K_GuamDPRI_PesticideSoilsManagementPlan_Nov2011.pdf

FINAL

PESTICIDE SOILS MANAGEMENT PLAN

FOR THE GUAM DPRI PROGRAM

Prepared for

NAVFAC PACIFIC

Honolulu, Hawaii

November 2011

Prepared by

PARSONS

1132 BISHOP STREET, SUITE 901 • HONOLULU • HAWAII 96813

Pesticide Soils Management Plan DPRI Program - Guam

TABLE OF CONTENTS

TABLE OF CONTENTS

1.0 INTRODUCTION AND BACKGROUND

2.0 TIER IB SOIL SCREENING LEVELS

3.0 SAMPLING AND REPORTING REQUIREMENTS

4.0 PESTICIDE-IMPACTED SOILS MANAGEMENT – DEMOLITION AND

CONSTRUCTION ACTIVITIES

5.0 PESTICIDE-IMPACTED SOILS MANAGEMENT – OPERATIONS AND

MAINTENANCE OF HOUSING AREAS

6.0 REFERENCES

LIST OF TABLES

TABLE 1....................................................................................................... Pesticides of Concern TABLE 2............................. Pesticide Partioning Coefficients and Groundwater Screening Levels TABLE 3a............................................................ Tier IB Soil Screening Levels – Resident (Child) TABLE 3b................................................................ Tier IB Soil Screening Levels – BEQ Housing TABLE 3c...................................................... Tier IB Soil Screening Levels – Recreational (Child) TABLE 3d.....................................Tier IB Soil Screening Levels – Commerical/Industrial Workers TABLE 3e................................................... Tier IB Soil Screening Levels – Construction Workers TABLE 4.................................................... Site Checklist for Use of Tier IB Soil Screening Levels TABLE 5............................................... Stage 2 Sampling – Total Number of Buildings to Sample TABLE 6.................. Target Compound List and Maximum Allowable Limits of Quantitation (Soil) TABLE 7.......................Target Compound List and Maximum Allowable Detection Limits (Water)

LIST OF FIGURES

FIGURE 1 ........................................... Generalized Sampling Locations and Sample Composites

LIST OF APPENDICES

APPENDIX A ....................Detailed Calculations for Human Health Tier IB Soil Screening Levels APPENDIX B ............................................................. Modeling of Potential Groundwater Impacts

1.0 INTRODUCTION AND BACKGROUND

This Pesticide Soils Management Plan (Plan) provides guidance on testing and management of pesticide-impacted soils at Department of Defense (DoD) facilities on Guam under the Defense Policy Review Initiative (DPRI) Program. This Plan provides soil screening levels based on planned land use; sampling, analytical, data analysis, and reporting procedures; and, guidelines for the safe handling and management of pesticide-impacted soils to minimize exposure to pesticide-impacted soils by future residents, commercial/industrial workers, construction workers, and maintenance workers.

This Plan is intended for use at DoD housing construction and redevelopment sites in Guam that are not being performed under Environmental Restoration, Navy (ER,N) or Base Realignment and Closure (BRAC).

This Plan contains 6 sections and 2 appendices. Section 2 provides the methodology used and the development of Tier IB Soil Screening Levels for the pesticides of concern. Section 3 discusses required soil sampling and data analysis procedures. Section 4 includes the requirements for management of pesticide-impacted soils during demolition and construction activities. Section 5 includes the requirements for management of pesticide-impacted soils during ongoing maintenance and operations. Section 6 lists references cited in this Plan.

Appendix A provides detailed calculations used in the development of the Tier IB Soil Screening Levels for Human Health. Appendix B provides details used in the evaluation of potential groundwater impacts.

1.1 BACKGROUND INFORMATION ON PESTICIDES OF CONCERN

Chlorinated pesticides are synthetic compounds that in the past were used as insecticides on food crops and as termiticides in and around buildings and homes. Since the mid- to late 1980’s, the use and commercial production of certain chlorinated pesticides (e.g., notably chlordane, heptachlor, aldrin, dieldrin, and endrin) have been prohibited in the United States and many other countries. However, residues from these and other chlorinated pesticides may still be present beneath and around structures that were treated prior to the mid- to late 1980’s. The following subsections describe the ten (10) organochlorine pesticides that may have been routinely applied to control termites and other pests in Guam. General information for these pesticides is summarized from information published by the U.S. Department of Health and Human Services (ATSDR, 2006).

This Plan addresses residual concentrations of these 10 pesticides in soil based on the historical, legally-applied use of these pesticides in Guam, along with Conceptual Site Model (CSM) and sampling results from previous pesticide investigations conducted in Guam and Hawaii, as discussed and summarized in Section 1.2.

1.1.1 Chlordane

Chlordane is a persistent organochlorine pesticide that was used as a pesticide in the United States (U.S.) from 1948 to 1988. Chlordane was historically applied under and around building foundations to protect the buildings from termite infestation. From 1983 to 1988, the only approved use of chlordane was to control termites in homes. Most exposures to chlordane are through inhalation, ingestion or absorption through the skin. Swallowing small amounts of chlordane or breathing air containing high concentrations of chlordane vapors can cause headaches, irritation, confusion, weakness, and vision problems. Based on animal studies, the

U.S. Environmental Protection Agency (USEPA) has determined that chlordane is a probable human carcinogen. Exposure to high enough amounts can cause adverse effects to the liver.

The specific formulation of chlordane used is known as “technical chlordane”, a mixture of pure chlordane and related chemicals, primarily heptachlor and heptachlor epoxide.

1.1.2 Heptachlor and Heptachlor Epoxide

Heptachlor is a persistent organochlorine pesticide that is structurally similar to chlordane and was used in the formulation of “technical chlordane”. Heptachlor was used until 1988 in the formulation of technical chlordane, and as an insecticide in homes, buildings, and on food crops.

