LOWER FOX RIVER 30 PERCENT DESIGN

LOWER FOX RIVER 30 PERCENT DESIGN

November 30, 2007

Prepared for:

Fort James Operating Company, Inc. NCR Corporation

For Submittal to:

Wisconsin Department of Natural Resources U.S. Environmental Protection Agency

Prepared by:

Shaw Environmental & Infrastructure, Inc.

and

Anchor Environmental, L.L.C.

Table of Contents

i

LIST OF TABLES....................................................................................................................................... V

LIST OF FIGURES..................................................................................................................................... V

ACRONYMS AND ABBREVIATIONS................................................................................................VII

1 INTRODUCTION................................................................................................................................ 1 1.1 Site Description .......................................................................................................................... 2 1.2 Site Characteristics ..................................................................................................................... 3 1.3 Summary of OUs 2 to 5 Remedy.............................................................................................. 4 1.4 Organization of this Document................................................................................................ 7

2 SITE CHARACTERISTICS ................................................................................................................. 8 2.1 Sampling and Analysis Data .................................................................................................... 8

2.1.1 Pre‐Design Data ................................................................................................................... 8 2.1.2 2004 Sampling and Analysis Program.............................................................................. 8 2.1.2.1 Data Validation ............................................................................................................ 9

2.1.3 2005 Sampling and Analysis Program.............................................................................. 9 2.1.3.1 Testing Methods (Chemical and Geotechnical) .................................................... 10 2.1.3.2 Data Validation .......................................................................................................... 10

2.1.4 2006 Sampling and Analysis Program............................................................................ 10 2.1.4.1 Testing Methods (Chemical and Geotechnical) .................................................... 10 2.1.4.2 Data Validation .......................................................................................................... 11

2.1.5 2007 Sampling and Analysis Program............................................................................ 11 2.1.5.1 Testing Methods (Chemical and Geotechnical) .................................................... 11 2.1.5.2 Data Validation .......................................................................................................... 11

2.2 Summary of Physical Site Characteristics ............................................................................ 11 2.2.1 Geotechnical Conditions................................................................................................... 12

2.3 Summary of Spatial Extent of PCBs ...................................................................................... 13 2.3.1 Improved Geostatistical Delineation of Remediation Boundaries ............................. 13 2.3.2 Spatial Extent of PCBs Exceeding 1.0 ppm..................................................................... 16

2.4 Characterization of Material for Beneficial Use and Disposal Purposes.......................... 16 2.4.1 Beneficial Use Opportunities ........................................................................................... 16 2.4.2 Sediments Subject to Non‐TSCA Disposal Requirements ........................................... 17 2.4.3 Sediments Subject to TSCA Disposal Requirements .................................................... 17

3 SITE PREPARATION AND STAGING AREA DEVELOPMENT.............................................. 19 3.1 Staging Area Requirements and Design............................................................................... 19 3.2 Potential Staging Area Locations........................................................................................... 20 3.3 Real Estate, Easement Requirements and Location‐Specific ARARs ............................... 22

3.3.1 Real Estate and Easement ................................................................................................. 22 3.3.2 Possible Location‐Specific ARARs .................................................................................. 22

4 SEDIMENT DREDGING.................................................................................................................. 27

Table of Contents

ii

4.1 Summary of Sediment Physical Properties .......................................................................... 27 4.1.1 Dredgeability ...................................................................................................................... 27 4.1.2 Seasonal Construction Windows and Weather‐Related Work Impacts .................... 28 4.1.3 Federal Navigation Channel Considerations................................................................. 28

4.2 Equipment Selection and Production Rates ......................................................................... 29 4.2.1 Equipment Selection Process............................................................................................ 29 4.2.2 Production Rate Considerations ...................................................................................... 29 4.2.3 Equipment Selection to Remove Prospective TSCA Sediments.................................. 30 4.2.3.1 Hydraulic Dredge Removal of Prospective TSCA Sediments ............................ 31 4.2.3.2 Mechanical Dredge Removal of Prospective TSCA Sediments .......................... 31

4.3 Methodology for Developing and Optimizing Dredge Prism Design ............................. 32 4.3.1 Define the Neatline Area .................................................................................................. 32 4.3.2 Specify Site and Project Design Criteria ......................................................................... 33 4.3.3 Iterative Design Refinements ........................................................................................... 34 4.3.4 Cost/Benefit Assessments and Contract Alternatives .................................................. 36

4.4 Dredge Plan Design for Sediments Potentially Subject to TSCA Disposal Requirements ........................................................................................................................................ 36 4.5 Dredge Plan Design Basis ....................................................................................................... 37

4.5.1 Sediment Volume Estimates............................................................................................. 37 4.5.2 PCB Mass Estimates .......................................................................................................... 37

4.6 Potential Impacts from Dredging .......................................................................................... 38 4.6.1 Slope and Structural Considerations .............................................................................. 38 4.6.2 Short‐term Water Quality Considerations ..................................................................... 39 4.6.3 Dredge Residual Management ........................................................................................ 40 4.6.4 Noise and Air Quality Considerations ........................................................................... 42

5 MATERIALS HANDLING, TRANSPORT AND DISPOSAL ..................................................... 44 5.1 Transport of Debris and Dredged Material.......................................................................... 44 5.2 Sediment Handling – Sediments Potentially Subject to TSCA Disposal Requirements 44 5.3 Sediment Handling ‐ Non‐TSCA Sediments........................................................................ 45

5.3.1 Mechanically Removed Sediment Transport in OUs 2, 3 & 4 ..................................... 45 5.3.2 Hydraulically Removed Sediment Transport in OU 3 ................................................. 46 5.3.3 Hydraulically Removed Sediment Transport in OU 4 ................................................. 47

5.4 Mechanical Dewatering Operations...................................................................................... 47 5.5 Water Treatment Operations.................................................................................................. 50 5.6 Equipment Selection and Production Rates ......................................................................... 51

5.6.1 Process Flow of Major Unit Operations.......................................................................... 51 5.6.2 Preliminary Mass Balance................................................................................................. 55

5.7 Beneficial Use Considerations................................................................................................ 55 5.7.1 Desanding Technologies................................................................................................... 55 5.7.2 Materials Potentially Suitable for Beneficial Use .......................................................... 56 5.7.3 Description for Potential Beneficial Use Alternatives .................................................. 56

5.8 Transport and Disposal of Dewatered Sediment and Debris ............................................ 57

Table of Contents

iii

5.9 Potential Upland Disposal Facilities...................................................................................... 57

6 ENGINEERED CAP DESIGN.......................................................................................................... 59 6.1 Cap Design Criteria ................................................................................................................. 60

6.1.1 Chemical Isolation Component ....................................................................................... 60 6.1.2 Bioturbation Component .................................................................................................. 61 6.1.3 Consolidation Component ............................................................................................... 62 6.1.4 Erosion Protection Component........................................................................................ 62 6.1.4.1 Supplemental Vessel‐Induced Propeller Wash Analysis..................................... 62 6.1.4.2 Supplemental Hydrodynamic Flow Analysis ....................................................... 64 6.1.4.3 Vessel Wake Analysis ............................................................................................... 65

6.1.5 Operational Component ................................................................................................... 68 6.2 Additional Cap Design Considerations................................................................................ 69

6.2.1 Federal Navigation Channel ............................................................................................ 69 6.2.2 Infrastructure and Utilities ............................................................................................... 70 6.2.3 Geotechnical Stability........................................................................................................ 71 6.2.4 Ebullition............................................................................................................................. 72 6.2.5 Post‐Cap Water Depth ...................................................................................................... 72

6.3 Capping Designs and Areas ................................................................................................... 72 6.3.1 Cap Designs ........................................................................................................................ 72 6.3.1.1 Engineered Shoreline Caps ...................................................................................... 74

6.3.2 Delineation of Cap Areas.................................................................................................. 76 6.4 Equipment Selection and Production Rates ......................................................................... 77

7 SAND COVER DESIGN ................................................................................................................... 79 7.1 Sand Cover Design and Areas................................................................................................ 79 7.2 Equipment Selection and Production Rates ......................................................................... 79

8 INSTITUTIONAL CONTROLS ....................................................................................................... 80 8.1 Institutional Control Definitions............................................................................................ 80 8.2 ROD Amendment Requirements........................................................................................... 81 8.3 Specific Institutional Controls under Consideration for OUs 2 to 5................................. 81

8.3.1 Water Use Restrictions ...................................................................................................... 82 8.3.2 Construction Limitations .................................................................................................. 84 8.3.3 Monitoring and Maintenance........................................................................................... 85 8.3.4 Public Information and Advisories ................................................................................. 86

9 CONSTRUCTION SCHEDULE AND SEQUENCING................................................................ 88 9.1 Operations Sequencing............................................................................................................ 88 9.2 Construction Schedule ............................................................................................................ 89 9.3 Contracting Strategy................................................................................................................ 92

10 MONITORING, MAINTENANCE, AND ADAPTIVE MANAGEMENT ................................ 94 10.1 Construction Monitoring ........................................................................................................ 94

Table of Contents

iv

10.1.1 Post‐Dredge Verification................................................................................................... 95 10.1.2 Engineered Cap and Sand Cover Placement Verification.......................................... 100

10.2 Post‐Construction Operation, Maintenance, and Monitoring ......................................... 105 10.2.1 Cap Performance Monitoring and Maintenance......................................................... 105 10.2.2 Natural Recovery Monitoring........................................................................................ 109

10.3 Long Term Monitoring.......................................................................................................... 110 10.3.1 LTMP Objectives .............................................................................................................. 110 10.3.2 Water Quality Monitoring Plan ..................................................................................... 111 10.3.3 Fish Tissue Monitoring Plan........................................................................................... 113

10.4 Adaptive Management.......................................................................................................... 114 10.5 Monitoring, Maintenance, and Adaptive Management Schedule .................................. 116

11 REFERENCES .................................................................................................................................. 117

Table of Contents

v

List of Tables Table 2‐1 Low Water Pool Elevations in OUs 2 to 5 Table 2‐2 Summary of all RD Geotechnical Data Table 2‐3 Summary of Kriging Cross‐Validation Metrics for OUs 3 to 4 Table 4‐1 Geotechnical Properties of Sediments Targeted For Dredging Table 4‐2 Summary of Dredge Volumes Table 4‐3 Lower Fox River PCB Mass Estimates Table 5‐1 Built‐in Over‐Capacity Based on the Current Equipment Design Criterion Table 5‐2 Effluent Discharge Monitoring Requirements Table 6‐1 Engineered Cap Designs Developed in BODR Table 6‐2 Summary of Cap Armor Recommendations for Recreational Propwash Table 6‐3 Summary of Design Vessels for Vessel Wake Analysis Table 6‐4 Summary of Cap Armor Recommendations for Vessel Wakes Table 6‐5 Potential Shoreline Remedial Design Considerations Table 6‐6 Summary Baseline Water Elevations Table 6‐7 Preliminary Estimate of Shoreline Capping Areas Table 6‐8 Summary of OUs 2 to 5 Engineered Cap Designs List of Figures Figure 1‐1 Fox River OU2 to 5 Project Area Figure 2‐1 Sample Location Map OUs 2 to 5 2004‐2007 Figure 2‐2 2006 Sample Location Map OUs 2 to 5 Figure 2‐3 2007 Sample Location Map OUs 2 to 5 Figure 2‐4 Spatial Distribution of PCB Mass OU 3 Figure 2‐5 Spatial Distribution of PCB Mass OUs 4/5 Figure 2‐6 Estimated Depth of PCB Contamination, OU3 Figure 2‐7 Estimated Depth of PCB Contamination, OU4 Figure 2‐8 Estimated Remediation Footprints OU 3 Figure 2‐9 Estimated Remediation Footprints OUs 4/5 Figure 3‐1 Sediment Dewatering & Process Water Treatment System Site Layout Figure 3‐2 Sediment Dewatering & Process Water Treatment System Equipment Layout Figure 4‐1 Sample Locations with Hard Substrate Immediately Underlying Contaminated

Sediments; OUs 2 and 3 Figure 4‐2 Sample Locations with Hard Substrate Immediately Underlying Contaminated

Sediments; OUs 2 and 3 Figure 4‐3 Example LOS Evaluation OU4 Figure 4‐4 Dredge Plan for OUs 2 and 3 Figure 4‐5 Dredge Plan for OUs 4 and 5 Figure 5‐1 Process Flow Diagram‐Sediment Dewatering System; Updated Mass Balance Figure 5‐2 Process Flow Diagram‐5000 GPM Water Treatment Figure 6‐1 Maximum Predicted Shear Stress For Extreme Flow Event; OUs 4 and 5

Table of Contents

vi

Figure 6‐2 Maximum Predicted Shear Stress During June 23, 1990 Flow Event; OU 3 Figure 6‐3 Maximum Predicted Shear Stress During June 23, 1990 Flow Event; OUs 4 and 5 Figure 6‐4 Engineered Cap and Sand Cover Plan for OUs 2 and 3 Figure 6‐5 Engineered Cap and Sand Cover Plan for OUs 4 and 5 Figure 9‐1 Preliminary Construction Schedule Figure 10‐1 Preliminary Example Layout of DMUs and DCUs for Example Area in OU 4 List of Appendices Appendix A Dredge Design Support Documentation Appendix B Cap Design Support Documentation Appendix C Engineered Plan Drawings Appendix D Outline of Technical Specifications

Acronyms and Abbreviations

vii

ACRONYMS AND ABBREVIATIONS Anchor Anchor Environmental, L.L.C. ACES Automated Coastal Engineering System AOC Administrative Order on Consent A/OT Agencies/Oversight Team ARAR Applicable or Relevant and Appropriate Requirements ARCS Assessment and Remediation of Contaminated Sediments ARS Agricultural Research Service ASNRI Areas of Special Natural Resource Interest BMPs best management practices BODR Basis of Design Report CCU cap certification unit CD Consent Decree CDF confined disposal facility CERCLA Comprehensive Environmental Response, Compensation and Liability Act CFR Code of Federal Regulations CHL Coastal and Hydraulics Laboratory cm centimeter CMU cap management unit CQAPP Construction Quality and Assurance Project Plan cy cubic yards DCU dredge certification unit DMU dredge management unit DOC depth of contamination DRET Dredging Elutriate Test ERDC Engineer Research and Development Center ESA Environmentally Sensitive Area FEMA Federal Emergency Management Agency FIK full indicator kriging Fort James Fort James Operating Company, Inc. FRVOR Fox River Valley Organic Recycling GAC granulated activated carbon GIS geographic information system gpm gallons per minute ICIAP Institutional Control Implementation and Assurance Plan LDD Land Development Desktop (by AutoDesk) LOS level of significance LTMP Long Term Monitoring Plan MAE Mean Absolute Error MEP Maximum Extent Practicable mg/L milligrams per liter μg/m3 micrograms per cubic meter

Table of Contents

viii

MGD million gallons per day MNR monitored natural recovery NCR NCR Corporation NOI Notice of Intent NOT Notice of Termination NPDES National Pollutant Discharge Elimination System NRCS Natural Resource Conservation Service NTU nephelometric turbidity units OM&M Operations, Maintenance, and Monitoring OMMP Operations, Maintenance and Monitoring Plan OU Operable Unit PCB polychlorinated biphenyl P&ID Piping and Instrumentation Diagrams ppm parts‐per‐million QA quality assurance QAPP Quality Assurance Project Plan QC quality control RAL remedial action level RAO Remedial Action Objective RMSE Root Mean Squared Error RNA regulated navigation areas RD Remedial Design RM river mile ROD Record of Decision RPD relative percent difference RTK Real Time Kinematic SAP sampling and analysis plan SAP/QAPP Lower Fox River Operable Units 2 to 5 Pre Design Sampling Plan SCCU sand cover certification units SCMU sand cover management units SEI Sea Engineering Inc. Shaw Shaw Environmental and Infrastructure, Inc. SICT seepage induced consolidation testing SLAMM Source Loading and Management Model Modeling SLOH State Lab of Hygiene SMU Sediment Management Unit SOW Statement of Work SPT Standard Penetration Test SQT sediment quality threshold SWAC surface weighted average concentration SWAT Soil and Water Assessment Tool TIN triangulated irregular network TSCA Toxic Substances Control Act

Table of Contents

ix

TSS total suspended solids µg microgram (or micron) UMR‐IWW Upper Mississippi River‐Illinois Water USACE U.S. Army Corps of Engineers U.S.C. United States Code USCG U.S. Coast Guard USCS Unified Soil Classification System USDA United States Department of Agriculture USEPA U.S. Environmental Protection Agency WDOC Wisconsin Department of Commerce WDOT‐PAL Wisconsin Department of Transportation Product Acceptability List WDNR Wisconsin Department of Natural Resources WRDA Water Resource Development Authority WST Waste Stream Technologies WWTP wastewater treatment plant ZID zone of initial dilution

Section 1 – Introduction

1

1 INTRODUCTION This document presents the Preliminary (30 Percent) Design for the remediation of polychlorinated biphenyls (PCBs) in Operable Units (OUs) 2, 3, 4 and 5 of the Lower Fox River and Green Bay Site (Site; Figure 1-1). Included in this document are summaries of remedial design (RD) analyses completed to date, along with engineering design plans, cross-sections, and drawings that describe the design in more detail. The ongoing RD is also addressing the sequencing of remedial actions to account for the multi-faceted and multi-year components of the OUs 2 to 5 PCB cleanup remedy.

The PCB cleanup remedy for the Lower Fox River was originally set forth in Records of Decision (RODs) for OUs 2 to 5 issued in December 2002 and June 2003 by the United States Environmental Protection Agency (USEPA) and the Wisconsin Department of Natural Resources (WDNR) under the authority of the Comprehensive Environmental Response, Compensation and Liability Act (CERCLA), as amended, 42 U.S.C. §§ 9601-9675. In order to support detailed RD analyses consistent with the RODs, intensive data collection was performed in 2004-2005, including analysis of approximately 10,000 sediment samples collected from more than 1,400 locations. Much of that new information was compiled and analyzed in the Basis of Design Report (BODR) for OUs 2 to 5 (Shaw and Anchor 2006), approved by USEPA and WDNR in July 2006. In June 2007, a ROD Amendment was issued by USEPA and WDNR that made changes to parts of the remedy described in the original RODs in response to new information obtained from the 2004-2005 data collection effort and analyzed in the BODR, and also from experience with prior remediation activities in OU 1 (USEPA and WDNR 2007). The design of remedial actions in OU 1 is being addressed under a separate agreement between USEPA, WDNR, and the WTM1 Company. This RD submittal addresses only OUs 2 to 5.

This 30 Percent Design submittal builds off of the BODR and ROD Amendment, and was prepared consistent with requirements set forth in the Administrative Order on Consent (AOC) and associated Statement of Work (SOW) for OUs 2 to 5 (USEPA 2004), executed in March 2004 by Fort James Operating Company, Inc.1 (Fort James) and NCR Corporation (NCR) (collectively the “Participating Companies”) in cooperation with the USEPA and WDNR (collectively the “Response Agencies”). USEPA and WDNR are overseeing the RD process, and design documents prepared by the Participating Companies are subject to review and approval by USEPA and WDNR. Throughout the RD process, the Response Agencies and Participating Companies have collaboratively sought to resolve key technical and implementation issues through the timely use of workgroups and other communications.

The requirements for the 30 Percent Design submittal are more specifically described in the Remedial Design Work Plan (RD Work Plan), approved by the Response Agencies on June 28, 2004. This 30 Percent Design submittal includes the following:

1 In January 2007, Fort James Operating Company, Inc was converted to Georgia‐Pacific Consumer Products LP

Section 1

2

• Determination of specific technologies for sediment capping and dredging, dewatering, transportation and disposal of dredged sediments and associated wastewaters;

• Design assumptions and parameters, including design restrictions, process performance criteria, appropriate unit processes for the treatment train, and expected removal or treatment efficiencies;

• Detailed plans, cross-sections, drawings, and sketches, including design calculations; • Outline of technical specifications; • Proposed siting/locations of processes/construction activities; • Proposed disposal locations based upon effectiveness, implementability and cost; • Preliminary construction schedule, including contracting strategy; • Preliminary sections of the Construction Quality Assurance Project Plan (CQAPP), including

draft sediment removal and capping verification plans; • Outline of the Operations, Maintenance and Monitoring Plan (OMMP), including expected long-

term monitoring and operation requirements; • Outline of the Adaptive Management Plan to modify the cleanup plan as appropriate in response

to new information and experience during initial remediation activities in OUs 2 to 5; • Outline of institutional control requirements; and • Significant new information from other projects and activities.

Subsequent design phases (e.g. 60, 90, and 100 percent) will include further development and refinement of the remedial design for OUs 2 to 5.

1.1 Site Description

The Lower Fox River Site defined by the Response Agencies extends 39 miles from the outlet of Lake Winnebago to the mouth of the river where it discharges into Green Bay (Figure 1-1). The Lower Fox River is the most industrialized river in Wisconsin; since the mid 1800s water quality has been degraded by expanding industries and communities discharging sewage and industrial wastes into the river as well as by agricultural activity (USEPA and WDNR 2003). PCBs were discovered in the Lower Fox River in the 1970s. As set forth in the RODs, PCBs are the focus of current RD efforts.

The Lower Fox River is divided into five operable units:

• OU 1 is also known as Little Lake Butte des Morts. The Neenah and Menasha Dams control the pool elevation of Lake Winnebago and the discharge to the upstream end of OU 1 at river mile (RM) 39. Remedial design of OU 1 is being addressed under a separate SOW and Consent Order.

• OU 2 extends from the Appleton Locks at RM 31.9 to the Little Rapids Dam at RM 13.1. This unit contains the majority of locks and dams in the Lower Fox River system and the greatest elevation drop and gradient. Sediments have a very patchy distribution in this reach with extensive intervening bedrock exposures. The OU 1 to 2 ROD calls for active remediation in Deposit DD only, while monitored natural recovery (MNR) is the selected remedy for the remainder of OU 2.

Section 1

3

• OU 3 extends from the Little Rapids Dam to the De Pere Dam at RM 7.1. Soft sediment covers most of this unit.

• OU 4 extends from the De Pere Dam to the river mouth at Green Bay. This operable unit contains a federal navigation channel, the northern portion of which is currently maintained by the U.S. Army Corps of Engineers (USACE). The area around OU 4 is highly urbanized, and includes the City of Green Bay.

• OU 5 begins at the river mouth, and includes the entire bay of Green Bay, which is approximately 119 miles long and is an average of 23 miles wide (USEPA and WDNR 2003). The OU 3 to 5 ROD specified MNR as the selected remedy for OU 5, with the exception of potential dredging and capping near the river mouth.

1.2 Site Characteristics

The new data and analyses presented in the BODR (Shaw and Anchor 2006), and summarized in the ROD Amendment (USEPA and WDNR 2007) demonstrated that:

• PCBs are not uniformly spread throughout OUs 2 to 5, but tend to be concentrated in smaller, definable areas.

• A 20-acre area, with PCB concentrations in near-surface sediments as high as 3,000 parts-per-million (ppm), the highest known PCB concentrations in the Lower Fox River, was found just downstream and west of the De Pere Dam. This area is being addressed as part of the separate Phase 1 remediation project (see Figure 1-1), with more than 145,000 cubic yards (cy) of PCB-contaminated sediment targeted for removal during 2007, and additional material to be removed in early 2008. This project includes removal of an estimated 26,000 cy of sediment with PCB concentrations greater than 50 ppm (Toxic Substances Control Act [TSCA] materials).

• Contaminated sediment as deep as 13 feet below the river bottom was found in the middle channel stretches of OU 4. Relatively less contaminated sediment now covers that deeply-buried sediment contamination. To remove the more highly-contaminated sediment and to maintain a stable river bottom in these areas, a significant volume of relatively uncontaminated sediment would also have to be removed and disposed.

• Approximately 210 acres out of a total 1,170 acres of the PCB contaminated sediment (roughly 18 percent by area and 0.5 percent of the PCB mass) have a relatively thin layer (i.e., less than six inches) of contamination, with relatively low PCB concentrations (between 1.0 and 2.0 ppm).

• Recent experience with dredging in OU 1 and other projects has shown that dredging equipment cannot completely remove contaminated sediment from dredged areas. Thus, residual contaminant concentrations often remain after dredging is completed in an area. Dredging alone would thus likely not achieve the PCB surface weighted average concentration (SWAC) goals established by the 2003 ROD and 2007 ROD Amendment.

• Dredging probably cannot be used to remove contaminated sediment in some areas near shoreline facilities and in-water structures because removal of the sediment could undermine and destabilize those facilities and structures.

Section 1

4

1.3 Summary of OUs 2 to 5 Remedy

The ROD Amendment requires remedial action for all sediment exceeding the 1.0 ppm PCB remedial action level (RAL). Consistent with the ROD Amendment, the OUs 2 to 5 remedy described in this 30 Percent Design report includes the following elements:

• Performance Standards. The ROD Amendment requires remediation of all contaminated sediment exceeding the 1.0 ppm PCB RAL in OU 2 (Deposit DD), OU 3, OU 4, and OU 5 (river mouth), excluding exceptional areas, either by the removal, engineered capping, or sand cover approaches discussed below. The ROD Amendment also establishes two standards that will be used to judge the completion of construction of the OUs 2 to 5 remedy in each OU: 1) a RAL performance standard; and 2) a SWAC goal. Construction of the remedy in an OU is deemed complete if the RAL performance standard is met throughout the OU. If the RAL performance standard is not met at the completion of construction, then the remedy is deemed complete if the SWAC meets the goal for the OU. The construction of the remedy is not deemed complete based on the SWAC goal unless all sediment exceeding the RAL is addressed using the remedial approaches outlined below.

