OPERABLE UNIT III
BROOKHAVEN NATIONAL LABORATORY
UPTON, NEW YORK

This record of decision (ROD) presents the selected remedial actions for Operable Unit (OU) III of the Brookhaven National Laboratory (BNL) site in Upton, New York. OU III was developed to address groundwater plumes emanating from the central and southern portion of the BNL site. The selected remedy addresses on and off-site groundwater contaminated with volatile organic compounds (VOCs), and tritium and strontium-90 in groundwater on-site. Thirteen areas of concern (AOCs) located in OU III, four AOCs from other OUs and two Additional Areas of Investigation (AAIs) were investigated and characterized in the Remedial Investigation Report for OU III.

These remedial actions were selected in accordance with the Comprehensive Environmental Response, Compensation and Liability Act (CERCLA) as amended by the Superfund Amendments and Reauthorization Act of 1986 (hereinafter jointly referred to as CERCLA), and is consistent, to the extent practicable, with the National Oil and Hazardous Substances Pollution Contingency Plan (National Contingency Plan). This decision is based on the Administrative Record for the BNL site.

The State of New York concurs with the selected remedial actions.

Actual or potential releases of hazardous substances, including chemical and radioactive materials from these areas, may present a threat to public health, welfare or the environment if they are not addressed by implementing the response actions selected in this ROD.

Operable Unit III is one of the six operable units at the BNL site for which remedies have been or will be selected. This ROD documents the selected remedial actions for groundwater contamination in OU III. Removal actions, which are either complete or on going are integrated into the final actions. Completed removal actions and source areas are addressed in Table 2. This ROD documents remedies that are consistent with the overall site cleanup strategy. The ROD includes a description of principal contaminants and their representative risks. Cleanup goals have been established to meet regulatory standards. The clean up objectives are: to meet the drinking water standards in groundwater for VOCs, strontium-90 and tritium; complete the cleanup of the groundwater in a timely manner, which for the Upper Glacial Aquifer goal is 30 years or less; and, prevent or minimize further migration of VOCs, Strontium-90 and tritium in groundwater. Current and future land uses were evaluated in this ROD. The costs of each remedy were estimated and are discussed in the ROD. The best balance of the Environmental Protection Agency's (EPA) remedy selection criteria was used to identify the following selected actions:

Volatile Organic Compounds (VOCs) Remedy: There is a large plume of groundwater contaminated with VOCs in the central and southern portion of the BNL Site and off-site. Several Interim Removal Actions (IRAs) have begun to address VOC contamination, including treatment systems at the southern site boundary and in an off-site, downgradient industrial park. Additionally, public water was provided in a large area south of the BNL Site, to protect public health while the groundwater cleanup is underway.

The selected remedy, Alternative V10c, involves active remediation of both on-site and off-site VOC contamination. It includes the following systems: operation of the on-site and off-site IRAs, including the On-Site Southern Boundary IRA and the Off-Site Industrial Complex IRA; installation of new remedial systems at the Long Island Power Authority (LIPA) right-of-way, North Street, the Brookhaven Airport, downgradient of North Street East, and the eastern portion of the industrial park; and an additional treatment system on-site at Middle Road. The remedy also includes either a new remedial system and/or expansion of the existing on-site pump and treat system to address lower levels of VOCs in the western part of the plume, and a source removal system using re-circulation wells with air stripping treatment near Building 96. Details of the specific number of treatment systems and locations needed to meet the cleanup objectives will be determined during the design process. The period of pumping needed to achieve cleanup objectives will be determined based on monitoring and operating data. Each treatment system will have a monitoring well network which will include downgradient sentry wells. These monitoring well networks will be used to help assess the effectiveness of achieving the clean up objectives. The exact number of monitoring wells will be determined during the design process. The assessment and evaluation of all treatment systems in achieving the clean up objectives will be performed annually. The details of the annual assessment and evaluation will be determined during the design process. If the annual assessments show that the treatment systems are not achieving the clean up objectives then the treatment systems will be modified and/or augmented to ensure that the clean up objectives are being met.

This selected remedy (V10c) is not the one that was proposed in the Proposed Remedial Action Plan (PRAP). The proposed remedy (V10b) did not include the treatment system located on-site for the western low-level VOC plume. The additional system was added in response to community and regulatory concerns about potential impacts to the Carmans River.

If, after source control is complete and effective, the annual assessment indicates that continued operation of the components of the selected remedy is not producing further reductions in the concentrations of contaminants in groundwater, DOE, NYSDEC, and EPA will evaluate whether discontinuance of the remedy is warranted or if modification and/or augmentation of the treatment systems is needed to ensure that the cleanup objectives are met.

Tritium Remedy: A pump and recharge system, which includes three pumping wells located on-site along Princeton Avenue, was installed in May 1997 to extract the tritium contaminated groundwater and discharge it further north to a recharge basin on-site. Pumping at the leading edge of the plume was taken as a precautionary measure to inhibit contaminated groundwater from advancing towards the site's boundary and allow more time for the tritium to decay. A carbon filtration unit also was included in the pump and recharge system to remove VOC's that are also present.

The selected remedy is a modification of alternative T4, as originally proposed in the PRAP. The remedy will combine extraction of groundwater in response to specific contingencies and extensive monitoring and reporting to assure that the cleanup objectives are met. Three specific contingencies were identified in the PRAP, and a fourth has been added in this ROD to address regulatory concerns. Other actions will be evaluated and implemented, as necessary, to ensure that the cleanup objectives are met. Additional monitoring wells will supplement the existing groundwater monitoring network downgradient of the High Flux Beam Reactor's (HFBR) spent fuel pool.

The first and second contingencies were developed to ensure that the tritium plume would migrate no further downgradient above drinking water standards. After an evaluation period established during design of the selected remedy, the tritium pump and recharge system on Princeton Avenue will be put on stand-by and later operated as needed as an integral component of these contingencies. The evaluation period will extend up to a maximum of one year after ROD finalization and will include an analysis of the data against the following two contingency criteria. These two specific contingencies identified are 1) to evaluate the need to reactivate the Princeton Avenue IRA if tritium concentrations exceed 25,000 pCi/l at the Chilled Water Plant Road, and/or 2) reactivate the Princeton Avenue IRA if tritium concentrations exceed 20,000 pCi/l at Weaver Drive.

A third contingency was developed to ensure that if the most concentrated part of the plume were to act as a source of continuing contamination, active remediation would remove this problem. This contingency proposed a low flow extraction system to be installed in the most concentrated area of tritium contamination near the HFBR and activated if concentrations exceed 2,000,000 pCi/l at the front of the reactor. This system then would be used to remove groundwater containing the highest concentrations of tritium from the aquifer. The extracted tritium contaminated water will be disposed of offsite. Technologies to reduce the volume of water that requires off-site disposal may be identified during design. Since the PRAP was issued to the public, groundwater near the HFBR has exceeded 2,000,000 pCi/l. DOE is currently in the process of constructing some of the wells for this low flow extraction system on Cornell Avenue and developing plans to extract the most concentrated part of the plume in front of the HFBR. These extraction wells are scheduled to begin operation no later than three months after execution of this ROD. The detailed operational parameters for this system will be developed during design.

In addition to the ones originally identified in Alternative T4 and proposed in the PRAP, a fourth contingency, an additional low flow extraction system will be installed and operated near Temple Place. This additional system was added in response to regulatory concerns about potential plume migration. The exact location, operational parameters and treatment and disposal options for the extracted water will be developed during design. Operation of the Temple Place extraction system will continue for up to one year. As these extraction wells operate, extensive monitoring will occur to evaluate the effect of extraction locally, as well as on the entire plume. Because of the inherent uncertainties of predicating plume behavior based on groundwater modeling, the actual monitoring data will be evaluated and used to help determine whether continued operation of this extraction system is needed to achieve the cleanup objectives. The criteria to continue system operation beyond one year will be developed during design and based on the attainment of the cleanup objectives.

Strontium-90 Remedy: There are concentrated areas of strontium-90 contamination in the groundwater at three on-site locations: the Chemical Holes Area, the Brookhaven Graphite Research Reactor (BGRR), and the Waste Concentration Facility.

The selected remedy, Alternative S5a, involves installing extraction wells and using ion exchange to remove strontium-90 from the extracted water. Details of the specific number of treatment systems and locations needed to meet the cleanup objectives will be determined during the design process. The period of pumping needed to achieve the cleanup objectives will be determined based on monitoring and operating data. Before implementation of the remedy, a pilot treatability study will be performed to evaluate the effectiveness of extraction and treatment. The final remedy may potentially be modified based on the results of this study. Clean water will be discharged on-site. Residual waste that contains strontium-90 will be disposed of at a licensed facility off-site.

If an assessment and evaluation indicates that continued operation of the components of the selected remedy is not producing further reductions in the concentrations of contaminants in groundwater, DOE, NYSDEC, and EPA will evaluate whether discontinuance of the remedy is warranted or if modification and/or augmentation of the treatment systems is needed to ensure that cleanup objectives are met

Source Areas: Some source areas and soil contaminants are, have been, or will be addressed in other RODS. Thirteen AOCs assigned to OU III were investigated as suspected source areas of groundwater contamination. Also, as the work for OU III was proceeding, groundwater contamination from other OUs and Additional Areas of Investigation (AAIs) was included in the investigation and assessment. Table 1 describes these AOCs and AAIs. Table 2 outlines the actions required for these suspected source areas. Many of the suspected source areas had completed and/or ongoing removal actions and no further action is required. The selected remedy requires a source removal system using re-circulation wells with air stripping treatment near Building 96; excavation and off site disposal of the PCB contaminated soils at Building 96 that are above the New York State cleanup levels; remediation of the groundwater near the Carbon Tetrachloride Tank Spill Area; completion of the Building 830 Underground Storage Tank Removal Action; and management of other suspected source areas as shown in Table 2. The final remedy for potential source areas in AOC-26B (Building 96), such as the anomalies discovered during the geophysical survey, will be documented in a subsequent Record of Decision.

Other Remedy Components: All of the groundwater plumes will require monitoring of new and existing wells and institutional control of the groundwater until completion of remediation. These wells will be located adjacent to the treatment systems and along the downgradient plumes. They will help determine the effectiveness of each treatment system in reducing the concentrations of contaminants over time. Long-term monitoring will also determine the ultimate duration for operation of the treatment systems and will support future decisions to make any changes to the final remedy. At the request of the homeowner, DOE can arrange for monitoring of private wells used for drinking water on properties that previously have declined DOE's offer of public water hookups. In addition, any sale or transfer of BNL property will meet the requirements of CERCLA 120(h) to ensure that future users will not be exposed to unacceptable levels of contamination in the groundwater.

Deferred Decisions: The final remedy for potential source areas in AOC-26B (Building 96), such as the anomalies discovered during the geophysical survey, will be documented in a subsequent Record of Decision. Also, the final remedy for AOC-9D, the Pile Fan Sump, will be documented in the Brookhaven Graphite Research Reactor (BGRR) Record of Decision.

The selected remedies are protective of human health and the environment, comply with federal and state requirements that are legally applicable or relevant and appropriate to the remedial actions, and are cost effective. These remedies utilize permanent solutions and alternative treatment technologies to the maximum extent practicable, and satisfy the statutory preference for remedies that employ treatment that reduces contaminant toxicity, mobility or volume as a principal element.

Should new information become available regarding disposal costs or the cost effectiveness of new technologies during the remedial design or remedial actions that could affect how the remedy selected in this ROD is implemented, the remedy may be modified and documented if such a change does not constitute a fundamental change in the remedy.

A five-year review of the remedial action pursuant to CERCLA §121(c), 42 U.S.C. §9621(c), will be necessary, since some of the selected remedies could result in hazardous substances remaining on site above health-based levels.

George J. Malosh
Manager, Brookhaven Group
U.S. Department of Energy

Robert P. Gordon
Contracting Officer, Brookhaven Group
U.S. Department of Energy

Jeanne Fox
Regional Administrator, Region 2
U.S. Environmental Protection Agency

I. DECLARATION OF THE RECORD OF DECISION

SITE NAME AND LOCATION

STATEMENT OF BASIS AND PURPOSE

ASSESSMENT OF THE SITE

DESCRIPTION OF THE SELECTED REMEDY

DECLARATION

TABLE OF CONTENTS

LIST OF ACRONYMNS

II. DECISION SUMMARY

1. SITE NAME, LOCATION AND DESCRIPTION

2. SITE HISTORY AND ENFORCEMENT ACTIVITIES

2.1 Site History

2.1.1 Previous Actions and Remedial Investigation/ Feasibility Study

2.1.2 History of OU III

2.2 Enforcement Activities

3. HIGHLIGHTS OF COMMUNITY PARTICIPATION

4. SCOPE AND ROLE OF OPERABLE UNIT AND RESPONSE ACTION

5. SUMMARY OF SITE CHARACTERISTICS

5.1 Identification of Contamination

5.2 Soil, Sediment, Groundwater, and Surface Water Investigations

5.2.1 Summary of Investigation of Study Area

5.2.2 Summary of Nature and Extent of Contamination

5.3 Action Summary for OU III

5.3.1 Source Removal

5.3.2 Facility Site Review

5.3.3 Removal Actions and Interim Removal Actions

5.3.4 Current Remedial Action Summary

6. SUMMARY OF SITE RISKS

6.1 Human Health Risks

6.1.1 Identification of Contaminants of Potential Concern

6.1.2 Exposure Assessment

6.1.3 Toxicity Assessment

6.1.4 Human Risk Characterization

6.2 Ecological Risk

6.3 Basis for Response/Remedial Action Objectives

7. DESCRIPTION OF ALTERNATIVES

7.1 Cleanup of VOC Contaminated Groundwater

7.2 Cleanup of Sr-90 Contaminated Groundwater

7.3 Cleanup of Tritium Contaminated Groundwater

8. SUMMARY OF COMPARATIVE ANALYSIS OF ALTERNATIVES

8.1 CERCLA Criteria

8.2 Comparative Analysis

8.2.1 TVOC Comparative Analysis

8.2.2 Strontium Comparative Analysis

8.2.3 Tritium Comparative Analysis

9. SELECTED REMEDIES

10. STATUTORY DETERMINATIONS

10.1 Protection of Human health and the Environment

10.2 Compliance with ARARs

10.3 Cost Effectiveness

10.4 Use of Permanent Solutions and Alternative Treatment Technologies to the Maximum Extent Practicable

10.5 Five-Year Review

III. RESPONSIVENESS SUMMARY

INTRODUCTION

1. RESPONSIVENESS SUMMARY OVERVIEW

Site History

Operable Unit III

Selected Remedial Alternatives for OU III

Level Of Community Support for the Preferred Alternative

Changes to the Proposed Alternative

2. BACKGROUND ON COMMUNITY INVOLVEMENT AND CONCERNS

Community Profile

History of Community Involvement

Summary of Community Participation Activities

3. COMPREHENSIVE SUMMARY OF MAJOR QUESTIONS, COMMENTS, CONCERNS AND DOE RESPONSES

Overview

Summary and Response to Questions and Comments

4. COMMUNITY RELATIONS ACTIVITIES

5. REFERENCES

IV. REFERENCES

V. TABLES

VI. FIGURES

APPENDIX A: COMMENT LETTERS

TABLES

Table 1 - OU III Areas of Concern and Additional Areas of Investigation

Table 2 - Summary of Selected Remedies and Previous Actions

Table 3 - Source Removal Actions

Table 4 - Specific Screening Criteria for Contaminants Exceeding Screening Criteria in any Media in OU III

Table 5 - Detected Concentration Range of the Constituents of Potential Concern in OU III

Table 6 - Exposure Scenarios Evaluated in the Chemical Baseline Human Health Risk Assessment

Table 7 - Exposure Scenarios Evaluated in the Radiological Baseline Human Health Risk Assessment

Table 8 - Non-Carcinogenic Effects: Toxicity Values and Effects of Constituents Of Potential Concern

Table 9 - Carcinogenic Effects: Toxicity Values and Effects of Constituents of Potential Concern

Table 10 - Cancer Risk Slope Factors for Radionuclides of Potential Concern

Table 11 - Chemical Risk Assessment: Cancer Risks and Hazard Indices for Reasonable Maximum Exposure (RME) Scenario

Table 12 - Radiological Risk Assessment: Cancer Risks for Reasonable Maximum Exposure (RME) Scenario

Table 13 - Ecological Constituents of Potential Concern in Environmental Media of OU III

Table 14 - TVOC Remedial Alternatives

Table 15 - Strontium-90 Remedial Alternatives

Table 16 - Tritium Remedial Alternatives

Table 17 - Summary of Comparative Analysis of TVOC Alternatives

Table 18 - Summary of Comparative Analysis of Strontium Alternatives

Table 19 - Summary of Comparative Analysis of Tritium Alternatives

Table 20 - Summary of Estimated Costs

Figure 1 - Regional Site Location Map

Figure 2 - Current Land Use at BNL

Figure 3 - BNL Operable Units

Figure 4 - AOCs and AAIs addressed in this Record of Decision

Figure 5 - Areal Extent of TVOC Contamination in Groundwater

Figure 6 - Areal Extent of Strontium-90 Contamination in Groundwater

Figure 7 - Areal Extent of Tritium Contamination in Groundwater

Figure 8 - Schematic showing the existing Operable Unit III South Boundary Pump-and-Treat System

Figure 9 - Southern Boundary and Industrial Complex Interim Removal Action TVOC Well Locations

Figure 10 - Public Water Hookup Areas

Figure 11 - Tritium Recirculation Flow Schematic of Existing Tritium Interim Removal Action System

Figure 12 - Schematic showing a typical in-well air-stripping system

Figure 13 - Schematic showing strontium-90 ion exchange system

Figure 14 - Areal extent of TVOC, Sr-90 and tritium contamination in groundwater with existing and planned pumping locations