Heptachlor is currently EPA-registered only to control fire ants in power transformers.

Heptachlor epoxide is an oxidation product of chlordane and heptachlor, but not produced commercially in the U.S. Bacteria and animals break down heptachlor to form heptachlor epoxide.

The most common sources of exposure to heptachlor and heptachlor epoxide are through inhalation, ingestion, and absorption through the skin. Based on animal studies, the USEPA has determined that heptachlor and heptachlor epoxide are probable human carcinogens.

Exposure to high enough amounts can cause adverse effects to the liver.

1.1.3 Aldrin and Dieldrin

Aldrin and dieldrin are the common names of two structurally similar compounds that were once used as insecticides and termiticides. From the 1950s to the 1970s, aldrin and dieldrin were used as insecticides on crops such as corn and cotton. From 1972 through 1987, aldrin and dieldrin were used to control termites. Aldrin readily breaks down into dieldrin once it enters the environment or the human body.

Most exposures to aldrin and dieldrin are through inhalation, ingestion or dermal exposure to impacted soil. Exposure to moderate levels of aldrin/dieldrin for a long time causes headaches, dizziness, irritability, vomiting, or uncontrollable muscle movements. Based on animal studies, the USEPA has determined that aldrin and dieldrin are probable human carcinogens. Exposure to high enough amounts can cause adverse effects to the liver.

1.1.4 Endrin

Endrin is a stereoisomer of dieldrin used to control insects, rodents, birds, and other pests on sugarcane, tobacco, and apple orchards. It has many similar chemical properties to aldrin and dieldrin, but was never used to control termites. Endrin has not been produced or sold for general use in the U.S. since 1986. Endrin ketone is a breakdown product of endrin when endrin is exposed to light. Endrin aldehyde is an impurity and a breakdown product of endrin.

Exposure to endrin mostly occurs through incidental ingestion, inhalation, or dermal exposure to impacted soil and eating contaminated food. Endrin has not been classified by the USEPA as a carcinogen, but exposure to high enough amounts can cause non-cancer health effects, such as kidney or liver damage.

1.1.5 DDT, DDD, and DDE

DDT is a persistent organochlorine pesticide once widely used to control insects and insects that carry diseases such as malaria and typhus. Its use in the United States was banned in 1972. DDD was also used to kill pests and its use has also been banned. DDD and DDE also can enter the environment from technical formulations of DDT and DDD or as breakdown products of DDT and DDD. Although these 3 chemicals breakdown in air rapidly, they adhere to soil strongly. DDT in soil is broken down slowly to DDE and DDD by microorganisms in 2 to 15 years, depending on the soil type.

Exposure to DDT, DDD and DDE is through ingestion and inhalation, may result in tremors and seizures, and may affect the nervous and reproductive systems. The USEPA has determined that DDT, DDD, and DDE are probable human carcinogens based on animal studies. Exposure to high enough amounts can cause adverse effects to the liver.

1.1.6 Endosulfan

Endosulfan is a manufactured organochlorine pesticide used to control insects on food and non-food crops, and it is also used as a wood preservative. Technical-grade endosulfan contains at least 94% α-endosulfan (Endosulfan I) and β-endosulfan (Endosulfan II) isomers, in the ratio of 7:3, respectively. Endosulfan sulfate is a reaction byproduct found in technical grade endosulfan as a result of oxidation, biotransformation, or photolysis. There is very little difference in toxicity between endosulfan and endosulfan sulfate; however, the α-isomer has been shown to be about three times as toxic as the β-isomer of endosulfan.

Exposure to endosulfan is through ingestion and inhalation. Neurotoxicity is the primary effect observed in humans following occupational exposure to endosulfan and in animal studies.

Endosulfan has not been classified by the USEPA as a carcinogen, but exposure to high enough amounts can cause non-cancer health effects.

1.2 SUMMARY OF PREVIOUS PESTICIDE INVESTIGATIONS AT NAVY

HOUSING PROPERTIES

Sampling results from previous pesticide investigations conducted in Guam and Hawaii (Parsons, 2008, 2010; Nexus Guam, 2006; IHP, 2007) were consistent with a conceptual site model (CSM) of historic use of chlorinated pesticides to control termites (as termiticides), using procedures consisting of applying the termiticides to the ground surface prior to slab construction and then subsequently around the immediate edge of the foundation. The previous pesticide investigations included sampling of a broad suite of organochlorine pesticides at 21 Navy housing communities in Hawaii and 3 Navy housing communities in Guam between the 2004 and 2010.

In these previous investigations, the following general observations were noted:

• Aldrin, dieldrin, and chlordane were the pesticides most commonly detected above screening levels.

• Although included in the analytical program, which investigated a broad suite of organochlorine pesticides, the following pesticides were not detected above screening levels at any site: methoxychlor, toxaphene, lindane (gamma-hexachlorocyclohexane), endrin ketone, endrin aldehyde, endosulfan sulfate.

• Although a majority of sites investigated had pesticide concentrations which exceeded screening levels for one or more pesticides, some sites did not have any exceedances.

• The majority of screening level exceedances were detected in samples collected from beneath building slabs.

• Screening level exceedances in samples collected from around the slab perimeters generally occurred in samples collected within 1.5 feet from the slab edge, but not in samples collected 2 feet or more from the slab edge or from common areas and yards.

• No significant exceedances above construction/maintenance worker screening levels occurred for any samples; therefore, no special construction/maintenance worker personal protection equipment (PPE) was required during construction.

The highest pesticide concentrations were found in soil samples collected beneath existing concrete slab foundations and immediately adjacent to the foundations of existing structures.

Concentrations decreased rapidly with distance from the slabs, generally below levels of concern at a distance of 2 feet from the foundation. Concentrations also decreased rapidly with depth from the ground surface, generally below levels of concern at a depth of 2 feet below ground surface.