• Staging Areas. Staging area(s) are required for facilities associated with sediment dewatering, sediment handling, water treatment, and cap/cover material staging. A single staging area in OU 4 was identified during the RD process and additional staging area(s) may be identified during subsequent RD. Site preparation at the staging area(s) will include collecting soil samples, securing the onshore property for equipment staging, and constructing necessary onshore facilities. Docking facilities for dredging and loading/offloading equipment and ancillary equipment will be constructed and multiple staging areas may be necessary. Preparation for remedial actions will also include obtaining needed access agreements and landfill disposal agreements.

• Sediment Removal. Sediment with PCB concentrations exceeding the 1.0 ppm RAL are targeted for removal in most areas within OU 2 (Deposit DD), OU 3, OU 4, and OU 5 (River Mouth). In areas targeted for sediment removal without subsequent placement of an engineered cap, sediment will be removed to a target elevation that: (1) encompasses all contaminated sediment exceeding the 1.0 ppm PCB RAL (as determined from RD sampling data and geostatistical data interpolation); (2) removes additional sediment to ensure that side slopes are stable for the remaining sediment; and (3) allows the contractor an overdredge allowance. Sediment removal will generally be conducted using a hydraulic dredge, although in certain circumstances (such as in areas that cannot be accessed by hydraulic dredging equipment) some sediment may be removed by mechanical dredging and transported by barge, or other appropriate sediment removal technologies. For hydraulic dredging, in-water pipelines or other transportation methods will carry the dredged sediment from the dredge to the staging area(s).

• Sediment De-sanding. The sand fraction of sediment that is removed from OUs 2 to 5 may be recovered, washed or otherwise treated, and beneficially reused. The PCB concentration of the recovered sand will generally need to be less than 1.0 ppm before it can be beneficially reused on-site at the Shell Property, and less than 0.25 ppm to be reused off-site, although an alternate concentration threshold may be used for particular uses. Some examples of potential beneficial uses are as partial fill for staging areas, road fill, or daily cover for a landfill.

• Sediment Dewatering and Disposal. Dewatering will be employed at staging facilities for dredged sediment. The dewatering will be accomplished using plate and frame presses or similar technologies to remove water from dredged sediment before disposal. Dewatered non-TSCA sediment will be transported by truck, rail, and/or barge to a dedicated engineered landfill or another suitable upland disposal facility, consistent with applicable federal and state

Section 1

5

requirements. Dewatered sediments subject to TSCA disposal requirements will be transported by truck, rail, and/or barge to a landfill facility appropriately permitted to receive TSCA waste. There currently are no TSCA-permitted landfills in Wisconsin.

• Water Treatment. Superfund cleanups are required to meet the substantive discharge requirements of the Clean Water Act, but National Pollutant Discharge Elimination System (NPDES) permits are not required for on-site work. Thus, water generated by dredging, de-sanding, and dewatering operations will be treated prior to discharge back to the river and will meet all state and federal water quality standards. This may include bag filters, sand filtration, and granulated activated carbon (GAC) treatment. Treated water will be sampled and analyzed to verify compliance with the appropriate discharge requirements according to plans developed during RD and approved by the Response Agencies.

• Post-removal Residuals Management. The ROD Amendment used the term “generated residuals” for sediment that is disturbed by dredging activities (e.g., debris removal or dredge operation) and re-deposited on the surface of a newly-dredged area (usually within the top six inches of the sediment), and also uses the term “undisturbed residuals” for sediment unaffected by dredging operations. Although it is possible for generated residuals to have more (or less) than six inches of thickness, the ROD Amendment considers all residuals present in the top six inches of post-dredge sediment to be generated residuals and all residuals below six inches to be undisturbed residuals. If confirmatory sampling in a sediment removal area reveals post-removal generated residuals or undisturbed residuals with PCB concentrations exceeding the 1.0 ppm PCB RAL, then the following management actions will occur:

o For management of generated residuals: Generated residuals with a PCB concentration equal to or greater than 10

ppm must either be: (1) re-dredged in accordance with the sediment removal requirements specified above; or (2) capped, if the eligibility criteria for that alternate remedial approach can be met, as specified below.

Generated residuals with a PCB concentration between 1.0 ppm and 10 ppm must be covered with at least 6 inches of clean sand from an off-site source (referred to as a “residual sand cover”) if placement of a residual sand cover in the area is necessary to meet the SWAC goal for the OU (i.e., a SWAC of 0.28 ppm PCBs in OU 3 and a SWAC of 0.25 ppm PCBs in OU 4).

o For management of undisturbed residuals: Undisturbed residuals with a PCB concentration exceeding the 1.0 ppm PCB

RAL must be remediated, typically in accordance with the sediment removal requirements specified above. However, a different residuals management approach (such as a cap or a sand cover) may be used for undisturbed residuals if the PCB levels in the undisturbed residuals are only slightly above the 1.0 ppm PCB RAL.

• Engineered Caps. An engineered cap consisting of a sand layer and an armor stone layer may be installed in an area if the following eligibility criteria are satisfied:

o Minimum water depth criteria for capping: Capping is allowed in areas below the federally-authorized navigation

channel if the top of the cap is at least 2 feet below the authorized navigation depth.

Capping is allowed in areas outside of the federally-authorized navigation channel if the top of the cap is at least 3 feet below the low water datum as defined in the BODR.

Section 1

6

o Engineered caps of at least 33 inches in thickness, including a surface armor layer comprised of quarry spall or equivalent materials, may be used to contain contaminated sediments in: (1) areas below the federally authorized navigation channel depth (sediment in specific areas may be dredged as necessary to meet this criterion before the cap is installed); (2) areas with deeply-buried sediment having PCB concentrations above 50 ppm; and (3) nearshore areas with sediment having PCB concentrations exceeding 50 ppm, if removal of such sediment would impair shoreline stability.

o Engineered caps of at least 16 inches in thickness, including a surface armor layer comprised of gravel materials, may be used in areas outside of the federally authorized navigational channel where sediment beneath the cap does not exceed 50 ppm PCBs at any depth within the sediment profile. Sediment in specific areas may be dredged as necessary to meet these criteria before the cap is installed.

o Engineered caps of at least 13 inches in thickness, including a surface armor layer comprised of gravel materials, may be used in areas outside of the federally authorized navigational channel where sediment PCB concentrations beneath the cap do not exceed 50 ppm at any depth within the sediment profile and PCB concentrations in the 6-inch layer immediately beneath the cap does not exceed 10 ppm. Sediment in specific areas may be dredged as necessary to meet these criteria before the cap is installed.

• Sand Covers. A cover composed of at least 6 inches of uncontaminated sand from an off-site source can be placed over certain undredged areas that have low PCB concentrations in a relatively thin layer of PCB-contaminated sediment exceeding the 1.0 ppm PCB RAL if both of the following criteria are met:

o The sediment beneath the sand cover must not exceed 2.0 ppm at any depth within the sediment profile; and

o The sediment profile shall contain only one 6-inch interval with PCB concentrations between 1.0 and 2.0 ppm.

• Exceptional Areas. Modified remedial approaches may be used in exceptional areas in OUs 2 to 5 when evaluation shows another remedial approach is sufficiently protective, more feasible and more cost effective than the approaches outlined above. The specific remedial approach for each exceptional area will be included in the final RD submittal.

• Demobilization and Restoration. Demobilization, site restoration, and decontamination of equipment will require removing all equipment from the staging and work areas and restoring the site to an acceptable condition.

• Natural Recovery after Remediation. Although the 1.0 ppm RAL performance standard or the SWAC goal (0.28 ppm in OU 3 and 0.25 ppm in OU 4) will be met before construction of the remedial action can be deemed complete in an OU, the Response Agencies have concluded that it will take additional time for natural recovery before some of the remedial action objectives (RAOs) are achieved. For example, it is estimated that a SWAC of approximately 0.28 ppm PCBs will be achieved in OU 3 after the completion of active remediation, but the sediment quality threshold (SQT) for unlimited walleye consumption is lower than the SWAC (i.e., 0.049 ppm PCBs), and the Response Agencies estimate that it will take an estimated 9 years to achieve that reduced sediment surface concentration in OU 3. Post-remediation natural recovery will occur before certain SQTs and other RAOs can be achieved.

• Long-term Monitoring of Surface Water and Biota. Long-term monitoring of surface water and biota will be performed to assess progress in achieving RAOs. Monitoring will continue until acceptable levels of PCBs are reached in surface water and fish. A detailed Long-Term

Section 1

7

Monitoring Plan (LTMP), specifying the types and frequency of monitoring, will be developed during the RD process.

• Long-term Cap Monitoring. Long-term monitoring will be performed on any caps that are installed in OUs 2 to 5 to ensure their long-term integrity and protectiveness. The long-term monitoring will include bathymetric surveys and core sampling. If monitoring or other information indicates that the cap in an area no longer meets its original as-built design criteria and that degradation of the cap in the area may result in an actual or threatened release of PCBs at or from the area at levels that preclude achieving the RAOs, additional response activities may be undertaken in the area.

1.4 Organization of this Document

Major design elements for this remedial action were developed during the 30 Percent design phase. The inclusion of a series of reviews and technical exchanges between the RD Team and the Response Agencies/Oversight Team (A/OT) during design activities was important in completing the 30 Percent Design. Specific collaborative work elements included:

• Final propeller wash analyses leading to cap armor size and thickness specifications for various portions of the river

• Refinement of dredging and capping plans, including transition zones

• Tie-in with the Phase 1 remediation project in upper OU 4 and prior remedial actions in Sediment Management Unit (SMU) 56/57

• Integration of 2006 and 2007 sampling data (as practicable)

• Development of preliminary design approaches in shoreline and utility crossing/structure areas (i.e., setback and stable slope assumptions). For this 30 Percent Design submittal, shoreline design is limited to general designs to accommodate structures.

To document the design effort, this report has been organized to provide the following: a summary of site characteristics from completed remedial design sampling and analysis events; updated dredge prism design since the BODR submittal; beneficial reuse opportunities and landfill disposal requirements for sediments; design criteria for staging area, sediment dredging, material handling, transportation and disposal of sediments, engineered capping, and sand covering; institutional controls; scheduling; and monitoring, maintenance, and adaptive management strategies.

In addition, attached to this report are the following appendices which support the document:

• Appendix A Dredge Design Support Documentation

• Appendix B Cap Design Support Documentation

• Appendix C Engineered Plan Drawings

• Appendix D Outline of Technical Specifications

Section 2 – Site Characteristics

8

2 SITE CHARACTERISTICS 2.1 Sampling and Analysis Data

2.1.1 Pre-Design Data

The RD sampling and analysis program includes data collection activities from 2004 through 2007 as described below. In addition, data collected prior to 2004 have been utilized, where appropriate, to support the remedial design. These data were compiled and summarized to provide an assessment of current information on the nature and extent of contamination, bathymetry and sub-bottom profiles of the river channel and side-slope areas, and the location of candidate areas for either active remediation or MNR, consistent with the ROD Amendment. Preliminary assessments of dredging, transport, upland landfill disposal, and MNR elements of the selected remedy, along with concurrent assessments of engineering capping and alternative disposal sites, focused initial (2004) sampling and analysis and initial RD efforts. Available information on land use plans, including future dredging and channel de-authorization plans, was also integrated into this analysis. The review and analysis of existing data focused on the portions of the OUs requiring active remediation as identified in the RODs as follows: OU 2 (Deposit DD), OU 3 and OU 4 (in their entirety), and OU 5 (immediately adjacent to the mouth of the Lower Fox River). The locations where samples were collected during the 2004 to 2007 RD field investigations are depicted in Figure 2-1, and included collection of the following:

• 1,696 Vibracore and surface sediment (0-10 centimeter [cm]) sampling locations; • 73 piston core locations and surface sediment (0-6 inch) sampling locations; • Approximately 130 in situ vane shear measurements from selected locations; • 855 sediment samples collected and analyzed for selected geotechnical parameters; • 9,328 sediment samples collected and analyzed for selected physical and chemical parameters; • 8 composite samples from different regions of the river tested for detailed chemical mobility and

desanding bench studies.

2.1.2 2004 Sampling and Analysis Program

Based on a review of the available data as outlined above, a detailed RD Sampling and Analysis Plan (SAP) and Quality Assurance Project Plan (QAPP) were prepared in 2004 (Shaw and Anchor 2004). The purpose of the 2004 Sampling and Analysis Program was to gather field and analytical data from OUs 2 to 5 as necessary to support RD. The sampling associated with the RD activities involved collecting a sufficient number of high-quality samples to define the extent of sediment exceeding the 1.0 ppm PCB RAL for future remedial activities associated with the sediment in OU 2 Deposit DD, OU 3, OU 4, and OU 5 adjacent to the Lower Fox River mouth. Sampling also included geotechnical characterization and determination of engineering properties for design purposes. In addition to sediment chemical and physical characterization, chemical mobility and treatability testing samples were also collected as part of the RD sampling. Data collection methods are described in detail in the agency-approved SAP and QAPP


9

and are not repeated herein. The 2004 sampling locations are depicted in Figure 2-1, and included collection of the following:

• 1,292 Vibracore and surface sediment (0-10 cm) sampling locations; • Approximately 130 in situ vane shear measurements from selected locations; • 726 sediment samples analyzed for selected geotechnical parameters; • 6,580 sediment samples analyzed for selected physical and chemical parameters; and • 8 composite samples from different regions of the river tested for detailed chemical mobility and

desanding bench studies.

The initial (2004) sampling grid provided a minimum coverage of 1 core per 6.2 acres, and 1 core per 1.6 acres in more critical areas, including areas with sediments potentially subject to TSCA disposal requirements. PCB core profiles were obtained on 0.5-foot vertical intervals (prior to compaction correction) to accurately determine the “neatline” depth of contamination, particularly relative to the 1 ppm RAL, and the delineation of deposits potentially subject to TSCA disposal requirements. Data collected during the 2004 Sampling and Analysis Program have been separately submitted to the Response Agencies; a brief overview of the program is provided in Section 2.1.2 of the BODR.

2.1.2.1 Data Validation

A data validation process was conducted to assess the reliability and usability of sampling and analysis data collected during the 2004 RD evaluation. The validation process is summarized in Section 2.1.2.4 of the BODR. The 2004 validation reports were provided to the Response Agencies in electronic format. The RD Data Report (Shaw/Anchor 2006a) presents the data validation report for the 2004 dataset. All data collected during this period were deemed acceptable for use in RD as qualified.


Following completion of the 2004 Sampling and Analysis Program, additional sample locations were needed in areas where increased definition of PCB distribution would better define the extent of contamination and allow for more accurate dredge prism design. In addition, more geotechnical information was needed for detailed design of the remedy. Subsequently, a detailed RD SAP and QAPP addendum was prepared to accomplish these objectives (Shaw and Anchor 2005). The 2005 Sampling and Analysis Program objectives were as follows:

• Further delineation of PCB distributions greater than 50 ppm along prospective TSCA boundaries in OU 4

• Further delineation of PCB distributions at 1 ppm within OUs 3 and 4

• Determination of PCB distributions within the upstream “off-limit” boundaries of the De Pere (OU 3) and Little Rapids (OU 2) Dams

• Geotechnical analysis within a prospective Cat Island habitat restoration and beneficial use area


10

The 2005 sampling locations are depicted in Figure 2-1, and included collection of the following:

• 191 Vibracore and surface sediment (0-10 cm) sampling locations; • 1,601 sediment samples analyzed for selected physical and chemical parameters; and • 98 focused geotechnical samples.

2.1.3.1 Testing Methods (Chemical and Geotechnical)

Testing methods used in 2005 were identical to those used in 2004 (see BODR Section 2.1.2.2).


A data validation process was conducted to assess the reliability and usability of sampling and analysis data collected during the 2005 RD evaluation. The validation process was equivalent to that used for the 2004 program, as summarized in Section 2.1.2.4 of the BODR. The 2005 data validation reports were provided to the Response Agencies in electronic format. The RD Data Report (Shaw and Anchor 2006a) presents the data validation report for the 2005 dataset. All data collected during this period were deemed acceptable for use in RD as qualified.


Following completion of the 2005 Supplemental Sampling and Analysis Program, and as design activities progressed, further data collection was needed in areas where increased definition of the PCB distribution would help define the dredge prism more accurately. In addition, more geotechnical information was needed to support evaluation of the engineered capping design. Subsequently, a detailed RD SAP and QAPP addendum was prepared (Shaw and Anchor 2006). The 2006 Sampling and Analysis Program objectives were as follows:

• Further delineation of sediment field data from OUs 2, 3 and 4 to support detailed design of the remedy along the edges of the river and adjacent in-water structures

• Data collection and survey work to support the design of sediment removal (dredging) and capping in the vicinity of shoreline and in-water features such as structures, slopes, and utility crossings.

The 2006 sampling locations are depicted in Figure 2-2, and included collection of the following:

• 180 Vibracore and surface sediment (0-10 cm) sampling locations; • 912 sediment samples analyzed for selected physical and chemical parameters; • 18 focused geotechnical samples; and • Physical observations and surveying of shoreline areas targeted for remedial action.




11


A data validation process was conducted to assess the reliability and usability of sampling and analysis data collected during the 2006 RD evaluation. The validation process was equivalent to that used for the 2004 program, as summarized in Section 2.1.2.4 of the BODR. The 2006 data validation reports were provided to the Response Agencies in electronic format. The RD Data Report (Shaw and Anchor 2006a) presents the data validation report for the 2006 dataset. All data collected during this period were deemed acceptable for use in RD as qualified.


The 2007 sampling involved collecting additional sediment core samples using vibracore or piston core methods to refine the 1-ppm PCB remediation prism in certain areas of OUs 2 to 5, and to obtain additional geotechnical information. In addition, at each vibracore location, a co-located surface sediment sample (top 10 cm) was obtained using a van Veen, Ponar, Eckman dredge, or equivalent “grab” sampling method. Surface grab samples were not collected at piston core locations since this sampling device was capable of retrieving a representative surface sample. Subsequently, a detailed RD SAP and QAPP addendum was prepared (Shaw/Anchor 2007). The 2007 sampling locations are depicted in Figure 2-3, and included collection of the following:

• 33 Vibracore and surface sediment (0-6 inch) sampling locations; • 73 piston core and surface sediment (0-6 inch) samples; • 235 sediment samples were analyzed for selected physical and chemical parameters; and • 13 focused geotechnical samples. • Physical observations and surveying of shoreline areas targeted for remedial action.




A data validation process will be conducted to assess the reliability and usability of sampling and analysis data collected during the 2007 RD evaluation. The validation process is equivalent to that used for the 2004 program, as summarized in Section 2.1.2.4 of the BODR. The 2007 data validation reports will be provided to the Response Agencies in electronic format.

2.2 Summary of Physical Site Characteristics

The BODR provides a summary of the physical characteristics of OUs 2 to 5 including:

• Site units and uses: usage (e.g. recreational, industrial, etc.) statistics; water depth and bathymetry; navigation channels; locks, and dams (see Table 2-1); infrastructure and utilities

• Regional geologic conditions


12

• Regional hydraulic conditions: Fox River flows and velocities, etc. • Geotechnical conditions: grain size, Atterberg Limits, etc.

The reader is referred to the BODR for a summary of each of these physical characteristics. In addition, the following section provides an updated summary of the geotechnical conditions in OUs 2 to 5, including the results of sampling conducted subsequent to the BODR.

Table 2-1. Low Water Pool Elevations in OUs 2 to 5

Pool Operable Unit River Mile Low Water Pool

Elevation (feet IGLD 85)

Lift (feet)

Green Bay OU 4/5 0.0 577.5

De Pere Dam OU 3 7.1 587.4 9.9

Little Rapids Dam OU 2 13.1 593.5 6.1

Note: Low Water Pool elevations from NOAA (2002)

2.2.1 Geotechnical Conditions

The BODR (Section 2.2) provides a detailed summary of the geotechnical properties of sediments sampled during the 2004 and 2005 RD field investigations. Table 2-2 provides an updated summary of the geotechnical properties for all samples collected during the 2004-2006 field investigations. Data from the 2007 field investigations will be incorporated into subsequent design deliverables.


13

Table 2-2. Summary of all RD Geotechnical Data a

Moisture Content b

[%]

Percent Solids b,c

[%]

Percent Sand/Gravel

[%]

Percent Silt/Clay

[%]

Liquid Limit [%]

Plasticity Index [%]

Organic Content

Specific Gravity

Dry Density b,c

[pcf]

OU 2/3 No. of

Samples 195 195 113 113 132 132 1 12 195

Avg. 159 35% 45 55 143 102 74 2.42 27 St. Dev. 103 20% 33 33 49 40 NA 0.15 28

OU 4/5 No. of

Samples 418 418 252 252 257 256 16 32 418

Avg. 134 32% 51 49 119 84 13.3 2.43 25 St. Dev. 102 19% 29 29 62 50 16.3 0.12 26

OU 2-5 No. of

Samples 613 613 365 365 389 388 17 44 613

Avg. 142 32% 49 51 127 90.2 16.8 2.43 25 St. Dev. 103 20% 30 30 59 47.2 21.5 0.13 27

a. Includes 2004, 2005, and 2006 sample results b. Corrected for core compaction c. Weighted average as discussed in BODR Appendix A (excludes 2006 shoreline samples)

2.3 Summary of Spatial Extent of PCBs

Extensive sampling efforts were conducted in 2004 and 2005 to characterize the nature and extent of PCBs in OUs 2 to 5. Geostatistical methods were used to delineate the depth of contamination (DOC) boundary in OUs 2 to 5, defined as the boundary beyond which sediment PCB concentrations are at or below the RAL of 1.0 ppm as specified in the ROD Amendment. Section 2.1 of this report discusses the additional sampling conducted in 2006 and 2007 to further delineate the spatial extent of PCBs within the shoreline areas of OUs 2 to 4. These data results were not included in the geostatistical analysis; however, improvements to the modeled neatline (the idealized representation of the extent of sediments exceeding the 1 ppm RAL) have been completed since publication of the BODR. The remainder of this section discusses the refinements (initially described in Geostatistics Technical Memorandum No. 4 [Anchor and LTI 2006d, Attachment A-1]) to the geostatistical model and the resulting updated neatline model surface.

2.3.1 Improved Geostatistical Delineation of Remediation Boundaries

The kriging model presented in the BODR was initially developed using the 2004 sampling data and evaluated with respect to a number of cross-validation metrics, which are discussed in detail in the BODR and technical memoranda (Anchor and LTI 2006a, 2006b, and 2006c). During subsequent collaborative Workgroup meetings, the kriging analysis was improved by including the 2005 RD sampling data and a series of refinements such as coordinate transformation (“river straightening”) based on shoreline


14

geometry, and adjustments to reflect historical channel features. Since the BODR, the following work has been completed:

• Inclusion of New Data. The updated kriging analysis incorporated additional sediment core data collected during a second phase of field work in May 2005. These additional cores were targeted towards areas where more precise delineation of depth to the RAL was needed, for example, in areas with steep concentration gradients. Several of the 2005 cores were located in areas where 2004 data indicated contamination extended all the way to refusal, to confirm that no further penetration of the sample coring device was possible.

• Channel Segregation. The federal navigation channel, including the recently reauthorized portion in OU 4A and the active portion in OU 4B, was segregated and kriged separately. This was done because of the distinct character of the channel and its past activities: previous interpolations, which did not segregate channel from out-of-channel locations, had consistently underestimated DOC in the channel, while DOC on the nearshore benches was being overestimated. In addition, the boundaries of the channel for geostatistical purposes were extended 22 feet beyond the actual channel line on either side. It was determined by inspection that the DOC in all cores within this distance on the channel margins was consistent with cores in the channel proper, whereas further widening would have included samples with much shallower DOC outside the influence of channel activities. This indicates some disturbance and sloughing of the sidewalls likely occurred during channel dredging as might be expected.

• River Straightening. In the previous kriging model, the primary correlation axis was fixed along the average direction of the OU 3 and 4 reaches. Along river bends, however, a fixed correlation axis will sometimes deviate from the local flow direction, generating interpolations of depositional features that are oblique to the direction of the river. These artifacts were corrected by performing a coordinate transformation (“river straightening”) based on shoreline geometry. This technique allows the correlation axis to align with the local flow direction, and interpolates between data points along paths that follow the bends in the river. This type of model also conforms better with geomorphological principles.

The supplemental kriging analysis was performed step-wise to evaluate the potential improvements associated with the new data and the “physical” modifications separately. The cross-validation metrics were updated for each reach and for OU 4, and are discussed in detail in Geostatistics Technical Memorandum No. 4 (Anchor and LTI 2006d). This verification process was also completed for OU 3 and presented to the Response Agencies in a series of Workgroup meetings.