AAI - Additional Areas of Investigation

AGS - Alternating Gradient Synchotron

AOC - Area of Concern

ARAR - Applicable or Relevant and Appropriate Requirement

AS - Air Stripping

ATSDR - Agency for Toxic Substances and Disease Registry

BER - Brookhaven Executive Roundtable

BGRR - Brookhaven Graphite Research Reactor

BHG - Brookhaven Group

BLIP - Brookhaven Linear Isotope Producer

BLS - Below Land Surface

BMRR - Brookhaven Medical Research Reactor

BMSL - Below Mean Sea Level

BNL - Brookhaven National Laboratory

CEDE - Committed Effective Dose Equivalent

CERCLA - Comprehensive Environmental Response Compensation & Liability Act

COPC - Chemicals of Potential Concern

CWF - Chilled Water Facility

DCE - 1,1 dichloroethene

DOE - United States Department of Energy

DOT - Department of Transportation

EE/CA - Engineering Evaluation/Cost Analysis

EDB - Ethylene dibromide

EP - Extraction Procedure

EPA - United States Environmental Protection Agency

ERD - Environmental Restoration Division

ERM - Effects Range Median

ES&HS - Environmental Safety and Health Services

FS - Feasibility Study

GRA - General Response Action

HFBR - High Flux Beam Reactor

HI - Hazard Index

HWMF - Hazardous Waste Management Facility

IAG - Interagency Agreement

ILCR - Individual Lifetime Cancer Risk

IRA - Interim Removal Action

LDL - Low Detection Limit

LINAC - Linear Accelerator

LIPA - Long Island Power Authority

MCL - Maximum Contaminant Level

mg/kg - milligrams per kilogram

NCP - National Contingency Plan

NEPA - National Environmental Policy Act

NPL - National Priorities List

NYCRR - New York State Codes, Rules and Regulations

NYS - New York State

NYSDEC - New York State Department of Environmental Conservation

OU - Operable Unit

PA/SI - Preliminary Assessment/Site Inspection

PCB - Polychlorinated biphenols

PCE - tetrachloroethene

PFS - Pile Fan Sump

PRAP - Proposed Remedial Actin Plan

PRG - Preliminary Remediation Goal

RAO - Remedial Action Objective

RAV - Remedial Action V

RCG - Remedial Capture Goal

RCRA - Resource Conservation and Recovery Act

RESRAD- Residual Radioactive Material Guideline Computer Code

RI - Remedial Investigation

ROD - Record of Decision

RS - Responsiveness Summary

SCDHS - Suffolk County Department of Health Services

SCWA - Suffolk County Water Authority

S&EP - Safety and Environmental Protection Division

SPDES - State Pollutant Discharge Elimination System

STP - Sewage Treatment Plant

SVE - Soil Vapor Extraction

TAL - Target Analyte List

TAGM - NYSDEC Technical Assistance Guidance Memorandum

TBC - To Be Considered

TCA - 1,1,1 trichloroethane

TCE - trichloroethylene

TCL - Target Compound List

TCLP - Toxicity Characteristic Leaching Procedure

TOC - Total Organic Carbon

TPH - Total Petroleum Hydrocarbons

TVOC - Total Volatile Organic Compound

mg/l - micrograms per liter

UST - Underground Storage Tank

VOC - Volatile Organic Compound

WCF - Waste Concentration Facility

Brookhaven National Laboratory (BNL) is a federal facility owned by the U.S. Department of Energy (DOE). BNL conducts research in physical, biomedical and environmental sciences and energy technologies.

BNL is located in Upton, Suffolk County, New York, about 60 miles east of New York City, near the geographic center of Long Island (Figure 1). The following are the distances to neighboring communities from BNL: Patchogue 10 miles west-southwest, Bellport 8 miles southwest, Center Moriches 7 miles southeast, Riverhead, 13 miles east; Wading River, 7 miles north-northeast; and Port Jefferson, 11 miles northwest.

The BNL property, consisting of 5,321 acres, forms an irregular polygon, and each side is approximately 2.5 miles long. Figure 2 is a current land use map of the BNL site. The developed portion of the site includes the principal facilities located near the center of the site, on relatively high ground. They are contained in an area of approximately 900 acres, 500 acres of which were originally developed by the Army. The remaining 400 acres are occupied mostly by various large research machine facilities. Outlying facilities occupy approximately 550 acres and include an apartment area, Biology Field, former Hazardous Waste Management Facility, Sewage Treatment Plant, firebreaks, and the Landfill Areas. The site's terrain is gently rolling, with elevations varying between 40 to 120 feet above sea level. The land lies on the western rim of the shallow Peconic River watershed, with a tributary of the river rising in marshy areas in the northern section of the tract.

The sole source aquifer beneath BNL has three water-bearing units: the moraine and outwash deposits, the Magothy Formation, and the Lloyd Sand Member of the Raritan Formation. These units are hydraulically connected and make up a single zone of saturation with varying physical properties extending from a depth of 45 to 1,500 feet below the land surface. These three water-bearing units are designated as a "sole-source aquifer" by the U.S. Environmental Protection Agency (EPA) and serve as the primary source of drinking-water for Nassau and Suffolk Counties.

To effectively manage remediation of the BNL site, 29 Areas of Concern (AOCs) were identified and divided into discrete groups called Operable Units (OUs), and Removal Actions. The BNL site is divided into six Operable Units (Figure 3).

Figure 4 shows the extent of OU III. It encompasses approximately 50 percent of the total area of the Laboratory. OU III was developed to address groundwater contamination in the central and southern portion of the site and in the off-site areas where groundwater contamination has migrated. Thirteen AOCs assigned to OU III were investigated as suspected source areas of groundwater contamination. Also, as the work for OU III was proceeding, groundwater contamination from other OUs and Additional Areas of Investigation (AAIs) was included in the investigation and assessment. Table 1 describes these AOCs and AAIs.

The BNL site, formerly Camp Upton, was occupied by the U.S. Army during World Wars I and II. Between the wars, the site was operated by the Civilian Conservation Corps. It was transferred to the Atomic Energy Commission in 1947, to the Energy Research and Development Administration in 1975, and to DOE in 1977.

In 1980, the BNL site was placed on New York State's Department of Environmental Conservation (NYSDEC) list of Inactive Hazardous Waste Sites. On December 21, 1989, the BNL site was included on EPA's National Priorities List because of soil and groundwater contamination that resulted from past operations of BNL. Subsequently, the EPA, NYSDEC, and DOE entered into a Federal Facilities Agreement (herein referred to as the IAG) that became effective in May, 1992 (Administrative Docket Number: II-CERCLA-FFA-00201) to coordinate cleanup activities. The IAG identified areas of concern that were grouped into operable units to be evaluated for response actions. The IAG requires a remedial investigation/feasibility study for OU III, pursuant to 42 U.S.C. 9601-9675, to meet the Comprehensive Environmental Response Compensation and Liability Act (CERCLA) requirements. The IAG also requires cleanup actions to address the identified concerns. Cleanup actions at the BNL site will be conducted pursuant to CERCLA, 40 CFR Part 300.

BNL's Final Response Strategy Document (SAIC, 1992) grouped the identified areas of concern into seven operable units. Several operable units were subsequently combined. Remedial investigations and risk assessments were conducted to evaluate the nature and extent of contamination, and potential risks associated with the areas of concern addressed in this Record of Decision. The Operable Unit III Feasibility Study Report (IT, 1999b) was prepared to evaluate the alternatives for remediating the contaminated groundwater addressed in this ROD.

Removal actions and a CERCLA-compliant Remedial Investigation/Feasibility Study (RI/FS) were identified and implemented for OU III. Removal Actions are accelerated actions to prevent, minimize, and mitigate damages to public health or the environment from a release or threatened release and/or be consistent with this final action. Table 2 summarizes these removal actions.

DOE took additional actions in OU III to remove sources of groundwater contamination. These actions include removal of contaminated soils and underground piping and cesspools and septic tanks. These actions are listed in Table 1 where each AOC in OU III is described and are also summarized in Table 3.

The Operable Unit III Remedial Investigation Report (IT, 1999a) includes an evaluation of the nature and extent of contamination, and the human-health and ecological risks associated with the contamination from thirteen AOCs in OU III, and groundwater contamination from four AOCs in OU II/VII. Two additional areas of investigation were characterized.

The Operable Unit III Feasibility Study Report (IT, 1999b) addresses the procedures used in identifying, developing, screening, and evaluating a range of remedial alternatives for the contamination in OU III.

Remedial action alternatives evaluated in the Operable Unit III Feasibility Study Report dealt with on- and off-site groundwater contaminated with VOCs (AOC 15, AOC 24A, AAI 1, and AAI 2), on-site groundwater contaminated with tritium (AOC 29), and on-site groundwater contaminated with strontium (AOC 9, AOC 10, AAI I, and AAI 2). The selected alternatives for groundwater contamination in OU III are described below and summarized in Table 2.

Several interim removal actions already have begun to address VOC contamination as part of the proposed remedy:

• A groundwater treatment system began operation in June 1997 through which VOC contaminated groundwater is extracted at the south boundary and treated by air-stripping. The goal is to prevent additional migration of the contaminated groundwater off the BNL site.
• Another groundwater treatment system began operation in September 1999 along the southern side of the Industrial Complex south of the Laboratory. This system will prevent further migration of the highest concentrations of the deep VOC plume using in-well air stripping.
• Public water was provided to an area south of BNL, and will protect public health while the groundwater cleanup is underway.
• Two underground storage tanks and contaminated soils, which are potential sources of groundwater contamination, have been removed from Building 830.

In addition to these activities, the selected remedy, Alternative V10c, includes a groundwater treatment system at BNL's Middle Road to prevent migration and further contamination of the deeper Magothy Aquifer, and to reduce the duration of remediation in the Upper Glacial Aquifer.

The selected remedy will also include a source removal system using re-circulation wells with air stripping treatment near Building 96. The final remedy for potential source areas in AOC-26B (Building 96), such as the anomalies discovered during the geophysical survey, will be documented in a subsequent Record of Decision. Finally, additional off-site groundwater treatment systems are planned to capture and treat VOCs; they will be located at the Long Island Power Authority (LIPA) right-of-way, North Street, the Brookhaven Airport, downgradient of North Street East, the eastern portion of the Industrial Park and in the western OU III low-level VOC plume. The Feasibility Study estimated approximate numbers and locations of treatment wells. However, details of the specific number of treatment systems and locations needed to meet the performance objective will be determined during the design process.

The exact number of years of active groundwater treatment needed to achieve Remedial Action Objectives will be determined based on monitoring and operating data. If, after source control is complete and effective, monitoring indicates that continued operation of the components of the selected remedy is not producing further reductions in the concentrations of contaminants in ground water, in accordance with the National Contingency Plan, DOE, NYSDEC, and EPA will evaluate whether discontinuance of the remedy is warranted. The criteria for discontinuation will include but not be limited to complete and effective source control, an evaluation of the operating conditions and parameters and a determination that the remedy has attained the feasible limit of contaminant reduction and that further reductions would be impractical.

In addition to the active groundwater treatment systems to remediate the VOCs in groundwater, this alternative requires: monitoring of new and existing wells; completion of the Building 830 Underground Storage Tank Removal Action; management of other potential source areas as shown in Table 2; and institutional control of the on-site groundwater until completion of remediation. At the request of the homeowner, DOE can arrange for monitoring of private wells used for drinking water on properties that previously have declined DOE's offer of public water hookups.

At present, limited characterization has been performed in the Magothy, so additional characterization and installation of groundwater monitoring wells are planned. This work will be done during the design of the remedy, and will be included in the site records. When this characterization and monitoring is completed, the need for a remedy for the Magothy Aquifer, will be evaluated by DOE, EPA and NYS DEC. If a remedy for the Magothy Aquifer is necessary, either this record of Decision will be modified or another decision document will establish the selected action. In either case, the public will have an opportunity to review and comment in accordance with CERCLA.

This selected remedy (V10c) is not the one proposed in the PRAP. The proposed remedy (V10b) did not include the treatment system located on-site for the western low-level VOC plume. The additional system was added in response to community and regulator concerns about potential impacts to the Carmans River.

A pump and recharge system, which includes three pumping wells located on-site along Princeton Avenue, was installed in May 1997 to extract the tritium contaminated groundwater and discharge it further north to a recharge basin on-site. Pumping at the leading edge of the plume was taken as a precautionary measure to inhibit contaminated groundwater from advancing towards the site's boundary and allow more time for the tritium to decay. A carbon filtration unit is included in the pump and recharge system to remove VOCs that are also present in the groundwater.

The selected remedy is a modification of alternative T4, as originally proposed in the PRAP. The remedy will combine extraction of groundwater in response to specific contingencies and extensive monitoring and reporting to assure that the cleanup objectives are met. Three specific contingencies were identified in the PRAP, and a fourth has been added in this ROD to address regulatory concerns. Other actions will be evaluated and implemented, as necessary, to ensure that the cleanup objectives are met. Additional monitoring wells will supplement the existing groundwater monitoring network downgradient of the High Flux Beam Reactor's (HFBR) spent fuel pool.

The first and second contingencies were developed to ensure that the tritium plume would migrate no further downgradient above drinking water standards. After an evaluation period established during design of the selected remedy, the tritium pump and recharge system on Princeton Avenue will be put on stand-by and later operated as needed as an integral component of these contingencies. The evaluation period will extend up to a maximum of one year after ROD finalization and will include an analysis of the data against the following two contingency criteria. These two specific contingencies identified are 1) to evaluate the need to reactivate the Princeton Avenue IRA if tritium concentrations exceed 25,000pCi/l at the Chilled Water Plant Road, and/or 2) reactivate the Princeton Avenue IRA if tritium concentrations exceed 20,000 pCi/l at Weaver Drive.

A third contingency was developed to ensure that if the most concentrated part of the plume were to act as a source of continuing contamination, active remediation would remove this problem. This contingency proposed a low flow extraction system to be installed in the most concentrated area of tritium contamination near the HFBR and activated if concentrations exceed 2,000,000 pCi/l at the front of the reactor. This system then would be used to remove groundwater containing the highest concentrations of tritium from the aquifer. The extracted tritium contaminated water will be disposed of offsite. Technologies to reduce the volume of water that requires off-site disposal may be identified during design. Since the PRAP was issued to the public, groundwater near the HFBR has exceeded 2,000,000 pCi/l. DOE is currently in the process of constructing some of the wells for this low flow extraction system on Cornell Avenue and developing plans to extract the most concentrated part of the plume in front of the HFBR. The detailed operational parameters for this system will be developed during design.

In addition to the ones originally identified in Alternative T4 and proposed in the PRAP, a fourth contingency, an additional low flow extraction system will be installed and operated near Temple Place. This additional system was added in response to regulatory concerns about potential plume migration. The exact location, operational parameters and treatment and disposal options for the extracted water will be developed during design. Operation of the Temple Place extraction system will continue for up to one year. As these extraction wells operate, extensive monitoring will occur to evaluate the effect of extraction locally, as well as on the entire plume. Because of the inherent uncertainties of predicating plume behavior based on groundwater modeling, the actual monitoring data will be evaluated and used to help determine whether continued operation of this extraction system is needed to achieve the cleanup objectives. The criteria to continue system operation beyond one year will be developed during design and based on the attainment of the cleanup objectives.

There are concentrated areas of strontium-90 contamination in the groundwater at three on-site locations: the Glass Holes area, the Brookhaven Graphite Research Reactor (BGRR), and the Waste Concentration Facility. Strontium-90 is a radioactive element with a half-life of 29.1 years.

The selected remedy, Alternative S5a, involves installing extraction wells and using ion exchange to remove the strontium-90 from the extracted water and on-site discharge of the clean water. Details of the specific number of treatment systems and locations needed to meet the cleanup objectives will be determined during the design process. Before implementation of the remedy, a pilot treatability study will be performed to evaluate the effectiveness of extraction and treatment. The final remedy may potentially be modified based on the results of this study. Residuals that contains strontium-90 will be disposed of off-site.

If, after source control is complete and effective, monitoring indicates that continued operation of the components of the selected remedy is not producing further reductions in the concentrations of contaminants in groundwater, in accordance with the NCP, DOE, NYSDEC, and EPA will evaluate whether discontinuance of the remedy is warranted. The criteria for discontinuation will include but not be limited to complete and effective source control, an evaluation of the operating conditions and parameters and a determination that the remedy has attained the feasible limit of contaminant reduction and that further reductions would be impractical.

Table 1 summarizes the AOCs and AAIs in OU III. A summary of inorganic, organic, and radiological contamination of groundwater, soil, and surface water before the Remedial Investigation is given in the Operable Unit III Remedial Investigation/Feasibility Study Work Plan (IT, 1994). More detailed descriptions and references are given in the Operable Unit III Remedial Investigation Report for OU III (IT 1999a).

In 1980, the BNL site was placed on NYSDEC's list of Inactive Hazardous Waste Sites. On December 21, 1989, the BNL site was included on EPA's National Priorities List (NPL). Inclusion on the NPL reflects the relative importance placed by the federal government on ensuring the expedient completion of environmental investigations and the resulting cleanup. Subsequently, the EPA, NYSDEC, and DOE entered into a Federal Facilities Agreement (herein referred to as the InterAgency Agreement; IAG) that became effective in May 1992 (Administrative Docket Number: II-CERCLA-FFA-00201). It identified AOCs to be evaluated for response actions at the BNL site. The IAG requires a Remedial Investigation/Feasibility Study to be conducted for OU III, pursuant to 42 U.S.C. 9601 et. seq., to meet CERCLA requirements. The IAG also requires the conduct of cleanup actions to address identified concerns.

BNL's Final Response Strategy Document (SAIC, 1992) grouped the identified AOCs into seven OUs; several of these were subsequently combined The OUs are in various stages of completion. Remediation at the BNL site will be conducted under CERCLA, 40 CFR Part 300.

After issuing the RODs for the remaining OUs, the necessity of a final assessment from a site-wide perspective will be determined to ensure that the ongoing or planned remedial actions will provide a comprehensive remedy for the BNL site, which is protective of human health and the environment.

A Community Relations Plan was finalized for the BNL site in September, 1991. In accordance with CERCLA Section 113 (k) (2)(B)(I-v) and 117, and the community relations plan, the community relations program focused on public information and involvement. A variety of activities provide information and seek public participation, including a stakeholders mailing list, community meetings, availability sessions, site tours, workshops, and fact sheets. An Administrative Record, documenting the basis for the selection of removal and remedial actions at the BNL site, was established and is maintained at the local libraries listed below. The Administrative Record also includes current site reports, press releases and fact sheets. The following libraries maintain the Administrative Record:

Longwood Public Library
800 Middle Country Road
Middle Island, NY 11953

Mastics-Moriches-Shirley Community Library
301 William Floyd Parkway
Shirley, NY 11967

Brookhaven National Laboratory
Research Library
Bldg. 477A
Upton, NY 11973

The Administrative Record also is kept at EPA's Region II Administrative Records Room, 290 Broadway, New York, NY, 10007-1866.