1.3 CONCEPTUAL SITE MODEL

A Conceptual Site Model (CSM) is an effective tool to define site dynamics, streamline the risk evaluation, and develop appropriate response actions. The CSM is a representation of the nature, extent, and fate of contamination that allows assessment of the potential exposures to contamination, so that the decision maker can evaluate strategies to reduce the risks from contamination. The CSM identifies complete exposure pathways between environmental media affected by site-related contamination and potential receptors.

The CSM is dynamic and should be updated as necessary. For example, if changes in site conditions occur, or additional site characterization information is collected, the CSM should be revised if needed to more accurately reflect the most current information. Understanding site conditions and land uses helps to accurately identify potential receptors under current and likely future scenarios, as well as the most appropriate corrective action(s), if necessary.

The fate and transport characteristics of the pesticides of concern (Section 1.1), previous investigation results (Section 1.2), and the existing and expected future land uses for housing redevelopment sites under the Guam DPRI Program were all used to develop the assumptions for the current CSM for this Plan:

Suspected sources and types of contaminants present. The 10 organochlorine pesticides that may have been routinely applied to control termites and other pests in Guam are provided in Section 1.1. These pesticides were legally applied, typically in soil under and around foundations before the foundations were poured, and then subsequently around the foundation perimeters during routine maintenance/prevention activities.

Contaminant release and transport mechanisms. Pesticide application is generally within the upper 2 feet of ground surface. Pesticides tend to strongly adsorb to soil and will not leach from soil at significant concentrations under most geologic and climatic conditions.

Affected media. Shallow soil is the directly affected media since pesticides were applied to surface and near-surface soils. The potential for pesticides to leach from soil into groundwater or surface water also exists under some physical conditions; therefore, soil leaching was considered during the development of soil screening levels for groundwater and surface water protection.

Potential receptors under current or future land use scenarios. Potential receptors that could contact site-related contaminants in affected media include: military residents (bachelor housing residents and military family residents), commercial/industrial workers, maintenance workers, and recreational users. Current and reasonably anticipated future land uses under the Guam DPRI Program were considered in identifying potential receptors; exposure parameter assumptions are provided in Section 2.2.1.

Exposure points and pathways for chemical intake by receptors. Exposure routes for receptors include: direct contact with soil, incidental soil ingestion, and inhalation of fugitive dust.

This CSM and the previous investigation results were used to form the basis for development of screening levels and recommended sampling procedures in this Plan, discussed in Sections 2 and 3, respectively.

2.0 TIER IB SOIL SCREENING LEVELS

The purpose of this section is to describe the methodology used to develop Tier IB Soil Screening Levels for the Guam DPRI Program, present in tabular form the Tier IB Soil Screening Levels based on proposed land use, and explain the process for using the tabulated screening levels after site-specific sampling results are obtained.

As used in this Plan, “Tier IB Soil Screening Levels” refer to the Navy’s Tier IB human health risk assessment process (USNAVY, 2008). Navy Tier IB Risk-Based Screening Concentrations (RBSCs) are site-specific, risk-based screening concentrations that are considered protective of human health and are determined based on proposed land use, exposure, toxicity assessments, and potential groundwater impacts. They incorporate site-specific exposure scenarios combined with USEPA toxicity values and target risk goals to determine acceptable concentrations of chemicals.

For the purposes of this Plan, the Tier IB Soil Screening Levels developed for the DPRI Program can be used to screen pesticide concentrations in soil, provide options for management and potential onsite re-use of pesticide-impacted soils during construction, and/or used to trigger further investigation. Tier IB Soil Screening Levels were developed for the expected land use scenarios under the DPRI Program, and the Tier IB Soil Screening Levels that correspond with the most stringent and plausible current/future land use should be selected (e.g., protection of expected military personnel resident at the site, industrial workers, recreational use, or unrestricted). Soil samples must be collected and analyzed from the site being evaluated, as described in more detail in Section 3.0.

2.1 PESTICIDES OF CONCERN

Based on the CSM and previous historical investigations (Section 1.0), the pesticides of concern for the DPRI Program in Guam are provided in Table 1. Tier IB Soil Screening Levels are limited to the pesticides included in Table 1 and the analytical program must include all of these pesticides, and no other pesticides are required to be analyzed.

TABLE 1

PESTICIDES OF CONCERN

Pesticide CAS No. Soil Saturation Limit (mg/kg) *

Aldrin 309-00-2 10.77 Chlordane (Technical) 12789-03-6 29.12

DDD 72-54-8 82.36

DDE 72-55-9 36.60

DDT 50-29-3 7.27

Dieldrin 60-57-1 15.93 Endosulfan 115-29-7 59.42 Endrin 72-20-8 15.93 Heptachlor 76-44-8 56.62 Heptachlor epoxide 1024-57-3 6.33

* Ref. Table G (HDOH, 2008).

All of these pesticides are included in the standard laboratory analytical method for organochlorine pesticides (EPA Method 8081A); however, this method also includes additional pesticides not included in Table 1 (e.g., lindane, methoxychlor, toxaphene). The laboratory should be instructed to include only the pesticides listed in Table 1 when performing their analysis. The additional pesticides included in EPA Method 8081A were not previously detected above levels of concern in Hawaii or Guam; therefore, they are not of concern for the Guam DPRI Program and Tier IB Soil Screening Levels were not developed for them.

2.2 DEVELOPMENT OF TIER IB SOIL SCREENING LEVELS

Screening levels for pesticides were calculated following USEPA (1989, 1990, 1991, 1996, 2002, 2003, 2004, 2009, 2011a) and Department of the Navy (USNAVY, 2008) guidance documents.

Each of the steps performed in the calculations are detailed in the following subsections and detailed in Appendix A.