The 2005 data were preferentially located in areas of uncertainty based on the 2004 data, and the greater difficulty of prediction in those areas is reflected in a slight deterioration of the cross-validation metrics when the 2005 data are added to the unstraightened model. With the full 2004-2005 dataset included, however, straightening improved most of the metrics for both OU 3 and OU 4. A key advantage of the new model was its ability to more accurately predict the DOC as indicated in the summary statistics presented in Table 2-3. For example in OU4A, this is reflected in the reduction in the Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE). This is at least partly attributed to more accurate predictions of DOC in the reauthorized OU 4A navigation channel and DePere turning basin. The DOC in these areas had been consistently underestimated in the previous model. The geostatistical metrics are discussed in detail in Geostatistics Technical Memorandum No. 4 (Anchor and LTI 2006d).


15

Table 2-3. Summary of Kriging Cross-Validation Metrics for OUs 3 and 4 2004 2004 and 2005 Data

Unstraightened

River Straightened River & Segregated Channel1 OU 3 Total

Significance Level: 0.5 0.5 0.5 0.4 0.3 0.2 0.1 False Positives 46% 40% 34% 38% 46% 55% 58% False Negatives 21% 22% 19% 15% 13% 11% 11%

Sensitivity 49% 51% 58% 70% 79% 87% 88% Specificity 83% 84% 86% 80% 67% 50% 43%

Percent Correct 73% 73% 77% 76% 71% 62% 57% RMSE 0.5 0.6 0.5 0.5 0.6 0.7 0.8 MAE 0.3 0.3 0.3 0.3 0.3 0.5 0.6 Bias -0.1 -0.1 -0.1 0.0 0.1 0.3 0.4

OU 4A Significance Level: 0.5 0.5 0.5 0.4 0.3 0.2 0.1

False Positives 15% 15% 13% 17% 19% 21% 24% False Negatives 22% 25% 25% 19% 13% 8% 5%


Percent Correct 83% 82% 83% 83% 82% 81% 79% RMSE 2.2 2.4 1.9 1.9 2 2.4 2.8 MAE 1.3 1.4 1.1 1.1 1.3 1.5 1.8 Bias -0.3 -0.3 -0.3 0.1 0.5 1.1 1.5

OU 4B Significance Level: 0.5 0.5 0.5 0.4 0.3 0.2 0.1



Percent Correct 80% 79% 77% 77% 77% 74% 72% RMSE 2.5 2.6 2.6 2.8 3 3.6 4.3 MAE 1.7 1.8 1.8 1.9 2.1 2.5 3.2 Bias -0.3 -0.2 -0.3 0.3 1 1.7 2.7

OU 4 Total Significance Level: 0.5 0.5 0.5 0.4 0.3 0.2 0.1



Percent Correct 82% 80% 80% 80% 79% 78% 76% RMSE 2.3 2.4 2.2 2.3 2.5 3 3.5 MAE 1.4 1.6 1.4 1.5 1.6 2 2.4 Bias -0.3 -0.3 -0.3 0.1 0.7 1.4 2

Note 1: Channels were only segregated in OU 4.


16

2.3.2 Spatial Extent of PCBs Exceeding 1.0 ppm

The spatial distributions of PCB mass in OUs 3 and 4 are presented on Figures 2-4 and 2-5, respectively. The depth of contamination in OUs 3 and 4 is presented on Figures 2-6 and 2-7, respectively, using the refined, full indicator kriging model with a significance level of 0.5. The surface represented in DOC maps were subtracted from the mudline elevation to generate an elevation surface of the bottom of contamination, and used to develop the dredge plans (see Section 4.3).

Plan-view maps of the spatial extent of PCB concentrations above the RAL in OUs 3 and 4 (i.e., the “footprint” of the remediation boundary) are presented on Figures 2-8 and 2-9, respectively, using the refined full indicator kriging model at significance levels of 0.5, 0.4, and 0.3. Figures 2-4 through 2-7 provide important information on the PCB mass inventory in the sediments which is not captured in the indicator kriging maps on Figures 2-8 and 2-9, because indicator kriging discretizes data in terms of whether or not the RAL is exceeded, but does not convey information on the magnitude of the exceedance (i.e., how high the PCB concentrations are relative to the RAL). Together, these various sets of maps characterize the spatial distribution of PCBs in the project area.

2.4 Characterization of Material for Beneficial Use and Disposal Purposes

The BODR provides a comprehensive review of the RD considerations relative to the characterization and quantification of dredge material for disposal including NR 500 landfills, TSCA-licensed facilities, and other facilities that provide dewatering and/or disposal. Potential beneficial use options also exist for sediments containing relatively low concentrations of PCBs. The section below describes the methodology for making such determinations, and summarizes the estimated extent and volume of sediments that may be subject to TSCA regulation when removed and disposed, consistent with Addendum No. 3 to the RD Work Plan.

2.4.1 Beneficial Use Opportunities

Based upon the results of the 2004 and 2005 RD investigations (Shaw and Anchor 2005, 2006), the sediments targeted for removal as part of the OU 2 to 5 remedial action consist mainly of sand and silt-sized particles (75 to 80 percent by weight, see Table 4-1), with the remaining percentage consisting mainly of clay (approximately 20 percent) and a trace to slight amount (less than 5 percent) of gravel. The sand and gravel comprise approximately 40 percent by weight of the OU 3 and OU 4 sediment samples collected within the dredge prism.

The PCBs in the sediments from OUs 2 to 5 are largely adsorbed onto the fine-grained (less than 200 mesh) soil fractions of the sediment. Bench-scale treatability studies conducted with representative composite samples collected from OUs 3 to 5 indicate that the coarse-grained fraction of the sediments (greater than 200 mesh), referred to in this document as the sand/gravel fraction, have PCB concentrations less than 1 ppm. The separated sand/gravel fraction may be considered for possible beneficial use. As


17

discussed in Section 5.8, roughly 30 percent (by weight) of the total sediment solids in the dredge slurry can potentially be separated as sand/gravel containing less than 1 ppm PCBs using physical separation technologies as part of the sediment dewatering process. As such, approximately 260,000 cy of sand containing less than 1 ppm PCBs are anticipated to be available for potential beneficial use if desanding technologies are applied to the entire 3.6 million cy of non-TSCA dredge volume in OUs 2 to 5 (see Section 4.4).

Under Wisconsin Statute 289.43, the WDNR can be petitioned for an exemption for the management of low-hazard waste, covering the non-TSCA dredged material for beneficial reuse. Typically, WDNR evaluates the beneficial reuse of dredge sediments using its NR 538 regulations.

Based on the volume (260,000 cy) of separated sand, several beneficial use options have been considered during the RD. Approximately 150,000 cy of the segregated sand could be used to restore on-site borrow pits from which material was obtained to fill in behind the sheet pile wall constructed at the former Shell Property (as described in Section 3.2). Opportunities for beneficial use of the remaining approximately 110,000 cy of segregated sand, as described in the BODR, include: manufactured soil, landfill daily cover/internal structures, confined geotechnical fill, and transportation embankment material. These alternatives will be considered at later stages of the project based on the rate of generation and the rate of potential use.

2.4.2 Sediments Subject to Non-TSCA Disposal Requirements

As discussed in Section 2.4.2 of the BODR (Shaw and Anchor 2006), designation of dredged material that is suitable for non-TSCA disposal was based on 6-inch sampling depth data from individual RD sediment cores. Once all cores were analyzed using this in situ designation methodology, the vertical and horizontal extent of sediments requiring disposal in a TSCA-licensed landfill was delineated (see Section 2.4.3 below - Sediments Subject to TSCA Disposal Requirements). USEPA and WDNR have determined that sediments designated for non-TSCA disposal using this in situ methodology (i.e., PCB concentrations < 50 ppm) meet the substantive requirements for PCB testing for receiving landfill facilities, obviating the need for further PCB verification testing.

The following landfill requirements for disposal of non-TSCA material currently apply to both existing and planned (but currently unconstructed) landfill facilities: 1) the dewatered sediment must have the ability to support its own weight; 2) support the weight of material placed over it; 3) be capable of holding a stable 3H:1V slope under dynamic conditions; 4) have a minimum unconfined compressive strength of 0.8 tons per square foot.

2.4.3 Sediments Subject to TSCA Disposal Requirements

As discussed in Section 2.4.2 of the BODR, for designating dredged material that may be subject to TSCA disposal requirements, sampling data from RD sediment cores were vertically composited across


18

non-overlapping 2.5-foot (30-inch) sediment intervals beginning at the mudline. The 2.5-foot vertical interval accounts for normal operational tolerances including consideration of constructable dredge prisms and overdredge allowances. The sediment volume range in OU 4 derived from this analysis that may potentially be subject to TSCA disposal requirements is approximately 140,000 to 170,000 cy, excluding the volume of sediment removed as part of the Phase 1 dredging project that was subject to TSCA disposal requirements. This analysis did not identify any sediments in OU 2, OU 3, or OU 5 that are anticipated to require disposal in a TSCA-licensed landfill.

Disposal sites for PCB impacted sediments classified as TSCA material were inventoried as part of the BODR. The two existing sites closest to OUs 2 to 5 are EQ Wayne Disposal in Bellevile, MI and Peoria Disposal Company in Peoria, IL. No additional sites have been identified, and there are currently no disposal sites in Wisconsin licensed to receive PCB impacted sediments greater than or equal to 50 ppm. It is possible that disposal site(s) in Wisconsin could be permitted in the future to accept sediments potentially subject to TSCA disposal requirements. If this alternative becomes available in the future, it will also be evaluated at that time.

The landfill requirements for disposal of TSCA material will be based on existing RD sampling data. Additional PCB analysis will not be required on an ongoing basis during the project. The data from the in situ RD sampling, as detailed in the BODR, is representative of materials targeted for shipment to TSCA-licensed landfills. The remedial contractor will certify these conditions through an initial waste profile sheet and ongoing manifests.

Section 3 – Site Preparation and Staging Area Development

19

3 SITE PREPARATION AND STAGING AREA DEVELOPMENT 3.1 Staging Area Requirements and Design

The remedial action will require an upland staging area for desanding, dewatering, and water treatment operations discussed in Section 5. These operations require sufficient space for desanding equipment, a large circular clarifier tank, mechanical plate and frame presses, and water collection and treatment systems, as well as support facilities such as reagent silos, air compressors, electrical switchgear, and an operator control station. Dedicated areas are also required for various stockpiles, including sand from the desanding operation, dewatered sediment fines from the filter presses destined for landfill disposal, and clean sand and gravel for sand cover and engineered capping. In addition, the upland staging area can be used by the contractor for equipment storage, and will provide the location for the field office.

The staging area also requires water access and sufficient berthing depth and length to accommodate floating equipment used for the remedial action. Capping materials may be delivered via supply vessels (e.g., barges or bulk cargo ships) and would need sufficient water depth and berthing length along a dock to tie up and offload.

Additional staging areas may be necessary to support the work in the furthest upstream reaches (OUs 2 and 3), but have not been specifically identified at this time.

Other site improvements at the primary staging area will be required to support large-scale sediment dewatering as follows:

• Containment Slab - An approximate 300-foot x 500-foot area is needed for installation of desanding, dewatering, and water treatment equipment. A reinforced concrete slab atop a compacted base will be constructed over most of this area to support the equipment. The slab also provides a stable surface for wheel loaders and other rolling stock used during operation and maintenance of the equipment. Site geotechnical investigations will determine if some equipment requires sub-grade footings and support piers. The concrete slab will have a perimeter curb which will provide liquid containment for operations. All construction joints will have water-stop gaskets to prevent leakage. The slab will be sloped toward in-ground sumps that collect process water run-off and rainwater. Water from the sumps will be pumped to the water treatment system and treated prior to discharge.

• Filter Press Structure - The filter press area of the dewatering system will be covered with a pre-engineered, open-sided steel structure to allow efficient press operation during periods of inclement weather.

• Treated Dredge Material Storage – Segregated staging areas are required for processed dredge materials (e.g., washed sand and filter press solids) awaiting transport. It is anticipated that five days of storage for both sand and filter press solids is needed to accommodate confirmatory testing and transportation logistics. Material from the processing area will be moved to the bins by large wheel loaders. The washed sand storage area and filter press solids storage areas will be separated to prevent cross contamination by transport trucks and material handling equipment.


20

The base of the material storage area will consist of at least two courses of asphalt on compacted fill with a welded membrane liner beneath. The asphalt will slope to one or more sumps such that pile drainage and/or rainwater is collected and pumped to the water treatment system. The sand and filter press solids storage areas will each be divided into five bins approximately 50 feet wide by 100 feet long by Jersey barriers. Each bin will accommodate approximately one day’s production of either sand or filter cake solids. The material pile may be covered with tarps if needed during inclement weather.

• Clean Material Storage – Storage areas are required for incoming sand and gravel used for capping. These materials will be placed in a segregated area near the shore on compacted crushed stone.

• Truck Scales – Two platform scales will be installed near the roadway, one for incoming trucks and one for outgoing trucks.

• Equipment Laydown Area – A laydown area will be provided for contractor equipment. It will consist of graded, compacted crushed stone.

• Equipment Decon Pad – A reinforced concrete pad with containment curb and sump will be provided for cleaning of equipment. This pad will be used for rolling stock, processing equipment and any other item being demobilized that has come in contact with PCB-contaminated material.

• Electrical Service – The local utility will be contacted to determine a tie-in point for electrical service. The utility tie-in will feed a pad-mounted step-down transformer which will be located near the processing plant in order to supply 460 V / 3 phase power for the equipment. Transformer rating will be determined once total system electrical requirements have been calculated. A modular or room-mounted motor control center will be installed immediately adjacent to the processing plant in order to distribute power to the individual motors and other electrical loads.

• Administrative Offices – Modular offices will be provided near the entrance roadway. If feasible, these offices will be tied to city sanitary sewer and water service.

• Security Fence – A security fence and signage will be installed around the perimeter of the property.

3.2 Potential Staging Area Locations

The BODR identified the former Shell Oil Property adjacent to the Georgia-Pacific West Mill in OU 4 as the most promising staging site for the OUs 2 to 5 work. The former Shell Property currently has approximately 22 acres of useable upland space, which should be sufficient to accommodate most, but not all, contractor operations on-site. The site will require expansion, by way of filling up to the existing bulkhead line, to accommodate the on shore operations. There are existing roadways and good accessibility to transportation corridors directly from the staging site. A portion of the property (as illustrated in Figure 3-1) will require some form of footings, pilings or foundations to support heavy processing equipment; filter presses, clarifiers, sand and carbon media vessels (Figure 3-2). A geotechnical survey of the staging site will be conducted in the spring of 2008 to confirm the load-bearing conditions used in foundation design. Several site improvements will be needed as outlined below in order for this site to be used as the staging area.


21

The existing shoreline along the eastern boundary of the Shell Property will be expanded and improved to generally match the shoreline of the adjacent northern property. This will involve installing approximately 1,500 lineal feet of steel sheetpile wall along the bulkhead line, creating a wharf area with 750 feet accessible for berthing along the face. It is anticipated that the sheetpile wall will consist of interlocking Z-type piles, driven up to 75 feet below the river bottom, based on similar depths of existing sheetpile installations in this area of the river. Geotechnical investigations will be performed during subsequent RD to further refine the required sheetpile design (and sheetpile depths). With the dredged river bottom elevation of roughly 560 feet IGLD 85 and the top of bank elevation approximately 585 feet; the total length of sheetpile will be up to 100 feet. Final design of this sheetpile wall will be determined based on a geotechnical survey results for the area.

A portion of the sheetpile could be installed from the shoreline, however, floating equipment will be required for installation of most of the sheetpile wall due to the distance from shore. Based on the existing shallow water conditions along the sheetpile wall alignment, its is expected that a limited amount of dredging (approximately 15,000 cy) will be performed to allow floating construction equipment to install the offshore portions of the sheetpile wall, and also to allow access to offload debris removed from within the wall alignment. Based on the PCB concentrations of the sediments in this area, it is expected that at least a portion of sediments dredged to facilitate sheetpile wall installation (approximately 1,500 cy) will be handled and disposed of in accordance with TSCA regulations. The access channel dredging and sheetpile installation are expected to take nearly a full construction season to complete (see Section 9).

Following sheetpile wall installation, mechanical dredging of targeted sediments within the sheetpile enclosure will be performed. While approximately 4,500 cy of non-TSCA sediment and 16,000 cy of sediment potentially subject to TSCA disposal requirements will be removed from within the sheetpile area as a part of site development, more deeply buried underlying contaminated sediment with lower PCB concentrations could be contained in place below the improved staging area fill. As discussed in the BODR, historical shoreline filling actions that occurred during the 1960s and 1970s in this area of OU 4 have similarly contained PCBs below the adjacent uplands. Because of the presence of considerable debris and other piling in this area, removal of such deeper sediment could be difficult and associated with relatively greater water quality and dredge residual impacts. These impacts will be obviated by installing the sheetpile wall prior to dredging of sediments potentially subject to TSCA disposal requirements, and by using containment/fill technologies to address the remaining, more deeply buried contaminants.

In addition to the dredging within the sheetpile enclosure, dredging will be performed immediately adjacent to the river side of the sheetpile wall to support suitable approach and berthing area functions (Figure 3-1). Dredging offshore of the sheetpile wall will provide 15 to 19.5 feet of water depth (i.e., roughly elevation 562.5 to 558 feet IGLD 85). Development of the wharf area will involve filling and conversion to uplands of approximately 5 acres of OU 4. This configuration will meet the staging area


22

criteria outlined above. Building out to the bulkhead line increases the upland acreage to approximately 28 acres, as necessary to support sediment processing and aggregate stockpiles. The backfill volume will be approximately 150,000 cy, assuming a top of bank elevation of approximately 585 feet IGLD 85. This upland backfill will use appropriate on-site “borrow” material and a portion of the separated sand (less than 1 ppm PCBs) from the desanding process.

It is currently anticipated that initial remediation efforts will include mechanical dredging of non-TSCA sediments from OU 2 and upper OU 3. This material will be loaded onto barges. However, it is not cost effective to deliver these materials to the Shell Property staging area due to long transit time and the small barge volume that the locks will accommodate. A separate staging area in OU 2/upper OU 3 is currently under consideration for this material. Dredging and capping actions in OU 2 and upper OU 3 will be part of the initial remediation efforts (see Section 9).

3.3 Real Estate, Easement Requirements and Location-Specific ARARs

3.3.1 Real Estate and Easement

At this time there are no real estate agreements or specific easements identified that are necessary for implementation of OUs 2 to 5 remedial actions, with the exception of: 1) negotiating long-term leasing arrangements for the former Shell Oil Property; 2) negotiating an easement to cross the DePere Dam with the dredge slurry pipeline; and 3) negotiating with one or more property owners for an agreement to lease or otherwise use shoreside property to support mechanical dredging and capping in OU 2 and upper OU 3.

3.3.2 Possible Location-Specific ARARs

The following lists location-specific, applicable or relevant and appropriate requirements (ARARs) which might apply to the various activities associated with the OUs 2 to 5 remedial action. Under CERCLA, the procedural requirements of federal, state and local permits are waived, but the substantive requirements of the permits must still be satisfied.

Federal – Substantive federal requirements may include:

• Protection of navigable streams and wetlands – These USACE permitting activities are usually handled through joint state/federal permitting submittals. Any modifications to navigable streams (including dredging), the banks of streams, or wetlands hydraulically connected to streams require ACOE permitting. Wetland delineations may be required to meet the substantive requirements of these regulations.

• Protection of maritime navigation – To ensure continued, safe navigation, the USCG oversees and must approve all private aids to navigation. Application submittals for beacons, buoys, and other such use must be submitted to USCG for approval.

• Protection of floodplain areas – Any activity, such as filling within a designated floodplain, which would cause increase to the regulatory floodplain elevations (100-year flood) are subject to FEMA regulations. Floodplain hydraulic evaluation can determine if such activities would cause


23

any backwater increases and if a submittal to FEMA would be required. Section 5.5.4 of the BODR presented an evaluation of the potential alterations to the 100-year flood plain as a result of the OU 2 to 5 remedial action and concluded that the remedy would not adversely impact the flood-plain.

• Protection of significant archeological or historic sites – The findings of significant archeological or historic artifacts on a site may require special review to comply with the substantive requirements of either federal or state programs. Documenting, mapping, and field inspection and survey of such finds may be required.

State – Given authority to regulate certain areas of environmental protection under federal jurisdiction, the WDNR requires permits (or equivalency thereof) under various regulatory programs. For protection of waters of the state, Chapter 30 permitting is the primary regulatory tool. For stormwater management, the WDNR Notice of Intent (NOI) program is supplemented with the Wisconsin Department of Commerce (WDOC) drainage and stormwater management regulations.

• Protection of navigable streams and wetlands – These WDNR permitting activities are usually handled through joint state/federal permitting submittals. Any modifications to navigable streams (including dredging), the banks of streams, or wetlands require WDNR permitting. Depending on the presence of wetlands, Areas of Special Natural Resource Interest (ASNRI), or Outstanding and Exceptional Resource Waters, wetland delineations and possible replanting may be required to meet the substantive requirements of these regulations.

• Drainage and Stormwater Management – The drainage and stormwater management of industrial and commercial establishments are regulated by the WDOC. Given the nature of the OUs 2 to 5 remediation efforts, it is unlikely that WDOC submittals would be required for remediation activities.

• Protection of significant archeological or historic sites – The findings of significant archeological or historic artifacts on a site may require special substantive review to comply with the substantive requirements of either federal or state programs. Documenting, mapping, and field inspection and survey of such finds may be required.

• Erosion Control and Stormwater Management – The WDNR’s stormwater management program (under NR 216 and NR 151) requires NOI submittals for any land disturbance greater than one acre (e.g., the Shell Property staging area or any additional staging areas used for the project). This does not apply to in-water dredging operations. Buffer areas around waters of the state are also included in these regulations. The stormwater management aspect of these regulations relate to post-construction features which would probably not apply to river remediation efforts. The associated erosion control efforts, however, would require submittal of the following:

o Erosion Control Plan (showing pollutant reductions of 80 percent)

o Site info, erosion prevention, flow control, monitoring plan

o Buffer Areas

o Best Management Practices (BMPs)

o The reduction of pollutants and erosion control to the Maximum Extent Practicable (MEP)


24

Under WDNR’s procedural requirements (which are not required under CERCLA), information is submitted to the WDNR in a NOI application which typically requires forms, a fee, and such additional information as:

• USGS/Location Map

• Wetland Inventory Map

o Site Photographs

o Narrative of Project

• Soils Information

• SLAMM (Source Loading and Management Model (water quality) Modeling)

• Dewatering Plan Checklist

• NOI Pre-Screen Correspondence

• WDOT PAL (Wisconsin Department of Transportation Product Acceptability List for Erosion Control) and Shear Calculations

• Materials Management Checklist

• Construction Specifications (Erosion Control Measures)

• Construction Drawings

• Sequence of Work Plan Checklist

• NOI Plan Review Checklist

• Monitoring (Inspection) is also required until the site is stabilized (solid vegetative cover). At that time a Notice of Termination (NOT) must be submitted to the WDNR.

In addition, while county and local requirements are not ARARs under CERCLA, the design will attempt to comply with the substantive provisions of these requirements to the extent reasonable. These requirements include:

County Shoreland Zoning - Counties have been delegated the responsibility to protect shorelands and floodplains and do so through their shoreland and floodplain zoning regulations. Additional investigation into Brown County specific requirements will be performed as part of the 60 Percent Design.

• Protection of floodplain areas – Any activity, such as filling within a designated floodplain, which would cause increase to the regulatory floodplain elevations (100-year flood) are subject to county shoreland regulations. This area used to be regulated under state Chapter 30 authority, but that authorization has now been delegated to the counties, and to local civil court control. Floodplain hydraulic evaluations, such as those presented in the BODR, can determine if such activities would cause any backwater increases.

• Protecting shoreland areas – Generally, these regulations apply to the banks of streams, steep slopes along the streams, and a zone which varies depending if an upland pond, floodway, or other factor is involved.


25

• Site remediation and redevelopment – These state regulations may apply where an upland site contains certain contaminated soils or materials. Site specific information would be needed to determine the applicability of these regulations.

County Planning - Under the Clean Water Act Section 208 review authority for sanitary regulations, county planning Response Agencies sometimes take on the responsibility to regulate Environmentally Sensitive Areas (ESAs). The 208 review authority refers to the federal program which grants authority to the state for water quality management. The authority is passed on to county planning agencies through Wisconsin Administrative Code NR 121. It pertains to sanitary sewer extensions and environmentally sensitive area controls. This is the case in Brown County, which have detailed stipulations for development or modifications within the ESAs. ESAs include streams, wetlands, and buffer areas surrounding these features. The buffers around ESAs vary along streams, floodplains, and wetlands.

• Protection of navigable streams and wetlands – Any modifications to navigable streams, the

banks of streams, wetlands, or lands within buffers associated with these features require Brown County Planning permitting. The modifications to ESAs must gain approval through the ESA amendment process which is often quite arduous.

Local Zoning, Stormwater Management, Planning, and Building Permits – Villages, cities, towns and counties have varying regulations and ordinances pertaining to a variety of activities.

• Protection of maritime navigation – Some communities adjacent to waters of the state oversee navigation aid installation within their jurisdiction in addition to the USCG.

• Erosion Control and Stormwater Management – The WDNR’s stormwater management program is moving into local control. These controls apply to land disturbance generally greater than one acre, though some communities are dropping that threshold. This does not apply to in-water dredging operations. Buffer areas around waters of the state are also sometimes included in these regulations. The stormwater management aspect of these regulations usually relate to post-construction features which would probably not apply to river remediation efforts. The associated erosion control efforts, however, would require submittal of erosion control plans and other site information.