A public comment period to review the proposed remedy (Proposed Remedial Action Plan, PRAP) and the Final Operable Unit III Remedial Investigation Report and Feasibility Study Report began on March 1, 1999 and was extended through April 30, 1999. A public meeting was held on March 24, 1999 in Berkner Hall at Brookhaven National Laboratory. The Responsiveness Summary section of this document summarizes written and oral comments and DOE responses on the preferred remedial alternatives.

From the comments received during the public-comment period, DOE and BNL believe that the public and local elected officials are in general agreement with the selected remedial alternatives.

During the sixty-day comment period, 28 written comments were received on the OU III documents. The majority of them focused on general concerns, such as the length of time required for cleanup, the length of the comment period, the volume and complexity of material, and the issue of property value. Concern was also voiced about the limited characterization of groundwater in the Magothy Aquifer and the potential for human exposures to VOCs transferred to air in the VOC air stripping treatment processes. Several commentors wanted more specific information on the location of treatment wells and on the location and frequency of monitoring. There was some concern about using natural attenuation as part of the remedy, and some people felt that more active treatment in a shorter time should be undertaken. Several commentors also requested more detailed information on performance standards for the proposed treatment systems.

The Responsiveness Summary summarizes community comments on the preferred remedial alternatives.

In response to requests by stakeholders, the comment period was extended an additional 30 days.

The following modifications were made to the preferred remedial alternative based on regulators' and the public's concerns and input:

• The selected remedy for VOC contamination in groundwater for OU III (V10c) is not the one proposed by DOE in the Proposed Remedial Action Plan (PRAP). The proposed remedy (V10b) did not include the treatment system to be located in the western low-level VOC plume. The additional system was added in response to community and regulator concerns about potential impacts to the Carmans River.

• The selected remedy is a modification of alternative T4, as originally proposed in the PRAP. The remedy will combine extraction of groundwater in response to specific contingencies and extensive monitoring and reporting to assure that the cleanup objectives are met. Three specific contingencies were identified in the PRAP, and a fourth has been added in this ROD to address regulatory concerns. In addition to the ones originally identified in Alternative T4 and proposed in the PRAP, a fourth contingency, an additional low flow extraction system will be installed and operated near Temple Place. This additional system was added in response to regulatory concerns about potential plume migration. In addition to the ones originally identified in Alternative T4 and proposed in the PRAP, a fourth contingency, an additional low flow extraction system will be installed and operated near Temple Place. This additional system was added in response to regulatory concerns about potential plume migration. The exact location, operational parameters and treatment and disposal options for the extracted water will be developed during design. Operation of the Temple Place extraction system will continue for up to one year. As these extraction wells operate, extensive monitoring will occur to evaluate the effect of extraction locally, as well as on the entire plume. Because of the inherent uncertainties of predicating plume behavior based on groundwater modeling, the actual monitoring data will be evaluated and used to help determine whether continued operation of this extraction system is needed to achieve the cleanup objectives. The criteria to continue system operation beyond one year will be developed during design and based on the attainment of the cleanup objectives. Other actions will be evaluated and implemented, as necessary, to ensure that the cleanup objectives are met. Additional monitoring wells will supplement the existing groundwater monitoring network downgradient of the High Flux Beam Reactor's (HFBR) spent fuel pool.

• Community and regulator concerns were raised on the adequacy of the proposed remedy for the Magothy aquifer. As a result of continued input, the proposed remedy for the Magothy aquifer that was contained in the FS/PRAP has been removed from the ROD. Additional characterization and installation of groundwater monitoring wells are planned. After the additional characterization of the Magothy aquifer has been completed the need for a remedy for the Magothy aquifer will be evaluated by DOE, EPA, and the NYS DEC. If a remedy for the Magothy aquifer is necessary, either this Record of Decision will be modified or another decision document will establish the selected action.

• The proposed remedy in the FS/PRAP for Building 96 was air sparging/soil vapor extraction (AS/SVE). Based upon additional technical evaluation, a source removal system using re-circulation wells with air stripping treatment near Building 96 was selected as the preferred remedy for the VOC groundwater contamination for Building 96.

DOE encourages public input to ensure that the preferred remedy for Operable Unit III effectively meets community needs and protects human health and the environment. To ensure early and effective community input into this process, DOE and BNL began reaching out to the community before the Proposed Plan was released. In August and September of 1998, stakeholders were invited to participate in Community Roundtables, and canvassing of residents was conducted. In October 1998, a Community Workshop on OU III cleanup options was held. These activities are summarized in the Final Report on OU III Early Community Input (BNL, 1998c).

Community members had the opportunity to discuss their concerns directly with the BNL and DOE staff. Some of their input was incorporated into the Feasibility Study. For example, stakeholders requested consideration of an option that would complete VOC cleanup faster (in approximately ten years). This alternative was added to the list of those evaluated in the Feasibility Study. Concern was also expressed about the impact of VOCs on the Carmans River, and additional groundwater modeling was done and a new cleanup alternative developed which included possible treatment systems for the western low-level VOC plume. Stakeholder support for leaving the tritium and strontium in the ground rather than extracting it was strong, and this also affected the alternatives recommended for cleanup. This input was used to help develop and evaluate cleanup alternatives in the Feasibility Study.

During the comment period on the Proposed Plan and Feasibility Study, information sessions were held. A public meeting was held on March 24, 1999 in Berkner Hall at Brookhaven National Laboratory. Additional community relations activities included briefings to elected officials and community groups, and articles in the BNL's Environmental Restoration Division's newsletter cleanupdate.

Over 2,300 people are on the BNL mailing list. They receive the newsletter cleanupdate along with frequent mailings about specific remediation activities. Invitations to roundtables, information sessions or public meetings are often included in the mailings. BNL employees and retirees (a combined total of nearly 5,000) also receive cleanupdate and articles in the Brookhaven Bulletin which update them on specific remediation topics. The recently formed Community Advisory Council is another avenue for stakeholder groups to have access to BNL and DOE management and to learn about BNL. While the public continues to be concerned about the contamination that BNL caused and is interested in tracking the progress of cleanup, trust appears to be growing that the contamination is being addressed appropriately.

The Responsiveness Summary gives an overview of all the community relations activities for OU III.

To adequately evaluate BNL's existing and potential environmental problems, and to group these problems into workable units that could be properly scheduled and managed, the 29 AOCs were grouped into six OUs and a number of Removal Actions.

The OU III Remedial Investigation/Feasibility Study, Proposed Plan, and ROD were completed and are in the Administrative Record. Pursuant to the findings documented in the Remedial Investigation/Risk Assessment Report, Feasibility Study, and the Proposed Plan, this ROD addresses remediation of contaminated groundwater in OU III, and documents earlier actions to remediate groundwater, remove cesspools and septic tanks, connect nearby residents to public water supplies and remove sources of groundwater contamination in OU III. Conducting these remedial actions under OU III is part of BNL's overall response strategy, and is expected to be consistent with any planned future actions.

The other OUs are currently in different phases of Remedial Investigation/Feasibility Study, or remedy implementation, and have been or will be addressed in separate RODs.

The main purposes of the Remedial Investigation were to determine the nature, magnitude and extent of contamination from the AOCs included in OU III, those AOCs in OU II/VII that may be associated with groundwater contamination in OU III, and additional areas of investigation, and also to characterize the potential health risks and environmental impacts of any contaminants present. Sampling and analyses conducted during the investigation consisted of geophysical logging, radiological surveys, Geoprobe soil sampling, monitoring well borings, Geoprobe groundwater sampling, monitoring well groundwater sampling, supply well sampling, surface water sampling, and sediment sampling.

Classification of the nature and extent of soil and groundwater contamination was based on screening criteria for chemicals and radiological constituents in the various sample media. Whenever possible, established regulatory criteria, known as chemical specific Applicable or Relevant and Appropriate Requirements (ARARs) were used to screen the analytical data. This was the case for groundwater, where state and/or federal Maximum Contaminant Levels (MCLs) exist for many chemicals. In the absence of ARARs, non-enforceable regulatory guidance values, known as "to be considered" criteria, or "TBCs" were used to screen the data. This was the case for soils, which have no established state or federal ARARs. Radionuclides in soils, for which there are no individual ARAR or TBC concentrations, were screened against site-specific levels calculated using a risk model (Manual for Implementing Residual Radioactive Material Guidelines Using RESRAD (ANL, 1993)) that allowed a dose limit of 15 mrem/year above background. For chemical contamination, State (NYSDEC, 1994a) and EPA (EPA, 1994) soil cleanup guidance was used.

The screening concentrations were used to identify potential contaminant source areas, evaluate contaminant distribution patterns, and assess potential contaminant migration pathways. Table 4 summarizes media-specific screening criteria for contaminants that exceeded any screening criteria in OU III.

The OU III Remedial Investigation (RI) characterized the nature and extent of contamination in OU III's sixteen AOCs and four other identified areas that pose an unacceptable risk to human health and the environment. The field investigation for the OU III RI was conducted from October 1995 through July 1997; sampling and analysis activities consisted of the following:

• Collection and analysis of 76 Geoprobe soil samples for one or more of the following analyses: target compound list (TCL) volatile organic compounds (VOCs), TCL semivolatile organics, TCL pesticides and polychlorinated biphenyls (PCBs), target analyte list (TAL) inorganics, gamma spectroscopy, strontium-89, strontium-90, tritium, isotopic americium and thorium, gross alpha/beta activity, geotechnical parameters, and total organic carbon (TOC).
• Collection and analysis of 406 Geoprobe groundwater samples for one or more of the following analytes: TCL low detection limit (LDL) VOCs, TCL semivolatile organics, TCL pesticides and PCBs, TAL inorganics, filtered TAL inorganics, gamma spectroscopy, LDL gamma spectroscopy, strontium-89, strontium-90, isotopic americium and thorium, tritium, gross alpha/beta activity, europium-154, europium-155, radium 226/228, vanadium-238, plutonium-238, plutonium-239/240, technetium-99, cyanide, filtered cyanide, and TOC.
• Collection and analysis of 123 groundwater samples from monitoring-wells for one or more of the following analytes: TCL VOCs, TCL semivolatile organics, TCL pesticides and PCBs, TAL inorganics, gamma spectroscopy, strontium-89, strontium-90, tritium, gross alpha/beta activity, and cyanide.
• Collection and analysis of four supply-well samples for one or more of the following analytes: TCLVOCs, TCL semivolatile organics, TCL pesticides and PCBs, TAL inorganics, gamma spectroscopy, strontium-89, strontium-90, tritium, gross alpha/beta activity, and cyanide.
• Collection and analysis of three surface-water samples for TCL VOCs, TCL semivolatile organics, TCL pesticides and PCBs, TAL inorganics, gamma spectroscopy, strontium-89, strontium-90, tritium, wet chemistry parameters, and cyanide.
• Collection and analysis of nine surface-sediment samples for TCL VOCs, TCL semivolatile organics, TCL pesticides and PCBs, TAL inorganics, gamma spectroscopy, strontium-89, strontium-90, tritium, gross alpha/beta activity, geotechnical parameters, and TOC.
• Collection and analysis of two sump-water samples for TCL VOCs, gamma spectroscopy, strontium-89, strontium-90, tritium, and gross alpha/beta activity.
• Installation and sampling of 182 temporary wells on-site and off-site.

The data collected during the OU III Remedial Investigation in conjunction with additional screening surveys at BNL, and the HFBR Tritium Plume Investigation were used to assess the nature and extent of contamination in the soils, groundwater, surface water, and sediments in the OU III study area. Significant findings on the types of contaminants identified, potential sources of contamination, and the horizontal and vertical extent of contamination is summarized for each medium in the following sections. Table 4 shows the contaminants identified as being of potential concern (i.e. elevated) based on a comparison to screening levels in each media and area of concern.

To evaluate the nature and extent of contamination in surface soils, samples were taken at the Building 830 Pipe Leak and Underground Storage Tanks, the TCE Spill Area, and the Process Supply Wells and Recharge Basins AOCs. Most inorganic analytes were detected at concentrations either slightly above or below screening concentrations. Thallium and mercury were elevated in samples collected from the Building 830 area. Elevated levels of copper and manganese were detected in the recharge basins in the Process Supply Wells and Recharge Basins AOC. Volatile organic compounds and pesticides were not detected above screening levels in surface soil. PCBs were detected in surface soils above screening levels in the Building 96 area (AOC-26B). Benzo(a)pyrene was the only semi-volatile organic compound detected at a concentration more than twice the screening level, in surface soils from the TCE Spill area. Polycyclic aromatic hydrocarbons, such as benzo(a)pyrene, are commonly encountered in commercial/industrial areas, and can enter the environment in releases from truck and automobile exhausts. Cesium-137 was the only radionuclide with an activity above the screening concentration, in two samples from the Building 830 area. These contaminated soils were removed as part of an OU III Removal Action, using the soil cleanup levels developed under OU I.

Subsurface soil was sampled to determine the horizontal and vertical extent of contamination in OU III. Subsurface soil samples were collected from the Paint Shop, the Building 830 area, the Bubble Chamber Spill Area, the TCE Spill Area, Leaking Sewer Pipes, the Old Firehouse, and the Process Supply Wells and Recharge Basins AOC. The average concentrations of most analytes in the subsurface soils were below the screening concentration. Analytes detected at concentrations above screening levels were manganese, nickel, thallium, benzo(a)pyrene, and cesium-137. Manganese was elevated in subsurface samples from the recharge basins in the Process Supply Wells and Recharge Basins AOC. Nickel was elevated in samples from the Building 830 area and the Bubble Chamber Spill Area. Thallium concentrations were elevated in subsurface soil from the Paint Shop, the Building 830 area, the Bubble Chamber Spill Area, TCE Spill Area, Leaking Sewer Pipes, and the Old Firehouse. Elevated concentrations of benzo(a)pyrene were found in subsurface samples collected from the Old Firehouse. Cesium-137 and Thorium-230 were detected above screening levels in a subsurface sample collected from the area of the Building 830 USTs. These contaminated soils were removed as part of an OU III Removal Action, using soil cleanup levels developed under OU I.

Three recharge basins were sampled as part of OU III: the two basins in AOC 24C, and the recharge basin in the Bubble Chamber Spill Area. There was no evidence of contamination of the Recharge Basins from radioactive wastewater discharges. The basin in the Bubble Chamber Spill Area had elevated levels of copper and benzo(a)pyrene. Iron and copper were elevated in the two basins in AOC 24C. Volatile organics, pesticides and PCBs were not elevated in OU III surface water.

Sediment samples were taken from the recharge basins in AOC 24, an inactive cesspool associated with the Paint Shop, a recharge basin in the Bubble Chamber Spill Area, and recharge basin HT at the North End of the LINAC. Contamination was found only in Recharge Basin HT at the North End of LINAC, with elevated levels of mercury, copper, lead, silver, and zinc. A separate sample contained elevated levels of PAHs and one pesticide, delta-BHC. Radionuclides were not detected in sediments in excess of screening levels. The contamination with petroleum hydrocarbon and pesticides may be related to storm water run-off containing oils and greases from nearby asphalt paved roads and parking lots, and run-off from the LINAC area.

Groundwater sampling was conducted to define the vertical and horizontal extent of contamination in groundwater. The groundwater investigation identified the following plumes of contamination: VOCs (carbon tetrachloride, tetrachloroethene, 1,1,1-trichloroethane, trichloroethene), strontium-90, and tritium.

Volatile Organic Compounds: Carbon tetrachloride was detected at elevated levels in the deep glacial zone (60-150 feet below sea level), in a north-south direction from an area south of Princeton Avenue to an area south of Moriches-Middle Island Road. The carbon tetrachloride plume is approximately 9,500 feet long and up to 900 feet wide. The highest concentrations of carbon tetrachloride, greater than 1,000 ppb, are located between the BNL South Boundary and Carlton Drive. The highest concentration detected to date was approximately 5,100 ppb. The 1,000 ppb plume is approximately 1,500 feet long by 200 feet wide. The exact source of the contamination has not been identified, but it is suspected that it no longer exists. Potential sources for carbon tetrachloride and other contaminants are being evaluated under BNL's Facility Review and PA/SI programs.

Tetrachloroethene (PCE) was found in the vicinity of Building 96 in the water-table zone and in the deep glacial zone near the site boundary. PCE in groundwater samples ranged from 10 to 15,000 ppb. The main source of the PCE is the area immediately south of Building 96, which had been used as a truck-wash station and drum-storage area. In the water-table zone, the PCE plume is approximately 1,600 feet long by 500 feet wide. In the mid-glacial it is about 4,400 feet long by 600 feet wide. There are high concentrations of PCE (greater than 1,000 ppb) in the deep glacial zone from an area north of Princeton Avenue to the southern portion of the Industrial Park.

1,1,1-trichloroethane (TCA) was found in groundwater samples above the MCL of 5 ppb at concentrations ranging from 6 to 1,600 ppb. The two areas with most of the elevated TCA concentrations are the area just south of Building 96 in the middle of the site and the area around the Waste Concentration Facility (WCF) and Alternating Gradient Synchrotron (AGS) in the northern portion of the site. The sources of the elevated levels near the WCF and AGS probably were the cesspools associated with the Bubble Chamber Area. These cesspools contained TCA and were remediated. The TCA in the mid-glacial zone can be described as two types of contamination: high (greater than 50 ppb) and low (less than 50 ppb). The high concentrations occur at three locations between Brookhaven Avenue and South Boundary Road.

Trichloroethene (TCE) was detected in wells above the MCL of 5 ppb at levels ranging from 7 to 27 ppb, primarily in the area between Princeton Avenue and the South Boundary Road.

Because of the similarities of the VOCs found in groundwater in OU III, the horizontal and vertical extent of total volatile organic compounds (TVOC) in groundwater also were assessed. In addition to the data collected as part of the OU III RI, groundwater data were collected for the OU I/IV and for Removal Action V (RA V) located in OU I.

Figure 5 shows the areal extent of Total Volatile Organic Compounds (TVOC) in groundwater. The TVOC contamination extends from the water table to 150 feet below mean sea level. However, the TVOC plume encompasses a larger area, due to the presence of other compounds, such as carbon tetrachloride and TCA. Elevated concentrations of TVOCs are located south of Building 96, in the AGS area, in the Supply and Material Area, and south of the former landfill. TVOCs in groundwater near the AGS and Supply and Material area are being monitored. Further migration of contamination will be prevented by the existing south boundary treatment system and planned systems at Middle Road. Contamination near the former landfill is addressed under Removal Action V. Contamination near Building 96 is addressed under this ROD.