2.2.1 Identification of Receptors and Exposure Parameters

The human receptors likely to be present at military bases in Guam were determined in consultation with the Navy based on land use and include the following:

• Military family residents

• Bachelor Enlisted Quarters (BEQ) residents

• Recreational users

• Commercial/industrial workers (referred to hereinafter as “industrial workers”)

• Construction workers

Since future land use may be different than current land use, the most recent version of the Guam Joint Military Master Plan must be used to determine the most appropriate of the 5 land use scenarios above to use for a particular site undergoing redevelopment. The Navy must be consulted if a particular site is not included in the Master Plan or if there are any concerns regarding which land use scenario is most appropriate.

BEQ residents were assumed to be unaccompanied military personnel with a standard 2-year length and a maximum of 2 tours in Guam. Other military residents were assumed to be accompanied military personnel with families and children with a standard 3-year length and a maximum of 2 tours in Guam. Industrial and construction workers were assumed to be civilians working at Navy facilities on Guam. For industrial workers, all of the exposure parameters used are standard default USEPA exposure parameters. Construction worker and recreational user exposure durations and exposure frequencies were developed in consultation with the Navy.

Both residents and recreational users were assumed to be the children of military residents, since children have a higher degree of exposure than adults. Detailed exposure parameters used are provided in Appendix A.

The particulate emissions factors (PEFs; i.e., the amount of dust emitted from a site into the atmosphere) were calculated following USEPA (2002) guidance (Appendix A). USEPA (2002) provides parameter values to calculate PEFs only for locations in the continental United States, and neither Guam nor HDOH (2009) provide methods for calculating PEFs for islands in the Pacific. To account for potential exposures to dusts, the PEF and dispersion factor (Q/C) equations from USEPA (2002) were used. Where region-specific values were required in the calculations, values from the only other low-lying tropical city (i.e., Miami, FL) available from USEPA (2002) were used. For the construction PEF, weather information for Guam was taken from NOAA (2008, 2011).

2.2.2 Toxicity Values

The toxicity values used to calculate the screening levels were compiled following USEPA (2003, 2011a) guidance from the following hierarchy of sources:

• USEPA’s (2011b) Integrated Risk Information System (IRIS) on-line database

• USEPA’s (2011c) Provisional Peer Reviewed Toxicity Values (PPRTVs)

• Agency for Toxic Substances and Disease Registry’s (ATSDR) minimal risk levels (MRLs;

ATSDR 2011)

• California EPA’s (2011) Toxicity Criteria Database

• USEPA’s (1997) Health Effects Assessment Tables (HEAST)

If inhalation toxicity values were unavailable, route-to-route extrapolations were not used to estimate inhalation toxicity values from oral toxicity values as this is not recommended under current USEPA (1989, 2009) guidance. Specific cancer and noncancer toxicity values are provided in Appendix A.

The primary toxicity sources listed above provide oral and inhalation toxicity values, but not dermal toxicity values. Dermal toxicity values were calculated from oral toxicity values as follows

(USEPA 2004):

Dermal RfD = Oral RfD x ABSGI

Dermal SF = Oral SF / ABSGI where:

RfD = noncancer reference dose SF = cancer slope factor ABSGI = gastrointestinal absorption fraction

2.2.3 Target Risk/Hazard Level

Target risk levels were determined according to USEPA (1990, 1991) guidance. USEPA (1990) guidance indicates that the target risk range for carcinogenic risks is 1 x 10-4 to 1 x 10-6. The screening levels included in this Plan were developed for a cancer risk goal (i.e., the risk associated with summing the individual risks for carcinogens) of 1 x 10-5.

USEPA uses a hazard index (HI) of 1.0 as the benchmark level of concern to indicate whether adverse noncarcinogenic health effects are likely to occur as a result of site-related exposures.

Hazard indexes greater than 1.0 indicate that adverse noncarcinogenic health effects may occur, whereas HIs less than 1.0 indicate that adverse noncarcinogenic health effects are unlikely to occur. The screening levels included in this Plan were developed for a cumulative noncancer risk goal (i.e., the hazard associated with summing the individual hazards for noncarcinogens) and HI of 1.0.

These goals are set to ensure that the cumulative risks/hazards associated with exposure to residual concentrations of pesticides in soil for the DPRI Program are less than cancer risk of 1 x 10-5 and a cumulative noncancer risk HI of 1.0. USEPA assumes that carcinogenic risks and noncancer hazards from exposures to chemicals at a site may be additive; i.e., the risks from exposure to two chemicals is equal to the sum of the risks from each chemical. To be protective of the total cancer risks and noncancer hazards, the screening levels were also developed based on the number of pesticides detected at site.

2.2.4 Calculation of Human Health Screening Levels

Human health screening levels were calculated using the equations from USEPA (2011a). The equations and calculations are detailed in Appendix A. Equations were created by rearranging the standard USEPA (1989, 2004, 2009) equations used to the calculate risks and hazards from exposures at a site, and then rearranged to solve for the concentration in soil given a target risk/hazard level.

2.2.5 Incorporation of Soil Saturation Limits

After the human health screening levels were developed for each land use scenario (Appendix A), they were then compared to the soil saturation limit for each of the pesticides of concern (Table 1).

The goals of this comparison were to prevent allowing saturated soils to be a potential source for leaching into groundwater or surface water (i.e., the model used to calculate soil concentrations protective of groundwater and surface water are not applicable to concentrations above the saturation limit). The lower of the soil saturation limit and the Human Health Screening Levels was used. In general, this analysis lowered the screening values used for chlordane, DDD, DDE, DDT, endosulfan, and endrin compared to the human health-derived values, depending on the receptors, the type of land use, and the number of pesticides detected. The soil saturation limits provided in Table G of the Pacific Basin Edition (PBE) of Evaluation of Environmental Hazards at Sites with Contaminated Soil and Groundwater (HDOH, 2008) were used to make this comparison and are included in Table 1 for reference.