• Building Permits – Local authorities control building construction through their building permitting process. Often, variances are needed to allow certain construction in certain land use zones.

• Planning – Local authority is sometimes granted to local planning Response Agencies to take on the responsibility to regulate certain environmental areas. Buffers around such areas may be in place which may require variances to perform certain activities.

The following are links to selected agency web sites with information relevant to ARARs that may be applicable to the OUs 2 to 5 remediation activities.

Wisconsin Department of Natural Resources (WDNR)

Waterway http://www.dnr.state.wi.us/org/water/fhp/waterway/


26

Stormwater Management http://www.dnr.state.wi.us/org/water/wm/nps/stormwater.htm

Areas of Special Natural Resource Interest (ASNRI) http://dnrmaps.wisconsin.gov/imf/imf.jsp?site=SurfaceWaterViewer.deswaters

Waterway Viewer http://dnr.wi.gov/maps/gis/appwebview.htm

Wetland Mapping http://dnr.wi.gov/org/water/fhp/wetlands/mapping.shtml

Outstanding and Exceptional Resource Waters http://dnr.wi.gov/org/water/wm/wqs/orwerw/

Natural Resource Conservation Service (NRCS)

Soils Maps http://websoilsurvey.nrcs.usda.gov/app/

Section 4 – Sediment Dredging

27

4 SEDIMENT DREDGING 4.1 Summary of Sediment Physical Properties

Table 4-1 presents a summary of the geotechnical properties for all samples collected within the limits and extents of the 30 Percent Design dredge prism in OUs 2 to 5 during the RD investigations. The sediments targeted for dredging in OUs 2 to 5 can be generally characterized as soft, silty, clayey sand with an average in situ percent solids content of approximately 32 percent by weight. The sediment within the target dredge prism is approximately 37 percent sand, 36 percent silt, and 25 percent clay by weight, with the remaining trace fraction being gravel-sized particles. The data presented in Table 4-1 has been corrected for coring-induced sample compaction. In addition, Table 4-1 presents a weighted average estimate of percent solids data to be used in dredge design and disposal volume calculations, which accounts for both the vertical and spatial representation of each sample. In some locations within the river, a layer of stiff native clay was identified beneath the soft sediment targeted for dredging. Appendix A, Attachment A-6 provides a complete summary of geotechnical data collected within OUs 2 to 5 as well as data collected specifically within the targeted dredge prism.

Table 4-1. Geotechnical Properties of Sediments Targeted For Dredging Parameter OU 2/3 OU 4/5 OU 2 to 5

Fines Content (% Finer than No. 200 Sieve)

Avg.(a) 77% 61% 62%

Std. Dev. 12% 23% 23%

Percent Solids by Wt. (%)

Avg.(b) 35% 32% 32%

Std. Dev. 14% 14% 14%

Dry Density (pcf)

Avg.(b) 27 25 25

Std. Dev. 15 18 17 Notes:

(a) Numerical average of measured data (b) Weighted average of measured data (See Appendix A of BODR)

4.1.1 Dredgeability

Dredgeability refers to the physical characteristics of the proposed dredge material and how readily the material can be dredged using different pieces of equipment. One typical measurement of dredgeability is the relative density of the in situ sediment, which can be measured using the Standard Penetration Test (SPT) and is expressed in blow counts (N-value). In general, the higher the SPT blow count, the harder the material is to dredge. Based on the results of RD investigations, including soft sediment poling data, the material within the dredge prism is expected to be very soft, with very low or even zero (i.e., “weight of rods”) blow count readings, with buried denser deposits at the native sediment contact interface. In


28

general, the buried native deposits have been characterized as stiff clay (additional geotechnical characterization of the encountered materials are provided in Appendix A, Attachment A-6). Limited areas of very hard materials or rock (i.e., “refusal”) immediately underlying sediment with >1 ppm PCBs were encountered during RD sampling activities in OUs 2 to 5, as shown on Figures 4-1 and 4-2, respectively.

Dredgeability of the material affects the type and size/power of equipment that the contractor will use for dredging. The material to be removed from OU 3 and OU 4 consists of sands and silts, with some clay component as summarized in Table 4-1. Using the available grain size data, a preliminary evaluation was performed and presented in the BODR (Section 3.2) to identify the general size and power range of equipment that will be used to hydraulically transport the material (e.g., horsepower required for booster pumps), the ability to cut and remove the sediments, and the potential for coarser-grained sediment to inhibit production.

4.1.2 Seasonal Construction Windows and Weather-Related Work Impacts

The BODR discusses the constraints of performing in-water construction activities during the cold-weather months. To alleviate winter construction impacts, a work window is established during which all in-water work will occur. Outside of this window, major dredging operations will not take place; however, maintenance and more routine, non-weather-dependent activities may occur. The work window for dredging and disposal is envisioned as a 180-day construction window between April/May and October/November to avoid seasonal and weather-related impacts. Non-weather dependent activities (mobilization, winterization, site preparation, etc.) may be completed during the several weeks immediately preceding or following this in-water work window. Scheduling and sequencing of remedial actions in OUs 2 to 5 is discussed further in Section 9 of this report.

In addition to the planned seasonal shut down of major operations during the winter, other seasonal weather patterns could affect the efficiency with which work is completed. Low water levels in the summer or storm events resulting in high wind or current velocities can disrupt dredging production. Operational procedures will be formulated to adjust for any large flow fluctuations and to secure any completed activities from damage or erosion of exposed contaminants. Therefore, the number of active, uncompleted dredging reaches will be limited to the extent possible to reduce the risk during these transient events.

4.1.3 Federal Navigation Channel Considerations

As discussed in Section 2.2, the federal navigation channel in the Lower Fox River extends 7.1 miles from the mouth of the river at Green Bay to the De Pere Dam. Upstream of the turning basin, in OU 4A (extending from the southern extent of the Fort Howard turning basin to the De Pere Dam), the federally authorized navigation channel was recently reauthorized in width and depth (see below). Several years ago, the OU 4A channel was placed in “caretaker” status and not actively maintained. As part of the


29

Water Resources Development Act (WRDA) of 2007 (Public Law 110-114), the width and depth of OU4A channel was re-authorized to 75 feet and 6 feet, respectively.

For the purposes of preparing the engineering drawings for this 30 Percent Design, the centerline of federal navigation channel was used as a baseline for referencing position within the river.

4.2 Equipment Selection and Production Rates

4.2.1 Equipment Selection Process

The primary method of removal for both TSCA and non-TSCA sediment material will be through hydraulic dredging and pipeline transfer. Mechanical dredging is planned for non-TSCA sediment in OU 2 and upper OU 3 due to site constraints (primarily the distance between these areas and the Shell Property staging area in OU 4). Mechanical dredging of TSCA and non-TSCA sediment is also planned in OU 4 adjacent to and inside the sheetpile wall (bulkhead) that will be constructed at the former Shell Property. Approximately 24,500 cy in OU 2, 20,000 cy in OU 3, and 35,500 cy (17,500 cy TSCA and 18,000 cy non-TSCA, including initial access channel dredging) in OU 4 are anticipated to be suitable for mechanical dredging. It should be noted that the mechanical dredge volume (and subsequent backfill quantity) anticipated in OU 4 adjacent to the former Shell Property may be refined during 60 Percent design pending further consideration of site constraints. Mechanical removal (via barge-mounted crane with clamshell bucket or barge-mounted excavator) will be used mainly for debris removal.

4.2.2 Production Rate Considerations

For this project, and for subsequent dewatering process flow considerations, a single 12-inch hydraulic dredge (as determined in the BODR) is assumed to have a production rate of 208 in situ cy/hour with an average pipeline slurry concentration of approximately 14 percent, by volume, based on analyses presented in the BODR. This pipeline slurry concentration is based upon a flow rate with multiple passes of lessening in-situ concentrations resulting in average concentrations of in situ sediments at 14 percent by volume. Preliminary confirmation of these production rate estimates was provided through discussions with experienced dredging contractors.

The in situ dredge rate of 208 cy/hour and 14 percent dredge volume concentration results in a volumetric dredge flow of just over 5,000 gallons per minute (gpm) as calculated below:

• 208 cy/hr x 1/0.14 x 1 hr/60 min x 27 cf/cy x 7.48 gal/cf = 5001 gpm dredge flow

The average dredge slurry concentration will be approximately 5.4 percent solids by weight based on an in situ concentration of 32 percent solids by weight for OUs 3 to 5, as summarized in Table 4-1. However, the dredge slurry concentration at any given time will vary somewhat depending on physical properties of the sediment deposit being dredged and on other operational factors including thickness of cut.


30

A dredge efficiency, or percent “uptime” (expressed as the percentage of the working day that dredging is active) of 64 percent, or 15.36 hours per 24-hour day, has been assumed, as presented in the BODR. The estimated dredge production rate of 208 in situ cy/hour and a dredge efficiency of 64 percent results in a daily in situ removal rate of approximately 3,200 cy/day:

• 208 cy/hr x 24 hr/day x 0.64 = 3,195 cy/day

The production rate for hydraulic dredging will vary both in daily uptime and slurry concentrations, and could be impacted by the capacity and mechanical reliability of the dewatering system. Therefore, the dewatering system has been designed with additional capacity, as discussed in Section 5, in order to mitigate these factors.

It should also be noted that contractors may utilize several different dredges (of varying sizes and types) in order to maximize efficiency of the dredging. This would also potentially impact the dewatering system.

As discussed in Section 4.1.2, in-water construction work will not be possible during winter months. Although the available in-water construction season may vary from year to year, an average working window of approximately 6 months (26 weeks) has been identified as reasonable for planning purposes based on recent work at the OU 1 and Phase 1 projects. In addition, a six-day per week work schedule is envisioned, with the sixth day planned for equipment maintenance, as necessary. With an average of 5 dredging days per week, there will be a total of 130 dredging days per year. This will yield approximately 416,000 cy (in situ) of dredge production per year.

Total in situ volume targeted for dredging is estimated to be 3,803,700 cy, as discussed in Section 4.5, including 3,633,700 cy of non-TSCA sediment and up to 170,000 cy of TSCA sediment. Of this total, approximately 80,000 cy is targeted for early mechanical dredging as discussed in Section 4.2.1 (24,500 cy in OU 2, 20,000 cy in OU 3, and 35,500 cy in OU 4). Therefore, the remaining in situ volume of 3,723,700 cy is targeted for hydraulic dredging. Based on this sediment volume and a production rate of 416,000 cy per year, a total hydraulic dredging duration of approximately 9 years is estimated, as presented in Section 9. With the numerous assumptions that were used to arrive at this calculation, it is reasonable to project that hydraulic dredging (excluding any potential re-dredging following post-dredge verification sampling) will be completed within a range of 8 to 10 years.

4.2.3 Equipment Selection to Remove Prospective TSCA Sediments

For removal of sediment potentially subject to TSCA disposal requirements, both hydraulic and mechanical dredge methods were evaluated. Based on varying dredge quantities of prospective TSCA material within specific dredge prisms, as well as production and logistical restraints, both methods of dredging will probably be utilized for OU 4.


31

4.2.3.1 Hydraulic Dredge Removal of Prospective TSCA Sediments

As discussed in Section 4.2.1, the primary method of removal for sediment potentially subject to TSCA disposal requirements will be through hydraulic dredging and pipeline transfer of the resultant dredge slurry to the existing desanding and mechanical dewatering system at the Shell Property staging area (described in Section 5.5). Of the 170,000 cy (maximum) of TSCA sediment present in OUs 2 to 5, approximately 152,500 cy are currently targeted for hydraulic dredging, with the remainder (17,500 cy) targeted for mechanical dredging (see below). The objective of the sediment dewatering process applied to hydraulically dredged material is to maximize the removal of sediments from the dredge slurry and produce a dewatered material that is suitable for overland transport via truck and disposal at a suitable TSCA-approved upland facility.

4.2.3.2 Mechanical Dredge Removal of Prospective TSCA Sediments

The mechanical dredge process is the most cost-effective and appropriate equipment for small quantities of sediment potentially subject to TSCA disposal requirements in shallow draft areas, or areas consisting of large amounts of debris. As discussed above, approximately 17,500 cy of TSCA sediments adjacent to and inside the sheetpile bulkhead area to be constructed at the former Shell Property, along with additional non-TSCA material, are currently targeted for early removal using mechanical dredging equipment. As discussed in the BODR, mechanical dredge production is limited somewhat by the necessity to more precisely remove the sediments while minimizing the entrained water and the associated need for dewatering amendments. Some of the face heights (i.e., cut thickness) of the sediment potentially subject to TSCA disposal requirements are expected to be small, so a large bucket is not desirable; therefore, a 5 cubic yard clamshell environmental bucket is envisioned. Cycle times (revised since the BODR) used in production estimates for the mechanical dredge reflect slower than typical cycles to account for precise placement of the bucket, lowering/raising of the bucket and cleanup passes to remove the material. Assuming that the contractor would likely employ two 12-hour shifts per day (6 days per week) in order to maintain the project schedule, and considering the production and logistical constraints, an average daily production rate of approximately 1,050 to 1,200 cy is estimated for the mechanical dredging of sediments potentially subject to TSCA disposal requirements. Once removed, the sediment will be placed for transport within dredge material transport barges (or perhaps roll-off boxes on smaller barges) sized to suit the water depths and physical restraints in accessing the materials. These dredge material transport barges will then be transported to an offloading operation at the Shell Property staging area.

Mechanical dredging requires that the dewatering takes place directly in the dredge material transport barges. In this case, decant water would first be removed and treated through the on-site dewatering facility. Following removal of decant water, the sediment remaining within the barge (or roll-off box) would likely still not be suitable for overland transport and disposal. It is anticipated that some form of dewatering amendment (e.g. quicklime, cement kiln dust, or other appropriate material), at approximately 15 percent by weight, would be added and mechanically mixed with the dredged sediment in order to


32

produce a consistency that is suitable for overland transport via truck and disposal at a suitable TSCA-approved upland facility.

4.3 Methodology for Developing and Optimizing Dredge Prism Design

When preparing an engineering design to dredge and dispose of sediment, a major component of the design is to define the dredge prism. The dredge prism presented in this 30 Percent Design Report for OUs 2 to 5 consists of a required dredge plan and an allowable overdepth. The required dredge plan represents the elevations, grades, and horizontal extents that a dredging contractor will be required to remove during remedial action implementation. Consistent with the ROD, the OUs 2 to 5 dredge prism has been designed to remove PCB contaminated sediment with concentrations greater than the 1 ppm RAL. The allowable overdepth is a constant thickness of sediment below the required dredge prism that engineers typically allow and pay the contractor for to account for dredging equipment accuracy and tolerances. The dredge prism design (including overdredge) reflects the fact that it is not possible for any dredge to excavate to an exact surface; in order to achieve a required elevation or grade, the dredge typically ends up removing excess material below the required dredge plan.

The dredge plan is developed to take into account many criteria. While the dredge prism design attempts to account for all critical criteria, the design is subjective and relies on dredge plan design experience, best professional judgment, and the quality and accuracy of the information available to the design team. Because the dredge plan design relies on multiple sets of data, the precision of each data set (e.g., bathymetry and the horizontal and vertical extent of contamination (i.e., the “neatline”) as defined by geostatistical methods) affects the level of certainty that the dredge prism encompasses all the contaminated sediments.

A summary of the overall steps in designing a dredge plan include: define the “neatline” area and depth to be remediated; specify site and project design criteria; prepare the dredge plan; and evaluate the benefit of adjusting the dredge plan based on cost versus benefit and engineering considerations. Each of these primary steps was presented in detail in the BODR; refinements to the approach are discussed in the following sections of this report.

4.3.1 Define the Neatline Area

The neatline area was defined through a multi-step, iterative process. Initially, a core-by-core evaluation was performed (as part of the BODR development) to determine preliminary dredge, cap, cover, and dredge and cap (D&C) boundaries using Thiessen polygons based on sediment PCB concentration profiles, comparisons of mudline elevations with stability benchmarks, and other relevant design information. At many locations, removal of contaminated sediments through dredging was determined in the BODR to be the optimal remedial action. Once the core-by-core evaluation was completed, a “mosaic” of remedial actions was developed and applied in the BODR across OUs 2 to 5 to identify and group areas of common remedial actions (e.g., dredge to 1 ppm, dredge-and-cap, etc). In this step, remedial actions and groupings were applied to the entire Thiessen polygon areas associated with each


33

core location. Then the mosaic was examined for apparently isolated remedial action “outliers,” and actions in some Thiessen polygons were adjusted to be more compatible with remedial actions in neighboring areas. For example, if the preliminary remedial action for a particular area was to apply a cap, but several neighboring areas were preliminarily designated for dredging such that a side slope would extend into the subject area, the final remedial action for that area might be dredging, rather than capping, in order to achieve a more uniform and constructable dredge surface.

For areas selected for dredging, the neatline area, depth, and associated volume was based on the geostatistical delineation of the horizontal and vertical distribution of PCB concentrations exceeding the 1.0 ppm RAL (see Section 2.3 and Figures 2-6 to 2-9). As discussed in the BODR, geostatistical analyses and cross-validation results indicate that a level of significance (LOS) of 0.5 will likely provide an optimum combination of maximum percent correct predictions and minimum overall bias, and thus was used as the primary method for defining the geostatistical neatline in OUs 3 to 5. Because of the relatively limited extent of sediments exceeding the RAL in OU 2, Thiessen polygons, rather than the geostatistical model, were used to define the neatline in this area of the site. Within OU 4, scatter plots of predicted versus observed remediation depths for full indicator kriging (FIK) identified a few isolated outliers with unusually high negative or positive biases (refer to Figure 2-7). Thiessen polygons were superimposed over the kriged surface at these outlier locations to adjust the depth of contamination and improve the accuracy of the neatline surface based on observed core information.

4.3.2 Specify Site and Project Design Criteria

Once the initial neatline areas within OUs 2 to 5 were defined, other design criteria were specified for the designer to evaluate during design. These criteria affect how the dredge plan is developed to encompass the neatline. The main objective in specifying design criteria for the dredge prism is to make sure that the dredge plan is constructable. Dredging equipment accuracies and tolerances limit the ability of a contractor to precisely remove sediment to a specified neatline, since a dredge generally works in a two-dimensional plane, either by dredging across a constant dredge elevation or constant defined slope over a specific area. Since the neatline is typically a variable surface that undulates in three dimensions (often paralleling the bathymetry), the dredging contractor is typically not capable of removing only sediments to the neatline cost effectively, but instead ends up removing additional (non-neatline) sediment. The quantity and extent of non-neatline sediment removed depends upon the complexity of the neatline area and how carefully the required dredge plan is designed and constructed to minimize non-neatline area removal.

As discussed above, the initial boundaries of dredging locations selected from the BODR core-by-core process were delineated using a Thiessen polygon approach. As the design progressed from the BODR conceptual level to the 30 Percent Design level, the boundaries were refined using the preliminary dredge plan and the spatial extents of the DOC at a LOS of 0.5. As shown on Figures 2-8 and 2-9, the spatial extent of the DOC resembles a curvilinear polygon. The preliminary dredge plan was developed by delineating a series of rectangles set at either a constant elevation or a constant slope to fully capture the


34

DOC, while minimizing the volume of sediment removed below the 1 ppm RAL. The specific criteria developed to guide the dredge plan designer through the process included:

• Maximum dredge slopes of 3H:1V for submerged areas of the site, but excluding shoreline areas. Slopes could be designed flatter, where necessary to accommodate the neatline.

• Maximum dredge slopes of 5H:1V in shoreline areas with nearby infrastructure, such as in the area of SMU 56/57, consistent with previous remedial actions at this location.

• Minimum dredge slope of 25H:1V, recognizing how a contractor will typically implement a flatter slope.

• Allowable overdepth of 0.5 feet will generally be achieved, but will not be relied upon to capture the neatline.

• Minimum dredging cut widths (i.e., “lanes”) ranging from 40 to 60 feet.

• Minimum dredging cut lengths of 200 feet.

• Preliminary infrastructure and structure setbacks of 10 feet or more, transitioning to the required dredge depth with a 3H:1V or shallower slope; more thorough examination of significant structures may be required as the design progresses (i.e., 60 Percent Design).

4.3.3 Iterative Design Refinements

Development of the engineered dredge plan design, which is presented in Appendix C, was an iterative process. Specifically, cross-sections were generated every 100 to 200 feet along the alignment of the river and at various locations where additional detail was required (e.g., areas where the channel alignment and shoreline are not parallel). Each cross-section was analyzed individually and preliminary target dredge cuts developed at or below the geostatistically modeled neatline (i.e., the LOS 0.5 surface). The elevation of each preliminary dredge cut was noted in plan view and a lane formed in nominal 40- to 60-foot-wide intervals and in minimum 2002-foot-long runs. As a starting point, it was assumed that the preliminary dredge cut elevations within each lane were valid to the midpoint between cross-sections. The extent of each lane elevation was verified by comparing the dredge plan (without overdredge allowances) to the neatline surface. During this step, if a significant area (e.g., a 40-foot by 2002-foot lane) of the dredge plan crossed above the neatline by more than 0.3 vertical feet, then the dredge plan was adjusted downward. Conversely, if the dredge plan/ neatline comparison indicated that excess (i.e., greater than 0.6 feet) sediment below the RAL was being removed, the dredge plan was either adjusted upward or lane extents modified to better capture the sediment deposit. During optimization, the final elevations of adjacent remedies (cap, cover, dredge, etc.) were assessed to ensure that implementation of the design will result in the construction of relatively consistent channel and river bottom elevations. For example, some isolated areas, initially designated in the BODR as “cap”, were converted to “dredge” areas to provide a consistent river bottom elevation (see Appendix A, Attachment A-2 for a summary of design adjustments since the BODR).

The dredge plan (Appendix C) was also optimized to balance between design conservatism, anticipated post-dredge PCB concentrations, and potential residuals management actions. Based on geostatistical


35

metrics, the final RAL neatline model used for dredge plan design has an estimated accuracy of about +/- 0.5 to 1.0 foot. The Workgroup concluded that the existing geostatistical model was a suitable basis for dredge plan design and that designing to an LOS of 0.5 was an appropriate starting point. The engineered design and overdepth allowances will likely result in removal of sediments with PCB concentrations less than the RAL below the neatline, further improving confidence that the dredge plan will encompass the in situ sediments exceeding the RAL. Based on the degree to which the design dredge plan achieved more stringent significance levels (e.g., LOS of 0.4 or 0.3), further adjustments and refinements of the original neatline model surface were performed during the 30 Percent Design to optimize the dredge plan. To support this process, the predicted post-dredge surface was evaluated with respect to the modeled DOC at several LOS values.

Using the AutoCAD software, a series of plots were developed to evaluate the effective LOS achieved by the dredge plan with and without an allowable overdepth (refer to Figure 4-3 for an example; a full set of analyses is provided in Appendix A; Attachment A-4). These comparisons were initially used to verify that the dredge plan will adequately remove sediment above the RAL throughout dredge-only areas without overdepth allowances at the 0.5 LOS. Generally, the analysis indicated that the average post-dredge effective LOS associated with the 30 Percent Design dredge plans is less than 0.5 for the required dredge cut and approximately 0.3 or less when the allowable overdepth is considered.

As a result of the LOS evaluation, potential risk-based adjustment areas were identified based on the dredge plan LOS evaluation, along with estimates of potential undisturbed residuals concentrations (based on RD sampling results; see Section 4.6.3) within specific areas of OUs 2 to 5. A set of LOS evaluation plots were prepared for the entire dredge plan with and without the allowable overdepth (0.5 feet). For substantially sized areas (minimum of 4 to 5 contiguous acres) that are in the upper range of target LOS after consideration of overdepth (i.e., 0.4 or greater), adjacent RD cores were evaluated to determine the probability of encountering relatively high concentration undisturbed residuals. Within such contiguous areas, potential undisturbed residual concentrations were estimated by pooling the RD core data within the immediate vicinity of the identified area and determining the concentrations at various intervals within one foot (above and below) of the anticipated final post-dredge elevation.

To the extent that higher PCB concentrations coincided with the high-end of the target LOS range, and/or lower PCB concentrations coincided with the low-end of the target LOS range, then a more detailed estimate of the undisturbed residuals was performed. Where such evaluations revealed that a significant (e.g., greater than 10 to 20 percent) probability of higher-concentration undisturbed residuals would be encountered on the post-dredge surface, the required dredge elevation was lowered to increase the confidence level of removing all sediment above the RAL. Conversely, in areas where the average LOS was predicted to be well below the target range (e.g., 0.3 or lower), and where there was also a low probability of higher-concentration undisturbed residuals, then the dredge plan was adjusted upwards, provided that the overdredge cut elevation still remained at or below the LOS 0.5 surface. Ultimately only one area was selected for upward adjustment using this analysis; however, the LOS evaluation plots


36

were useful in identifying a number of locations where downward adjustment of the dredge plan was beneficial in achieving the RAL within dredge areas. The engineered dredge plans are presented in Appendix C.

4.3.4 Cost/Benefit Assessments and Contract Alternatives

A formal cost versus benefit evaluation was not performed during development of the dredge plan; however, the general concept of cost versus benefit was fundamental to the OUs 2 to 5 design process. In general, a more complex dredge prism design typically results in a less constructable and less efficient dredging operation, but with less volume. Less efficiency results in a higher unit price for dredging the sediment. A simplified dredge prism typically results in a more constructable and efficient dredging operation, but with an associated increase in volume. Higher efficiency results in a lower unit price to dredge the sediment.