The highest concentrations of TVOC in the mid-glacial zone (greater than 50 ppb) appear as slugs, or discrete areas of contamination, at three locations between Brookhaven Avenue and the South Boundary Road. The high concentration TVOC plume in the deep-glacial zone extends from north of Rowland Street to the downgradient extent of the carbon tetrachloride plume located between Moriches-Middle Island Road and Crestwood Avenue. The deep-glacial plume is approximately 14,000 feet long and up to 2,000 feet wide.

There are two locations of OU I/IV VOC plumes in the mid-glacial zone. The first is on-site, south of Brookhaven Avenue and North of Princeton Avenue; the second is off BNL with elevated concentrations within a localized area along Sleepy Hollow Drive. The Removal Action V (RAV) VOC plume within the mid-glacial zone is located off site just south of the RAV extraction wells. The plume is approximately 3,000 feet long and extends from the Long Island Expressway to the south. The highest concentration of VOCs detected was 258 ppb of TCA.

Strontium-90: Strontium-90 was detected above the MCL of 8 picoCuries per liter (pCi/l) at concentrations ranging from 8.45 to 566 pCi/l. The highest activities (i.e. 566 pCi/l) were observed during the Pile Fan Sump (PFS) groundwater sampling. Most strontium-90 in groundwater is associated with two areas on-site: the Brookhaven Graphite Research Reactor (BGRR), and the Waste Concentration Facility (WCF). There are two distinct strontium-90-contaminated plumes (Figure 6), one around the BGRR, WCF, and PFS, and the other around the Glass Holes.

The plume south of the BGRR is approximately 1,000 feet long and 500 feet wide. The larger of the two Strontium-90 plumes actually is composed of two plumes, the northern half composed of Strontium-90 originating from the WCF and associated tanks and pipelines, and the southern originating from the BGRR Pile Fan Sump area. The larger WCF/Pile Fan Sump Plume is approximately 2,000 feet long and 500 feet wide.

Tritium: Elevated concentrations of tritium were detected downgradient of the High Flux Beam Reactor (HFBR). The source of this tritium was the HFBR Spent Fuel Pool, which was emptied in December 1997. The highest activity was 2,290,000 pCi/l in a monitoring well directly in front of the HFBR (IT, 1999c) ; tritium activity at the downgradient edge of the plume is between 1,000 and 5,000 pCi/l. The tritium plume is located entirely within the boundaries of the Laboratory. The portion of the plume that exceeds the MCL for tritium (20,000 pCi/l) extends approximately 4,500 feet north of BNL's southern boundary at depths from 40 to 150 feet below land surface. The dimensions of the 1,000 pCi/l plume are approximately 3,200 feet long and 625 feet wide. The 20,000 pCi/l plume is approximately 2,600 feet long and 250 feet wide. A second area immediately north of the HFBR stack has tritium concentrations greater than the drinking water. Figure 7 shows the extent of the tritium plume on-site.

Two separate groups of contaminants were identified as potentially of concern in OU III: the groundwater contaminants, and the sediment, soil, and surface-water contaminants. In general, the contaminants in groundwater at OU III are relatively mobile, having moderate to high water-solubility and/or low K_OC values. The majority of the contaminants identified as potentially of concern in soils, sediments, and surface waters at OU III exhibit relatively low- water-solubilities and/or high K_OC values and, therefore, have low leachabilities and low mobilities in groundwater. Also, most of the contaminants detected in near surface areas (i.e., surface soils, surface water) are not highly volatile. Almost all of the contaminants exhibit a strong tendency to adsorb to soil particles and remain relatively immobile in the soils as demonstrated by their high K_OC and K_d values.

The fate of a constituent in the environment is a function of its chemical properties and the physical nature of the site. The potential for environmental transport was examined based on a review of the topographic and hydrogeologic characteristics of the site and a review of the available physical constants and chemical characteristics of each constituent. The most significant fate and transport processes for the study area are summarized below:

• The greatest potential for transport of contaminants at the OU III site is via groundwater transport. Volatile organic compounds, including PCE, TCA, and carbon tetrachloride, have been detected in groundwater plumes indicating their ongoing transport. Also, the radionuclides strontium-90 (which has limited mobility) and tritium (which is mobile and moves with water) were detected in groundwater.
• Volatilization, dust generation, and air transport are considered insignificant based on the extremely limited surface-soil and surface-water contamination.
• Most constituents present in vadose soils and sediments on site are relatively insoluble and have a greater tendency to remain adsorbed. Thus, leaching to groundwater is not expected to be important. Those contaminants considered leachable have already been detected in groundwater.
• From the limited biota expected to occur in the recharge basins or in association with the recharge basins, bioaccumulation of constituents in the food chain is expected to be an insignificant transport pathway.

Several actions have been taken to remove sources of groundwater contamination (Table 3).

Additional actions were taken to remove potential sources of groundwater contamination at other locations on-site; these include the landfills removal action, removal of cesspools and cesspool contents, removal of underground storage-tanks, and replacement of leaking sewer-pipes.

BNL has embarked on an extensive Facility Site Review to identify potential release- points of contaminants from BNL's facilities to the environment. The review began in April 1997 and is an important element of BNL's comprehensive plan to delineate and characterize environmental issues at the site and to develop strategies for cleanup and remediation. The purpose of the project was to review all BNL facilities to identify equipment, operations and activities that have the potential to degrade groundwater. The Facility Site Review categorized facilities as either Priority I or Priority II, based upon previous uses and the age of the facility.

Priority I facilities are those that used or generated significant quantities of radioactive material during the 1950s and 1960s. In addition, facilities that have a history of major programmatic changes during operational periods are considered Priority I facilities. Facilities that do not meet the criteria for Priority I status were designated Priority II.

Twelve Priority I and eight Priority II action items were identified for the Environmental Restoration Division (ERD) during the Facility Site Review. All action items identified in the Facility Site Review will be tracked by representatives of BNL's Environmental Safety and Health Services (ES&HS) until closeout reports have been prepared.

Preliminary Assessment /Site Inspection (PA/SI) investigation were developed to evaluate areas of interest identified in the April 1997 Facility Review. The PA/SI consisted of a field investigation that included collecting and analyzing soil and groundwater samples. The results of this investigation will be used to determine if an identified area should be considered an AOC. Follow-up activities from the Facility Review are continuing.

The following interim removal actions (IRAs) and Removal Actions have been or are being undertaken to immediately reduce concentrations, migration, or exposure to groundwater contaminants:

• On-site OU III Southern Boundary Groundwater Interim Removal Action (IRA): This IRA was implemented in June 1997 in response to the detection of a plume of VOCs in groundwater both on- and off-site. The goal is to prevent additional off-site migration of VOCs in the most concentrated part of the plume at the southern boundary. The IRA consists of a groundwater recovery system at the southern boundary (Figure 8), extraction of groundwater through six wells, and treatment through air-stripping. The clean water then is discharged to a recharge basin. The locations of wells for this pump-and-treat system are shown in Figure 9.
• Off-site OU III Industrial Complex Groundwater Interim Removal Action: This IRA is being implemented to address the off-site migration of the highest concentrations of the deep VOC plume beyond the industrial complex located south of OU III. Its objective is to hydraulically control, extract, and treat groundwater through in-well air-stripping, using an array of seven wells along the southern side of the industrial complex. This interim removal action began in September 1999.
• Off-site Public Water Hookup Interim Action: To ensure protection of the health of the residents located downgradient of OU III and OU I, a residential public-water hookup was established (Figure 10). Public water was provided to homes potentially in the path of contaminated groundwater associated with BNL. Long-term monitoring of groundwater off-site will be conducted.
• Tritium Groundwater Interim Removal Action: This IRA was implemented in response to elevated levels of tritium detected downgradient of the HFBR. It consists of (1) removing of spent fuel from the pool and installation of a stainless steel liner, (2) elimination of potential sources of leakage, and (3) pumping groundwater at the leading edge of the plume. At the time, tritium concentrations were expected to decrease over time and not cross the BNL boundary, three wells were installed along Princeton Avenue as a precautionary measure to extract tritium-contaminated groundwater. The groundwater is treated for chemical contamination by carbon adsorption and discharged to a recharge basin to allow additional time for the tritium to decay. This system began operation in May 1997. A schematic of the re-circulation system is shown in Figure 11.
• Building 830 Underground Storage Tanks Removal Action: This action has removed two out-of-service underground storage tanks, a concrete valve pit, associated piping and contaminated soils and vegetation. The tanks have been removed and the soils have been excavated. Contaminated wastes are in the process of being shipped off site for disposal.
• Interim Removal Action V: This Removal Action addresses on-site groundwater associated with the Current Landfill and former Hazardous Waste Management Facility, both located in OU I. A pump-and-treat system was installed at the south boundary in December 1996 to intercept groundwater containing VOCs migrating from the two source areas and prevent them from moving off-site. The system includes two extraction wells and an air-stripping tower. The clean water is recharged via the RA V recharge basin in the center of the BNL site.
• Carbon Tetrachloride Tank Removal Action. The tank was removed and a removal action is underway to pump-and-treat carbon tetrachloride in groundwater in the immediate vicinity of the former tank.

Based on the results of the OU III RI, the primary concerns associated with the OU III study area are groundwater contamination by VOCs, tritium, and strontium-90. A detailed analysis of alternatives was conducted in the OU III Feasibility Study for onsite groundwater contamination by strontium, on- and off-site groundwater contamination by TVOCs, and on-site groundwater contamination by tritium. Soil contamination with Cesium-137 found in AOC 11/12 (Building 830 Pipe Leak and Tanks) was addressed under an Interim Removal Action.

A baseline risk assessment was done to estimate the human health and ecological risks that could result from exposure to contaminants in OU III if no remediation is performed beyond that accomplished to date. Present and future potential exposures to chemical and radiological contaminants in groundwater, surface water, sediment, soil and subsurface soil were evaluated. The risk assessment is documented in the OU III Remedial Investigation Report (IT, 1999a).

Data collected from the four AAIs were not included in the risk assessment because cleanup actions are underway (Table 2).

A four-step process was used to assess site-related human health risks assuming a reasonable maximum exposure scenario:

• Hazard Identification: identifies the contaminants of concern based upon factors such as toxicity, frequency of occurrence, and concentration.
• Exposure Assessment: estimates the magnitude of actual and/or potential human exposures, their frequency and duration, and exposure pathways (e.g., external exposure from gamma radiation, ingestion of contaminated well water) by which humans could be exposed.
• Toxicity Assessment: determines the types of adverse health effects associated with exposures and the relationship between the magnitude of exposure (dose) and severity of adverse effects (response).
• Risk Characterization: summarizes and combines outputs of the exposure and toxicity assessments to quantitatively assess site-related risks.

Two kinds of human health hazards were addressed in the risk assessment for Operable Unit III: cancer induction and non-carcinogenic toxicity.

Cancer Risk is expressed in terms of the probability that a given human receptor will develop cancer due to estimated exposures over a 70-year lifetime. The current federal acceptable risk range for individual lifetime excess carcinogenic risk is one-in-ten-thousand to one-in-one-million.

Non-carcinogenic effect risks due to Operable Unit III contaminants were estimated by dividing the intake of a chemical by the acceptable intake over the period of exposure. These non-carcinogenic effects are expressed as Hazard Indices (HI). A Hazard Index greater than 1.0 indicates a potential for non-carcinogenic health effects. The maximum acceptable HI is 1.0.

The baseline risk assessment evaluated the health effects that could result from exposure to chemical and radiological contamination in groundwater, surface water, soil and sediment as a result of dermal contact, inhalation, and ingestion associated with current and potential future land uses.

The risk assessment focused on contaminants that are likely to pose significant risks to human health; they are summarized in Table 5. Six inorganic constituents, 16 radionuclides, and 8 organics were identified as chemicals of potential concern.

The baseline risk assessment addressed potential risks to human health by identifying potential pathways by which people may be exposed to contaminants at the site under current and future land-use conditions. Tables 6 and 7 summarize the exposure scenarios evaluated in this baseline risk assessment. The reasonable maximum exposure scenario was evaluated.

The populations exposed under the current land-use scenario were assumed to be on-site industrial workers and an on-site trespasser. The current on-site worker was assumed to perform routine daily activities in OU III, and soil-related exposure pathways were analyzed (inhalation of resuspended soil, incidental ingestion of soil, and dermal contact with soil). Potential exposure of onsite workers or trespassers to subsurface soil was not considered because there is no construction work involving excavation currently in progress in OU III. Occupational exposures to surface water and sediment were not considered since BNL personnel are not routinely exposed to surface water and sediment in the recharge basins during their daily work assignments.

For an older child on-site trespasser, five exposure pathways were evaluated in the current land use exposure assessment: inhalation of resuspended soil; incidental ingestion of soil; direct dermal contact with soil; direct dermal contact with surface water; and, direct dermal contact with sediment. Trespassers were assumed not to ingest surface water or sediment since the recharge basins are too shallow for swimming or wading. The risk assessment for radionuclides did not include the pathways for dermal contact with sediment and surface water because the exposures were orders of magnitude smaller than those for ingestion and inhalation.

Ingestion of on-site groundwater also was not included in the exposure assessment for current use. Wells on-site are constantly monitored for contamination and, if necessary, the groundwater is treated to remove it.

Off-site residences were offered connections to the public-water supply, but a few have elected not to make this connection. The baseline risk assessment evaluated risk to off-site populations from exposure to contaminants in groundwater.

Three hypothetically exposed populations were identified for potential future exposures: a short-term construction worker, an industrial worker, and a resident. The future land use scenario was conservatively assumed to occur after 30 years. The radionuclide risk assessment also estimated risks at 50, 100 and 1000 years.

The short-term construction worker was assumed to be exposed through inhalation of soil particulates and dust, incidental ingestion of soil, and dermal contact with soil. Three soil-related exposure pathways were assumed for the future industrial worker: inhalation of particulates and dust; incidental ingestion of soil; and dermal contact with soil.

Residential exposures were evaluated for both an adult and a young child (age 0-6 years). This conservative scenario assumed that a resident would live on-site for 30 years and use on-site groundwater for all domestic water needs. The exposure pathways were: inhalation of soil; incidental ingestion of soil; dermal contact with soil; ingestion, dermal contact (bathing; chemicals only) and inhalation (showering; chemicals only) of groundwater. Because radionuclides may bioaccumulate in plants and animals, the radiological risk assessment included ingestion of home-grown vegetables and of game and livestock as a potential exposure pathway.

Two human health hazards were addressed in the risk assessment for Operable Unit III: cancer induction and non-carcinogenic toxicity. Tables 8, 9 and 10 summarize the non-carcinogenic and carcinogenic toxicity values for the contaminants of concern.

EPA developed reference doses for indicating the potential for adverse health effects. Reference Doses (RfDs), expressed in units of milligrams/kilogram-day (mg/kg-day), are estimates of daily exposure levels for humans that are thought to be safe over a lifetime.

Cancer slope-factors were developed by EPA for estimating excess lifetime cancer risk associated with exposure to potentially carcinogenic chemicals. Slope factors are expressed in units of (mg/kg-day)^-1.

In the toxicity assessment the toxicological properties of the selected chemicals of potential concern were summarized. Many carcinogenic slope-factors and reference doses were obtained from EPA's Integrated Risk Information System database. Slope-factors and reference doses/concentrations not available in that database were obtained from EPA's second most current source of toxicity information, the Health Effects Assessment Summary Tables. When toxicity values were not available for a specific chemical, the chemical was evaluated qualitatively. Uncertainties related to the chemical toxicity data were also addressed. Some toxicity values in the risk assessment are extremely conservative estimates and include uncertainty factors that may reduce the estimated safe exposure concentrations by up to 1000 times.

Non-carcinogenic risks were assessed using a hazard index (HI) approach, based on a comparison of expected intakes of the contaminant and safe levels of intake (RfD, Reference Doses, Table 8). Estimated intakes of chemicals from environmental media (e.g. the amount of a chemical ingested from contaminated drinking water) are compared to the RfD to derive the hazard quotient for the contaminant in the particular medium. The HI is obtained by adding the hazard quotients for all compounds across all media that impact a particular population. An HI greater than 1.0 indicates that the potential exists for noncarcinogenic health effects to occur from site-related exposures. The HI is a useful reference point for gauging the potential significance of multiple contaminant exposures within a single medium or across media.

Potential carcinogenic risks were evaluated using the cancer slope-factors (Tables 9 and 10) developed by EPA. Slope-factors are multiplied by the estimated intake of a potential carcinogen to generate an upper-bound estimate of the excess lifetime cancer risk associated with exposure to the compound at that level of intake.

For known or suspected carcinogens, EPA considers excess upper-bound individual lifetime cancer risks of between 10^-4 to 10^-6 to be acceptable. This level indicates that an individual has not greater than a one-in-ten-thousand to one-in-one-million chance of developing cancer as a result of site-related exposures to a carcinogen over a 70-year period under specific exposure conditions.

The non-carcinogenic and carcinogenic risks associated with the chemical contaminants of concern at the site that exceed EPA's acceptable levels are summarized in Table 11.

Under current land use, the cumulative carcinogenic risk is 2 x 10^-6 for both an on-site worker and an older child as an on-site trespasser. These risks are within the EPA's acceptable cancer risk range (1 x 10^-4 to 1 x 10^-6). The total cumulative non-carcinogenic hazards to the on-site worker and on-site trespasser were negligible (0.08 and <0.01, respectively) compared to the acceptable HI value of 1.

The carcinogenic risk from carbon tetrachloride to the current adult and young child off-site resident exposed to the maximum concentrations measured in groundwater were 8 x 10^-3 and 4 x 10^-3, respectively. These values exceed the acceptable cancer risk range. The non-carcinogenic health hazard from carbon tetrachloride for the adult and young child off-site resident exposed to the maximum concentrations measured in groundwater were 200 and 470, respectively, both of which exceed EPA guidance levels. TCA is not a human carcinogen and there is no EPA published value for non-carcinogenic risk; thus, the risks associated with current land use exposure cannot be quantitatively estimated for off-site residents. However, the maximum concentration of TCA measured off-site (100 &micro;g/l) is 20 times the maximum contaminant level (5 &micro;g/l). Thus, the presence of TCA and carbon tetrachloride plumes in off-site groundwater could present a public health concern to the few off-site residents who declined access to publicly supplied water.