2.2.6 Incorporation of Groundwater Leaching Concerns

After the human health screening levels were adjusted for soil saturation limits, an additional analysis was performed to determine the potential for pesticide concentrations below the soil saturation limit or the human health screening levels to leach into groundwater. The goal of this analysis was to determine the maximum concentrations which could be left in soil without potentially impacting groundwater or surface water receptors under conservative, site-specific geologic and weather conditions specific to Guam.

Many organic pesticides tend to strongly adsorb to soil and do not leach through vadose zone soil to groundwater. Fate and transport modeling performed by the Hawaii Department of Health (HDOH) demonstrated that constituents with a organic carbon-water partitioning coefficient (Koc) greater than 30,000 cm3/g are essentially immobile in the subsurface under normal conditions (HDOH, 1996), conditions that are also consistent with Guam rainfall amounts less than 110 inches per year for sites expected to be within the DPRI Program (Appendix B). For the pesticides of concern in Table 1, only the four (4) pesticides listed in Table 2 have Koc values near to or less than a Koc of 30,000 cm3/g (HDOH, 2008).

TABLE 2

PESTICIDE PARTIONING COEFFICIENTS AND

GROUNDWATER SCREENING LEVELS

Pesticide Koc (cm3/g)

PBE Tier 1 ESL (ug/L)

Basis

Dieldrin 10,600 0.0019 Aquatic Habitat Goal Endosulfan 21,990 0.0087 Aquatic Habitat Goal Endrin 10,600 0.0023 Aquatic Habitat Goal Heptachlor epoxide 5,260 0.0036 Aquatic Habitat Goal

* Ref. Table F-1a and Table J (HDOH, 2008).

Fate and transport modeling performed by HDOH and incorporated into the default Tier 1 Screening Levels in the PBE (HDOH, 2008) is based on a generic linear regression algorithm used for all chemicals and is overly conservative for most pesticides.

Therefore, constituent-specific SESOIL/AT123D modeling was performed in order to estimate the potential for the 4 pesticides in Table 2 to leach from soil to groundwater, then migrate in groundwater to an assumed drinking water well located downgradient of the pesticide source at a concentration equal to or greater than the groundwater environmental screening level (ESL) provided in Table F-1a of the PBE guidance document (also provided in Table 2). Geologic and weather conditions specific to Guam were used in the model. Details on the modeling inputs and results are provided in Appendix B, including excerpts from inputs used in previous HDOH SESOIL modeling (HDOH, 1996). The model input parameters used are applicable to a variety of environmental conditions that may be encountered in Guam and are almost identical to the inputs used in previous HDOH SESOIL modeling.

The constituent-specific modeling for dieldrin, endosulfan, and endrin demonstrated that these three (3) constituents are essentially immobile and will not leach to groundwater, not unexpected since these 3 constituents have the highest Koc values for the 4 pesticides evaluated (Table 2), and only slightly lower than the 30,000 cm3/g rule-of-thumb HDOH developed to determine contaminant mobility. Therefore, the Tier IB Soil Screening Levels for these 3 constituents were not modified from the previous analysis using the minimum of the soil saturation limit and the Human Health Screening Level, as described in Section 2.2.5.

The constituent-specific modeling for heptachlor epoxide (which has a much lower Koc value than the 30,000 cm3/g rule-of-thumb) demonstrated that this constituent is more mobile than other pesticides of concern and could impact groundwater above the PBE Tier 1 ESL of 0.0036 ug/L under the conservative modeled conditions when the soil concentration was greater than

0.11 mg/kg (Appendix B). Therefore, the Tier IB Soil Screening Levels for heptachlor epoxide under all land use scenarios were adjusted to the minimum of the constituent-specific modeling result of 0.11 mg/kg modeled in Appendix B (which is less than the soil saturation limit

6.33 mg/kg) and the human health-only screening levels (Appendix A) developed for the 5 land use and exposure scenarios (Section 2.2.1).

2.4 FINAL TIER IB SOIL SCREENING LEVELS

The Final Tier IB Soil Screening Levels are provided in Tables 3a through 3e, one for each of the five (5) land use scenarios (Section 2.2.1). Since future land use determines which of the Tier IB Soil Screening Level tables is most appropriate, the most recent version of the Guam Joint Military Master Plan must be used to determine the most appropriate land use scenario for a particular site undergoing redevelopment. The Navy must be consulted if a particular site is not included in the Master Plan or if there are any concerns regarding which land use scenario and Tier IB Soil Screening Level table is most appropriate. Before applying the Tier IB Soil Screening Levels presented in these tables, the user must also consider whether or not the exposure pathways and exposure scenarios at a particular site are fully accounted for in the Tier IB Soil Screening Level calculations.

If a construction project is planned at a site which does not meet the conditions used in the model assumptions, or the site does not meet the conditions and assumptions used for the 5 land use scenarios, then the Navy must be consulted to determine the applicability of the Tier IB Soil Screening Levels included in this Plan and the potential need for site-specific modeling and/or development of site-specific screening levels. A checklist of site criteria that must be met in order to use the Tier IB Soil Screening Levels included in this Plan is provided in Table 4.

TABLE 4

SITE CHECKLIST FOR USE OF TIER IB SOIL SCREENING LEVELS

Site-Specific Information Criteria Source / Basis

Annual Rainfall <= 110 inches/year

Assumption used in groundwater leaching model. Refer to Guam Annual Rainfall figure and model results in Appendix B.

