model, and ACLs for seasonal visitor adults were developed using the ALM, as described below.

Children Seasonal Visitors. Potential child seasonal visitors are assumed to be present at the light station for 3 months (90 days, see Table 2) per year. A 90-day period is also considered by EPA to be the shortest period under which their lead models most reliably predict steady-state

TECH MEMO_ACL_FINAL_UPDATE.DOC 15

blood lead levels. For seasonal visitor children, the same three-step process described above for the lighthouse worker/residents was used to calculate a time-weighted average risk-based soil lead concentration to accommodate adjustment of exposure frequency for the IEUBK model.

The first two steps are the same as previously described. The third step was conducted as follows:

• Using the onsite (TConsite = 822 mg/kg) and offsite (TCoffsite = 341 mg/kg) target risk-based soil lead concentrations previously calculated, a time-weighted average target soil lead concentration (TWC = 460 mg/kg) was computed (including both onsite and offsite exposure) using the following equation:

/365

/275()* /365

/90( OffsiteOnsite TC yrdays offsiteyrdaysTC yrdays onsiteyrdaysTWC +=

This concentration represents the continuous arithmetic average soil lead concentration resulting in a predicted blood-lead level of 10 μg/dL in greater than 5 percent of the potentially exposed population. Since offsite background soil (Cbkgd = 13.2 mg/kg) is significantly lower than 460 mg/kg, the average onsite concentration can be significantly greater than 460 mg/kg in order to retain a long-term average (both onsite and offsite) of 460 mg/kg. Therefore, the final ACL for onsite lead in soil is calculated using the following equation:

/365

/90(

/365

/275([ yrdays onsiteyrdays

C yrdays offsiteyrdaysTWC

ACL

Bkgd−

Adult Seasonal Visitors. For seasonal visitor adults, USEPA’s Adult Lead Model (ALM) was used to calculate ACLs. The ALM guidance and spreadsheet are available from the USEPA website at: http://www.epa.gov/superfund/programs/lead/index.htm. This guidance presents a method for developing a risk-based concentration for lead that is protective of fetuses carried by women who may be exposed to lead in soil. The ALM uses a biokinetic slope factor (BKSF) to represent lead biokinetics and a relatively simple exposure model in which all exposure pathways, except the soil ingestion pathway, are represented by a baseline blood lead concentration (PbBadult,0). The baseline, or background blood lead level represents the typical blood lead concentration in women of child-bearing age in the absence of exposures at the site being assessed. Developing fetuses of adult women workers of child-bearing age that have site exposures to lead are considered the focus for ACL development in potential future workers at the light stations.

The soil lead ACL was calculated using the following equation:

EFAFIRBKSF

ATPbBPbB

ACL

s adultgoalcentraladult lead ×××

= 0,,, Where:

ACL = alternative cleanup level for lead (mg/kg); the soil lead concentration that would be estimated to result in a central tendency blood lead concentration in adults at the site (i.e., women of child-bearing age) (PbBadult,central,goal) and the

TECH MEMO_ACL_FINAL_UPDATE.DOC 16

corresponding 95th percentile fetal blood lead concentration (PbBfetal,0.95,goal)

PbBadult,central,goal = goal for central estimate of blood-lead concentration (μg/dL) in adults (i.e., women of child-bearing age) that have site exposures;

the goal is intended to ensure that PbBfetal,0.95,goal does not exceed 10 μg/dL

PbBadult,0 = typical blood lead concentration (1.40 μg/dL) in women of child-bearing age at the site in the absence of exposures to the site;

baseline blood-lead level is based on data for the western quadrant of the U.S. (which includes Alaska) (USEPA, 2002).

AT = averaging time; the total number of days per year over which soil exposure is averaged (365 days/yr)

BKSF = ratio of (quasi-steady state) increase in typical adult blood lead concentration to average daily lead uptake ( 0.4 μg/dL PbB increase per μg/day lead uptake)

IRs = intake rate of soil, includes both outdoor soil and indoor soil-derived dust (50 mg/d)

AFs = absolute gastrointestinal absorption fraction for ingested lead in soil and lead in dust derived from soil (default = 0.12)

EFs = exposure frequency for contact with assessed soils and/or dust derived in part from these soils (default = 219 d/yr for worker;

site-specific = 90 d/yr for adult seasonal visitor)

This equation describes the empirical relationship between lead intake (from site soil and other non-site related sources) and the blood lead concentration in adult pregnant women. As part of the relationship described by this equation, the term PbBadult, central, goal defines the relationship between the maternal blood lead concentration and the corresponding 95th percentile fetal blood lead concentration, as follows:

materialfetaladulti goalfetal goalcentraladult RGSD

PbB PbB

645.1

,95.0, Where:

PbBfetal,0.95,goal = goal for the 95th percentile blood lead concentration (μg/dL) among fetuses of women having exposures to the specified site soil concentration; this is interpreted to mean that there is a 95 percent likelihood that a fetus, in a woman who experiences such exposures, would have a blood lead concentration no greater than PbBfetal,0.95,goal (i.e. the likelihood of a blood lead concentration greater than 10 μg/dL would be less than 5 percent using the

ALM)

GSDi,adult = estimated value of the individual geometric standard deviation (2.11); the GSD among adults based on data for the western

TECH MEMO_ACL_FINAL_UPDATE.DOC 17

quadrant of the U.S. (which includes Alaska) (USEPA, 2002).

Rfetal/maternal = constant of proportionality (0.9) between fetal PbB concentration at birth and maternal PbB concentration

All of the default model assumptions and parameters were used, with the exception of the baseline blood lead (PbBadult,0 ) and the geometric standard deviation (GSDi,adult), which were based on data for the western quadrant of the U.S. (which includes Alaska) from the NHANES III study (USEPA, 2002), and the exposure frequency, which was reduced to 90 days per year (three Summer months). Since the default bioavailability in the ALM (12%) is about equal to the estimated site-specific level, no bioavailability adjustment was used for the adult seasonal visitor exposure scenario.

Results for Lead

The ACLs estimated for lead using the methods described in the previous subsections are summarized in Table 6.

TABLE 6

ACLSa FOR LEAD IN SOIL (MG/KG)

Exposure Scenario

Without Site-Specific Bioavailability Adjustment

With Site-Specific Bioavailability Adjustment

Remote Lighthouse Worker/Resident

583 1,064

Seasonal Visitor (Adult) 3,132b 3,132b,c

Seasonal Visitor (Child) 1,343 1,824 a ACLs are a time-weighted average that accounts for time offsite and onsite (see text).

b Includes GM and GSD blood lead levels for West Region based on NHANES III (EPA 2002).

c The default GI bioavailability in the ALM (12%) is about equal to the estimated site-specific level, so no bioavailability adjustment was used for this case.

TECH MEMO_ACL_FINAL_UPDATE.DOC 18

References

USGS 1984. Chemical Analyses of Soils and Other Surficial Materials, Alaska. L.P. Gough, J.L.

Peard, R.C., Severson, H.T. Shacklette, M.L. Tompkins, K.C. Stewart, and P.H. Briggs. U.S.

Geological Survey Open-File Report 84–423.

USEPA, 2002. Blood Concentrations of U.S. Adult Females: Summary Statistics from Phases 1 and 2 of the National Health and Nutrition Evaluation Survey (NHANES III), Technical Review Workgroup for Lead, March.

TECH MEMO_ACL_FINAL_UPDATE.DOC 19

Appendix A – Bioavailability Evaluation Memo

LEAD EXTRACTIONS MEMO_DS.DOC 1

M E M O R A N D U M

Data Report for Bioavailability Evaluation of Soil at USCG Southeast Alaska Light Stations TO: Mark Ridgway/USCG Alaska COPIES: Jane Gendron/CH2M HILL Juneau FROM: Dennis Shelton/CH2M HILL Corvallis DATE: September 6, 2002

This memorandum provides the results of the lead extraction tests conducted to estimate the bioavailable fraction of lead in soil present at Coast Guard SE Alaska Light Stations.

Thirteen soil samples collected at the Tree Point Light Station on May 2002 were extracted to provide a representative assessment of bioavailability of lead-based paint used for the SE Alaska Light Stations. The rationale, methods, and results are provided in the following sections.

1.0 Rationale

Oral bioavailability is a measure of the amount of a constituent that is absorbed into the body after ingestion exposure. Some constituents are absorbed almost completely (100 percent bioavailability) when ingested in their pure, soluble form. Others may pass through the body largely unabsorbed. Oral bioavailability of soil-bound constituents largely depends on the rate at which constituents dissociate or solubilize from the soil matrix in the gastrointestinal (GI) tract. Soil-bound constituents are usually absorbed by the GI tract to a lesser degree than constituents in their pure, soluble form (Paustenbach, 1987). This reduced absorption results from the affinity between the constituent and soil matrix, the low solubility of the constituent form associated with the soil, or both. Thus, the bioavailability of lead from soil is expected to be low for constituents that are tightly bound within the soil matrix and/or are in a form that is insoluble in the GI tract under physiological conditions.