Under the future land-use conditions, the total chemical carcinogenic risks for a future on-site industrial or construction fell within or below the EPA acceptable risk range of 1 x 10^-4 to 1 x 10^-6, and the acceptable HI of one. The risks to the future residential child and adult were slightly above the EPA's target risk range. This risk is driven by arsenic, for which the risks are over-estimated. The slope factor for arsenic (1.5 mg/kg-day)^-1 is an overestimate. Uncertainties in the study used to derive this value include the likelihood of a non-linear dose-response relationship, problems with exposure estimates, and differences in protein intake levels which may result in a differential susceptibility to arsenic. Several epidemiological studies in the United States have found no association between skin cancer and arsenic in drinking water.

The non-carcinogenic hazard index for the hypothetical future on-site resident adult and young child were estimated to be 3.4 and 8.5, respectively. Ingestion of manganese in groundwater contributed the most hazard to the HI.

Manganese (Mn) is a ubiquitous element that is essential for normal physiologic functioning in all animal species, including humans. The National Research Council recommends a provisional daily dietary Mn intake for adults of 2.0 to 5.0 mg. The EPA established reference dose for Mn is 10 mg/day (0.14 mg/kg-day for a 70-kg adult) for chronic human consumption of Mn in the diet with an uncertainty factor of 3.

If conservative assumptions were made for OU III that a hypothetical future resident uses the groundwater at OU III as the sole water supply and drinks 2 L/day of water from wells, then, based on the 95% UCL of 1,173 &micro;g/L, the Mn intake can be calculated to be 0.034 mg/kg-day. This Mn intake of 0.034 mg/kg-day is much less than the EPA established RfD of 0.14 mg/kg-day for Mn. Even if based on the maximum detected Mn concentration, the calculated Mn intake is 0.195 mg/kg-day, which is only slightly higher than the EPA established reference dose of 0.14 mg/kg-day; this should not be a concern. The reference dose is estimated to be an intake for the general population, including sensitive subpopulations, that is likely to be without an appreciable risk of deleterious effects during a lifetime. In addition, the reported Mn concentrations were obtained from the unfiltered groundwater samples that contain more Mn than the filtered groundwater samples. The filtered groundwater would be more representative of drinking water conditions. Mn in groundwater is, therefore, not considered a concern for human health.

An additional risk assessment was done for the future receptors, assuming exposure to the VOC groundwater plumes identified in OU III (TCA, PCE and carbon tetrachloride). The conservative assumption was made that in the future (30 years) houses would be built near the highest detected concentrations of these on-site plumes, and the residents would use the residential wells as the sole water supply for domestic uses. The risk to a future resident using groundwater at the highest concentration of carbon tetrachloride and PCE exceeds the acceptable risk range. Estimated risks to an adult from exposure to carbon tetrachloride and PCE in groundwater were 6 x 10^-4 and 5 x 10^-3 respectively. Estimated risks to a child from exposure to carbon tetrachloride and PCE in groundwater were 3 x 10^-4 and 2 x 10^-3. Under this highly unlikely scenario, the presence of TCA, PCE and carbon tetrachloride plumes in groundwater on-site could pose a potential health concern for a future resident.

The non-carcinogenic HI for a future on-site residential adult who would be exposed to carbon tetrachloride and PCE at the maximum detected concentrations was estimated to be 14 and 20, respectively. The HI's for the future on-site residential child are 33 and 48. These calculated non-carcinogenic HIs exceed EPA's acceptable HI of 1. TCA risks to a future resident using water from the on-site plumes were not calculated quantitatively because there are no EPA established toxicity values for TCA. However, the maximum concentration of TCA in the on-site plume was 920 &micro;g/l, which is almost 200 times the MCL (5 &micro;g/l). Under this highly unlikely scenario, the presence of TCA, PCE and carbon tetrachloride plumes in groundwater on-site could pose a potential health concern for a future on-site resident.

The carcinogenic risk from carbon tetrachloride for the adult and young child off-site resident exposed to the maximum concentrations measured in groundwater in the future were assumed to be the same as for the current off-site risk assessment (8 x 10^-3 adult, 4 x 10^-3 child). The non-carcinogenic HI's for the future off-site adult and child were 200 and 470, respectively. Thus, the presence of TCA and carbon tetrachloride plumes in off-site groundwater in the future could present a public health concern to the few off-site residents who declined access to publicly supplied water.

Table 12 summarizes the results of the radiological baseline risk assessment for contaminants that exceeded the acceptable risk range.

The radiological risk analyses conducted found that under current land-use conditions, cancer risks for industrial workers at 1, 30 and 50 years from now were 4 x 10^-4, 3 x 10^-4 , and 1 x 10^-4, respectively. These risks are slightly above the acceptable risk range of 1 x 10^-4 to 1 x 10^-6. For the on-site trespasser, risks at 1, 30, and 50 years from now were 4 x 10^-5, 1 x 10^-5, and 6 x 10^-6, which fall below the acceptable risk range. External gamma exposure was the dominant pathway, and the major contributing radionuclides were Cs-137 and Co-60.

The conservative future land-use scenario assumed an on-site resident who was nearly self-sufficient in terms of raising or harvesting a significant portion of their diet from the OU III site. The calculated risk for this unlikely scenario suggests that OU III would pose potential cancer risks slightly above the acceptable risk range to a future on-site population (3 x 10^-4 at year 30 and 1 x 10^-4 at year 50). The major contributing pathway is exposure to external gamma from radionuclides in soil. For the future industrial worker, risk at year 30, is 1 x 10^-4. Risks to industrial workers at years 50 and 100 were below the acceptable risk range. The risk to a short-term construction worker involved in excavation activities in year 30 and beyond was very small (2 x 10^-7 in year 30, 8 x 10^-8 in year 50).

An additional risk assessment was done for the future on-site risk assessment, assuming exposure to the highest concentrations of tritium and strontium-90 measured in groundwater in OU III . The conservative assumption was made that future (30 years) residential houses would be built near the highest detected concentrations of these on-site plumes, and the residents would use the residential wells as the sole water supply for domestic uses. Cancer risks to an on-site resident via the groundwater ingestion pathway for strontium-90 was 1 x 10^-4, and for tritium 2 x 10^-3, which are at or above the acceptable risk range.

Because a few residents off-site elected not to be connected to the public-water supply, the risks to an off-site resident were evaluated. The calculated risk for an off-site resident exceeded EPA's recommended level.

The Ecological Risk Assessment determined whether historical activities at Operable Unit III resulted in levels of chemical and radiological contamination that could adversely affect the ecosystems there.

A standard ecological risk assessment (as prescribed by the EPA) consists of a four-step process used for assessing related ecological risks for a reasonable maximum exposure scenario:

• Problem Formulation: a qualitative evaluation of a contaminant's release, migration, and fate; identification of contaminants of concern, receptors, exposure pathways, and known ecological effects of the contaminants; and, selection of endpoints for further study.
• Exposure Assessment: a quantitative evaluation of the release, migration, and fate of the contaminant; characterization of exposure pathways, and receptors; and measurement or estimation of concentrations at exposure points.
• Ecological Effects Assessment: literature reviews, field studies and toxicity tests linking contaminant concentrations to effects on ecological receptors.
• Risk Characterization: measurement or estimation of current and future adverse effects.

Table 13 shows the potential chemicals of concern for the ecological risk assessment.

Unlike assessments of human-health risk which are concerned with effects on individuals, assessments of ecological risk focus on wildlife population and ecosystem-level effects. Because there is little toxicity data relevant to wildlife, it is difficult to draw inferences at the population-and ecosystems-level. Thus, the ecological assessment for OU III was largely qualitative.

The soil contamination to which terrestrial organisms could be exposed was limited to two small areas: one area at the TCE Soil Area is in a building courtyard that is virtually inaccessible to wildlife, and the other area occupies very limited surface area within the developed portions of OU III at the Building 830 Underground Storage Tank area. Therefore, the exposure of terrestrial wildlife to soil contaminants is insignificant.

From comparing surface-water concentrations in the Recharge Basins to available New York State surface water standards, the screening risk assessment indicated that the most significant potential risks to aquatic communities are due to copper in all three recharge basins investigated (HT at the North End of LINAC, HN01, and HN02). In addition, cadmium concentrations in Recharge basin HN01 were elevated. This analysis is very conservative. The risk was estimated by comparing criteria for dissolved metals to a total measured metal concentration, which will necessarily overestimate risk. In addition, New York State Class D surface water-body standards were used as a screening benchmark. The habitat potential of the recharge basins is very limited due to low water levels, the intermittent presence of water, high temperatures and low dissolved oxygen. Recharge basins are not expected to function as Class D water bodies, and therefore, the risk to aquatic biota is not significant.

The potential risk to the benthic community was most significant in Recharge Basin HT, located at the north end of the LINAC. Mercury, copper, silver and several PAHs were more than an order of magnitude greater than the sediment quality criteria applied. Mercury posed a marginal risk in all other recharge basins. However, the benthic community expected in recharge basins is limited by the habitat. Applying sediment criteria to recharge basins overestimates the risk to the community that could occur there, and risk is expected to be minimal.

Consumption of surface water from the recharge basins by terrestrial animals was also evaluated. Surface water concentrations of contaminants were orders-of-magnitude less than the target species (cottontail rabbit) drinking water no-observed effect level.

Remedial action objectives (RAOs), or "cleanup objectives," are specific goals to protect human health and the environment. These objectives are based on available information standards, such as applicable or relevant and appropriate requirements (ARARs), and risk-based levels established in the risk assessment. After evaluating the nature and extent of contamination in soils, groundwater, surface water, and sediment, and assessing the chemical and radiological risks associated with exposure to contaminants of potential concern, the following RAOs were developed:

• Meet the drinking water standards in groundwater for VOCs, strontium-90 and tritium.
• Complete cleanup of the groundwater in a timely manner. For the Upper Glacial Aquifer, this goal is 30 years or less.
• Prevent or minimize further migration of VOC, Sr-90 and tritium in groundwater.

The selected remedies will prevent further migration of high concentrations of contaminants in groundwater.

If, after source control is complete and effective, monitoring indicates that continued operation of the components of the selected remedy is not producing further reductions in the concentrations of contaminants in ground water, in accordance with the National Contingency Plan, DOE, NYSDEC, and EPA will evaluate whether discontinuance of the remedy is warranted. The criteria for discontinuation will include complete and effective source control, an evaluation of the operating conditions and parameters as well as a determination that the remedy has attained the feasible limit of contaminant reduction and that further reductions would be impractical.

From the results of the Remedial Investigation and Baseline Risk Assessment, it was determined that contaminants in all environmental media, except groundwater, posed minimal risk to human health and the environment. Soil contamination that exceeded screening levels in the Remedial Investigation study did not present important risks to human or ecological health with one exception; the soils contaminated with Cesium-137 at Building 830's underground storage tanks. This soil has been excavated under a Removal Action. It should be noted that many sources of contaminated soil and sediment not included under OU III already have been remediated.

Residents immediately south of the Laboratory were offered a hookup to public water supplies, eliminating the potential source of exposure to, and risk from groundwater contaminants. However, some residents elected not to be connected to public water, or still use well-water for various purposes, like watering a garden and filling a swimming pool. The human-health risk assessment found that VOCs in groundwater could present a public health concern for the few off-site residents who declined publicly supplied water. These homes will be monitored at the request of the homeowner.

The following contaminated groundwater plumes were identified to be of concern:

• On-site groundwater contaminated with strontium;
• On-site groundwater contaminated with tritium; and
• On-site and off-site groundwater contaminated with VOCs.

The remedial action alternatives presented in the Feasibility Study and this Proposal Plan address these plumes. In addition, seven interim removal actions (IRAs) were undertaken to immediately reduce concentrations, impact, migration, or exposure to groundwater contaminants.

The primary contaminants identified in groundwater were carbon tetrachloride, tetrachloroethene, trichloroethane, strontium-90, and tritium. Groundwater contamination in OU III was separated into four areas according to the type and location of contaminants.

These four areas are 1) the on-site TVOC area which includes the TVOC present in the water table and Upper Glacial aquifer on BNL; 2) the off-site TVOC area which includes contamination in the water table, Upper Glacial aquifer, and Magothy aquifer off-site and south of BNL; 3) the strontium-90 contamination in the water-table zone present at the BGRR/WCF and the Glass Holes area; and, 4) the tritium plume in the vicinity of the HFBR.

CERCLA requires that each site remedy be protective of human health and the environment, be cost effective, comply with other statutory laws, and utilize permanent solutions and alternative treatment technologies and resource recovery alternatives to the maximum extent practicable. In addition, the statute includes a preference for using treatment as a principal element for reducing the toxicity, mobility and volume of hazardous substances.

Remedial action alternatives evaluated in the Operable Unit III Feasibility Study Report addressed on-site groundwater contaminated with strontium, on-site groundwater contaminated with tritium, and on- and off-site groundwater contaminated with VOCs. The following alternatives were retained for detailed analysis in the Feasibility Study Report.

Most alternatives to remediate VOCs in groundwater use in-well air-stripping systems or other appropriate technologies in combinations of different locations. Figure 12 is a schematic of a typical in-well air-stripping system. Possible locations for off-site treatment systems include the Long Island Power Authority (LIPA) right of way, North Street, the Eastern Portion of the Industrial Park, and two locations at the northern end of the Brookhaven Airport.

All alternatives (except the No Action Alternative) also assume a groundwater treatment system on the BNL site at Middle Road, and at Building 96, continued operation of the south boundary pump-and-treat system, and completion and operation of the Industrial Park in-well air-stripping system, all of which will help prevent further migration of contaminants into the deeper Magothy Aquifer. All the alternatives rely on natural attenuation to reduce concentrations and include extensive monitoring and modeling of the plume over time.

The number of wells selected for each alternative was based on available characterization and hydrogeological data. The actual number of wells used in the selected alternative will be identified during the design phase. Alternatives investigated in detail to remediate VOCs in groundwater are described below. Table 14 summarizes the costs and time to meet Remedial Action Objectives. Because not all of the alternatives originally identified in the Feasibility Study were evaluated in detail, the alternatives listed below are not all numbered sequentially.

The no action alternative includes no remedial activities for site-wide VOC contamination. In accordance with the National Contingency Plan, the No Action Alternative must be assessed and compared to the other alternatives.

Under this alternative, VOC contamination in groundwater will be remediated through the continued operation of three IRAs: the Southern Boundary IRA treatment system; the Off-site Industrial Complex IRA; and, the Off-site Public Water Hookup Interim Action. This alternative also includes a source removal system using re-circulation wells with air stripping treatment near Building 96. Additional reductions in on- and off-site concentrations of VOCs in groundwater will be achieved through natural attenuation. Natural attenuation occurs when physical, chemical and biological processes reduce the mass, toxicity and mobility of subsurface contamination in a way that reduces risk to human health and the environment to acceptable levels. Installing new monitoring wells, and groundwater monitoring and modeling will be required to evaluate the possibility of impacting potential receptors, such as surface-water bodies, supply wells, and potable wells.

This alternative involves actively remediating on-site and off-site VOC contamination. It includes the on-site systems in alternative V3: the operation of the on-site and off-site IRAs, installation of an in-well air-stripping system at Middle Road and a source removal system in the vicinity of Building 96. This alternative also involves installing in-well air-stripping systems at five locations off-site: the Long Island Power Authority (LIPA) Right-of-Way, Brookhaven Airport, North Street/Sleepy Hollow Drive, near North Street in the OU I RAV plume, and within the western OU III low-level VOC plume. Based on the installation, system operation, modeling, and pre-design data, the specific number of treatment systems and locations needed to meet the performance objective may be modified during the design process. The exact number of years of pumping needed to achieve Remedial Action Objectives will be determined based on monitoring and operating data. Additional monitoring wells are planned and sampling and analysis will be conducted. The goal of this alternative is to reduce further migration of the VOC plume south of the off-site sub-systems.

This alternative includes all the components of Alternative V7, with an additional well in the OUs I/IV Industrial Complex (East) and without the in-well air-stripping wells in the western OU III low-level VOC plume. This alternative involves actively remediating both on-site and off-site VOC contamination. It includes the following on-site systems: operation of the on-site and off-site IRAs, installation of an in-well air-stripping system at Middle Road, and installation of a source-removal system near Building 96. This alternative also involves installing in-well air-stripping systems at five locations off-site: 1 well in the industrial park east, 3 in-well air-stripping well at the LIPA Right-of-Way, 7 wells at Brookhaven Airport, 4 at North Street/Sleepy Hollow Drive, and 1 near North Street in the OU I RAV plume. The goal is to reduce further VOC plume migration south of the off-site sub-systems.

This alternative involves active remediation of both on-site and off-site VOC contamination. It includes the following systems: operation of the on-site and off-site IRAs, including the On-Site Southern Boundary IRA and the Off-Site Industrial Complex IRA; installation of new in-well air stripping systems at the LIPA right-of-way, North Street, the Brookhaven Airport, downgradient of North Street East, the eastern portion of the industrial park; additional treatment systems on-site at Middle Road and in the western OU III low-level VOC plume; and a source removal system using re-circulation wells with air stripping treatment near Building 96. The Brookhaven Airport containment systems, and the OU III and OUI/IV hot spot containment systems will be identical to the Alternatives V10b. The objective of this alternative is to capture and contain the OU III, OUI/IV, and RAV plume in a similar well configuration as alternative V10b in addition to capturing and containing of the western low level VOC plume. The purpose is prevent or reduce the levels at which this low level VOC plume migrates and discharges to the Carmans River. Details of the specific number of treatment systems and locations needed to meet the performance objective will be determined during the design process. The exact number of years of pumping needed to achieve Remedial Action Objectives will be determined based on monitoring and operating data.

This alternative involves active remediation of both on-site and off-site VOC contamination. It includes the following on-site systems: operation of the on-site and off-site IRAs, installation of an in-well air-stripping system at Middle Road, and the installation of a source-removal system near Building 96. This alternative also involves the installation of in-well air-stripping systems off-site: 1 well in the industrial park east, 10 wells at Brookhaven Airport, and 4 at North Street/Sleepy Hollow Drive. The goal is to reduce further migration of the VOC plume south of the off-site sub-systems. This alternative has no treatment at the LIPA right-of-way and, therefore, has more treatment wells located down-gradient at the Airport.