To develop the preliminary construction plan set for the OUs 2 to 5 project (Appendix C), an engineered dredge plan was selected as a more optimal design approach over a neatline contracting approach where the contractor would just be provided the geostatistically modeled RAL surface. While there are some advantages to the neatline approach, on balance this approach introduces more unknowns with respect to final dredge volumes, construction schedule, and ability to measure and accept the work as compared to an engineered dredge plan. By providing the contractors with a uniform basis (i.e., fewer unknowns) to develop costs, the implementing parties will be able to better evaluate each bid in an ‘apples to apples’ comparison. In addition, experience on other similar dredging projects has shown that a simplified set of plans and specifications based on a sound, engineered dredge plan with appropriate design contingencies will result in the greatest opportunity for a successful remedial action with minimal delays, cost overruns, and potential contract disputes.

4.4 Dredge Plan Design for Sediments Potentially Subject to TSCA Disposal Requirements

As discussed in Section 2.4.3, sediment PCB concentrations in some areas of OU 4 exceed 50 ppm and may become subject to TSCA-imposed management and disposal requirements. An initial dredge plan analysis, based on Thiessen polygons, was conducted as part of the BODR to determine the volume of sediments potentially subject to TSCA disposal requirements. Section 2.4.3 of the BODR summarizes the methodology for making such a determination. Based on the preliminary dredge plans developed using Thiessen polygon analysis, up to 170,000 cy of sediments will require TSCA management following completion of the Phase 1 Remedial Action in fall 2007. The engineered dredge plan for these TSCA areas will be completed as part of the 60 Percent Design.


37

4.5 Dredge Plan Design Basis

4.5.1 Sediment Volume Estimates

Figures 4-4 and 4-5 illustrate the aerial extent of the dredge plan for OUs 2/3 and 4/5, respectively, designed as described above. Detailed design plans are provided in Appendix C of this 30 Percent Design report. Volumes associated with the 30 Percent Design dredge plan were calculated using AutoDesk’s Land Development Desktop (LDD) software. A three-dimensional surface was created in AutoCAD v. 2004 for both the existing bathymetry and the required dredge prism, accounting for design side slopes. These surfaces each consisted of a set of contiguous, non-overlapping triangles known as a triangulated irregular network (TIN). Using LDD, the volume between these two TINs was calculated to represent the required dredge volume. An allowable overdepth surface was developed by lowering the required dredge plan by 0.5 feet in elevation and a total required and overdepth allowance volume was computed.

Table 4-2 summarizes the updated volume calculations based on the 30 Percent Design dredge plan. Overall dredge volumes estimated in this 30 Percent Design (excluding Phase 1 Project volumes) are approximately seven percent higher than those estimated in the BODR. The volume increases are largely attributable to additional dredging (beyond that identified in the BODR) to comply with ROD Amendment requirements and also reflect more detailed engineering refinements of the overall dredge plan. These total dredge volumes include sediment potentially subject to TSCA disposal requirements, which was estimated at approximately 140,000 to 170,000 cy, as discussed above.

Table 4-2. Summary of Dredge Volumes BODR 30 Percent Design

Operable Unit Dredge Volume with 6-inch

Overdepth Allowance (cy)

Required Dredge Volume (cy)

Required Dredge Volume with 6-inch Overdepth Allowance

(cy) TSCA

OU 2 0 0 0 OU 3 0 0 0

OU 4/5 170,000 170,000 170,000 Non-TSCA

OU 2 24,000 19,900 24,500 OU 3 204,000 160,400 221,200

OU 4/5 3,143,000 2,915,300 3,388,000 Sub-Total Non-TSCA 3,371,000 3,095,600 3,633,700 Total OUs 2-5 Volume 3,541,000 3,265,600 3,803,700

Note: The BODR volume summary excludes approximately 145,000 cy of sediment removed as part of the Phase 1 project.

4.5.2 PCB Mass Estimates

The BODR provided an estimate of the total mass of PCBs within the Lower Fox River based on the results of the 2004 and 2005 RD sampling and analysis programs using the equations shown below for each core location delineated on a Thiessen polygon basis. For the 30 Percent Design, the approach to calculating the volume of sediment associated with each core was refined to better represent the modeled DOC using a geographic information system (GIS) based sediment volume calculation. The Thiessen


38

polygon approach simplifies the dredge prism neatline by assuming a constant bottom elevation (i.e., simplified to length times area as shown in the equation below). The GIS approach calculates volume by integrating within the required dredge plan the geostatistically modeled neatline (i.e., the LOS 0.5 surface) within a specified area of influence for each core location.

AlPCB ⋅⋅⋅= ρ coreper Mass PCB where:

PCB = Sample PCB concentration, mg/kg (dry weight basis) ρ = dry density of sediment, g/cm3 l = sample length, cm A = Thiessen polygon area represented by core.

Additional details of the GIS-based approach to computing the PCB mass to be removed by the OUs 2 to 5 remedial action is presented in Attachment A-3 of Appendix A. Table 4-3 presents BODR estimates of the total PCB mass in OUs 2 to 5 using the 2004 and 2005 RD data on a Thiessen polygon basis and the updated GIS estimates for the total PCB deposit. The 60 Percent Design will include an estimate of the PCB mass targted for removal as part of the OUs 2 to 5 remedy. .

Table 4-3. Lower Fox River PCB Mass Estimates Estimated PCB Mass

BODR (Polygon Based

Estimate)

30 Percent Design Total PCB Mass Estimate Operable Unit

(kg) (kg) OU 2 (a) 130 130 OU 3 1,000 723 OU 4 19,950 18,404

OU 5 (b) 360 206 Site Total 21,440 19,463

(a) Deposit DD only – always calculated using Thiessen polygon-based spreadsheet (BODR) method.

(b) Portion near mouth of Fox River only

4.6 Potential Impacts from Dredging

4.6.1 Slope and Structural Considerations

As the design progresses through more detailed final iterations, the infrastructure and obstructions identified in the project area from bathymetric surveys, side-scan sonar surveys, sub-bottom profiling and site surveys will be superimposed onto the dredge prism. Shoreline structures and areas containing submerged features such as pipelines, cables, or ruins may limit the use of a dredge in that area. In addition, rock and debris may also inhibit dredging and may require removal prior to dredging when feasible. In areas of excessive debris and obstructions, dredging may not be possible and capping may be required. In addition to evaluating the WDNR (2003) side-scan sonar survey results, the contractor may elect to perform their own pre-construction debris surveys to identify any new obstructions prior to construction.


39

As discussed in Section 2.1, supplemental sampling and field work was performed during the summers of 2006 and 2007 to support the remedial design in the vicinity of shoreline and in-water features such as structures, slopes, and utility crossings. Extensive efforts have been undertaken to locate and obtain structural surveys and as-built record drawings for the numerous shoreline and in-river structural features, utility crossings, and overhead obstructions. These data were necessary to develop required setbacks protective of the structures during construction and over the long-term.

Along the shoreline, slope setbacks may be necessary to prevent undermining the shoreline, or removing lateral earth support for shoreline structures. For the preliminary dredge designs developed to date (Appendix C), a 50-foot offset from the shoreline was incorporated into the RD to prevent undermining of existing slopes by dredging. In areas where greater than approximately 2 feet of sediments exceeding the RAL was identified within this shoreline zone, shoreline capping may be necessary (see Appendix B, Attachment B-6). Section 6.3.1 of this report provides additional details regarding shoreline cap design. Final evaluation and design of these transition areas will be provided with the 60 Percent Design submittal and will include a location-specific review of shoreline stability and potential structural impacts.

4.6.2 Short-term Water Quality Considerations

During in-water construction activities, the contractor will be required to meet the substantive requirements of applicable water quality standards that will be specified as part of a 401 Water Quality Certification for the project. Consistent with the requirements of NR 102.05(3) and with the approved design for the OU 1 and Phase 1 projects, it is anticipated that a total suspended solids (TSS) limit of no more than an 80 mg/L incremental increase above ambient conditions will be permitted outside of a 500 foot mixing zone extending from the point of dredging (or form the boundary of the dredge area if a silt curtain enclosure is utilized).

Section 3.3.6 of the BODR summarized an evaluation of potential short-term water quality impacts associated with the anticipated dredging and cap/cover placement activities relative to the anticipated water quality compliance criteria. This evaluation utilized the dredge plume models developed by the USACE (e.g., DREDGE) in conjunction with the results of site-specific Dredge Elutriate Tests (DRET) performed on representative samples from OUs 2 to 5 to simulate the dissipation and attenuation of the dredge plume through the mixing zone. Under the various scenarios modeled in the BODR, including average and worst-case conditions, TSS was predicted to meet the water quality standard between 50 and 230 feet of the dredge. Thus, dredging operations are predicted to comply with the water quality standard well before the mixing zone boundary is reached at 500 feet downstream from the dredge area. Furthermore, based on the DREDGE model results, it is not expected that additional BMPs beyond the standard dredging practices (e.g., limiting cut depths to minimize slope failure) will be needed to control water quality throughout most of the OU 2 to 5.

Operational BMPs and controls will be implemented, as necessary, to minimize the potential for deviations from water quality standards.


40

4.6.3 Dredge Residual Management

The presence of residual contaminants is inevitable when dredging contaminated sediments due to the inability of any dredging equipment to completely remove all sediment within a dredge prism. A review of numerous recently completed environmental dredging projects demonstrates that post-dredge residuals can be expected in all dredging projects to differing degrees, and can result in post-remediation contaminant exposure within and immediately beyond the dredge prism if not adequately addressed (Patmont and Palermo 2007).

A workshop recently held at the U.S. Army Engineers Research and Development Center (ERDC) on Relating the “4 Rs” of Environmental Dredging: Resuspension, Release, Residual, and Risk (Bridges et al in press) focused in part on dredging residuals. Based on this work, dredging residuals can be generally defined as follows:

• Undisturbed Residuals: Contaminated sediments (at concentrations above the action level) found at the post-dredge sediment surface that have been uncovered but not fully removed as a result of the dredging operation.

• Generated Residuals: Contaminated post-dredge surface sediments (at concentrations above the action level) that are dislodged or suspended by the dredging operation and are subsequently re-deposited on the bottom either within or adjacent to the dredging footprint.

Patmont and Palermo (2007) presented data from a series of eleven relatively well-documented environmental dredging projects (pilots and full-scale) providing a bounding-level range of expected generated residuals ranging from approximately 2 to 9 percent (average of 4 percent) of the mass of contaminated sediment dredged during the last production cut of the dredging. Based on this evaluation, the BODR presented an estimate of the post-dredge SWAC, prior to residuals management, in OUs 3 and 4 of approximately 0.57 ppm in OU 3 and 3.7 ppm in OU 4. (Based on the RD sampling data, the existing SWACs in OU 3 and 4 are 2.0 and 3.2 ppm, respectively.) The post-dredge SWAC estimates, particularly within OU 4, are well above the ROD target of 0.25 ppm, and underscore the need for effective dredge residual management.

In order to accurately characterize the nature and extent of post-dredge generated and undisturbed residuals in OU 2 to 5, sediment samples will be collected following dredging and submitted for chemical (PCB) and physical (primarily percent solids and/or density) testing. Section 10.1 presents an overview of the CQAPP, which contains a draft sediment removal verification plan for OU 2 to 5, as briefly summarized below.

If concentrations in the post-dredging surface sediments are found to exceed the 1.0 ppm RAL, an initial screening for assessing the suitability of a sand cover (see Section 7) as a residuals management technique will be performed. In accordance with the ROD Amendment, sand cover will be considered a suitable for management of post-dredge residuals meeting the following criteria:


41

• Arithmetic mean of all 0 to 6-inch samples within a given dredge area is equal to or less than 10 ppm

• Arithmetic mean of all samples within a given dredge area for layers below the upper 0- to 6-inch interval is equal to or less than 1.0 ppm

If the post-dredge sediment concentrations exceed the sand cover screening criteria outlined above, additional sampling and/or analysis may be performed to determine the appropriate extent of areas requiring additional response. An engineering evaluation will be conducted to determine the most appropriate residual management action(s). The engineering evaluation will consider:

• Calculation of the percent PCB mass removed to date and remaining PCB mass per unit areas within a given dredge area

• Practicability, technical feasibility, cost-effectiveness, and implementability factors (e.g., layer thickness, PCB concentration, and density)

• Consideration of the residual management (if any) in adjacent dredge areas.

The determination of post-dredge contingency response decisions appropriate within all or a portion of a given dredge certification area will be performed as a collaborative undertaking between the entities performing the remedy and the Response Agencies. As discussed in Section 10.1, such contingency response decisions will need to be made on an expedited basis.

Possible residuals management actions include: additional production dredging passes or completion of a cleanup dredging pass; placement of an engineered isolation cap (see Section 6); placement of a residuals sand cover (see Section 7); or monitored natural recovery. Preliminary evaluations of the RD data set and the anticipated range of residuals presented above were initially presented in the BODR, suggesting that approximately 60 to 65 percent of the OU 2 to 5 dredge areas may require some form of residuals management. These estimates were updated and refined for this 30 Percent Design report, assuming generated residuals ranging from approximately 2 to 9 percent of the mass of contaminated sediment dredged during the last production cut (Patmont and Palermo 2007). As discussed in Section 4.3.3, the updated analysis also considered that the final RAL neatline model used for dredge plan design has an estimated accuracy of about +/- 0.5 to 1.0 foot, with corresponding uncertainties associated with the delineation of undisturbed residuals. These parameter ranges were input to a relatively simple Monte Carlo-based probabilistic analysis that also assumed use of a verification sampling design as described in Section 10.1. Based on this analysis, updated (and rounded) estimates of dredge residuals management areas in OUs 2 to 5 are summarized as follows:

• No further action in approximately 25 percent of post-dredge areas.

• Post-dredge sand cover or engineered capping in approximately 50 percent of dredge areas.


42

• Redredging in approximately 25 percent of dredge areas. Considering uncertainties in the RAL neatline model and dredge operation efficiencies, a typical 1-foot redredge cut (including overdredge allowances) is anticipated in OUs 2 to 5, resulting in a total estimated redredging volume of approximately 270,000 cy (roughly 7 to 8 percent of the target volume before redredging).

4.6.4 Noise and Air Quality Considerations

Noise. As discussed in the BODR (Section 3.6.5), noise emanating from industrial operations and other activities is generally regulated at the local level. Noise is regulated in the City of Green Bay under City Code Chapter 27, Subchapter II, Section 27.201, Regulation of Noise. Brown County regulates noise under County Code Chapter 39, Section 39.01, Regulation of Noise. A review of these two ordinances indicates that the noise control requirements are essentially the same with set noise levels based on zoning and time of day with special exemptions for construction sites. The contractor’s remedial action work plans may need to demonstrate how work within Brown and Outagamie counties will meet the substantive requirements of any applicable ordinances (e.g., equipment modifications).

Air Quality. Under WDNR regulations, NR 406 for Construction Air Permits and NR 407 for Operation Air Permits, air permitting could be triggered if emissions of either particulate matter and/or PCBs are above certain levels. Air emissions of PCBs are also regulated by the WDNR under NR 445 for control of state hazardous air pollutants. More specifically, air emissions of PCBs are required to meet substantive requirements of Table A to NR 445.07 for PCB emissions. Emission rates specified by the WDNR are specified to ensure ambient air concentrations do not exceed 12 micrograms per cubic meter (µg/m3), as a 24-hour average. While these substantive requirements will normally need to be considered under the air permitting program, air permits will not need to be pursued for the OU 2 to 5 cleanup due to the permit pre-emption under the CERCLA process.

Air monitoring was conducted during previous dredging projects on the Fox River as detailed in the BODR:

• Deposit N demonstration project (1998-1999): real-time monitoring for particulates on all four sides of the on-shore treatment facility where mechanical presses were operated and sediment loading occurred. Results showed no exceedances of the particulate threshold of 96 µg/m3 (Foth & Van Dyke 2001).

• SMU 56/57 demonstration project (1999): sampling for PCBs was conducted in 1999 at several locations both adjacent to and more distant from active sediment handling operations. Elevated concentrations above baseline levels were primarily associated with monitors that were within about 200 to 250 feet of the sediment handling operations. Samples collected at monitors situated at distances beyond this range approached background levels. Three samplers were located near the landfill area, ranging from 840 to 1,240 feet from the landfill site. All samples collected from these monitor locations had measured concentrations at or below background concentrations (WDNR 2000). Based on these results, air monitoring was not required during follow-on SMU 56/57 actions in 2000.


43

• OU 1 dredging project (2004): four air monitors were located around the sediment dewatering and load out pad with the closest monitor approximately 100 feet from the active operations. Concentrations of PCBs in the air were not detected at any of the locations (Foth & Van Dyke 2005).

• OU 1 Dredging Project (2005): four air monitors were located around the sediment dewatering and load out pad with the closest monitor approximately 100 feet from the active operations. Concentrations of PCBs in the air were not detected at any of the locations (Foth & Van Dyke 2006).

• OU 1 Dredging Project (2006): four air monitors were located around the sediment dewatering and load out pad with the closest monitor approximately 100 feet from the active operations. Concentrations of PCBs in the air were not detected at any of the locations (Foth, 2007).

• OU 1 Dredging Project (2007): four air monitors were located around the sediment dewatering and load out pad with the closest monitor approximately 100 feet from the active operations. Concentrations of PCBs in the air were detected ranging from 0.0007 to 0.0012 µg/m3.. The values detected during 2007 were well below the 24 hour average standard listed in WDNR NR 445.07 which is 12 µg/m3. (Note: The results presented here are DRAFT and preliminary. These data have not undergone QA procedures and should not be viewed as final).

• Phase 1 (OU 4A) Dredging Project (2007): four high volume air samplers were located adjacent to the sediment dewatering and load out pad, based on location of residential receptors, site topography, site operations, and prevailing wind directions. Concentrations of PCBs in the air were detected at ranging from 0.0002 to 0.0236 µg/m3. The values detected during 2007 were well below the 24 hour average standard listed in WDNR NR 445.07 which is 12 µg/m3.(Note: The results presented here are DRAFT and preliminary. These data have not undergone QA procedures and should not be viewed as final).

Section 5 – Material Handling, Transport, and Disposal

44

5 MATERIALS HANDLING, TRANSPORT AND DISPOSAL This section presents a review of material handling, transport, and disposal options conducted as a part of 30 Percent Design and identifies options specific to this current level of remedial design.

5.1 Transport of Debris and Dredged Material

As described in the BODR, prior to and during dredging operations in OUs 2 to 5, large debris within the dredge prisms will be mechanically removed and transported by barge to the staging area and offloaded using mechanical equipment. It is anticipated that the contractor will use a barge mounted excavator (or crane) equipped with a rake, “orange-peel” grapple, or perforated bucket attachment to perform debris removal operations. The rake, orange-peel grapple, or perforated bucket will allow sediments to “pass through”, for subsequent hydraulic dredging, while the debris and oversized materials remain in the rake or bucket for placement in a material transport barge, or scow. The barge will be transported to the Shell property staging area dock (or other appropriate offloading facility) where the debris will be mechanically offloaded utilizing an excavator or crane, stockpiled on shore, then placed in transport trucks for disposal at the appropriately permitted disposal facility.

As discussed earlier, mechanical dredging is planned for non-TSCA sediment in OU 2 and upper OU 3 due to site constraints (primarily the distance between these areas and the Shell Property staging area in OU 4). Approximately 24,500 cy in OU 2, 20,000 cy in OU 3, and 35,500 cy in OU 4 are anticipated to be suitable for mechanical dredging. This is described in more detail in Section 5.4.1.

The hydraulically removed dredge material (dredge slurry) from the remaining areas within OU 3 and OU 4 will be transported through a HDPE floating pipeline system to multiple in-water floating booster pump stations which, in turn, will pump the slurry to the land-based dewatering facility described in Section 5.5. The mechanical dewatering process will produce a material that is suitable for overland transport via truck and disposal at a suitable non-TSCA permitted facility.

5.2 Sediment Handling – Sediments Potentially Subject to TSCA Disposal

Requirements

As previously described, removal of approximately 140,000 to 170,000 cy of sediment potentially subject to TSCA disposal requirements will be performed using both hydraulic and mechanical dredge methods, based on the dredge quantity within a specific dredge prism, production and logistical restraints. Most of the existing TSCA sediment (up to 152,500 cy) will be dredged using hydraulic equipment, as discussed below. Sediment will be either mechanically dewatered or solidified by addition of an amendment (e.g., quicklime, cement kiln dust, or other appropriate material) to meet the permit requirements at the TSCA landfills identified for this project; EQ Wayne Disposal in Belleville, MI and Peoria Disposal Company in Peoria, IL.


45

As discussed in Section 4.2.3, for small isolated quantities of sediment potentially subject to TSCA disposal requirements, including approximately 17,500 cy of material adjacent to the Shell Property, mechanical equipment is the cost-effective and appropriate equipment. This method is envisioned primarily for the TSCA dredging within the confines of the sheetpile wall extension at the Shell Property staging area. Because production is limited by the necessity to more precisely remove the sediments while minimizing the entrained water and the associated need for dewatering amendments, a 5 cubic yard clamshell environmental bucket is envisioned. Material will be placed in appropriately sized water-tight transport barges. As described in Section 5.3.1, dewatering of these materials will likely take place directly in the dredge material transport barges. Following dewatering and amendment the material will be transported via truck and disposed of at a suitable TSCA permitted facility.

5.3 Sediment Handling - Non-TSCA Sediments

The following section describes the sediment handling procedures for non-TSCA sediments dredged both mechanically and hydraulically in OUs 2 to 5.

5.3.1 Mechanically Removed Sediment Transport in OUs 2, 3 & 4

Sediment removal during the OUs 2 to 5 remedial action will generally occur in an upstream-to-downstream sequence to minimize the potential for recontamination of completed remediation areas. Because of their relatively small volume, difficult shoreline dredging logistics, and the considerable distance from the primary downstream PCB deposits, target sediments within OU 2 and the furthest upstream portions of OU 3 are anticipated to be dredged under a scenario using mechanical equipment.

Remedial design plans developed to date for the 30 Percent Design submittal include a dredge-and-cap remedy for nearly all of the OU 2 deposit and a dredge-only remedy for the PCB deposit at the mouth of an unnamed tributary in OU 3. Approximately 24,500 cy of non-TSCA sediment in OU 2 are targeted for removal (including overdredge allowances). In addition to the OU 2 mechanical dredging, approximately 20,000 cy of nearshore deposits in the farthest upper reaches of OU 3 (i.e., the delta at the mouth of the tributary at the western end of transect 3008) and 18,000 cy of non-TSCA sediment adjacent to the Shell Property are also targeted for dredging using mechanical equipment. As discussed in Section 9, these actions would be performed during the first two years of full-scale in-water operations.

Sediments from OU 2 and upper OU 3 will be mechanically dredged and placed onto shallow draft barges (approximately 200 to 300 cy capacity). The dredging production rate in OU 2 and upper OU 3 will be limited, relative to rates assumed for the larger downstream remedial work, due to the dredging method (mechanical versus hydraulic) and the need to use relatively smaller equipment (due to water depths, lock size, river access, and other logistical constraints).


46

Dewatering will likely take place directly in the dredge material transport barges. In this case, decant water would first be removed and treated through the on-site dewatering equipment or containerized for disposal. Following removal of decant water, the sediment remaining within the barge still will not be suitable for overland transport and disposal. It is anticipated that some form of dewatering amendment (e.g. quicklime, cement kiln dust, or other appropriate material), at approximately 15 percent by weight, will be added and mechanically mixed with the dredged sediment in order to produce a consistency that is suitable for overland transport via truck and disposal at a suitable non-TSCA permitted facility.

5.3.2 Hydraulically Removed Sediment Transport in OU 3

Similar to the BODR design, a single 12-inch hydraulic cutterhead dredge is envisioned for removal of the remaining approximately 201,200 cy of non-TSCA sediment in OU 3. A single dredge was chosen for OU 3 due to the narrow navigation channel and the potential boating hazard if more than one dredge (and the accompanying pipelines) was used. The dredge will remove the sediment within the designed dredge prism and pump the material through a floating pipeline and accompanying floating booster stations to an upstream DePere Dam easement, crossing the dam and proceeding through OU 4 to the dewatering facility at the Shell Property staging area.

Dredging in OU 3 will begin in the upstream sections of this reach, proceeding downstream towards the DePere Dam. Based on the linear pumping distance to the DePere Dam and OU 3 sediment characteristics, from one to three floating booster pump stations will be required to maintain flow velocities. These are in addition to the three floating booster stations within OU 4 to maintain the flow velocities all the way to the dewatering facility at the Shell Property staging area. As the pumping distance decreases (moving downstream towards the DePere Dam), one, and then two of the OU 3 booster pumps will be removed. Calculations since the BODR indicate that a minimum of one floating booster pump, in tandem with the dredge will have to be used to negotiate the DePere Dam easement and maintain the proper flow velocities.