A physiologically-relevant extraction was used to estimate the site-specific bioavailability of lead in surface soil collected from the Tree Point Light Station. The protocol for this study (described below) has been modified from the physiologically relevant extraction procedure (PREP) (BC Environment, 1993) and the physiologically-based extraction test (PBET) (Ruby et al., 1993 and 1996).

The protocol used is expected to yield a conservative screening estimate of the bioavailable fraction of inorganic compounds from site soils because it is assumed that the total portion of acid-extractable compound would be absorbed in the GI tract. Under real physiological conditions, a portion of the solubilized portion might not be absorbed into the bloodstream, for the following reasons:

• It could bind to food in the GI tract.

DATA REPORT FOR BIOAVAILABILITY EVALUATION OF SOIL AT USCG SOUTHEAST ALASKA LIGHT STATIONS

LEAD EXTRACTIONS MEMO_DS.DOC 2

• Gastric contents entering the intestine are subject to a pH increase to about 7.0, with consequent precipitation of lead to a less bioavailable form.

• Some absorption mechanisms are saturable.

As a result, the bioavailable fraction identified by the method described here is to be viewed as a rough but conservative estimation of the amount contributing to potential exposure from lead in light station soil.

2.0 Physiologically-Based Extraction Methodology

Sample Preparation and Extraction Soil samples for analysis were selected to represent a relatively large range of total lead concentrations. Soil samples were dried at 105 °C and then sieved prior to analysis. The fraction passing through a 250 µm sieve was utilized for the physiologically-based extraction. Soil sample aliquots were added to plastic centrifuge tubes and extracted for 1 hour at 37°C with 0.1 molar hydrochloric acid (HCl) buffered to pH 2.0 with phosphate buffer at a 30:1 ratio of extraction fluid to soil. The acid concentration and pH were selected to simulate stomach conditions. Following extraction, the samples were centrifuged to remove suspended solids. The supernatant was isolated for analysis and preserved by adding HNO3. Each sieved soil sample was also analyzed for total recoverable lead (using the conventional digestion procedure) to identify the total amount of lead in the sample.

The percentage of extractable lead (relative to the total lead) was calculated as the estimated bioavailable fraction.

To provide a positive control, a spiked sample was prepared by adding solid lead carbonate (PbCO3) powder to a sieved portion of soil. The sample for spiking (i.e., the soil blank) was selected based on the low levels of lead it contained relative to the other samples. After addition of the PbCO3 powder, the sample was well mixed using a vortex mixer prior to removing aliquots for analysis. Lead carbonate (known as “white lead”) is the white pigment extensively used as a pigment in white oil based paints in the first part of the 20th century.

Chemical Analysis The metal extracts were analyzed by ICP-OES following EPA method SW6010B.

3.0 Summary of Results

Table 1 provides a summary of the extraction results for the light station soil samples. The raw data package is provided as an attachment to this memorandum. The results indicate that the average extractable fraction for the thirteen samples was 13 percent. Consistent with these results, the results for the positive control (PbCO3) indicate that this compound also has relatively low extractability (and bioavailability). Duplicate soil extractions for three samples demonstrated excellent reproducibility in the procedure.

These results will be considered when developing alternate cleanup levels (ACLs) for lead at the Coast Guard SE Alaska Light Stations. The average soil bioavailability estimate of 13 percent is lower than the default bioavailable fraction of 30 percent included in USEPA’s

DATA REPORT FOR BIOAVAILABILITY EVALUATION OF SOIL AT USCG SOUTHEAST ALASKA LIGHT STATIONS

LEAD EXTRACTIONS MEMO_DS.DOC 3

Integrated Exposure Uptake Biokinetic (IEUBK) model used to estimate risk from exposure to lead in soil.

4.0 References

BC Environment. Development of a Physiologically Relevant Extraction Procedure. CB Research

International Corporation. Prepared for BC Ministry of Environment, Lands and Parks Environmental Protection Division. April 28, 1993.

Paustenbach, D.J. Assessing the Potential Environmental and Human Health Risks of Contaminated Soil. Comments Toxicol. 1(3-4):185-226. 1987.

Ruby, M.V., A. Davis, T.E. Link, R. Schoof, R.L. Chaney, G.B. Freeman, and P. Bergstrom.

Development of an in Vitro Screening Test to Evaluate the in Vivo Bioaccessibility of Ingested Mine-Waste Lead. Environmental Science and Technology 27(13):2870-2877.

1993.

Ruby, M.V., A. Davis, R. Schoof, S. Eberle, and C.M. Sellstone. Estimation of Lead and Arsenic Bioavailability Using a Physiologically Based Extraction Test. Environmental Science and Technology 30(2):422-430. 1996.

TECH MEMO_ACL_FINAL_UPDATE.DOC 20

Appendix B – IEUBK and ALM Model Output