The configuration for this alternative is identical to that of Alternative V10b. Groundwater collected by all the extraction wells will be pumped via piping to a treatment system located on-site, treated by an air-stripper to remove volatiles, and discharged to the OU III basin. This alternative includes the following on-site systems: operation of the on-site and off-site IRAs, installation of extraction wells at Middle Road, and installation of a source-removal system near Building 96. This alternative also involves installing extraction wells at locations off-site: 1 well in the industrial park east, 1 well at the LIPA Right-of-Way, 7 wells at Brookhaven Airport, and 4 at North Street/Sleepy Hollow Drive. The goal is to reduce further VOC-plume migration south of the off-site sub-systems.

Alternatives investigated in detail to remediate strontium-90 in groundwater are described below. Table 15 summarizes the costs and time to meet Remedial Action Objectives. Because not all of the alternatives originally identified in the Feasibility Study were evaluated in detail, the alternatives listed below are not all numbered sequentially.

The no action alternative has no remedial activities. In accordance with the National Contingency Plan, the No Action Alternative must be assessed for comparison to the other alternatives.

Under this alternative, the Sr-90 contamination in the water-table zone near the Brookhaven Graphite Research Reactor, Waste Concentration Facility, and Pile Fan Sump (BGRR/WCF/PFS) is slowly reduced through natural attenuation without any control, removal, treatment, or other mitigating actions. Modeling and monitoring of groundwater is required to evaluate the possibility of impacting potential receptors, such as surface-water bodies, supply-and potable-wells. The monitoring program involves installing new wells to monitor the extent and boundaries of the plumes.

In this innovative alternative, a two step in-situ chemical precipitation process is used to contain the strontium-90 plume. In the first step, solutions containing dissolved phosphate are forced through the groundwater and soil, via injection wells, to react with the strontium contaminants, and convert them to more insoluble compounds. Phosphate salts of strontium are very insoluble. In the second step, solutions of lime are injected into the aquifer. This forms calcium hydroxyapatite (a calcium phosphate), which can co-precipitate or adsorb the strontium. Continued groundwater monitoring would be a part of this alternative.

This alternative includes extracting groundwater from two wells within the BGRR/WCF/PFS plume, and one well downgradient of well 106-16 located south of the Glass Holes area. At each location (BGRR area and Glass Holes area), a system will be installed to treat the using ion-exchange before recharge to an on-site recharge basin. Figure 13 shows a schematic of the proposed Sr-90 ion-exchange system. Ion-exchange resin will be disposed of off-site. The BGRR and WCF pumps would operate for 25 to 30 years and the Glass Holes pumps for 8 years. Continued groundwater monitoring also would be a part of this alternative.

Under this alternative, the WCF/BGRR/PFS strontium plume will be remediated utilizing two extraction wells with groundwater treatment via ion exchange, similar to Alternative S5a. However, the Glass Holes strontium plume remediation will be accomplished using a permeable reactive barrier. The permeable reactive walls will consist of a 3-foot-thick bed of granular clinoptilolite. As the groundwater flows through this zeolite mineral, strontium will be absorbed on the bed. Continued groundwater monitoring would also be a part of this alternative.

Remedial alternatives are being developed for different sections of the tritium plume. Of special interest is the "hot-spot" area of the plume, located along the downgradient edge of the HFBR Building footprint. Several alternatives address containment or removal of this highly contaminated groundwater, including ones that address the leading edge of the 20,000 pCi/l tritium plume.

A tritium Interim Removal Action (IRA) system is operating that recovers approximately 120 gallons per minute from three wells located along Princeton Avenue. The groundwater is treated by carbon adsorption to remove VOCs and discharged to the RA V recharge basin. Because the HFBR spent-fuel pool was emptied, no additional source of tritium exists.

Alternatives investigated in detail to remediate tritium in groundwater are described below, and the costs and time to meet Remedial Action Objectives are summarized in Table 16.

The No Action alternative provides a comparative baseline against which to evaluate other alternatives. Under this alternative, no remedial action will occur and the contamination will be left "as is," without any control, removal, treatment, or other mitigating actions. Long term monitoring and modeling will not be performed for the No Action alternative.

This alternative will consist of natural attenuation with the deactivation of the tritium IRA at Princeton Avenue. Natural attenuation is the process by which concentrations of tritium decrease in the groundwater by diffusion, dilution, and radioactive decay. The natural attenuation process can effectively reduce the contaminant's toxicity, mobility, or volume to levels that are protective of human health and the environment. This option requires groundwater modeling, and evaluating the contaminant's degradation rates and pathways. The primary objective of modeling is to demonstrate that natural processes of decay can reduce concentrations to levels below regulatory standards. Sampling and analyses must be conducted throughout the natural attenuation process to confirm that degradation is proceeding at rates consistent with those predicted through groundwater modeling. The monitoring program will involve at a minimum, 88 existing monitoring wells. Additional monitoring wells are being planned. The wells will be sampled and analyzed for tritium quarterly for five years and annually for the following 20 years. The 20-year time frame is a conservative estimate.

This alternative is the same as Alternative T2, except it includes the continued operation of the tritium IRA. This option requires modeling, and evaluating the contaminant degradation rates. Sampling and analyses must be conducted throughout the natural attenuation process to confirm that degradation is proceeding at rates consistent with those predicted through groundwater modeling. The monitoring program will involve, at a minimum, 88 existing monitoring wells. Additional monitoring wells are currently being planned. The wells will be sampled and analyzed for tritium quarterly for five years and annually for the following 15 years.

This alternative includes monitored natural attenuation with a contingency remedy to address tritium contamination in groundwater. Additional monitoring wells will supplement the existing groundwater monitoring network downgradient of the High Flux Beam Reactor's spent-fuel pool. After an evaluation period established during design of the selected remedy and consultation with EPA and NYSDEC, the tritium pump and recharge system on Princeton Avenue will be put on stand-by and operated as needed as an integral component of the contingency remedy for the tritium plume. The evaluation period will extend up to a maximum of one year after ROD finalization. The tritium plume will be monitored to ensure that natural attenuation is achieving the remedial action objectives. If the tritium plume is not attenuating as expected, one or more contingencies will be implemented to assure the remedial action objectives. Specific contingencies identified are 1) to evaluate the need to reactivate the Princeton Avenue IRA if tritium concentrations exceed 25,000 pCi/l at the Chilled Water Plant Road, and/or 2) reactivate the Princeton Avenue IRA if tritium concentrations exceed 20,000 pCi/l at Weaver Drive. The exact method of determining when these levels have been exceeded, including the number of confirmation samples, will be determined during the design phase. A low-flow extraction system will be installed in the most concentrated area of tritium contamination near the HFBR and activated if concentrations exceed 2,000,000 pCi/l at the front of the reactor. The 2,000,000 pCi/l value incorporates a 25 percent safety factor over the maximum value of 1,600,000 pCi/l that was detected during the remedial investigation to account for uncertainties in sampling and analysis. This system then would be used to remove groundwater containing the highest concentrations of tritium from the aquifer. The exact method of determining when these levels have been exceeded, including the number of confirmation samples, will be determined during the design phase. The extracted tritiated water will be disposed of off-site. Additional monitoring wells will be installed at the HFBR and included in the existing network.

This alternative will actively contain the tritium plume with concentrations above 20,000 pCi/l. It includes extracting groundwater at the furthest downgradient portion of the 20,000 pCi/l plume and recirculating the extracted groundwater to the RA V recharge basin. This alternative is similar to the current tritium IRA, except for the location of the extraction wells. It assumes that the tritium IRA will be placed in standby mode.

This alternative uses two extraction wells, pumping at very low rates, to contain and capture the highest concentrations of tritium at the downgradient edge of the plume. The goal is to decrease the extent of the entire tritium plume, its migration, and the duration of time to achieve 20,000 pCi/l concentration, given a one-year focused tritium hot-spot removal action. Two extraction wells will be installed directly downgradient of the HFBR pumping 1 gpm each. The extraction wells will operate for one year and will remove a total of 1.05 million gallons of groundwater. The recovered groundwater will be pumped and stored in a 1.2 million-gallon above-ground tank for approximately 50 years, until the concentration of tritium naturally decays to activities below drinking water standards (20,000 pCi/l). The groundwater then will be pumped to recharge basin RA V where it will percolate through the soil column into the water table. The monitoring program will involve, at a minimum, 88 existing monitoring wells. Additional monitoring wells are being planned. This alternative assumes that the tritium IRA will be placed in standby mode.

This alternative includes installing the same groundwater extraction system discussed in Alternative T6. However, instead of on-site storage, the tritiated groundwater will be evaporated off-site. The extracted groundwater will be transferred directly to a 20,000 gallon feed-tank, and then into tanker trucks that will be transported to a treatment facility for evaporation. No residuals will result from this treatment. This alternative assumes that the tritium IRA will be placed in standby mode.

This alternative includes the installation of the same groundwater extraction system as discussed in Alternative T6. However, instead of on-site storage, tritium will be evaporated into the atmosphere using an existing evaporator. It will evaporate the tritiated water to the atmosphere from a stack 70 feet from the base of its skid. No residuals will be produced from this process. This alternative assumes that the tritium IRA will be placed in standby mode.

CERCLA guidance requires that each remedial alternative identified in the Feasibility Study be compared according to nine criteria. Those criteria are subdivided into three categories:

1. threshold criteria that relate directly to statutory findings and must be satisfied by each chosen alternative;
2. primary balancing criteria that include long- and short-term effectiveness, implementability, reduction of toxicity, mobility, volume, and cost; and
3. modifying criteria that measure the acceptability of the alternatives to state agencies and the community.

DOE identified its preferred remedy by evaluating all of the alternatives against EPA's nine evaluation criteria.

Each alternative was evaluated against the following seven criteria (1) overall protection of human health and the environment, (2) compliance with ARARs, (3) long-term effectiveness and permanence, (4) reduction of toxicity, mobility, and volume, (5) short-term effectiveness, (6) implementability, (7) cost, (8) state acceptance and (9) community acceptance. To the maximum extent practical, CERCLA requires that remedial action alternatives must 1) be protective of human health and the environment, (2) attain ARARs, (3) be cost effective, (4) utilize permanent solutions and alternative treatment technologies, and (5) reduce toxicity, mobility, or volume.

Overall Protection of Human Health and the Environment addresses whether an alternative provides adequate protection, and describes how risks are eliminated, reduced, or controlled through treatment, engineering controls, or institutional controls.

Compliance with Applicable or Relevant and Appropriate Requirements (ARARs) considers if a remedy meets all federal and state ARARs, including provisions for invoking a waiver.

Once an alternative satisfies the threshold criteria, five balancing criteria are used to evaluate other aspects of the remedial alternatives. The balancing criteria are 1) long-term effectiveness and permanence; (2) reduction of toxicity, mobility, or volume through treatment; (3) short-term effectiveness; (4) implementability; and (5) cost.

Long-Term Effectiveness addresses the amount of remaining risk, and the ability of an alternative to protect human health and the environment over time, once cleanup goals have been met.

Reduction of Toxicity, Mobility, or Volume addresses the anticipated performance of treatment that permanently and significantly reduces the toxicity, mobility, or volume of waste.

Short-Term Effectiveness and Environmental Impacts addresses the impact to the community and site-workers during construction or implementation, and includes the time needed to finish work.

Implementability addresses the technical and administrative feasibility of an alternative, including the availability of materials and services required for cleanup.

Cost compares the differences in cost, including capital, operation, and maintenance costs. Estimates are based on present-day costs and are highly uncertain.

The modifying criteria are used in the final evaluation of remedial alternatives: state and community acceptance. For both, the factors that are considered include the elements of the alternatives that are supported, the elements those that are not supported, and elements of the alternatives that have strong opposition.

State Acceptance addresses whether the State agrees with, opposes, or has no comment on the preferred alternative.

Community Acceptance addresses the issues and concerns the public may have regarding each of the alternatives.

A detailed comparative analysis of all alternatives is provided in the Feasibility Study. A summary of comparative analysis of alternatives, based upon the evaluation criteria noted above, is given below. This detailed evaluation of alternatives was done only for strontium contamination in groundwater; tritium contamination in groundwater, and TVOCs in groundwater.

Once each of the alternatives was individually evaluated against the seven criteria, a comparative analysis of alternatives was conducted. A brief summary of the comparative analysis of alternatives is provided below. Tables 17, 18, and 19 summarize the comparative analyses of alternatives. Cost estimates are given in Table 20.

For groundwater contaminated with VOCs, seven alternatives were evaluated in detail. The alternatives include natural attenuation to address all or portions of plume which might not be directly influenced by an active remedial system. This remedial approach is cost-effective and efficient for restoring VOC-contaminated groundwater.

All alternatives except V1 (No Action) include operating the Southern Boundary System, Industrial Complex IRA system, and a source removal near Building 96. Capital- and operating-costs for these three items also were included for each alternative to represent the total cost of remediation of the VOCs. The majority of alternatives that include additional treatment use in-well air-stripping to further treat VOC contaminated groundwater. Alternative V13 uses traditional pump-and-treat technology to capture, contain, and treat groundwater on- and off-site.

Due to the depth to contaminants in the groundwater, the type of contaminants, and type of geology, only two types of groundwater extraction technologies were used to develop alternatives; groundwater extraction wells and in-situ in-well air-stripping. Treatment technologies evaluated included air-stripping, carbon-adsorption, and UV-oxidation.

If, after source control is complete and effective, monitoring indicates that continued operation of the components of the selected remedy is not producing further reductions in the concentrations of contaminants in ground water, in accordance with the NCP, DOE, NYSDEC, and EPA will evaluate whether discontinuance of the remedy is warranted. The criteria for discontinuation will include complete and effective source control, an evaluation of the operating conditions and parameters as well as a determination that the remedy has attained the feasible limit of contaminant reduction and that further reductions would be impractical.

The alternatives (except V1, No Action) focus on restoring the Upper Glacial aquifer due to the higher velocity of groundwater, more potential receptors, and increased potential for plume growth and migration. The remediation of the Upper Glacial aquifer will also reduce VOCs migration into the Magothy resulting in faster cleanup of the deeper aquifer. Additional characterization and monitoring of the Magothy aquifer will be conducted to allow evaluation of the need for a remedy for the Magothy aquifer.

Several alternatives (V7, V10b, V10c, V11, V13) include installing treatment wells at the downgradient edge of the VOC plume at Brookhaven Airport. These wells reduce the plume's migration south of Flower Hill Drive.

Two alternatives (V7 and V10c) also have remedial subsystems, which address the low-level VOC plume, present to the west of the main plume. They attempt to reduce the migration and plume growth of the low levels of VOC which eventually discharge to the Carmans River.

All alternatives with the exception of V1 include the operation of the southern boundary treatment system, off-site Industrial Complex IRA, the hookup of residential homes to public water downgradient of the BNL site, installation and monitoring of additional on-site and off-site wells and VOC source removal at Building 96. Therefore, all of the alternatives with the exception of V1 will provide a degree of protection of human health and the environment by minimizing exposure pathways.

Alternatives with off-site treatment (V7, V10b, V10c, V11, V13) provide for the protection of human health and the environment because they offer a high reduction in contaminant concentrations and mobility. These alternatives improve overall protection of human health and the environment by removing the contaminants from off-site groundwater to RAOs and by allowing contaminant levels in the aquifer to reach MCLs over time by natural attenuation. Alternatives V10b and V10c provide the greatest amount of protection through the reduction of contaminants both on and off-site of BNL and result in compliance with ARARs in 30 years.

All treatment alternatives require long periods of time to remediate (25 years to greater than 30 years). In the case of Alternative V1 and V2, contaminated groundwater will continue to migrate, and protection of human health and the environment will not be achieved. However, through the implementation of a risk management program including groundwater monitoring, residential well monitoring, public water hookups, and a natural attenuation remedial plan, risks posed by the VOCs to human health and the environment may be minimized.

For those alternatives that implement off-site remediation, groundwater modeling indicates that the VOC contaminants will discharge to the Carmans River at low concentrations (i.e., less than 5 &micro;g/l). Carmans River discharges as a result of the OU III and OU I/IV plumes are less than 1 &micro;g/l. Carmans River discharges as a result of the low-level VOC plume is less than 5 &micro;g/l. The VOC discharge levels to the Carmans River should likely be reduced or prevented as part of Alternatives V7 and V10c which incorporate a treatment system for the OU III low level VOC plume.

Alternatives with air emissions will be evaluated for compliance with appropriate air regulations (Air Guide-1). On-site treatment systems not passing air discharge screening will include off-gas treatment prior to discharge. All off-site treatment systems will include off-gas treatment using carbon filters.

Alternatives V1 and V2 do not achieve chemical-specific ARARs for the on-site and off-site VOC contamination since Federal MCLs and state groundwater standards will be exceeded for the next 30 years. Alternatives V10b, V10c, and V13 remediate the groundwater in the Upper Glacial aquifer within 30 years, and are therefore in compliance with ARARs and RAOs. Alternative V7 does not comply with chemical-specific ARARs because VOC concentrations exceeding the MCL still persist within the Upper Glacial aquifer after 30 years.

Alternatives requiring discharge of water or air will comply with chemical-specific and action-specific ARARs at the discharge point through engineering controls and monitoring.

Alternative V1 will not significantly reduce concentrations of contaminants nor limit the mobility of the contaminated groundwater migration. All other alternatives actively treat the groundwater. Contaminant migration, plume growth, and VOC discharge levels to the Carmans River are the highest in this alternative compared to all other alternatives.

Alternative V2 includes operating the on-site and off-site IRAs along with an air-stripping/soil vapor extraction system for the VOC source-area. This baseline alternative results in no further impacts to the aquifer from VOCs at the source area, as well controlling the migration of the VOC plume at the boundary and at the Industrial Complex. Groundwater modeling shows that the MCL is not reached on-site and off-site of BNL within 30 years. However, significant reductions in the plume's extent are observed.

Model simulations indicate that Alternatives V7, V10b, V10c, V11 and V13 will prove very effective in long-term reduction of the contaminant's concentrations and mobility due to the intensive remedial effort applied. These alternatives are the most effective in removing and reducing VOC concentrations in the aquifers.

Alternative V11, which involves treatment through in-well stripping wells placed in non-residential areas, requires longer to achieve the MCL than other alternatives (except for V7) that include off-site wells within residential areas. Alternatives that include remediation wells sited within plume's hot spots, regardless of residential areas, have accelerated schedules and an effective remediation of the Upper Glacial aquifer in 30 years.