Dredging Best Management Practices (BMPs), will be utilized to maximize dredging efficiency while concurrently preventing damage to sensitive ecological areas. The following are some typical BMPs that will be followed during dredging operations:

• Operate cutterhead and ladder swing speeds appropriately to minimize sediment resuspension;

• Work in a upstream to downstream manner, except as necessary to provide proper blending of coarse- and fine-grained sediments;

• Overlap dredge cuts to avoid leaving ridges or windrows of contaminated sediment between adjacent cuts;

• Survey on a daily basis to determine the effectiveness of the dredging;

• Inspect and log observations of the dredge pipe daily for leaks and other problems; and


47

• Provide clear direction for response actions to various emergency situations (i.e., pipeline rupture, mechanical problems, etc.)

In addition, a marking plan for the dredge pipeline will be developed by the contractor and patterned after methods used by the USACE, and approved by the U.S. Coast Guard, to prevent the boating public from entering areas where operations are ongoing.

5.3.3 Hydraulically Removed Sediment Transport in OU 4

Similar to the BODR design, a single 12-inch hydraulic cutterhead dredge is envisioned for removal of approximately 3,370,100 cy of non-TSCA sediment and 152,500 cy of TSCA sediment in OU 4. A single dredge operating in OU 4 is appropriate due to the potential boating hazards if more than one dredge (and the accompanying pipelines) were to be used in this reach. The dredge will remove the sediment within the designed dredge prisms and pump the material through a floating pipeline and accompanying floating booster stations to the dewatering facility at the Shell Property staging area.

Dredging in OU 4 will begin in the upstream sections of this reach (just downstream of the DePere Dam), proceeding downstream towards the mouth. Based on the linear pumping distance from the DePere Dam to the dewatering facility at the Shell Property staging area, from one to three floating booster pump stations will be required to maintain flow velocities. As the pumping distance decreases moving downstream, one to three of the booster pumps will be removed then re-added as dredging progresses between the staging area and the mouth of the river.

As discussed above, dredging BMPs will be utilized to prevent damage to sensitive ecological areas and to prevent the boating public from entering areas where operations are ongoing.

5.4 Mechanical Dewatering Operations

As discussed in Section 4.2.1, the ROD Amendment envisioned the use of mechanical dewatering techniques. The dredge and booster pump(s) would transport dredge slurry from the river to a desanding and mechanical dewatering system at the Shell Property staging area. The sediment processing system is depicted on the process flow diagram on Figure 5-1. The incoming slurry will initially enter double-deck grizzly screens to remove coarse material and debris. The dredge slurry will then be directed through a vibratory wash screen (larger than the U.S. No. 200 sieve or 0.0029 inches) and spiral washer, followed by attrition scrubbers. This will liberate sand fractions from PCB-contaminated fractions. The attrition-scrubbed sand fraction will then be processed through traditional floatation technology (e.g., DAF) to remove the remaining humic matter. A hydrocyclone and dewatering screens will separate out the sand (anticipated to contain less than 1 ppm PCBs). The resultant sand fraction will be targeted for beneficial use. A front-end loader (or other earth moving equipment) will be used to move the sand to the


48

designated on-site storage area. Figure 5-1 presents a mass balance of the sediment and water processed with the proposed equipment.

The remaining slurry consisting of PCB-contaminated fractions, fines (finer than the No. 325 sieve) and humic material will be pumped to a circular clarifier / thickener tank. This tank provides a quiescent zone for solids settling and a rotating rake to move solids to a bottom discharge nozzle. Thickened slurry will be pumped to multiple 20,000-gallon agitated mix tanks. The clarified liquid from the clarifier / thickener tank overflows by means of a circular weir to a small head tank and then is pumped to the water treatment system. A large side stream of clarified liquid is recirculated for use in the desanding system.

Polymer will be added to the slurry in the mix tanks upstream of the plate and frame mechanical filter presses. Plate and frame mechanical filter presses were chosen in order to process the sediment into a “filter cake solid” that meets landfill specifications of 50 to 55 percent solids (by weight), with a compressive strength of at least 0.8 tons per square foot. After the appropriate “cycle time” is reached within the filter presses for the sediment to meet specifications, the dewatered sediment will be discharged onto a conveyor system that extends from the back of the press. A front-end loader will be used to move the filter cake to the designated on-site storage area. Figures 5-2 illustrate the overall process schematic and dewatering operation layout.

The layout and number of dewatering units was developed to accommodate the approximate 3,200 cy/day production rate of a single 12-inch hydraulic dredge. The cycle times, or production rates, were calculated based on an incoming dredge slurry rate of 5,000 gpm, and a solids concentration of 5.4 percent by weight, as discussed in Section 4.2.2. The dewatering operation is sized to handle the dredge running at full production for approximately 15.4 hours per day, based on the assumed dredge efficiency (i.e., “uptime”) of 64 percent and 24 working hours per day.

The dewatering (and water treatment) system(s) are sized to accommodate the nominal flow case presented in Figure 5-1 on a routine basis. But the process designs have also considered how to accommodate maximum and minimum flow conditions in order to prevent the dewatering operation from affecting the productivity (or efficiency) of the dredge operation; there is additional capacity inherent in each of the major types of equipment that will allow a maximum operating rate that is higher than the nominal flow case, and there are multiples of most equipment situated in parallel configuration that allows for units to be shut down in order to accommodate minimum flow conditions that are much lower than the nominal flow case. Also, whereas desanding operations must be capable of handling any surges in dredge flow directly, there is inherent surge capacity built into the circular clarifier, sludge mix tanks, and water treatment system storage tanks that allow that equipment to operate over a longer time period than the dredge operation in order to “catch up” if necessary. It should be noted that the flow case presented in Figure 5-1 is based on dredging, desanding, dewatering, and water treatment operations during 64% of the day, equivalent to 15.4 hours/day. It is estimated that the filter press operating period could be extended to at least 18 hours/day if necessary based on the storage capacity in the bottom of the


49

clarifier and slurry storage tanks. The water treatment system operating period could similarly be increased to 18 hours/day, based on the storage capacity of the water tanks (based on starting with empty tanks at the beginning of the day). This provides an inherent over-capacity of 17% for the dewatering system and 10% for the water treatment system.

Table 5-1 Built-in Over-Capacity Based on the Current Equipment Design Criterion

Item Quantity Fox River Nominal Rate

Equipment Rating (total for

all units)

Percent Over

Capacity Desanders 5 5,000 gpm 10,000 gpm 100%

Circular Clarifier 5 8,000 gpm 23%

Circular Clarifier 1 8,000 gpm 9,200 gpm1 15%

Filter Presses 10 1,248 cy/day 1,780 cy/day2 43%

WWTP Sand & Carbon Filters

10 4,662 gpm 7,000 gpm 50%

1 Based on 130 ft. diameter clarifier, nominal sizing factor of 1,000 gal/day/ft2. 2 Based on approx. 2 hour press cycle, 8 cycles/press/day, 600 ft3 press capacity.

Final Selection and sizing of pumps, piping and other ancillary equipment will be determined during 60 Percent Design to accommodate realistic over-design goals, which will be somewhere within the range of the design factors listed in the above table. For the purposes of filter press sizing, it is assumed that 30 percent of the total dry solids from the sediments will be coarse enough (greater than 100 microns) to be captured by desanding operations. The remaining 70 percent of the dry solids will be treated by the filter presses. The total dry solids from dredging operations, based on 3,200 cy/day in situ production, will be approximately 1,212 tons/day based on revised calculations since the BODR, as summarized below:

• 3,200 cy/day x 27 cf / cy x 28.1 pcf dry density / 2000 lb/ton = 1212 dry ton/day

[Note: the dry density of the sediment used in theses calculations are representative of OU 3-5 sediments targeted for hydraulic dredging]

The dry solids fraction that requires filter press operation is 1212 x 0.7 = 849 dry ton/day.

In July 2006, Waste Stream Technologies (WST) performed a treatability study to determine the most efficient and cost effective treatment regimen for dewatering dredged sediments from the Phase 1 Project, which targeted a relatively fine-grained sediment deposit in upper OU 4. Specifically, the study focused on recessed chamber plate and frame filter press technology, and investigated the use of various pre-treatment chemicals and dosages, along with varied samples and feed solids, cycle times, and pressures in enhancing mechanical dewatering while maximizing solids recovery. 2 Based on approx. 2 hour press cycle, 8 cycles/press/day, 600 ft3 press capacity.


50

Sediment cores from various locations of OUs 2 to 5 were sampled on June 23 through June 26, 2006, and transported to the treatability laboratory at WST. Samples were analyzed for percent solids, specific gravity, moisture content, pH and particle size. In addition, filter press tests were performed primarily on screened sample material. For these tests, “as received” sediment was manually passed through a #200 U.S. Standard Sieve with site water to remove particles greater than 75 micrometers. This process resulted in a diluted desanded sample that was representative of the anticipated overflow obtained from dredged sediment after being processed through a hydrocyclone. Sediment was diluted with site water to the specified percent feed solids, to simulate what is typically obtained through dredging operations in the field.

Based on the July 2006 Treatability Report for the Phase 1 Project, the filter cake may average approximately 55 percent solids content with a bulk density of 1.5 (ton/cy):

• 849 dry ton/day x 1/.55 x 2000 lb/ton x 1 cf / (62.4 cf x 1.5) = 32,984 cf / day filter cake

In order to accommodate this volume of filter cake, a total of 10 filter presses may be required, with a rated capacity of 600 cf per cycle. Approximately 55 press cycles per day would be required to accommodate this cake volume. Based on the treatability report by WST, a duration of 75 minutes provides adequate filtration time in order to achieve 55 percent solids (by weight) in the filter cake. Therefore, it was conservatively estimated that the total required press cycle time will be 2 hours. At this rate, nine presses will be capable of 80 press cycles in a 16 hour operating day, providing a conservative safety margin. However, in order to prevent equipment-related project delays, an additional press was assumed in the design to accommodate periods of relatively high production/efficiency, bringing the total filter presses required for the remedial action to 10.

5.5 Water Treatment Operations

The water treatment system is designed and installed to adequately treat the required volumes of water (approximately 5,000 gpm or 7.2 million gallons per day [MGD] for a 24-hour day) generated during dredging, sediment dewatering, and decontamination operations. The water treatment system will also treat all precipitation that comes into contact with the concrete and asphalt work pads and is conveyed to a common collection area.

The water treatment system consists of two water surge tanks (approximately 250,000 gallon capacity each) for temporary storage of untreated water. Treatment consists of sand filters, carbon absorption units, and bag filters. Treated water will be sampled to verify compliance with the anticipated discharge requirements, and then discharged back to the Fox River through a submerged outfall in accordance with State of Wisconsin water quality requirements (NR 105 and 106). Discharge requirements are anticipated to be similar to those in the Phase 1 project and are set forth in Table 5-1.


51

Table 5-2. Effluent Discharge Monitoring Requirements

Parameter Sampling Frequency Performance Standard Low Level Mercury Once per week Less than 0.5 ng/L PCBs Less than 0.5 µg/L BOD 50 lbs/day pH 6.0 to 9.0 SU

TSS Monthly average of 5 mg/L Daily maximum of 10 mg/L

Ammonia

Daily sampling using ISCO (or equivalent) automatic sampler. Sample made up of automatic composites based on discharge flow rate. Discrete samples may also be used in combination with flow proportional sampling.

*Based on a influent pH of 8.6 or lower, an ammonia concentration less than 82 mg/L

Notes: ng/L – nanograms per liter µg/L – micrograms per liter mg/L – milligrams per liter SU – standard units (pH) * The performance standard is based on a pH of 8.6 or lower, which would be 82 mg/l (assuming a dilution of 15:1) at the treatment plant outlet, as measured at the ISCO automatic sampler prior to the diffuser.

Figure 5-2 illustrates the dewatering/water treatment facility process flow diagram. The facility is similar to the process used during the SMU 56/57 project (Foth & Van Dyke et al. 2001) and the Phase 1 Project.


5.6.1 Process Flow of Major Unit Operations

As discussed in Section 4.2.2, the dredging operation will be staffed 24 hours per day, 5 days per week. The dewatering and water treatment equipment will also be staffed 24 hours per day, 5 days per week, as needed to meet the needs of dredging. It is expected that the overall operating efficiency (uptime) of the dredging, dewatering, and water treatment systems will approach 85 percent. However, for purposes of estimating daily production rates, a more conservative number of 64 percent was used.

It is anticipated that a single 12-inch cutterhead dredge will be in operation at any given time. This dredge will yield roughly 208 cy/hr in situ dredge volume. The weighted average solids content of the in situ sediment for OUs 3 to 5 is 32.2 percent (dry weight basis), with an in situ dry density of approximately 25 pcf. This is based on a weighted average of 112 geotechnical samples collected from areas targeted for dredging (see Table 4-1). This results in an average filter press dewatering production rate of 70.2 ton/hr dry solids (208 cy/hr x 27 x 25 pcf / 2000) while the dredge is in operation. However, a higher solids content (35.6 percent) and dry density (28.1 pcf) were used in design calculations in order to conservatively size the dewatering system, while accounting for data variability. The resultant design


52

production rate of 78.9 ton/hr dry solids is 12.4 percent higher than the anticipated average rate of 70.2 ton/hr. This 12.4 percent factor for dry solids production rate is essentially one standard deviation above the mean for the 112 geotechnical sample data set, and provides a suitable safety margin for the 30 Percent Design.

A dredge in situ volume-based concentration of 14 percent is assumed as discussed above, based on discussions with dredge manufacturers. This results in a dredge flow rate of approximately 5000 gpm (208 cy/hr in situ / 0.14 x 27 x 7.48 / 60 = 5001 gpm). This dredge flow rate is the basis for sizing the dewatering and water treatment equipment.

The dredged slurry will first be pumped through desander units which will be designed to separate the slurry into three streams: 1) debris and coarse gravel; 2) sand; and 3) slurry with fine particles (less than 100 microns). The sediment processing system will include five of these desanding units, each sized to accommodate 1,250 gpm. Normally four units will be in operation and one spare unit will be idle when operating at the design flow rate of 5,000 gpm of dredged slurry. The incoming flow will be evenly split between the operating desanders using a hydraulic splitter box. The following description applies to each of the five desanders.

The desander consists of a V-bottom tank with top-mounted shaker screens. The sediment slurry will first flow through a one course (3/8- to 2-inch) vibrating shaker screen mounted on top of the tank. The screen will remove debris, stones, zebra mussels, large wood chips, and some gravel and sand from the slurry. Screened material will discharge onto the processing pad, and after analysis for PCBs, will be transported to the stockpile pad at the process treatment area. The screened slurry gravity feeds into the V-bottom tank, where one of two pumps (one primary and one standby) will pump the slurry through multiple desanding hydrocyclones mounted on top of the tank. The hydrocyclones separate the heavier sand particles from the lighter silt fraction by centrifugal force. The underflow from the hydrocyclones is directed to dual vibrating linear motion shakers for sand removal. The overflow from the desander, carrying fine silt and clay particles (less than 100 microns), will flow into a 20,000 gallon agitated pump tank. The slurry will be pumped from the tank to a circular clarifier.

The sand from the desanders will be washed in hydrosanders designed to dislodge silt and organic material from the sand. Three hydrosanders are anticipated, with each one capable of processing the sand from two desanders. Two of the hydrosanders will be deployed such that a pair of desanders is positioned to discharge sand to a single hydrosander. The third hydrosander will be connected to the spare desander. The sand from the desander’s dual vibrating linear motion shakers will fall onto a belt conveyor and subsequently fed to the hydrosander. Clean water from the clarifier overflow will be pumped into the hydrosander to wash the sand. The cleaned sand will move from the hydrosander chamber via a large bucket wheel and will discharge onto the processing pad for transport to the stockpile pad for staging. Dirty wash water will overflow the hydrosander and will be pumped to a circular clarifier.


53

The large circular clarifier is designed to separate the de-sanded dredge slurry into two streams; thickened underflow slurry suitable for further dewatering via the filter press and a clarified supernatant that can be further treated in the water treatment system. Polymer flocculent will be added to the clarifier influent upstream of a static mixer in order to grow particles and aid settling of solids. The single clarifier is a 130’diameter, open-top steel tank that provides a quiescent zone to settle the solid particles to the bottom of the tank while water is discharged from the tank via a circumferential overflow weir. The clarifier will be equipped with a motor-driven rake that slowly rotates, moving the settled solids toward a bottom-center discharge tap where the thickened slurry will be pumped to the slurry mix tanks. The clarifier overflow liquid will be pumped to the water treatment system, with a large slip stream returned to the hydrosanders. The clarifier will be sized for the de-sanded dredge liquid flow, and also for the hydrosander return water and the filter press filtrate. The nominal design influent rate is 8,000 gpm.

The clarifier underflow slurry is pumped into ten 20,000 gallon agitated mix tanks. A full capacity spare standby pump will also likely be provided. The total slurry storage capacity will be 200,000 gallons, which equates to about 2.5 hours of capacity at the design slurry flow rate (1343 gpm at 15 percent solids). From the mix tanks the slurry is pumped into the filter press system. The filter press fast feed centrifugal pumps draw from the agitated mix tanks to fill ten 600 cubic foot filter presses. Each filter press will have a dedicated centrifugal fast feed pump, and a high pressure, low volume pump to complete the press cycle. In addition, there are two stand-by fast feed centrifugal pumps for any of the ten presses.

Polymer is added to the dredge slurry upstream of the press through a flow meter, static mixer, and polymer injection system to improve the filterability of the solids. Polymer is fed from a chemical storage tank or individual totes (250 gallons) via a polyclone system through an in-line static mixer.

For purposes of filter press sizing, it has been assumed that 30 percent of the total dry solids from the sediments will be coarse enough (greater than 100 microns) to be captured by debris removal and desanding operations. This compares conservatively with geotechnical data collected from OUs 3 to 5 dredge areas, which show an average of 38 percent sand/gravel. The 30 percent assumption leaves 70 percent of the dry solids to be treated by the filter presses. The total dry solids from dredging operations, based on 3,200 cy/day in situ production, and the design basis of 35.6 percent solids (28.1 pcf dry density) is approximately 1,212 tons/day based on BODR appendix calculations as summarized below:

• 3,200 cy/day x 27 cf / cy x 28.1 pcf dry density / 2000 lb/ton = 1,212 dry ton/day

The dry solids fraction that requires filter press operation is 1,212 x 0.7 = 849 dry ton/day. Based on the July 2006 Treatability Report prepared by WST, the filter cake averaged 55 percent solids content with a specific gravity of 1.5. This results in a cake production volume of approximately 1,222 cubic yards per day:

• 849 dry ton/day x 1/.55 x 2000 lb/ton x 1 cf / (62.4 cf x 1.5) = 32,984 cf / day filter cake


54

As previously discussed, 10 filter presses were selected, with a rated capacity of 600 cf per cycle, to accommodate this volume of filter cake.

Filter press filtrate is discharged to a head tank and pumped back into the clarifier. At the end of each filter run a core blow is conducted on the press to remove residual solids from the press feed lines. Any discharge from this process is returned to the agitated mix tanks. After discharge from the presses, the filter cake is conveyed to the end of the press and moved by a wheel loader to stockpiles on the processing pad.

Additional waste streams entering the solids removal portion of the process are storm water and filter backwash streams, clarifier sludge, decontamination water, and leachate from the storage cells. These streams are discharged into the agitated mix tanks. With the exception of storm water, these flows are expected to be minor when compared to the total flow through the system and have a minimal effect on the hydraulics of the dewatering operation. Storm water flow could be substantial during a heavy downpour and may need to be stored and treated. If necessary, dredging operations can be reduced to allow the system to process storm water. This is not expected to increase dredging time significantly since it is likely that those periods of high storm flow will correspond with poor conditions for dredging.

The water treatment system is sized for nominal capacity of 5,000 gpm. The projected feed rate from the dredge dewatering system is 4,662 gpm. Stormwater, decontamination water, and leachate run-off from the storage cells is also routed to the water treatment system. The system is designed to address PCBs, suspended solids, mercury, and ammonia. Since the PCBs are largely associated with solids, removal of suspended solids is a major focus of the wastewater treatment train, prior to carbon adsorption to further reduce PCB concentrations. Also, mercury present in the sediment is typically insoluble, meaning that removal of suspended solids will be the principal method for removing mercury from the wastewater. An effluent diffuser will provide sufficient initial dilution to address acute ammonia toxicity.

The clarifier overflow from the dewatering system discharges into two 250,000-gallon equalization tanks. These tanks provide a nominal capacity of 100 minutes at the design flow rate of 5,000 gpm. The tanks serve to equalize flow and load variations from the dewatering system and storm water. The equalization/filtrate tank also receives clarified spent backwash flows from the downstream sand filters and GAC reactors.

The water from the equalization tanks is pumped to the filtration process, designed to remove suspended solids and most PCBs prior to passing through the GAC reactors. There will be two 50 percent capacity pumps with a third 50 percent spare. The pump discharge flow will be measured using a magnetic flow meter. The first step in the process is ten sand filters, each sized for 500 to 600 gpm. Each filter is provided with mono-media filter sand. The sand filters are backwashed as required (e.g., when the operating pressure indicates backwashing is required) to remove any solids buildup and “fluff up” the sand media to reduce head loss through the filters. Backwash water is pumped to a settling basin. The


55

settled solids are periodically pumped to the filter press feed tanks. The water is returned to the equalization tanks.

The flow proceeds under pressure through three bag filters. Each bag filter unit is equipped with multiple filter bags, and is intended to protect the downstream GAC units. The bags in the filters are changed when the operating pressure becomes high enough to warrant it. Used bags are disposed of with the filter cake and desanded materials.

Following the sand filters and bag filters, the flow is pumped through ten GAC vessels. The purpose of the GAC vessels is to remove any remaining PCBs and other carbon-compatible organics and metals that may be present. Sufficient carbon is provided so that, at expected organics loadings, the carbon’s effectiveness (at a minimum) will last throughout a project season. As with the sand filters, the GAC vessels have backwash capability to remove any trapped solids and to “fluff up” the carbon media, reducing head loss.

After the GAC vessels, the water flows through three bag filter units, each with 0.5- to 5-micron bag filters. The purpose of the secondary bag filters is to remove any carbon fines and adsorbed PCBs that may have carried over from the GAC vessels.

The effluent from the secondary bag filters is pumped to the outfall diffuser. The outfall diffuser is designed to achieve the necessary initial dilution to comply with water quality performance standards for the project using a zone of initial dilution (ZID) as defined by Wisconsin regulations, which allows the use of Best Demonstrated Treatment Technology Reasonable Achievable (BDTTRA). The projected performance of the diffuser is modeled using CORMIX2 software, as summarized in the BODR. Dilution calculations for the proposed diffuser design show the dilution for the diffuser to be 31.1:1 at 10ft/sec. Based on an anticipated maximum ammonia concentration, the ZID should be capable of achieving the desired dilution.

5.6.2 Preliminary Mass Balance

Figure 5-22 of the BODR (Shaw et al. 2006) presented a mass balance for the sediment and water processed with the equipment proposed for use under the Optimized Remedy. Figure 5-1 presents an update mass balance incorporating the changes and modifications for the 30 Percent Design effort.

5.7 Beneficial Use Considerations

5.7.1 Desanding Technologies

The treatability testing report for the separation of sand fractions from PCB-containing sediments collected from OUs 2 to 5 is presented in Appendix C of the BODR (Shaw 2005). These tests revealed that physical separation technologies, specifically desanding with organic flotation and attrition scrubbing can be employed to separate relatively uncontaminated (less than 1 ppm) sand fractions (+200 mesh or 0.0029 inches) from sediment fractions containing higher PCB concentrations. Using this technology,


56

roughly 30 percent (by weight) of the total sediment solids in the dredge slurry can be separated as sand containing less than 1 ppm PCBs. This sand fraction is available for beneficial use.

As determined from the RD bench-scale tests (Appendix C of the BODR), the most effective means to accomplish separation of sands containing less than 1 ppm PCBs from the other OU 2 to 5 sediments is to first separate greater than 200 mesh material from the rest of the dredged slurry. The incoming slurry initially enters double-deck grizzly screens to remove coarse material and debris. The dredge slurry then goes through a vibratory wash screen (greater than the No. 200 sieve or 0.0029 inches) and spiral washer, followed by attrition scrubbers. This liberates sand fractions from PCB-contaminated fractions. The attrition-scrubbed sand fraction goes through traditional floatation technology (e.g. DAF) for removal of any remaining humic matter, then through a hydrocyclone and dewatering screen to separate out any remaining fines from the sand. This resultant sand fraction is the material slated for beneficial use. A front-end loader is used to move the sand to the designated on-site storage area.

The remaining slurry consisting of PCB-contaminated fractions (finer than No. 200 sieve) and humic material is pumped to the passive dewatering facility for further processing.

5.7.2 Materials Potentially Suitable for Beneficial Use

Based upon the results of the 2004 and 2005 RD investigations (Shaw 2005, Shaw 2006), the sediments targeted for removal under the ROD Remedy consist mainly of sand and silt-sized particles (75 to 80 percent by weight), with the remaining percentage consisting mainly of clay (approximately 20 percent) and a trace to slight amount (less than 5 percent) of gravel. The sand and gravel comprised approximately 30 to 40 percent by weight of the sediment samples from OU 3, OU 4 and OU 5 within the dredge prism.

The PCBs in the sediments from OUs 2 to 5 are largely adsorbed onto the fine-grained (less than 200 mesh) fractions of the sediment. The coarse-grained sediments (greater than 200 mesh), referred to in the BODR as the sand/gravel fraction, have PCB concentrations less than 1 mg/kg and are the materials under consideration for beneficial use. Approximately 260,000 cy of sand containing less than 1 ppm PCBs are anticipated to be available if desanding technologies are applied to the entire 3.6 million cy non-TSCA dredge volume in OUs 2 to 5.