Future Land Use

Must be either:

- Military Housing (BEQ, families)

- Commercial/Industrial

- Recreational

Only receptors for these types of land uses and construction/maintenance workers were evaluated (Appendix A)

Unaccompanied Military Personnel (Bachelor) Duration Time

<= 4 years total on Guam (2 tours at 2 years/tour)

Assumption used for exposure durations for BEQ Housing residents (Appendix A)

Military Family Duration Time

<= 6 years total on Guam (2 tours at 3 years/tour)

Assumption used for exposure durations for military family housing residents (Appendix A)

Non-Military Residents not allowed Land use categories only include military residents (Section 2.2.1)

If a soil pesticide concentration exceeds its Tier IB Soil Screening Level, the Navy must be consulted prior to final selection and commencement of management and remedial actions.

To use the tables:

1) Determine which land use scenario is most appropriate for the site based on assumptions used for the 5 land use scenarios and receptors (Section 2.2.1). The Navy must be consulted to determine the most appropriate land use scenario to use for a particular site undergoing redevelopment since future land use may be different than current land use.

2) Select the table of Tier IB Soil Screening Levels that represents the most conservative, plausible exposure scenario for the site.

3) Count the number of detected pesticides from the maximum total of 10 pesticides of concern using all laboratory results for the site.

4) Use the column that corresponds with the number of detected pesticides.

5) Compare all detected pesticide concentrations against the Tier IB Soil Screening Levels from the correct column (only one table and one column will be used for each site).

6) If all detected concentrations are below the Tier IB Soil Screening Levels in the correct column and appropriate table, then no further action is required; and/or,

7) If any detected concentration is above its Tier IB Soil Screening Level, then the soil management actions described in Sections 4.0 and 5.0 must be followed, and the Navy must be consulted prior to final selection and commencement of management and remedial actions.

Table 3a Tier IB Soil Screening Levels (mg/kg)

Resident (child)

Number of Detected Pesticides Pesticide 1 2 3 4 5 6 7 8 9 10 Aldrin 0.42 0.21 0.14 0.10 0.08 0.07 0.06 0.05 0.05 0.04 Chlordane 23.45 11.72 7.82 5.86 4.69 3.91 3.35 2.93 2.61 2.34

DDD 35.07 17.54 11.69 8.77 7.01 5.85 5.01 4.38 3.90 3.51

DDE 24.76 12.38 8.25 6.19 4.95 4.13 3.54 3.09 2.75 2.48

DDT 7.27 7.27 7.27 6.19 4.95 4.13 3.54 3.09 2.75 2.48

Dieldrin 0.45 0.22 0.15 0.11 0.09 0.07 0.06 0.06 0.05 0.04 Endosulfan 59.42 59.42 59.42 59.42 59.42 59.42 52.38 45.83 40.74 36.66 Endrin 9.78 4.89 3.26 2.44 1.96 1.63 1.40 1.22 1.09 0.98 Heptachlor 1.58 0.79 0.53 0.40 0.32 0.26 0.23 0.20 0.18 0.16 Heptachlor epoxide 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.10 0.09 0.08

Table 3b Tier IB Soil Screening Levels (mg/kg)

BEQ Housing

Number of Detected Pesticides Pesticide 1 2 3 4 5 6 7 8 9 10 Aldrin 5.37 2.69 1.79 1.34 1.07 0.90 0.77 0.67 0.60 0.54 Chlordane 29.12 29.12 29.12 29.12 29.12 29.12 29.12 29.12 29.12 29.12

DDD 82.36 82.36 82.36 81.50 65.20 54.33 46.57 40.75 36.22 32.60

DDE 36.60 36.60 36.60 36.60 36.60 36.60 36.60 36.60 36.22 32.60

DDT 7.27 7.27 7.27 7.27 7.27 7.27 7.27 7.27 7.27 7.27

Dieldrin 5.71 2.85 1.90 1.43 1.14 0.95 0.82 0.71 0.63 0.57 Endosulfan 59.42 59.42 59.42 59.42 59.42 59.42 59.42 59.42 59.42 59.42 Endrin 15.93 15.93 15.93 15.93 14.62 12.19 10.45 9.14 8.12 7.31 Heptachlor 20.29 10.15 6.76 5.07 4.06 3.38 2.90 2.54 2.25 2.03 Heptachlor epoxide 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.11

Table 3c Tier IB Soil Screening Levels (mg/kg)

Recreational (child)

Number of Detected Pesticides Pesticide 1 2 3 4 5 6 7 8 9 10 Aldrin 2.82 1.41 0.94 0.71 0.56 0.47 0.40 0.35 0.31 0.28 Chlordane 29.12 29.12 29.12 29.12 29.12 26.30 22.54 19.73 17.53 15.78

DDD 82.36 82.36 78.69 59.02 47.22 39.35 33.73 29.51 26.23 23.61

DDE 36.60 36.60 36.60 36.60 33.33 27.77 23.81 20.83 18.52 16.66

DDT 7.27 7.27 7.27 7.27 7.27 7.27 7.27 7.27 7.27 7.27

Dieldrin 3.00 1.50 1.00 0.75 0.60 0.50 0.43 0.37 0.33 0.30 Endosulfan 59.42 59.42 59.42 59.42 59.42 59.42 59.42 59.42 59.42 59.42 Endrin 15.93 15.93 15.93 15.93 13.16 10.97 9.40 8.23 7.31 6.58 Heptachlor 10.66 5.33 3.55 2.67 2.13 1.78 1.52 1.33 1.18 1.07 Heptachlor epoxide 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.11

Table 3d Tier IB Soil Screening Levels (mg/kg)

Commerical/Industrial Workers

Number of Detected Pesticides Pesticide 1 2 3 4 5 6 7 8 9 10 Aldrin 0.79 0.39 0.26 0.20 0.16 0.13 0.11 0.10 0.09 0.08 Chlordane 29.12 28.08 18.72 14.04 11.23 9.36 8.02 7.02 6.24 5.62