Alternative V7 is the least aggressive. This alternative includes the reduction and capture of VOCs within the OU III low-level VOC plume and RA V plume. It also includes installing Brookhaven Airport wells for prevention of migration beyond Flower Hill Drive. It still results in the greatest migration of VOC contaminants for the OU III and OU I/IV off-site plumes, with levels above the MCL persisting within the Upper Glacial aquifer after 30 years. Although not simulated, Alternatives V7 and V10c would likely result in the lowest levels of VOC discharge to the Carmans River due to the OU III low-level plume systems.

All alternatives except V1 include operating the southern boundary treatment system, off-site Industrial Complex IRA, installing and monitoring additional on-site and off-site monitoring wells for natural attenuation and source removal at Building 96. Groundwater modeling showed that these remedial components alone reduce contaminant volume and mobility and will prevent further migration of high concentrations of VOCs past the property line. Natural attenuation will significantly reduce contaminant concentrations in the aquifer for the on-site VOC plume during the 30-years of operation of the southern boundary treatment system. However, by placing additional recovery wells on-site and off-site in addition to the IRAs, the remedial strategy is accelerated.

Off-site treatment in Alternatives V7, V10b, V10c, V11 and V13 effectively reduces the toxicity, mobility, and volume of off-site VOCs, and prevents significant migration. Potential discharges to the Carmans River are also reduced. However, even with the aggressive off-site treatment in Alternatives V10b, V11 and V13, small discharges (less than 5 &micro;g/l) are simulated to occur within 30 years. These discharges may be further reduced or prevented by installing of an OU III low-level VOC plume treatment system, as in Alternatives V7 and V10c. If the alternatives provided no off-site treatment, groundwater modeling has shown that VOCs at concentrations between 5 &micro;g/l and 15 &micro;g/l may enter the Carmans River.

Natural attenuation, a component of all the alternatives, reduces contaminants by natural means over a period of time.

Alternatives V10b, V10c, and V13 restore the Upper Glacial aquifer to the MCL in approximately 30 years, and result in the greatest extent of reduction in the contaminant's toxicity, mobility, and volume. Alternative V10c may result in lower levels of VOC discharge to the Carmans River than Alternatives V10b and V13 because of the OU III low level VOC plume treatment system. Therefore, these alternatives comply with the RAOs discussed in this ROD. The amount of time required for Alternative V7 to restore the Upper Glacial aquifer exceeds 30 years.

Alternative V1 does not include any major active remediation and, therefore, presents the least risk to the community or workers.

Alternative V2, natural attenuation, represents the baseline for the VOC alternatives. It contains the installation of an air-stripping/soil vapor extraction system at building 96, a suspected source-area for VOCs, and installing and monitoring of additional on-site and off-site wells to assess natural attenuation. Alternative V2 also includes the installation and operation of an off-site IRA along with operation and maintenance of the Southern Boundary IRA treatment system. The operation of the IRAs, air-stripping/soil vapor extraction system, and a natural attenuation program are remedial components in all the alternatives except for Alternative V1. These components pose some risks of exposing on-site workers to VOCs through dermal contact, ingestion, and inhalation during construction activities and system operation. However, exposure can be prevented by using proper personal protection equipment. All alternatives except V1 produce process residuals, such as spent carbon requiring proper handling.

The alternatives involving in-well stripping (V7, V10b, and V10c, and V11) provide the least short-term risk to workers and to the community during installation because in-well air-stripping systems require less extensive construction, minimal contact with groundwater, and generate fewer process residuals. However, potential emissions and noise from off-site air-stripping systems located in residential areas may be a concern for the community. These impacts will be minimized by engineering controls such as off-gas treatment and enclosures.

Alternative V13 includes extracting groundwater for treatment, and its discharge. This alternative has some risk to on-site workers through dermal contact, ingestion, and inhalation of contaminants during construction and system operation. These risks can be reduced by using proper personal protection equipment and trained personnel.

From a technical standpoint, all of the alternatives can be implemented. Pump-and-treat and in-well stripping technologies have been demonstrated either on-site or at other contaminated sites. Equipment, contractors, and venders required to implement the alternatives are available. In-well air-stripping was demonstrated by field pilot tests at BNL to be effective in reducing contaminants to discharge standards.

Administratively, implementation of off-site alternatives will be difficult due to regulatory approval, public acceptance, and the requirements for property access for installing the off-site treatment systems. Alternative V13 will be the most difficult to implement administratively because it will involve installing underground piping through major roadways (e.g., Long Island Expressway), residential areas and industrial areas. The in-well air-stripping systems will require LIPA approval for implementation.

All remedial alternatives will require compliance with Air-Guide-1 air-discharge limits for the air-strippers, the in-well stripping systems, and the air-stripping/soil vapor extraction system. Compliance can be easily demonstrated by the use of off-gas treatment where appropriate. Off-gas treatment system (carbon adsorption) has been proposed at systems with high VOC contamination for wells located in residential areas.

Alternative V1, No Action, is the easiest to implement because it requires no construction, remedial or monitoring actions.

All VOC alternatives (except for V1) include costs for installing and operating the southern boundary and Industrial Complex IRA systems, and source removal at Building 96, as well as for natural attenuation/groundwater monitoring.

Southern boundary costs were included because on-site recovery at Middle Road affected the net present-worth cost of the southern boundary treatment system by influencing operating time frames. Implementation of on-site recovery at Middle Road reduced the cost of operating the southern boundary treatment system by its system's operating duration from 30 years to 15 years. However, the total remedial cost, including additional on-site treatment at Middle Road, was higher than the total cost for operating the southern boundary system alone for 30 years (alternative V2). Table 20 summarizes the capital, operation, and maintenance, and total net present worth cost for each of the alternatives.

The costs of alternatives V10b and V10c, are comparable. Alternatives V10b and V13, are similar in well configuration to one another. The VOC alternative with the lowest cost is alternative V2 (natural attenuation), and the alternative with the highest cost is V7.

New York State, based on its review of the Feasibility Study and the Proposed Plan, has concurred with the selected alternative (V10c).

Written and verbal comments received from the community during the public comment period and at the public meeting held on March 24, 1999 have been evaluated. From the comments received during the public-comment period, DOE and BNL believe that the public and local elected officials are in general agreement with the selected remedial alternatives. The Responsiveness Summary Section of this Record of Decision contains the comments from the community and DOE's responses. In response to community concerns, the selected remedy (V10c) includes treatment systems to be located in the western low-level VOC plume that were not part of the originally proposed remedy (V10b). These additional systems were added in response to the concerns of the community and regulators about potential impacts to the Carmans River.

Groundwater strontium contamination was detected around the BGRR, WCF, PFS, and the Glass Holes area. Five remedial alternatives were evaluated in detail to address the groundwater strontium contamination. In-situ technologies included in situ chemical precipitation (Alternative S4), and reactive permeable barrier (Alternative S7). Other remedial technologies evaluated included ion exchange (Alternative S5a).

The natural attenuation alternative S2 is protective of human health and environment because of the slow migration rate of the strontium in groundwater. No receptors are impacted for the duration of the remedial alternative, 60 years. However, this alternative does not result in compliance with RAOs within 30 years. Pump and treat alternative S5a is effective in removing strontium from the aquifer, and results in compliance with RAOs within 30 years.

In-situ technologies use containment as means of addressing the strontium contamination in the groundwater. These technologies prevent any further migration and rely on radioactive decay to comply with RAOs. However, because strontium is not very mobile in the aquifer and because of the flat groundwater gradient around the BGRR, these technologies are not cost effective and do not result in compliance with RAOs in a timely manner.

The Natural Attenuation alternative is protective of human health or the environment, over the long-term since groundwater modeling simulations show no impact to any potential receptors. However, the Natural Attenuation alternative requires 60 years to naturally decay to ARARs. The No Action alternative cannot address the future protection of human health and the environment due to the lack of long-term monitoring and modeling data.

The In situ Precipitation Alternative S4 provides added protection to the environment since the mobility of the Sr-90 is reduced.

Alternatives S5a and S7 are also protective of human health and the environment by remediating the Sr-90 contaminated groundwater to the MCL within 30 years.

Protection of human health can be measured by both the impact of the remediation scheme to the aquifer and the environment and the impact of the consequences of the remedial alternative. Although Alternatives S5a and S7 result in restoration of the aquifer, the potential exposure to contaminants has been increased. These alternatives also result in the generation of radiological waste that must be managed, transported, and disposed off-site which also increases potential exposure. Because of the low mobility of Sr-90 at both the Glass Holes and WCF/PFS areas, no potential impacts to receptors is anticipated, and all Sr-90 contamination remains within the boundaries of BNL.

Alternatives S5a and S7, as well as the In situ Precipitation Alternative S4 will comply with chemical specific ARARs within 30 years at all locations impacted by Sr-90; therefore, these alternatives comply with RAOs for the restoration of the aquifer within 30 years. In the Natural Attenuation Alternative S2 and the No Action Alternative S1, ARARs are not achieved within 30 years. However, the Glass Holes area reaches the MCL in approximately 40 to 50 years, and the WCF/PFS area reaches the MCL in approximately 60 to 70 years. However, plume mobility and growth are negligible at all locations during this time period.

Through proper design and permitting of Alternatives S5a and S7, the discharge of the treated Sr-90 will comply with all applicable chemical-specific and action-specific ARARs through the use of proven technology and proper design.

Alternatives S1 and S2 will reduce contaminant concentrations through natural processes of decay, dilution and adsorption. Alternative S4 enhances adsorption to prevent migration. The groundwater transport modeling shows no impact on potable water wells or BNL supply wells over a 60-year period. The concentration of Sr-90 within the plumes are, over the long-term, reduced.

For Alternatives S5a and S7, concentrations of Sr-90 are reduced to the MCL within 30 years at WCF/PFS area. In addition for Alternative S5a cleanup of the Glass Holes area will occur within 8 years. The treatment systems will generate residual waste as a result of the ion exchange technology. The residuals will be managed and disposed as low level radioactive waste. Exposure to waste can be minimized through proper training of personal and the use of personal protection equipment to reduce long-term risk of exposure. For Alternative S7, the PRB will adsorb Sr-90 from the plume for at least 25 years. It will remove Sr-90 from the groundwater and immobilize the adsorbed Sr-90 within the wall until it radioactively decays.

Alternative S1 does not use treatment or containment to reduce toxicity, mobility or volume of the Sr-90 in the groundwater. In Alternative S2, the Sr-90 concentrations will be reduced by natural processes of decay, dilution and adsorption. Minimal contaminant migration is supported by groundwater modeling and long-term monitoring in Alternative S2. Transport through the aquifer is minimal and the bulk of the contamination remains within the same area in the water table zone of the Upper Glacial aquifer.

Alternative S4 and the permeable reactive wall component Alternative S7 reduce the mobility of the Sr-90 by treatment. Reduction in toxicity and volume are achieved by natural decay.

Alternative S5a and the WCF/PFS component of the Alternative S7 use extraction and treatment systems to reduce the toxicity, mobility and volume of contaminated groundwater to below the 8 pCi/l MCL. The Sr-90 removed from the groundwater will be adsorbed on the zeolite unit within the groundwater treatment system and will be transported off-site for disposal. Since this material is disposed of off-site rather than into an on-site landfill, the Sr-90 is permanently removed from the site.

No significant advantages are observed in plume migration for Alternative S5a over Alternative S2 or S4 because of the low mobility of the Sr-90.

The No Action Alternative (S1) does not include any disturbance, site access or use of the site and therefore does not present the community or on-site workers with any additional risks resulting from potential release of contaminants.

Alternative S2, Natural Attenuation, consists of allowing the natural processes of radioactive decay, diffusion, dilution and adsorption to reduce the concentration of Sr-90 in the groundwater to the acceptable level. Short-term risks are limited to possible worker exposure to contaminated soil and groundwater during installation of monitoring wells and groundwater sampling. In the short term, this alternative contains minimal exposure to Sr-90 contamination to the construction worker or the community.

Alternative S4, the In situ Precipitation alternative, will require drilling of 55 injection wells and mixing and injection of immobilization chemicals into the aquifer to trap the Sr-90. Some risk exists from the construction activities and from the drill cuttings. Accidents and exposure can be prevented with proper training, and appropriate protective equipment.

Alternative S5a, Groundwater Extraction/Ion Exchange/On-site Discharge, the "pump and treat" remedy, involves extraction of groundwater from two areas of high Sr-90 concentration (WCF/PFS and Glass Holes), and treatment and discharge to on-site recharge basins. This alternative results in immediate control of the migration of the highest Sr-90 concentration areas and reduces Sr-90 concentrations in the aquifer. Installation of this system presents some risk to on-site workers through dermal contact, ingestion or inhalation of groundwater and/or soils during construction and O&M activities. However, exposure can be prevented by using proper personal protection equipment.

Alternative S7, Groundwater Extraction/Ion Exchange/On-site Discharge/Permeable Reactive Wall, the "Hybrid" alternative, includes the "pump and treat" system at the WCF/PFS area and the installation of a PRB at the Glass Holes area. This alternative requires the management of over 2,000 cubic yards of excavated soil, including up to 1,000 yards of soil from the aquifer that may contain some radioactivity. This soil must be managed to prevent exposure to construction workers and to prevent migration of dust.

The No Action Alternative S1 will be easily implemented since no action is required. The Natural Attenuation Alternative S2 will require a public awareness and monitoring program, both of which can be easily implemented. The In situ Precipitation Alternative S4 requires drilling of injection wells and mixing and injection of chemicals to immobilize the Sr-90. All activities associated with this alternative are readily available and proven, although avoiding underground utilities and pipelines in the WCF/PFS area will require extensive planning and some survey activities. A treatability study will be conducted to confirm the parameters necessary for stabilization of the Sr-90.

Alternative S5a will require the construction of treatment systems, extraction systems and discharge lines. The treatment equipment required are readily proven and commercially available. A treatability study will be conducted during the remedial design activities to confirm that Sr-90 loading and removal kinetics are as expected and that the 8 pCi/l MCL can be met. Sampling for treatment effectiveness and groundwater monitoring will also be required and can be easily implemented.

Installing the permeable reactive wall for Alternative S7 will be difficult. This alternative requires not only the excavation of an 80 foot deep trench under slurry and the placement of 1,039 cubic yards of clinoptilolite in the trench, but also management of over 2,000 cubic yards of excavated soil, including up to 1,000 yards of soil from the aquifer that may contain some radioactivity. This soil must be managed to prevent exposure to construction workers and to prevent release of dust to the atmosphere.

Costs for these alternatives are summarized in Table 20. There are no costs associated with Alternative S1. Installation of monitoring wells, groundwater sampling and analysis, and groundwater modeling are included in all other alternatives.

Although Alternatives S4 and S5a cost an additional $1,000,000 to $5,000,000 over Alternative S2, and restore the groundwater resource faster by reducing the mass of contaminants, no significant advantages are observed in implementing these alternatives over natural attenuation due to the low mobility of Sr-90 and lack of receptor impact.

New York State, based on its review of the Feasibility Study and the Proposed Plan, has concurred with the selected alternative (S5a).

Written and verbal comments received from the community during the public comment period and at the public meeting held on March 24, 1999 have been evaluated. From the comments received during the public-comment period, DOE and BNL believe that the public and local elected officials are in general agreement with the selected remedial alternatives. The Responsiveness Summary Section of this Record of Decision contains the comments from the community and DOE's responses.

The HFBR spent fuel pool tritium plume extends from the HFBR to Princeton Avenue and is currently being remediated with the Princeton Avenue IRA system. A total of eight remedial alternatives for the tritium plume were evaluated in detail. They include variations of natural attenuation (T2, T3, T4), and hot spot removal at the reactor (T6, T7, T8).

Hot spot extraction alternatives presents three methods of managing the tritium contaminated water: 1) on-site evaporation 2) off-site evaporation 3) on-site storage. No cost effective treatment technologies are available for the removal of tritium from groundwater. Therefore, no treatment alternatives were carried forward for a detailed analysis.

Groundwater modeling results for natural attenuation without the current IRA system indicated that the current IRA has little to no effect on the tritium plume and does not result in a shorter remediation timeframe for the plume. Therefore, most of the alternatives assumed that the Princeton IRA system would not be in operation.

Alternative T4 is based on natural attenuation of the tritium plume with contingency pumping based remediation at the HFBR and at Princeton Avenue. The contingencies were developed to address concerns regarding potential migration of tritium in excess of the simulated results and potential high levels of tritium which have not been detected at the HFBR. In the event that 1) the tritium plume in excess of 25,000 pCi/l reaches the Chilled Water Plant, an evaluation of the need to reactivate the Princeton Avenue IRA will be conducted, and/or 2) In the event the tritium plume in excess of 20,000 pCi/l migrates to Weaver Drive, the Princeton Avenue IRA system will be reactivated. Alternative T4 statesthat if tritium levels at the HFBR exceed 2,000,000 pCi/l, selective hot spot pumping will take place at the reactor.

Alternatives T7 and T8 involve on- and off-site evaporation of the extracted tritium, which introduces an additional risk to the public.

Based on the results of groundwater modeling presented, the no action alternative (T1) will not ensure the protection of human health and environment because even though the contaminant plume is predicted to attenuate to below chemical-specific ARARs before migrating off the BNL property, however, confirmation of this is not available, due to the lack of groundwater modeling or monitoring in this alternative. Therefore, overall protection of human health and the environment is not achieved.

Alternatives T2 through T8 are protective of human health and the environment. Tritium concentrations are rapidly reduced by dispersion, dilution, and decay. For the natural attenuation alternative, 20 to 25 years is required for the aquifer concentrations to meet the 20,000 pCi/l MCL. No impact to potential receptors is predicted, and groundwater with tritium levels higher than the MCL will not pass Princeton Avenue for any of the alternatives.

Alternatives T4, T5, T6, T7 and T8 require the extraction and handling of tritiated groundwater which can potentially increase the exposure rate to tritium. Transportation alternatives T4 and T7 also increase the chance for exposure due to the large distance of travel required for final disposition of the tritium. Alternatives T4, T7, T8 involve on- and off-site evaporation of the extracted tritium, which introduces an additional risk to the public.

The No Action Alternative T1 cannot prove compliance with the chemical-specific ARARs. Groundwater tritium quality is projected to be in compliance with ARARs after the 20 to 25 year period. However, confirmation of this is not available, due to the lack of groundwater monitoring in this alternative. Therefore, compliance with ARARs are not achieved for this alternative. The Natural Attenuation alternatives as well as Alternatives T4 through T8 eventually, within 20 years, comply with ARARs. Dilution and decay reduce tritium concentrations to below the MCL of 20,000 pCi/l.