The chemical criterion established for most beneficial uses of dredge material is less than 1 ppm total PCBs. Desanding, with organic flotation and attrition scrubbing, has shown that total PCB levels of less than 1 ppm can be achieved.

5.7.3 Description for Potential Beneficial Use Alternatives

As previously described in Section 2.4.1, several beneficial use options have been considered during the RD. It is assumed that most (approximately 150,000 cy) of the beneficial reuse material will be utilized at the former Shell Property for development of the staging facility (e.g., backfill material behind the expanded sheetpile dock wall). The remainder (110,000 cy) will be utilized in available and appropriate


57

off-site locations, as discussed in the BODR. These alternatives will be considered at later stages of the project based on the rate of generation and the opportunities for potential use.

5.8 Transport and Disposal of Dewatered Sediment and Debris

This 30 Percent Design provides a review of RD considerations relative to the characterization and quantification of dredge material for disposal including NR 500 (non-TSCA) landfills and TSCA-licensed facilities. The remedial action for OUs 2 to 5 will result in approximately 1,801,000 tons (1,450,000 cy) of filter cake and 76,475 tons (66,100 cy) mechanically dredgedand amended material requiring non-TSCA landfill disposal. In addition, the remedial action for OUs 2 to 5 will result in approximately 223,575 tons (180,000 cy) of filter cake and 21,415 tons (18,500 cy) mechanically dredged and amended material requiring TSCA landfill disposal capacity. These volumes refer to volumes after dewatering, and are based on the anticipated use of mechanical dewatering, of hydraulically dredged material, at the staging facility as the main dewatering technology, followed by trucking of the sediment to a regional landfill, as discussed below.

Based on conventional monofill characteristics for low strength solids material (such as paper mill sludge or sediments, the typical acreage requirement for disposal of this non-TSCA volume of material is 30 to 40 acres with an average fill height of 35 to 45 feet. Greater fill depths are possible, provided adequate engineering measures are taken.

5.9 Potential Upland Disposal Facilities

As discussed in Sections 5.2 to 5.4, sediment removal, processing, transport and disposal includes mechanically dredging and amending approximately 62,500 cy of non-TSCA sediment from portions of OUs 2, 3 and 4, and hydraulically dredging and mechanically dewatering approximately 3.6 million cy of non-TSCA sediment in OU 3 and 4, and then trucking the dewatered sediment to one of the two landfills identified below. . Hydraulic and mechanical dredging of 140,000 to 170,000 cy of sediment potentially subject to TSCA disposal requirements would also be performed. The prospective TSCA material would be dredged as a separate activity, but hydraulically dredged materials would be dewatered in the same fashion and at the same location as the non-TSCA sediments. As discussed above, dredging activities would be sequenced and carefully monitored to isolate the prospective TSCA materials from the non-TSCA material during dewatering, staging and transport.

The implementability evaluations presented in the BODR identified three potential disposal sites for non-TSCA material; the 30 Percent Design has narrowed this down to two:

• Brown County South • Veolia, Hickory Meadows Landfill


58

Since both potential sites each have adequate volume for disposal of dewatered sediments from OUs 2 to 5, other factors including socio-political acceptance, the ability to finish permitting and construction within the given timeframe, and cost will be considered prior to final selection of the disposal site(s).

Brown County South Site. The Brown County South site currently has a completed Feasibility Determination from WDNR, and the Plan of Operation (required for permitting of the facility) is currently under development. The site would have potential capacity for 3.7 million cy of material under the current Feasibility Determination, to accommodate the disposal of dewatered sediments from OUs 2 to 5. Future stakeholder interactions would be necessary to determine the path forward.

Veolia; Hickory Meadows Landfill Site. The Hickory Meadows Landfill site is located approximately 30 miles from OU 4 (RM 3.5), and is also projected to have capacity for 3.7 million cy of airspace remaining by the targeted start of dredging (2009). The local agreement for the existing Veolia site allows for disposal of PCB contaminated sediment and they are currently receiving dewatered sediment from Fox River OU 1, the Fox River Phase 1 Project, and potentially will receive similar material from the Sheboygan River Superfund site.

In addition to the two potential disposal options discussed above, the RD will continue to evaluate other potential disposal sites for non-TSCA sediment that either have accepted PCB-containing sediment in the past or could do so in the future.

Sediments determined to be subject to TSCA disposal requirements will be transported to either EQ Wayne Disposal in Belleville, Michigan; or Peoria Disposal Company in Peoria, Illinois; or to another TSCA-permitted disposal location identified at the time of remediation.

Section 6 – Engineered Cap Design

59

6 ENGINEERED CAP DESIGN As described in the BODR, designs for engineered sediment caps in OUs 2 to 5 were developed in accordance with the following detailed guidance for in situ capping developed by USEPA and the USACE:

• Contaminated Sediment Remediation Guidance for Hazardous Waste Sites (USEPA 2005a)

• Guidance for Subaqueous Dredged Material Capping (Palermo et al. 1998a)

• Assessment and Remediation of Contaminated Sediments (ARCS) Program Guidance for In Situ Subaqueous Capping of Contaminated Sediments (Palermo et al. 1998b)

These documents provide detailed procedures for site and sediment characterization, cap design, cap placement operations, and monitoring for subaqueous capping. Caps designed according to the USEPA and USACE guidance have been demonstrated to be protective of human health and the environment (USEPA 2005a).

Consistent with the above referenced guidance, the thickness and other design specifications for in situ engineered caps in OUs 2 to 5 consider the following five components:

• Chemical isolation of contaminants (Ti)

• Bioturbation (Tb)

• Consolidation (Tc)

• Erosion (Te)

• Operational considerations (i.e., gas generation, placement inaccuracies, and other pertinent processes) (To)

As discussed in the BODR and below in Section 6.1, an appropriate thickness of cap was determined individually for each component based on site-specific design parameters. The total cap thickness that satisfies all design components is equal to the summation of the individual component thicknesses listed above. However, the technical Workgroup concluded that the erosion component and the bioturbation component may be a concurrent thickness and not independent thickness requirements, consistent with White Paper 6B (Palermo et al. 2002). Therefore, a set thickness of an armor layer can serve to resist erosion as well as accommodate bioturbation. Given the variability of site conditions (PCB concentrations, erosion potential, etc.) throughout OUs 2 to 5, three general cap designs were developed for the BODR primarily based on PCB concentrations, as summarized in Table 6-1 and described in detail in Appendix D of the BODR. Subsequent refinements of these cap designs, including considerations of location-specific erosive forces and PCB concentrations, are presented in the sections below and summarized in Section 6.3.


60

Table 6-1. Engineered Cap Designs Developed in BODR

Cap Description PCB Concentration

in 0 to 6-inch Interval

Comments

Cap A - Minimum 13-inch Sand Cap with Gravel Armor

< 10 ppm Used in low concentration areas where mixing zone of clean sand provides necessary chemical isolation.

Cap B - Minimum 16-inch Sand Cap with Gravel Armor

10-50 ppm Used in areas where 3 inches of uncompromised chemical isolation layer is necessary for protection.

Cap C - 33-inch Sand/Quarry Spall Cap 10-100 ppm Used only in OU 4 navigation channel and/or where PCB

concentrations exceed 50 ppm at any depth.

Numerous technical memoranda and engineering evaluations (related to the cap design) were performed as part of the 30 Percent Design, as documented in Appendix B. Attachment B-1 contains an interim technical memorandum summarizing the overall design of the engineered caps. The remainder of this section presents a summary of the engineered cap design analyses. Specific cap components, additional design considerations, and 30 Percent cap designs and delineations are described in Sections 6.1, 6.2, and 6.3, respectively.

6.1 Cap Design Criteria

6.1.1 Chemical Isolation Component

As described in the BODR, a series of calculations were performed using location-specific conditions (PCB concentrations, vertical groundwater velocity, sediment total organic carbon [TOC], consolidation-induced porewater flux, etc.) to evaluate the chemical isolation component of a subaqueous cap for PCB containment. Specifically, chemical isolation modeling included the use of a transient model described in Appendix B of the ARCS Program Guidance for In Situ Subaqueous Capping of Contaminated Sediments (Palermo et al. 1988b) to estimate contaminant flux through the chemical isolation layer and the time to achieve steady state chemical flux conditions in the isolation layer of the cap. In addition, the steady state model of Reible et al. (2004) was used to estimate chemical concentrations in the surficial (bioturbation) sediment layers of the cap once steady state conditions are achieved. The results of the steady state model indicated that, with the cap designs presented herein, once steady state conditions are achieved, there is greater than a 99 percent probability that sediment PCB concentrations in the cap bioturbation zone would be maintained (in perpetuity) below the 1.0 ppm RAL.

The thickness of the chemical isolation layer for each cap design was preliminarily based on the average sediment PCB concentration in the top 1.5 feet of sediment immediately underlying the prospective cap layer for a given location. As such, cap designs were developed with varying chemical isolation thicknesses to contain sediment with between 1 and 10 ppm, 10 and 50 ppm, and greater than 50 ppm in the top 1.5 feet of sediment. Subsequent to the BODR, the delineation of caps with the appropriate chemical isolation layer thickness was refined based on the PCB concentration in the top 6 inches of sediment immediately underlying the cap, consistent with the criteria specified in the ROD Amendment:


61

• Engineered caps of at least 6 inches of sand for chemical isolation: PCB concentrations will not exceed 50 ppm within the sediment profile and PCBs in the top six inches of sediment immediately beneath the cap will be less than 10 ppm.

• Engineered caps of at least 9 inches of sand for chemical isolation: PCB concentrations will not exceed 50 ppm within the sediment profile.

• Engineered caps of at least 15 inches of sand (and filter material) for chemical isolation: PCB concentrations exceeding 50 ppm buried within the sediment profile or in shoreline areas where dredging would result in instability.

For dredge-and-cap (“hybrid”) areas of OUs 2 to 5, appropriate sand layer thicknesses for chemical isolation of PCBs (i.e., 6-, 9-, or 15-inch-thick layers) were developed for this 30 Percent Design based on estimates of the maximum PCB concentration anticipated in the top six inches of sediment immediately beneath the cap, considering the following:

a) The upper-bound generated residual PCB concentration and thickness/density anticipated following dredging (e.g., assuming a range of generated residual releases using the framework outlined by Patmont and Palermo 2007); and

b) The upper-bound undisturbed residual PCB concentration in the 6-inch interval immediately beneath the dredge elevation.

Consistent with Agency recommendations, the consideration of RD core sampling data along with anticipated residuals concentrations was used as a means of forecasting appropriate chemical isolation layer thickness, subject to field verification at the time of construction. Appendix B (Attachment B-2) includes a table summarizing the designed cap thickness for each cap area, based on the considerations outlined above, as well as other location-specific conditions (i.e., position relative to the Federal navigation channel), consistent with the ROD Amendment.

Field verification of designed cap thicknesses will include the collection of samples for PCB analysis in dredge and cap areas following dredging but prior to cap placement in order to verify RD forecasts, and confirm the appropriate cap thickness is applied based on the measured concentration of residuals. Sampling frequencies may be reduced over time if the RD forecasts are consistently verified.

6.1.2 Bioturbation Component

The BODR stated that the bioturbation depth is expected to be limited to the upper 5 to 10 centimeters (2 to 4 inches). As mentioned above and as discussed in the BODR, the cap designs presented herein provide an erosion protection layer component (Te) of the cap that is sufficient for both physical isolation and bioturbation (Tb).


62

6.1.3 Consolidation Component

An evaluation of cap-included consolidation was presented in the BODR, including seepage induced consolidation testing (SICT) to estimate the magnitude of consolidation anticipated to result from cap placement and the resulting volume of porewater expulsion that can contribute to the advective flux of contaminants into the cap for a given cap thickness. Post-consolidation permeability and porosity of the existing sediments were also estimated from the SICT results and used in the contaminant transport modeling to determine the chemical isolation thickness as discussed above in Section 6.1.1. The cap material itself will be granular and is expected to undergo elastic settlement within the period of construction. Therefore, no additional cap thickness is included to account for cap consolidation.

6.1.4 Erosion Protection Component

The erosion protection component of an in situ cap prevents external forces from disturbing the cap or the underlying contaminated sediments. Several potential forms of erosion including hydrodynamic flows, ice scour, wind-induced waves, and vessel-induced propeller wash were evaluated for the preliminary cap design, as detailed in Appendix D in the BODR. Refinements to and additional considerations regarding the erosion component of the cap design have been conducted since the submittal of the BODR, specifically related to hydrodynamic flows, vessel-induced propeller wash, and vessel wakes, as described in the sections below.

6.1.4.1 Supplemental Vessel-Induced Propeller Wash Analysis

As part of the BODR, the potential impacts of vessel-induced propeller wash were evaluated consistent with USEPA/USACE guidance documents (USACE 1998a) and technical literature (Verhey 1983, Blaauw and van de Kaa, 1978). The USEPA/ USACE guidance presents a modeling procedure and engineering framework that is appropriate for propwash/cap armor evaluations of large ocean-going vessels operating at very slow speeds (e.g. maneuvering operations), typical of the operations in the Fox River OU 4B navigation channel. Therefore, the available guidance was used in the BODR to design a protective cap for the OU 4B channel consisting of a 33-inch-thick sand, gravel, and quarry spall (6- to 9-inch diameter stones) cap. This propwash and associated cap armor design analysis specific to the OU 4B channel areas remain unchanged from the BODR, except as noted in Section 6.2.1 relative to the armor design for the navigation channel side slopes.

As discussed above, certain components of the available guidance and technical literature are based on large, maneuvering vessels which only operate within the limits of the OU 4B navigation channel. However, other portions of OUs 2 to 5 are subject to propwash from small, moving recreational vessels, for which the USEPA/USACE guidance may not be applicable. Therefore, to evaluate these recreational vessels as part of the BODR, the predictive equations developed for the larger vessels were adapted to address smaller recreational vessels under moving conditions based on a field study conducted in October 2005 where bottom-mounted current meters were used to measure actual bottom velocities of maneuvering and passing recreational vessels in the Fox River. Based on engineering evaluations


63

performed for the BODR, a median stone size of 1.5 inches (i.e., gravel) was shown to resist the reasonable worst-case hydrodynamic condition in all areas of OU 3 and 4 and to also resist erosion expected when subjected to the propeller wash of a range of characteristic vessels (e.g., Foxy Lady Tour boat or recreational boats) passing over an in situ cap under relatively shallow water conditions.

Since submittal of the BODR in May 2006, more detailed analyses of the propwash from recreational vessels was performed to refine and optimize cap designs to further ensure long-term stability and performance by developing recommendations for the size of armor stone that would be necessary to resist the erosive forces from the propeller wash generated by recreational boats operating on the Lower Fox River. As part of these more detailed analyses, available site-specific vessel information was reviewed to develop a refined propwash modeling framework specifically for evaluating recreational propwash on the Lower Fox River while taking into account modeling results, engineering considerations (e.g., material gradations, implementability, cost, etc.), and best professional judgment (e.g. consideration of other natural and engineered aquatic sites).

The JETWASH model used for the OUs 2 to 5 cap armor design is very similar to that recommended in the USEPA/USACE guidance, but includes additive velocities to account for the propeller shaft pitch relative to the bottom (a critical factor for recreational boats) and the reflection of the propeller jet with the bottom. The OUs 2 to 5 modeling framework utilizes a momentum-based particle stability evaluation to account for the transient nature of the recreational propwash. A series of technical memoranda were developed and submitted summarizing the technical basis for the Fox River propwash modeling framework and illustrating an example computation (see Attachment B-3 in Appendix B). The technical memoranda were reviewed and accepted by USEPA and WDNR on September 10, 2007.

Due to the wide variety of recreational vessels and modes of operation on the Lower Fox River, the refined propwash computations utilized a Monte Carlo simulation for a range of appropriately conservative input variables (e.g., engine and operational characteristics) based on site-specific data. The Monte Carlo simulations were used to generate a probability of occurrence for each particular combination of input parameters. Furthermore, the cap armor design framework utilized five levels of conservatism including model conservatism, conservative input parameters, low probability of occurrence (i.e., generally less than 5 or 10 percent), conservative cap gradation, and self-armoring. The modeling framework and approach are described in detail in the Cap Design Tech Memo dated September 24, 2007 (see Appendix B, Attachment B-1).

A total of approximately 2,500 combinations of parameters for each of several representative water depths (3, 5, 7, and 10 feet) were generated through the Monte Carlo simulation. As part of the model input development, consideration was given to combinations of engine and operational characteristics that are technically feasible and consistent with safe boat operation in the Fox River (e.g., consideration of vessel depth relative to shallow water present in portions of OUs 3 and 4). The JETWASH model and particle stability evaluation were then performed for each of the combinations.


64

Table 6-2 summarizes the general cap armor recommendations necessary to resist the erosive forces expected to be generated by recreational vessels operating in various water depths of the Lower Fox River. These recommendations were developed through the technical Workgroup process with the Agency/Oversight Team based on an engineering evaluation utilizing Monte Carlo model output, engineering considerations, and best professional judgment. A detailed summary of the refined propwash analyses is provided in Attachment B-3 of Appendix B.

Table 6-2. Summary of Cap Armor Recommendations for Recreational Propwash

Post-cap Water Depth

Median Stone Size, D50 [inches]

Maximum Stone Size,

D100 [inches] Classification

3 to 4 feet 3 6 Gravel/Cobble

4 to 6 feet 1.5 3 Gravel

>6 feet 0.25 1 Sand/Gravel

Note: This table presents recommended armoring to resist propwash from recreational vessels operating in OUs 2 to 5. Propwash armor designs for the OU 4B channel, where large ocean-going vessels operate, are not included here.

The general recommendations for cap armor materials are used in conjunction with the results of other hydrodynamic analyses relative to cap design (i.e., wind wave, vessel wake, river flows, etc.) to preliminarily delineate the extents of various cap armor designs within OUs 2 to 5 for the 30 Percent Design (see Section 6.3). Appropriateness of the general cap armor designs in specific localized areas of the river are being refined as necessary based on site-specific conditions such as proximity to storm water outfalls, tributaries, boat launch and/or marine facilities, and the frequency and probability of actual vessel use. Such refinements will continue in subsequent phases of the design.

6.1.4.2 Supplemental Hydrodynamic Flow Analysis

Sea Engineering Inc. (SEI) developed and calibrated a detailed 2-dimensional hydrodynamic model using an extensive data set from USGS in OUs 3 and 4 to predict bottom shear stresses during a design level flow event (24,200 cfs [685 m3/s]) with a recurrence interval of approximately 100 years. Initial model runs were conducted using existing bathymetry and results were used to evaluate the applicability of various cap designs. Following initial design, the detailed hydrodynamic model was run again using post-remedy design bathymetry and the remedial actions refined as necessary to accommodate predicted hydrodynamic erosion forces. These analyses are described in detail in Appendix D of the BODR. In summary, relative to existing conditions, the hydrodynamic model predicted only minor changes from existing conditions in shear stresses throughout OUs 3 and 4 under post-cap bathymetric conditions and corresponding to the reasonable worst-case hydrodynamic design condition (i.e., simultaneous 100-year flows, historical low water levels, and maximum seiche amplitude). A conservative maximum bottom shear stress of 100 dynes/cm2 was selected for design and was correlated to minimum thickness of 4 inches of 1.5-inch armor stone, consistent with USEPA/USACE guidance.


65

Recent reviews of the available hydrodynamic information available for the Lower Fox River suggests that local runoff (e.g., from the East River) may discharge into OU 4 at rates higher than the values assumed during the BODR hydrodynamic modeling. To further inform sensitivity analyses of RD parameters, additional supplemental model simulations were performed in March 2007 using a range of extreme flow assumptions, including hindcasting from a record rainfall event that occurred on June 22-23, 1990 (Shaw and Anchor 2007). Figure 6-1 presents the predicted bottom shear stresses for this extreme flow event modeling. Additional details are provided in Attachment B-4 of Appendix B. The results of this sensitivity analysis modeling further confirmed that the engineered cap designs presented in the BODR will adequately protect against disturbance from extreme river flows.

To supplement previous modeling of extreme river flows and to further ensure that appropriately conservative cap design are specified for localized areas of OU 4, the 2-dimensional hydrodynamic model was revised in October 2007 as part of the 30 Percent Design to evaluate the localized effects of tributary inflows at their specific geographic location during the peak discharges measured during the June 22-23, 1990 event discussed above. Tributary flow contributions to the Lower Fox River during the extreme rainfall event were determined using the Soil and Water Assessment Tool (SWAT) model, a physically-based river basin scale model developed for the USDA Agricultural Research Service (ARS) to quantify the impact of land management practices in large, complex watersheds (Technical Memorandum 2a, 1998). These tributary loadings were then applied to the hydrodynamic model developed for OUs 3 and 4 to simulate the June 22-23, 1990 flow conditions.

For both OUs 3 and 4, resulting shear stresses in the majority of the reaches were predicted to be significantly less than the maximum bottom stress of 100 dynes/cm2 selected for armor stone design in the BODR (see Figures 6-2 and 6-3). Localized shear stresses in excess of the original design shear stress (100 dynes/cm2) were observed in only two areas: in OU 3 immediately below the Little Rapids Dam, and in OU 4 just downstream of the East River. However, these areas have not been targeted for capping as part of the OUs 2 to 5 remedial action. Therefore, this supplemental modeling further confirms that the engineered cap designs presented in the BODR will adequately protect against disturbance from extreme river flows. The supplemental hydrodynamic modeling approach and results are described in further detail in Attachment B-4 of Appendix B

6.1.4.3 Vessel Wake Analysis

As part of the 30 Percent Design phase, additional engineering analyses were performed to further evaluate the erosive forces in shoreline areas designated for capping (i.e., engineered cap or dredge-and-cap). Specifically, impacts from vessel-generated waves were preliminarily evaluated for representative cap areas in OUs 3 and 4 considering typical design vessels passing through areas targeted for capping to design cap armor stones to resist the predicted design wave(s). The approach and preliminary results for this evaluation are generally summarized in this section and explained in further detail in Attachment B-5 of Appendix B. The results presented herein are considered preliminary and the design of shoreline caps will be more fully developed during the 60 Percent Design.


66

Classification of design vessels for this analysis were based on a comprehensive evaluation of data complied from several resources including ship arrival records from the Port of Green Bay, reported bridge openings on the Lower Fox River within OU 4, and information compiled for the propeller wash analysis discussed in Section 6.1.4.1. Table 6-3 provides a summary of vessels considered for this analysis and corresponding design parameters obtained from available data (e.g., vessel length, beam width, weight, draft, engine power, entrance length, and typical cruising speed). It should be noted that the “cargo” vessels evaluated include the deep-draft (approximately 15 to 24 feet) class of vessels that exclusively operate within the OU 4B federal navigation channel due to draft limitations in the upper reaches of the river, whereas recreational vessels and tug-boats were assumed to navigate in all reaches of OU 3 and OU 4.

A comprehensive evaluation and summary of available empirical models for vessel wave prediction was developed by the USACE Coastal and Hydraulics Laboratory (CHL) as part of the Upper Mississippi River-Illinois Water (UMR-IWW) System Navigation Study (Sorensen 1997). This report also discusses the applicability of these models for various vessel types. An evaluation of the models presented in this document and personal correspondence with Dr. Robert Sorensen led to the selection of four models applicable for predicting vessel wakes generated by the vessels characteristic of the Fox River remediation areas:

• Sorensen and Weggel (1984, 1986)

• Permanent International Association of Navigation Congresses (PIANC) (1987)

• Kriebel and Seelig (2005)

• Bhowmik (1991)


67

Table 6-3. Summary of Design Vessels for Vessel Wake Analysis

Site and Water Depth Vessel Name (type)

Draft

[ft]

Length

[ft]

Beam

[ft]

Capacity

[Gross Tons] or Deadweight

Bulk Carrier 12 650 60

Alpena 26.42 503 67 G.T. 13,900

J.A.W. Iglehart 27.5 502 68 G.T. 12,500

Paul H Townsend 22.17 432 50 G.T. 7,850

OU 4B 22 to 26 feet Recreation –Yacht 50

Foxy Lady (yacht) 4 65 G.T. 85

D.W. 40 (loaded)

Gopher (tug) 3.5 30 9 D.W. 7

Badger (tug) 3 30 10 D.W. 8.5

OU 4A 3 to 20 feet

Mary Jane (tug) 3 32 30 D.W. 8

Recreation – Small Fishing Boat 1.5 16

Recreation – Water ski boat 2 20

OU 3/4A 3 to 20 feet Recreation – Water ski

boat 3 24

G.T. = Gross tons D.W. = Dead weight (tons)

It should be noted that each of these models is applicable to a given range of vessel types and sizes and not all models are applicable to the full range of vessels expected to operate on the Fox River. These models were used to estimate the critical wave height generated by a given design vessel passing through representative sections of the Lower Fox River where dredging and/or shoreline capping are anticipated along the river bank. Appendix B, Attachment B-5 provides additional details of the critical wave predictions.