DDD 82.36 44.42 29.62 22.21 17.77 14.81 12.69 11.11 9.87 8.88

DDE 36.60 31.36 20.91 15.68 12.54 10.45 8.96 7.84 6.97 6.27

DDT 7.27 7.27 7.27 7.27 7.27 7.27 7.27 7.27 6.97 6.27

Dieldrin 0.84 0.42 0.28 0.21 0.17 0.14 0.12 0.10 0.09 0.08 Endosulfan 59.42 59.42 59.42 59.42 59.42 59.42 59.42 59.42 59.42 59.42 Endrin 15.93 15.93 15.25 11.44 9.15 7.63 6.54 5.72 5.08 4.58 Heptachlor 2.97 1.49 0.99 0.74 0.59 0.50 0.42 0.37 0.33 0.30 Heptachlor epoxide 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.11

Table 3e Tier IB Soil Screening Levels (mg/kg)

Construction Workers

Number of Detected Pesticides Pesticide 1 2 3 4 5 6 7 8 9 10 Aldrin 9.96 4.98 3.32 2.49 1.99 1.66 1.42 1.25 1.11 1.00 Chlordane 29.12 29.12 29.12 29.12 29.12 29.12 29.12 29.12 29.12 29.12

DDD 82.36 82.36 82.36 82.36 82.36 82.36 82.36 82.36 82.36 82.36

DDE 36.60 36.60 36.60 36.60 36.60 36.60 36.60 36.60 36.60 36.60

DDT 7.27 7.27 7.27 7.27 7.27 7.27 7.27 7.27 7.27 7.27

Dieldrin 10.59 5.29 3.53 2.65 2.12 1.76 1.51 1.32 1.18 1.06 Endosulfan 59.42 59.42 59.42 59.42 59.42 59.42 59.42 59.42 59.42 59.42 Endrin 15.93 15.93 15.93 15.93 15.93 15.93 15.93 15.93 15.93 15.93 Heptachlor 37.64 18.82 12.55 9.41 7.53 6.27 5.38 4.70 4.18 3.76 Heptachlor epoxide 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.11 0.11

3.0 SAMPLING AND REPORTING REQUIREMENTS

3.1 INTRODUCTION AND OBJECTIVES

This section describes the recommended sampling strategies, sampling methods, sample handling procedures, and general field procedures and requirements in order to compare pesticide soil concentrations against the Tier IB Soil Screening Levels provided in Section 2.0, and then implement the required soil management options described in Sections 4.0 and 5.0. A Site-Specific Work Plan (SSWP) must be developed that is consistent with the general guidance and requirements of this Section, and the SSWP must be approved by the Department of the Navy prior to implementation.

3.2 SAMPLING STRATEGY

3.2.1 Decision Units

The sampling strategy is to collect composite shallow subsurface soil samples from decision units (DUs) that are representative of similar previous use (e.g., either current housing, dormitories, or demolished housing) and history (e.g., construction date, demolition date) for the project site. An overall DU boundary could be as small one stand-alone structure being evaluated, or as large as an entire housing neighborhood where all structures were built at the same time and are still in place. A housing neighborhood should be subdivided into separate DUs if any characteristic of housing use, pesticide application, history of the neighborhood, and/or date of construction or demolition are different, characteristics which potentially affect the expected pesticide concentrations or their fate and transport in the subsurface. For example, the Finegayan neighborhood on Guam, although built at generally the same time period, consists of both existing housing structures and previously-demolished housing structures – each of these two areas should consist of a separate DU because the fate and transport of pesticides will be different under infiltration conditions with slabs (existing structures) vs.

infiltration conditions without slabs (demolished structures).

Distinction for DU boundaries should be made if there is a significant difference in the time of construction for portions of a neighborhood due to potential differences and changes in specific pesticide use and application methods. A difference of more than five years between the time of construction is considered a reasonable criteria based on a comparison of neighborhood construction timeframes and the respective pesticides detected for previous investigations conducted at 21 Navy neighborhoods in Hawaii (Parsons, 2006; 2007a; 2007b; 2008). A knowledge of the site history, an understanding of the current site layout and use, and professional judgment are all needed to establish good DU boundaries.

Justification for DU boundaries must be included in the SSWP.

3.2.2 Representativeness

Within each DU, sampling should be conducted at a representative number of buildings, locations, and depths to characterize the soils under and around existing or former slabs.

Sampling must conducted within both soil and potential underlying coralline materials to a depth of at least 2.0 feet below ground surface (bgs) at each sampled location. Results from this representative sampling approach and depth profile will ultimately be used to make decisions for the entire DU with regard to pesticide-impacted soils management and remediation, if needed.

A conventional composite, biased sampling approach for the collection and analysis of samples is generally recommended for most sites, based on considerations of feasibility, practicability of access, and cost constraints (e.g., difficulty of drilling a large number of boreholes beneath existing slabs, uncertainty with respect to exact locations of previously-demolished slabs).

Although sample compositing is primarily based on individual buildings and locations (i.e. under slabs, around slab perimeters), management and remediation decisions must be based on overall results, maximum concentrations, and evaluations of data from all samples collected within a DU. Because not all buildings will typically be sampled, management and remediation decisions cannot be implemented on an individual building basis. Rather, management and remediation decisions must be implemented for the entire DU, using the maximum concentrations in that DU, while taking into account any proposed differences in future land use areas within that DU.

Generally, it is recommended that sampling be conducted in two stages to avoid unnecessary additional field work and analytical costs when pesticides are present above screening levels for the proposed future land use. Stage 1 consists of sampling a minimum of 5% of the total number of buildings in each DU. The number of buildings and sample results collected during Stage 1 are used to determine if pesticides are generally present in soils above screening levels.

If Stage 1 results indicate pesticides are generally present in soils above screening levels, a conservative decision can optionally be made, without any additional sampling, to mitigate soils at all buildings in the DU (whether sampled or not sampled). This conservative approach is usually used due to the logistics, cost, and schedule needed to sample a large number of potentially occupied buildings. Based on the CSM and previous similar work, any additional sampling is unlikely to lead to a no mitigation decision and is, therefore, not expected to be cost-effective.