Groundwater discharge standards (chemical specific and action specific ARARs) for tritium and VOCs will be attained by all alternatives utilizing extraction.

The on-site evaporation alternative will require approval from regulators due to the discharge of tritium to the atmosphere. However, the discharge concentrations will be substantially below the existing limit for the HFBR Stack.

Alternative T1, Natural Attenuation Alternative T2 and Natural Attenuation Alternative T3 which includes the continuation of the tritium IRA will reduce contaminant concentrations in groundwater via decay, dilution and diffusion. For natural attenuation, groundwater transport modeling predicts no impact on potable water wells or BNL supply wells. The concentrations of tritium within the plumes are, over 20 to 25 years, attenuated by decay and dispersive processes to below the drinking water standard. Continuation of the IRA re-circulation system does not enhance the natural attenuation process and results in a second, low concentration (less than 2,000 pCi/l) plume south of the RA V basin.

Since the No Action Alternative T1 does not include modeling or monitoring, the long-term effectiveness of Alternative T1 cannot be verified.

Alternatives T6, T7 and T8, hot spot extraction of tritium groundwater and long-term storage for radioactive decay, off-site evaporation, or on-site evaporation will remove significant amounts of tritium. However, a significant reduction in cleanup duration was not observed from these alternatives when compared to Alternative T2. The cleanup time is reduced by only 3 to 5 years with no significant reduction in the overall plume migration distance. The same observation is true of the Contingency Based Remediation Alternative T4.

For Alternatives T4, T7 and T8, the long-term risks due to possible exposure tritium in the atmosphere are increased as a result of discharge of tritium to the atmosphere. This risk is not significant.

Neither the No Action Alternative (T1), continuation of the IRA (T3) nor Natural Attenuation (T2) uses treatment or containment to reduce toxicity, mobility or volume of the tritium in the groundwater. The tritium concentrations will be gradually reduced or attenuated by natural processes of decay, dilution and dispersion. Slight contaminant migration will continue because groundwater is not contained or treated. Alternatives T4, T6, T7 and T8 use extraction of the hot spot to reduce the toxicity, mobility and volume of contaminated groundwater. The tritium in the extracted groundwater is permanently removed from the aquifer.

None of the extraction alternatives (T3 through T8) have noticeable impacts on the migration of tritium as compared to the Natural Attenuation Alternative, T2. In all alternatives, tritium greater than the MCL will not migrate past Princeton Avenue.

Alternative T1 does not include any disturbance, site access or use of the site and therefore does not present the community or on-site workers with any additional risks resulting from potential release of contaminants.

Alternative T2, Natural Attenuation, allows the natural processes of radioactive decay, diffusion, dilution and adsorption to reduce the concentrations of tritium in the groundwater to acceptable levels. Short-term risks are limited to possible worker exposure to contaminated soil and groundwater during groundwater sampling. In the short term, this alternative allows further migration of the tritium plumes, although it is rapidly attenuated by decay and dispersive processes.

Alternative T3, Natural Attenuation with the operation of the tritium IRA is similar to T2; however, the operation of the IRA may increase potential exposure to workers since the groundwater is extracted and then discharged to an on-site recharge basin.

In Alternative T5, containment by recirculation, groundwater from the southern edge of the 20,000 pCi/l contour is extracted and pumped upgradient to the RA V recharge basins. This limits the migration of this part of the plume and minimizes the volume of the aquifer that exceeds the MCL for tritium.

Alternatives T4, T6, T7 and T8, the "hot spot" remedies, involve extraction of groundwater from the areas of high tritium concentration within the plume, treatment (by on-site storage, off-site evaporation, or on-site evaporation). All of these alternatives present some risk to on-site workers through dermal contact, ingestion and inhalation from construction activities and regular operation and maintenance activities. However, exposure can be minimized by using proper work practices and procedures.

In all alternatives, the time to remediate to MCLs within the aquifer is equal to or less than 20 to 25 years. No significant reductions are observed in the time remediate to reach MCLs when active remediation is implemented (T5, T6, T7 and T8).

Alternatives T1 and T2 will require a public awareness program and natural attenuation will require monitoring, both of which can be easily implemented. Alternatives T4 through T8 will require the construction of storage tanks, carbon units, extraction systems and discharge lines. The treatment equipment required is readily proven and commercially available. Sampling for treatment effectiveness and groundwater monitoring can be easily implemented.

The alternatives associated with off-site disposal, T4 and T7 may encounter some difficulty in obtaining approvals for transportation and off-site evaporation activities, which could lead to delays in the implementation of this alternative. Additionally, the on-site storage and on-site evaporation alternatives may also have community acceptance problems that could complicate or delay implementation.

There are no costs associated with the no action alternative. Costs associated with the installation of monitoring wells, groundwater sampling and analysis, and groundwater modeling are required for all alternatives.

Table 20 summarizes the costs for the evaluated alternatives. The alternative with the lowest capital cost is Alternative T2 since all monitoring wells required for natural attenuation monitoring have already been installed. The alternative with the highest capital cost is Alternative T6, groundwater recovery with on-site storage. This cost is mostly associated with construction of the large storage tank for this alternative. The cost for Alternative T4 that was originally proposed in the PRAP will be increased since the selected remedy is a modification of Alternative T4 which contains an additional low flow extraction system that will be installed and operated near Temple Place.

The alternative with the lowest operation and maintenance costs is alternative T2. Alternative T7 has the highest costs because of the expense of the transportation and off-site disposal of the approximately one million gallons of tritiated groundwater at $20 per gallon.

New York State, based on its review of the Feasibility Study and the Proposed Plan, has concurred with a modification of alternative (T4) which includes a fourth contingency. An additional system was added in response to regulatory concerns about potential plume growth.

Written and verbal comments received from the community during the public comment period and at the public meeting held on March 24, 1999 have been evaluated. From the comments received during the public-comment period, DOE and BNL believe that the public and local elected officials are in general agreement with the selected remedial alternatives. The Responsiveness Summary Section of this Record of Decision contains the comments from the community and DOE's responses. In response to community concerns, the proposed remedy for tritium contamination in groundwater (T4) was modified to be more specific about when the existing pump-and-recharge system would be put on stand-by. The selected remedy keeps the pump-and-treat system running for up to a maximum of one year after the ROD is signed.

This ROD documents the selected remedial actions for OU III. Figure 14 shows the areal extent of the TVOC, strontium-90 and tritium contamination in groundwater along with planned and existing pumping locations. The best balance of EPA's remedy selection criteria was used to identify the following selected actions:

Several Interim Removal Actions (IRAs) have begun to address VOC contamination, including treatment systems at the southern site boundary and in an off-site, downgradient industrial park. Additionally, public water was provided in a large area south of the Site, to protect public health while the groundwater cleanup is underway.

The selected remedy, Alternative V10c, involves active remediation of both on-site and off-site VOC contamination. It includes the following systems: operation of the on-site and off-site IRAs, including the On-Site Southern Boundary IRA and the Off-Site Industrial Complex IRA; installation of new in-well air stripping systems at the LIPA right-of-way, North Street, the Brookhaven Airport, downgradient of North Street East, and the eastern portion of the industrial park; and an additional treatment system on-site at Middle Road. The remedy also includes either in-well air stripping and/or expansion of the existing on-site pump and treat system to address lower levels of VOCs in the western part of the plume. The remedy will also include a source removal system using re-circulation wells with air stripping treatment near Building 96. The PCB contaminated soils at Building 96 that are above the New York State cleanup levels (1 ppm) will be excavated and sent to an off site disposal facility. The final remedy for potential source areas in AOC-26B (Building 96), such as the anomalies discovered during the geophysical survey, will be documented in a subsequent Record of Decision. Details of the specific number of treatment systems and locations needed to meet the cleanup objectives will be determined during the design process. The exact number of years of pumping needed to achieve cleanup objectives will be determined based on monitoring and operating data.

Additional surface soil sampling was conducted for the Building 96 area in accordance with the addendum to the Building 96 Scrapyard Predesign Characterization Work Plan to define the extent of PCB contaminated soils in the Building 96 Scrapyard. Surface soil samples were collected from twelve locations from zero to two inches and analyzed for PCBs. The results from the PCB sampling effort indicated concentrations of PCBs ranging from 5.6 ppm to 710 ppm. Additional sampling will be conducted to further define the area of contaminated soil. Results from these sampling events will be included in the Building 96 Final Design Report or Design Addendum. Based on the currently available data the remediation of the PCB soils will consist of excavation and off site disposal in a licensed facility.

If, after source control is complete and effective, monitoring indicates that continued operation of the components of the selected remedy is not producing further reductions in the concentrations of contaminants in ground water, in accordance with the National Contingency Plan, DOE, NYSDEC, and EPA will evaluate whether discontinuance of the remedy is warranted. The criteria for discontinuation will include complete and effective source control, an evaluation of the operating conditions and parameters as well as a determination that the remedy has attained the feasible limit of contaminant reduction and that further reductions would be impractical.

At present, limited characterization has been performed in the Magothy, so additional characterization and installation of groundwater monitoring wells are planned. This work will be done during the design of the remedy, and will be included in the site records. When this characterization and monitoring is completed, the need for a remedy for the Magothy Aquifer, will be evaluated by DOE, EPA, and NYS DEC. If a remedy for the Magothy Aquifer is necessary, either this record of Decision will be modified or another decision document will establish the selected action. In either case, the public will have an opportunity to review and comment in accordance with CERCLA.

There are concentrated areas of strontium-90 contamination in the groundwater at three on-site locations: the Glass Holes Area, the Brookhaven Graphite Research Reactor (BGRR), and the Waste Concentration Facility.

The selected remedy, Alternative S5a, involves installing extraction wells and using ion-exchange to remove strontium-90 from the extracted water. Details of the specific number of treatment systems and locations needed to meet the performance objective will be determined during the design process. The exact number of years of pumping needed to achieve Remedial Action Objectives will be determined based on monitoring and operating data. Before implementation of the remedy, a pilot treatability study will be performed to evaluate the effectiveness of extraction and treatment. The final remedy may potentially be modified based on the results of this study. Treated water will be discharged on-site. Residual waste that contains strontium-90 will be disposed off-site.

If, after source control is complete and effective, monitoring indicates that continued operation of the components of the selected remedy is not producing further reductions in the concentrations of contaminants in ground water, in accordance with the National Contingency Plan, DOE, NYSDEC, and EPA will evaluate whether discontinuance of the remedy is warranted. The criteria for discontinuation will include complete and effective source control, an evaluation of the operating conditions and parameters as well as a determination that the remedy has attained the feasible limit of contaminant reduction and that further reductions would be impractical.

A pump and recharge system, which includes three pumping wells located on-site along Princeton Avenue, was installed in May 1997 to extract the tritium contaminated groundwater and discharge it further north to a recharge basin on-site. Pumping at the leading edge of the plume was taken as a precautionary measure to inhibit contaminated groundwater from advancing towards the site's boundary and allow more time for the tritium to decay. A carbon filtration unit is included in the pump and recharge system to remove VOCs that are also present in the groundwater.

The selected remedy is a modification of alternative T4, as originally proposed in the PRAP. The remedy will combine extraction of groundwater in response to specific contingencies and extensive monitoring and reporting to assure that the cleanup objectives are met. Three specific contingencies were identified in the PRAP, and a fourth has been added in this ROD to address regulatory concerns. Other actions will be evaluated and implemented, as necessary, to ensure that the cleanup objectives are met. Additional monitoring wells will supplement the existing groundwater monitoring network downgradient of the High Flux Beam Reactor's (HFBR) spent fuel pool.

The first and second contingencies were developed to ensure that the tritium plume would migrate no further downgradient above drinking water standards. After an evaluation period established during design of the selected remedy, the tritium pump and recharge system on Princeton Avenue will be put on stand-by and later operated as needed as an integral component of these contingencies. The evaluation period will extend up to a maximum of one year after ROD finalization and will include an analysis of the data against the following two contingency criteria. These two specific contingencies identified are 1) to evaluate the need to reactivate the Princeton Avenue IRA if tritium concentrations exceed 25,000pCi/l at the Chilled Water Plant Road, and/or 2) reactivate the Princeton Avenue IRA if tritium concentrations exceed 20,000 pCi/l at Weaver Drive.

A third contingency was developed to ensure that if the most concentrated part of the plume were to act as a source of continuing contamination, active remediation would remove this problem. This contingency proposed a low flow extraction system to be installed in the most concentrated area of tritium contamination near the HFBR and activated if concentrations exceed 2,000,000 pCi/l at the front of the reactor. This system then would be used to remove groundwater containing the highest concentrations of tritium from the aquifer. The extracted tritium contaminated water will be disposed of offsite. Technologies to reduce the volume of water that requires off-site disposal may be identified during design. Since the PRAP was issued to the public, groundwater near the HFBR has exceeded 2,000,000 pCi/l. DOE is currently in the process of constructing some of the wells for this low flow extraction system on Cornell Avenue and developing plans to extract the most concentrated part of the plume in front of the HFBR. The detailed operational parameters for this system will be developed during design.

In addition to the ones originally identified in Alternative T4 and proposed in the PRAP, a fourth contingency, an additional low flow extraction system will be installed and operated near Temple Place. This additional system was added in response to regulatory concerns about potential plume migration. The exact location, operational parameters and treatment and disposal options for the extracted water will be developed during design. Operation of the Temple Place extraction system will continue for up to one year. As these extraction wells operate, extensive monitoring will occur to evaluate the effect of extraction locally, as well as on the entire plume. Because of the inherent uncertainties of predicating plume behavior based on groundwater modeling, the actual monitoring data will be evaluated and used to help determine whether continued operation of this extraction system is needed to achieve the cleanup objectives. The criteria to continue system operation beyond one year will be developed during design and based on the attainment of the cleanup objectives.

Thirteen AOCs assigned to OU III were investigated as suspected source areas of groundwater contamination. Also, as the work for OU III was proceeding, groundwater contamination from other OUs and Additional Areas of Investigation (AAIs) was included in the investigation and assessment. Table 1 describes these AOCs and AAIs. Table 2 outlines the actions required for these suspected source areas. Many of the suspected source areas had completed and/or ongoing removal actions, and no further action is required. Table 3 outlines source removal actions to date. This remedy requires a source removal system using in-well air stripping near Building 96; completion of the Building 830 Underground Storage Tank Removal Action; remediation of the groundwater near the Carbon Tetrachloride Tank Spill Area; and management of other suspected source areas as shown in Table 2.

All of the groundwater plumes will require monitoring of new and existing wells and institutional control of the groundwater until completion of remediation. These wells will be located adjacent to the treatment systems and along the downgradient plumes. They will help determine the effectiveness of each treatment system in reducing the concentrations of contaminants over time. Long-term monitoring will also determine the ultimate duration for operation of the treatment systems and will support future decisions to make any changes to the final remedy. At the request of the homeowner, DOE can arrange for monitoring of private wells used for drinking water on properties that previously have declined DOE's offer of public water hookups.

Remedy selection is based on CERCLA, and its amendments, and the regulations contained in the National Contingency Plan. All remedies must meet the threshold criteria; protection of human health and the environment, and compliance with ARARs. CERCLA also requires that the remedy use permanent solutions and alternative treatment technologies to the maximum extent practicable and that the implemented action must be cost effective. Finally, the statute includes a preference for remedies that employ treatment that permanently and significantly reduces the volume, toxicity or mobility of hazardous wastes as their principal element. The following sections discuss how the selected remedy meets these statutory requirements.

TVOCs in groundwater: Alternative V10c protects human health and the environment because it minimizes potential exposure pathways, offers a large reduction in contaminant toxicity, mobility, and volume, and meets ARARs in the Upper Glacial Aquifer in 30 years.

Strontium in Groundwater: Alternative S5a will remove Sr-90 from the aquifer and will prevent migration of the contaminated groundwater and restore the aquifer within 30 years. This alternative is protective of human health and the environment as the aquifer is restored to the Sr-90 MCL within 30 years. No receptors will be impacted during treatment.

Tritium in Groundwater: Alternative T4, which has been modified from what was originally proposed in the PRAP, will protect human health and the environment because the contaminant plume naturally attenuates to below chemical-specific ARARs within a reasonable period of time (20-25 years). No potential receptors are identified within the path of the plume for the duration of the time required for the plume to naturally attenuate to MCLs. Discharge of tritium at the off-site disposal facility will result in a short-term temporary increase in exposure to tritium at and near the facility. This increase will not present a significant human health risk.

The National Contingency Plan Section 300.430 (P) (5) (ii) (B) requires that the selected remedy attain the federal and state ARARs or obtains a waiver of an ARAR.

TVOCs in groundwater: Alternative V10c will meet ARARs for the Glacial Aquifer within 30 years. Discharges to water or air will comply with chemical-specific ARARs through engineering controls and monitoring.

Strontium in Groundwater: Alternative S5a will comply with chemical-specific ARARs within 30 years at all locations impacted by Sr-90. Discharges to water or air will comply with chemical-specific ARARs through engineering controls and monitoring.

Tritium in Groundwater: Alternative T4, which has been modified from what was originally proposed in the PRAP, will comply with the chemical-specific ARARs since Federal MCL and the New York State groundwater standard will not be exceeded after 20-25 years. The tritium will, through natural decay, dispersion and dilution reach the MCL within 20-25 years. If necessary, groundwater extraction and re-circulation or hot spot removal will be use to augument the natural attenuation process. Discharge of tritium at the off-site facility will be in accordance with the air permit for that facility.

No location-specific ARARs were identified for the proposed alternatives.

Remedies requiring discharge of water or air will comply with chemical-specific and action-specific ARARs at the discharge point through engineering controls, monitoring and acquisition of appropriate permits.

No to be considered guidance was identified for the selected remedies.

Based on the expected performance standards, the selected remedies have been determined to be cost effective because they provide overall protection of human health and the environment, long- and short-term effectiveness, and compliance with ARARs, at an acceptable cost. Table 20 provides a comparison of costs for all alternatives evaluated for strontium contaminated groundwater, tritium contaminated groundwater, and TVOC contaminated groundwater.

The National Contingency Plan prefers a permanent solution whenever possible. The recommended selected remedy is a final action that utilize permanent solutions to the maximum extent practicable.

Five-years review will be conducted until cleanup goals are met throughout the aquifer and to determine the effectiveness of the groundwater surveillance programs.

III. RESPONSIVENESS SUMMARY