If Stage 1 results indicate pesticides are not present in soils above screening levels, then additional Stage 2 sampling must be conducted to evaluate representativeness prior to making a no mitigation decision. The objective of the Stage 2 sampling is to increase the number of buildings sampled in order to provide a robust, statistically representative sampling population prior to making a no mitigation decision. Stage 2 sampling consists of sampling additional buildings within the same DU to meet the requirement for total number of buildings to sample provided in Table 5. As shown in Table 5, the total number of buildings to sample that is required to meet the minimum representativeness guidelines after both Stage 1 and Stage 2 sampling is based on the total number of buildings within the DU. The quantities and percentages in Table 5 are based on guidance provided by the Department of Housing and Urban Development (HUD, 1995).

When choosing which buildings to sample within a DU, they should be selected randomly.

Specific sampling locations must be included in the SSWP.

Prior to concluding that the soils have not been impacted and making a no mitigation decision, results obtained from both Stage 1 and Stage 2 sampling must be used to statistically determine if results are representative of all buildings in the DU.

Table 5

STAGE 2 SAMPLING – TOTAL NUMBER OF BUILDINGS TO SAMPLE

Number of Similar Buildings in the DU

Total Number of Buildings to Sample *

Calculated Percentage Range

1–4 All (100%) 100% 5–20 4 buildings or 50%, whichever is greater 50% to 80% (4 of 5) 21–75 10 buildings or 20%, whichever is greater 20% to 47% (10 of 21) 76–125 17 buildings 14% to 22% 126–175 19 buildings 11% to 15% 176–225 20 buildings 9% to 11% 226–300 21 buildings 7% to 9% 301+ 22 buildings + 1 building for each additional increment of 100 buildings or less 6% to 7%

* includes the Stage 1 total of 5% plus the additional Stage 2 total. For example, if the total number of buildings in the DU is 100 and 5 buildings were sampled in Stage 1 (5%), then 12 additional buildings would be sampled in Stage 2 to meet the requirement shown of 17 buildings.

3.3 SAMPLING LOCATIONS

For each building selected for sampling, soil samples should be collected from two (2) group locations (Figure 1):

1) under the foundation (“sub-slab samples”)

2) adjacent to the foundation (“near-slab samples”), within a distance of 2 feet from the foundation

These 2 group locations were selected to target the areas expected to have similar concentrations of pesticides based on the CSM presented in Section 1.2.

At a typical single-family home or duplex size building (less than 5,000 ft2), soil samples should be collected from a total of approximately 7 individual sampling locations within the above 2 group locations, as follows:

• 3 locations under the foundation (borings 1 through 3)

• 4 locations adjacent to the foundation (borings 4 through 7)

Larger size buildings (multi-family apartments, dormitories, commercial structures) will require a larger number of sampling locations, depending on size, and must be specified in the SSWP.

At each individual sampling location, discrete subsamples must be collected, at a minimum, of one sample between 0.0 and 2.0 feet (ft) bgs. These depths were selected to target the areas expected to have similar concentrations of pesticides based on the CSM presented in Section 1.2. Therefore, in this example a total of 7 subsamples would be collected from each building (7 subsamples, all from a single depth profile of 0.0 to 2.0 ft bgs).

Sub-Slab Boring

Near-Slab Boring

LEGEND

FIGURE 1

Generalized Sampling Locations

Guam DPRI Pesticide Soils Management Plan

Parsons

B01

B02

B03

B04

B05

B06

B07

- 18 -composite all sub-slab field samples into one (1) composite analytical sample composite all near-slab field samples into one (1) composite analytical sample and Sample Composites

Optionally, depth profiling may also be conducted to evaluate differences in pesticide concentrations between near surface soils (0.0 to 1.0 ft bgs) and deeper soils (1.0 to 2.0 ft bgs), differences between soil and coralline materials, soils deeper than 2.0 ft bgs, and/or other site-specific factors. For example, a total of 14 subsamples would be collected from each building if optional depth profiling is conducted to evaluate differences in pesticide concentrations in soil from 0.0 to 1.0 ft bgs compared to soil from 1.0 to 2.0 ft bgs (7 subsamples from each of 2 depths).

The depths proposed for sampling and the number of subsamples must be specified in the

SSWP.

Sampling tools which collect an identical volume of soil should used for each subsample to avoid bias (i.e. different soil volume) between subsampling locations. A small, disposable scoop should be used to scrape a uniform section down the length of the core into a container of known volume (typically, the 4-oz. sample jars provided by the laboratory). The sampling scoop should be used to cut a core-shaped “wedge” running throughout the entire length of the targeted sampling interval to ensure that the consistent volume of soil is collected that is representative of the entire depth interval. The separate subsamples are then placed in separate 4-oz. jars for each unique physical location and depth, assigned a unique sample ID, and shipped to the laboratory.

Subsamples must be composited, either in the field or in the laboratory, based on group location and depth interval. Subsamples from the same group location and depth interval should be composited together to ensure that the sample that is analyzed is representative of the soil from the entire composite group. As a result, in the primary example, a total of 2 samples per building would be analyzed — 1 sample from between 0.0 and 2.0 ft bgs from each of the 2 group locations — from the total of 7 subsamples at that building. The 3 sub-slab subsamples from the same building are composited together to comprise 1 of the 2 analyzed samples, and the 4 perimeter subsamples from the same building are composited together to comprise the second analyzed sample.

The buildings and locations proposed for sampling must be included in the SSWP.

3.4 ANALYTICAL METHODS AND LABORATORY REQUIREMENTS

All soil samples must be analyzed for (and only for) the pesticides of concern (Table 1) by EPA method 8081A.