The USACE Automated Coastal Engineering System (ACES 1992) was used to transform the critical wave height for each condition modeled to simulate the change in wave characteristics as the design wave propagates towards shore (e.g., accounting for the interaction of the wave with the bottom in shallower water depths). Furthermore, the ACES software was utilized to design an appropriate shoreline cap armor design including consideration of armor stone size(s) and layer thicknesses (i.e. revetment design). Table 6-4 presents a summary of the armor layer design necessary to resist vessel wakes anticipated for representative transects in OU 4A and 4B. Subsequent phases of design (e.g. 60 Percent Design) will include a review of the applicability of these designs to specific shoreline caps areas within OUs 2 to 5.


68

Table 6-4. Summary of Preliminary Cap Armor Recommendations for Vessel Wakes Cap Design

Armor Layer Filter Layer Representative Capping Area

Critical Design Wave Height [feet]

D50 D100 Thickness D50 D100 Thickness Cap Armor within Surf Zonea

OU 4A (Transect 4044)

3.2 0.83 ft. 1.32 ft. 1.7 ft. 0.10 ft. 0.18 ft. 1.0 ft.

OU 4B (Transect 4061)

3.7 0.99 ft. 1.57 ft. 2.0 ft. 0.12 ft. 0.22 ft. 1.0 ft.

Capping Armor below Surf Zonea D50 [Inches]

OU 4A (Transect 4044)

3.2 0.53 inches (Medium Gravel)

OU 4B (Transect 4061)

3.7 1.27 inches (Coarse Gravel)

Surf zone defined herein as water depth range subject to breaking waves, which may extend from the top of bank to approximately 1 to 2 times the wave breaking depth.

D50 = median particle diameter in gradation D100 = maximum particle diameter in gradation

6.1.5 Operational Component

As discussed in the BODR, given the inherent difficulties in achieving accurate placement tolerances for in-water construction, an additional thickness (“over-placement allowance”) is typically specified in the capping contract. For OUs 2 to 5 the over-placement amount is expected to vary between 0 to 6 inches with an average of less than 3 inches for sand and gravel layers and an average of 6 inches for quarry spall armor layers. This is based on anticipated cap placement equipment, experience at other similar capping projects, and considerations of likely contractor incentives to limit the amount of excess thickness. Therefore, for subsequent engineering evaluations (e.g., evaluation of post-cap water depth, geotechnical stability evaluations, construction schedule, etc.), the following additional thickness have been assumed, as summarized in Table 6-8:

• Sand cap material: 3 inches

• Gravel armor material (up to D50 of 1.5 inches): 3 inches

• Large gravel (D50 of 3 inches) and quarry spall armor material (where applicable): 6 inches.

In accordance with Workgroup discussions, specification language will be developed as part of the CQAPP which will detail the method to be used to verify that minimum cap thicknesses have been achieved over a certain area with a specified level of confidence.


69

6.2 Additional Cap Design Considerations

6.2.1 Federal Navigation Channel

The extent of engineered caps has been delineated to avoid interference with the navigation and maintenance of the federal channel. As such, the horizontal extent of caps were offset at least 10 feet horizontally from the lateral boundaries of the federal channel, while the top of the cap (with target overplacement allowance) was offset at least 2 feet below the vertical boundary of the navigation channel (i.e. 2 feet below the authorized channel depth). The boundaries of the federal navigation channel in OU 4A were based on the reauthorization language included in the Water Resources Development Act of 2007 (Pub. L. 110-114) (see Section 4.1.3).

Cap design evaluations conducted as part of the BODR indicated that 4- to 9-inch diameter armor stones (e.g., quarry spalls) would be appropriate for resisting propeller wash generated by large cargo ships operating in the OU 4B federal navigation channel. This armoring was estimated to be necessary primarily along the base of the navigation channel. Calculations completed as part of the 30 Percent Design indicate that smaller armoring (typically less than 3-inch diameter) would be appropriate to resist propwash along most of the side slopes of the navigation channel (designed at 3H:1V extending outside of the limits of the authorized channel) where the propwash would be generally directed along the centerline of the channel (i.e., parallel to the side slopes rather than perpendicular to it). In this case, the distance between the propeller and the side slope is typically in excess of several hundred feet, resulting in significant reductions in the erosion potential due to the radial spread and dissipation of energy within the propwash jet. Attachment B-3 of Appendix B presents a summary of the propwash calculations for the cap armor design on the slopes of the OU 4B navigation channel.

Within the Fort Howard Turning Basin it is possible that vessel maneuvering operations could result in the propeller being directed perpendicular to the side slopes. Furthermore, given the relatively limited turning radius with this area, the distance between the propeller and the side slope could be limited (200 feet or less in extreme cases). Therefore, side slopes of the Fort Howard Basin that require capping will be designed for the 33-inch-thick cap with quarry spall armoring.

Other limited stretches of the navigation channel side slopes (aside from the Fort Howard Turning Basin) may be subject to propwash flows at an incident angle to the side slope (i.e., not parallel or perpendicular). This would primarily occur on the outside of a bend in the navigation channel. Within the potential capping areas delineated as part of the 30 Percent Design, only the eastern side slope of the channel between transect 4050 and 4051 and the western side slope near transect 4056 could be subject to propwash flows at an incident angle. However, based on a preliminary propwash evaluation for these areas, it is estimated that propwash effects will be minor and 3-inch diameter cap armoring will be sufficient. The armor design for these side slopes will be verified during the 60 Percent Design.

In addition to the erosion protection provided by the large armor stone placed within the navigation channel as discussed above, it will also serve as a physical marker of the top of the cap if future


70

maintenance dredging inadvertently excavates well below the authorized depth in OU 4A or 4B channel. Therefore, consistent with the ROD Amendment, engineered cap designs within the limits of the navigation channel will be a minimum of 33 inches thick with 4- to 9-inch diameter armor stones.

6.2.2 Infrastructure and Utilities

The river banks along much of the OUs 2 to 5 site have been developed as either commercial (primarily in OU 4) or residential (primarily in OU 4A, OU 3 and OU 2). Along these banks and crossing the river, numerous structures (e.g. bulkhead walls, shore protection, etc.), docks, piers, bridges, and utility crossings have been identified through desktop surveys and field reconnaissance. For the purposes of the 30 Percent Design, remedial action plans (i.e., dredging, capping, and sand cover) were developed incorporating a nominal (10- to 50-foot) offset from shoreline structures such as boat docks. However, the 30 Percent Design plans presented herein do not include offsets from other in-water structures, including utility crossings, bridges, etc. Future design analyses as part of the 60 Percent Design will include a case-by-case evaluation of appropriate remedial actions in close proximity to each of these structures. Table 6-5 presents a preliminary set of potential shoreline remedial design considerations, as developed during the BODR. Design evaluations conducted during the 60 Percent Design phase may include assessing the extent of contamination, potential environmental risk posed by the contaminated sediment, and practicability and risks of performing remedial actions. Furthermore, the 60 Percent Design will include discussions with property owners, as appropriate, to support design.


71

Table 6-5. Potential Shoreline Remedial Design Considerations Shoreline Condition Potential Remedial Design (a,b)

Shoreline deposits If shoreline DOC < 2 ft, dredge with partial removal of uplands or cap if appropriate. Otherwise, cap along shoreline if dredging would impact stability.

Sheet pile wall Site-specific review of wall design relative to potential dredge cut; cap along shoreline if dredging would impact stability.

Riprap or armored slope Additional sampling in nearshore slope areas to refine extent of sediments > 1 ppm RAL; adjust dredging and capping plan accordingly.

Pile-supported wharf Site-specific review to address impacts of dredging and/or capping

Floating dock with guide piles Site-specific review to address impacts of dredging and/or capping

Outfall Site-specific review to address potential options including: dredge around outfall; cap above outfall; relocate outfall; and extend outfall through shoreline cap

Shoreline building Cap or dredge along shoreline depending on stability evaluation.

Shoreline or in-river bridge support Cap along shoreline with review of potential dragdown forces on support

Utility crossings Dredge offset from utility location; prospective capping area

Boat launch/ramp Potential options include armored cap and dredge/armored cap.

a. Final remedial design for these areas to be determined as part of the 60 Percent Design based on the results of detailed shoreline surveys and associated engineering analyses.

b. DOC = Depth of contamination, as determined through geostatistical modeling.

6.2.3 Geotechnical Stability

The BODR presents several geotechnical evaluations relative to the stability of engineered caps, including:

• Bearing capacity of existing sediments - A maximum cap layer thickness (i.e., critical height differential) of 10 to 12 inches that could be placed in a single application was calculated in general accordance with the EPA/Corps guidance (Palermo et al. 1998b). However, to minimize mixing of the cap into underlying sediments, a maximum 6-inch initial cap lift thickness was assumed during construction. Caps thicker than 6 inches will require multiple lifts, providing a consolidation period between lifts to increase bearing strength. The results of initial and ongoing cap placement monitoring may be used to adjust this maximum lift thickness as construction proceeds.

• Slope Stability – Analyses indicate that caps placed on slopes up to 2.75H:1V are predicted to be stable, with a factor of safety of 1.3 or better. More detailed evaluations of nearshore cap requirements will be included in subsequent design submittals.

• Cap Punch Through Analysis – Analyses were conducted consistent with USEPA/USACE cap design guidance (Palermo et al. 1998b), to ensure that caps would support the weight of an individual walking on the surface, assuming that top of cap could be in shallow water (e.g. 3 to 5


72

feet deep). This analysis concluded that the cap designs have a safety factor of at least 1.3 under this condition, and thus will be stable under worst-case bearing loads.

6.2.4 Ebullition

As discussed in the BODR, cap designs considered the potential for gas generation and its possible effect on cap stability. Based on the design analysis, sand and gravel materials incorporated into the cap design will dissipate any gas (e.g., methane) that may be produced in the underlying sediments. The design analysis did not identify gas ebullition as a short- or long-term pathway of potential concern that would affect performance of the cap system.

6.2.5 Post-Cap Water Depth

Consistent with the ROD Amendment, all engineered caps (including capping with and without prior dredging) are designed such that a minimum post-cap water depth is maintained under historic low water elevations summarized in Table 6-6, as developed in the BODR. In addition, detailed armor designs are developed for given ranges of post-cap water depth to accommodate specific erosion characteristics, including propwash, vessel wakes, etc, as detailed above.

Table 6-6. Summary Baseline Water Elevations

Operable Unit Baseline Water Elevation Dynamic Height (IGLD85) Basis for Selection

OU 2 593.5 feet NOAA Low Water Datum above Little Kaukauna Dam

OU 3 587.3 feet Crest of De Pere Dam (and NOAA Low Water Datum)

OU 4

576.5 feet

Lower 1% occurrence frequency of hourly summer data from NOAA gage at Green Bay (adjusted for long-term data record through 1953)

6.3 Capping Designs and Areas

6.3.1 Cap Designs

As discussed in Section 6.1, the BODR developed three general cap designs based on preliminary engineering analyses. These cap designs were subsequently adopted in the ROD Amendment, which specified minimum thickness criteria based on PCB concentration and the preliminary erosion analyses presented in the BODR as summarized below.

• Cap A - Sand and gravel cap for PCBs < 10 ppm– consisting of a minimum 3 inches of placed sand (equivalent to a targeted average thickness of 6 inches within the placement area considering normal overplacement allowances), overlain by a minimum 4 inches of placed armor material (7 inches with overplacement allowances). Note, thickness and size of armor layer refined during 30 Percent Design based on localized conditions, as detailed above.


73

• Cap B - Sand and gravel cap for PCBs > 10 ppm and < 50 ppm – consisting of a minimum 6 inches of placed sand (9 inches with overplacement allowances) overlain by a minimum 4 inches of placed armor material (7 inches with overplacement allowances). Note, thickness and size of armor layer refined during 30 Percent Design based on localized conditions, as detailed above.

• Cap C - Sand and quarry spall cap for PCBs > 50 ppm and in federal navigation channels– consisting of a minimum 6 inches of placed sand (9 inches with overplacement allowances) overlain by a filter layer of gravel (6 inches with overplacement allowances) and finally overlain by a minimum 12-inch-thick placed layer of suitably sized armor material (18 inches with overplacement allowances). Within the OU 4A and 4B navigation channel, four to nine-inch quarry spall material will be required for the armor layer.

In addition to the general cap designs summarized above, the BODR also identified the potential for shoreline capping in limited areas of the river (see Section 6.3.1.1 for additional details). Site-specific shoreline cap designs will be evaluated during subsequent design phases, but are generally anticipated to include the following:

• Shoreline Cap – consisting of 3 to 6 inches of placed sand (thickness depending on PCB concentrations) overlain by a filter layer (if necessary) and armor stone (size and thickness dependent on erosive forces). See Section 6.3.1.1 for additional details.

Table 6-7 presents a summary of the engineered cap areas delineated as part of the 30 Percent Design in comparison to that from the BODR.

Table 6-7. Summary of Cap Delineation 30 Percent Design

OU 2 [acres]

OU 3 [acres]

OU 4/5 [acres]

Total OU 2-5 [acres]

BODR [acres]

Cap A 6 54 155 215 350 Cap B 1 16 37 54 25 Cap C 0 0 103 103 25

All Engineered Caps 7 70 295 372 400 Shoreline Caps (a) 0 1 40 41 41

(a) Shoreline capping may be necessary in those areas where dredging will adversely impact the stability of existing slopes. Areas presented above are preliminary estimates, subject to further RD engineering evaluations, including a location-specific review of these areas during 60 Percent Design.

As discussed in the sections above, these cap designs were refined as part of the 30 Percent Design based on detailed erosion analyses. These analyses resulted in the expansion of the general cap designs included in the ROD Amendment, to include armor stone sizing and thickness based on location and depth-specific erosional conditions, a summarized in Table 6-8.


74

6.3.1.1 Engineered Shoreline Caps

Shoreline caps will be installed where subsequent remedial design engineering evaluations (conducted as part of the 60 or 90 Percent Design) conclude that dredging would adversely affect the stability of the existing slopes. Building upon the BODR conceptual design, the 30 Percent cap design plans preliminarily identify a nominal 50-foot-wide zone of potential shoreline capping along the river banks where greater than 1 to 2 feet of sediments exceeding the 1 ppm RAL was estimated at the river’s edge. The remedy within these shoreline zones will be evaluated on a case-by-case basis during later stages of the design using the results of shoreline investigations completed in 2006 and 2007 and detailed engineering evaluations, where necessary. Where shoreline capping is deemed necessary, appropriate armor stone sizes and thicknesses will be designed based on the results of wind wave analyses presented in the BODR along with vessel wake analyses presented in Section 6.1.4.3. Based on the preliminary vessel wake analyses, shoreline caps within OU 4B are expected to require larger armor stone (up to 1-foot median stone diameter) than those in OU 4A and OU 3. Due to their proximity to the shoreline, the engineered caps placed in these areas may result in post-construction water depths less than 3 feet. These design refinements will be integrated into the forthcoming 60 Percent Design.


75

Table 6-8 Summary of OUs 2 to 5 Engineered Cap Designs

Layer Thickness (b) (inches)

Placed Layer Thickness (c)

Cap Type

Post-Cap Water

Depth (a)

Median Stone Size of Gravel

Armor, D50 (inches)

Sand Gravel Armor

Quarry Spall Armor (D50 = 6-9”)

Cumulative Layer

Thickness (b) (inches)

Sand Gravel Armor

Quarry Spall Armor

(D50 = 6-9”)

Placed Total Cap

Thickness (c)

(inches)

Cap A : PCB in top 6” below cap < 10 ppm and < 50 anywhere in depth profile Cap A1 3 to 4 feet 3 3 6 0 9 6 12 0 18 Cap A2 4 to 6 feet 1.5 3 4 0 7 6 7 0 13 Cap A3 > 6 feet 0.25 3 4 0 7 6 7 0 13

Cap B: PCB in top 6” below cap > 10 ppm and < 50 anywhere in depth profile Cap B1 3 to 4 feet 3 6 6 0 12 9 12 0 21 Cap B2 4 to 6 feet 1.5 6 4 0 10 9 7 0 16 Cap B3 > 6 feet 0.25 6 4 0 10 9 7 0 16

Cap C: PCB > 50 ppm or in federal navigation channel Cap C1 > 3 feet 1.5 6 3 12 21 9 6 18 33

Shoreline caps d

N/A To be designed during subsequent design phases

Notes: a. Caps will not be placed in locations such that the minimum low river stage water depth is less than 3 feet above the top elevation of the constructed cap. b. Minimum required thickness based on USEPA/USACE design guidance. c. Contractor will be required to place enough material (as measured by placement logs) to achieve target thickness on average within a certification area (see

CQAPP). d. Shoreline cap design to be completed as part of future design phase with consideration of local erosion evaluations (propwash, vessel wakes, wind waves,

etc.)


76

6.3.2 Delineation of Cap Areas

As noted in Section 4.3, the BODR discussed the core-by-core process in which preliminary dredge, cap, cover, and dredge and cap (D&C) boundaries were delineated. Cap and D&C areas are generally sited in localized areas with deep, stable deposits of contaminated sediment that are not currently bioavailable, that do not contribute to current or future site risks, and/or that will pose considerable difficulties in a dredging-only remedy. Detailed hydrodynamic analyses were performed to evaluate potential erosion from a wide range of natural and anthropogenic forces at each location. Caps are incorporated into the design within areas where permanent stability and performance can be assured. In situ capping of sediments exceeding the 1.0 ppm RAL will also be performed along shoreline areas where RD evaluations conclude that dredging will adversely affect the stability of the existing slopes. The use of shoreline capping will be evaluated on a case-by-case during later phases of the design.

The initial boundaries of capping locations selected from the BODR core-by-core process were delineated using a Thiessen polygon approach. As the design progressed from conceptual level to the 30 Percent Design level, the boundaries were refined using the preliminary dredge plan and the spatial extent of the DOC at a LOS of 0.5. As shown on Figures 2-8 and 2-9, the spatial extent of the DOC resembles a curvilinear polygon. The preliminary cap plan was developed by delineating a series of rectangles that fully capture the DOC, transition to any adjacent dredge or cover areas, and are constructable (i.e., designated at regular 10-foot grid intervals). During this process, elevation was tracked to ensure a uniform surface was maintained between adjacent dredge and cap areas.

After the preliminary cap plan was defined, the cap criteria described in Section 6.3.1 were evaluated for each general area to determine the required cap type based on the underlying chemistry. For locations where the cap will be placed without prior dredging, the upper 6-inch sample from the nearest core location was evaluated to determine the appropriate chemical isolation layer thickness. For D&C areas, the underlying chemistry was based on an estimate of the generated residuals and predicted post-dredge concentration. This estimate was used to determine a preliminary cap type (i.e., Cap A, Cap B, etc.) for the area; although final cap type designation will be based on post-dredge confirmation sampling (see Section 10 for additional information).

Each location within the cap plan was then re-evaluated to determine if the associated core contained concentrations of PCBs greater than 50 ppm at any depth interval. In the event a location contained a core sample greater than 50 ppm, either a Cap C section was designed, or a dredge or D&C alternative was evaluated. Appendix B, Attachment B-2 contains a comprehensive design spreadsheet used to track these evaluations. After the final delineation of cap type based on chemical criteria was complete, the entire cap plan was reviewed to ensure the appropriate armoring layer was designated based on estimated post-cap water depths. Where the preliminary cap plan resulted in water depths less than 3 feet during low water conditions (i.e., above elevation 573.5 feet IGLD), these caps were converted to either dredge or D&C alternatives. Caps located within the authorized navigation channels and turning basins are designated with the most protective armor layer (Cap C). Additional analyses will be performed during


77

the 60 Percent Design to identify other areas that may require additional armoring, such as active commercial boat slips and slopes adjacent to navigation channels.

Figures 6-3 and 6-4 illustrate the aerial extents of engineered caps delineated for this 30 Percent Design. As discussed in Section 6.3.1.1, shoreline caps are preliminarily delineated for the 30 Percent Design as a nominal 50-foot-wide zone along the river banks where greater than 1 to 2 feet of sediments exceeding the 1 ppm RAL was estimated at the river’s edge. The delineation of these shoreline caps area will be refined during the 60 Percent Design. Similarly, cap designs required to protect utilities, infrastructure, and other sensitive structures will be completed during the 60 Percent Design.


Several methods of placing capping materials have been successfully implemented on previous projects including:

• Direct placement with a mechanical clamshell bucket;

• Surface release from a barge or hopper;

• Spreading with hydraulic pipeline and baffle box or plate;

• Submerged diffuser or tremie; and

• Washing off barge with high powered jet.

Selection of the most appropriate placement method will depend on numerous factors including but not limited to:

• Site conditions (e.g. water depth, water currents)

• Distance between material stockpile and placement location

• Site access limitations (e.g. presence of locks/dams, shallow water, pilings)

• Grain size and volume of material being placed

• Site-specific placement requirements (e.g. production rates, lift thicknesses, etc.)

• Availability of placement equipment

Based on a review of site conditions in OUs 2 to 5, evaluation of capping projects performed in similar environments, and discussions with regional contractors, the BODR assumed that caps within OUs 2 to 5 would be placed primarily using mechanical equipment (e.g., barge-mounted clamshell bucket). While mechanical placement will still be necessary for at least a portion of the capping areas (e.g., for the larger armor stones that cannot be pumped hydraulically), hydraulic placement may also be feasible and cost effective in portions of the site for placement of the sand chemical isolation layer and potentially for


78

placement of the armor stone in the deeper water recreational areas (e.g. D50 of 0.25 inches; see Table 6-2). Currently, a single large upland staging area (i.e., former Shell Property in OU 4) has been identified to support the OUs 2 to 5 remedial action. Although centrally located in OU 4, this property is located nearly 10 miles from the furthest upstream reach of the site (OU 2) making hydraulic transport from the Shell Property infeasible for all reaches. However, if other strategically located shore-side properties become available at some time in the future expanded use of hydraulic placement methods may be feasible.

Production rates will vary between mechanical and hydraulic placement methods as well as due to varying material types (sand versus gravel) and site conditions (water depth, size of contiguous capping area, etc.). In difficult to access areas, mechanical placement rates are expected to range from approximately 30 to 50 cy per operating hour (average of 15 to 25 cy per hour over entire working day), assuming a small (2 to 3 cy capacity) bucket. In unrestricted areas, mechanical cap placement rates are expected to range up to 150 cy per operating hour (average of 75 to 90 cy per hour over entire working day) assuming a 5 cy bucket. Hydraulic placement rates for sand and small armor stones are expected to range widely based on location from 50 to more than 100 cy per operating hour (average of 25 to 75 cy per hour over an entire working day).

Section 7 – Sand Cover Design

79

7 SAND COVER DESIGN 7.1 Sand Cover Design and Areas

As discussed in the BODR, a substantial area of OUs 2 to 5 contains a veneer (up to 6-inches-thick) of sediments with PCB concentrations marginally above the 1.0 ppm RAL. These surficial sediments, which contain maximum PCB concentrations of up to 2 ppm, overlie cleaner sediments with PCB concentrations well below 1 ppm. Additional sediment areas within OUs 2 to 5 contain a similarly thin (6-inch) subsurface layer of sediment with concentrations between 1 and 2 ppm underlying an existing surface layer of sediment with concentrations below the 1.0 ppm RAL. Consistent with the ROD Amendment, the remedial action plans presented in this 30 Percent Design include placement of 6-inch-thick sand covers to address low risk deposits that have the following characteristics:

• Maximum PCB concentration no greater than 2 ppm in any core sample interval

• Maximum of one sampled interval (6 inch thickness of sediment) in the core with concentrations exceeding the 1.0 ppm RAL

• All other sediment in the core equal to or less than the 1.0 ppm RAL

• Other exceptional areas as approved by the Response Agencies

Figures 6-4 and 6-5 depict the sand cover areas delineated for OUs 2/3 and 4/5, respectively, for this 30 Percent Design, which includes an approximate 207 acres within OUs 2 to 5. In comparison, the BODR estimated approximately 213 acres of sand cover.


Similar to that described in Section 6.4 for caps, several methods of sand cover placement are feasible depending on site conditions. Based on the site conditions (primarily the varying distances between likely upland stockpile locations and in-water placement locations), it is expected that sand cover placement will be accomplished through a combination of mechanical, hydraulic, and pneumatic equipment. Pneumatic placement is typically feasible in small areas that can not be easily accessed with floating equipment. Mechanical placement is expected to occur in OU 2 and a large portion of OU 3 due to the extended hydraulic pipeline distances that would be required from the only currently identified upland stockpile area (e.g., former Shell Property in OU 4B). Sand cover placement via hydraulic means would likely be feasible in large portions of OU 4. Sand cover production rates for mechanical and hydraulic placement are expected to be equivalent to those estimated for the sand layer of engineered caps in Section 6.4.

Section 8 – Institutional Controls

80

8 INSTITUTIONAL CONTROLS Within OUs 2 to 5, institutional controls are necessary to reduce potential PCB exposure in specific sediment areas where residual contamination will remain following completion of remedial actions. Interim institutional controls may also be necessary to prevent exposures to contaminants which may be released during construction, such as during dredging. Long-term protectiveness requirements of the ROD Amendment require compliance with effective institutional controls, which must be effectively implemented, monitored and maintained. Detailed descriptions of required institutional controls will be provided in the Institutional Control Implementation and Assurance Plan (ICIAP), to be included as part of the forthcoming 60 Percent Design submittal. This section presents an overview of the anticipated contents of the ICIAP, including definitions of institutional controls, a summary of ROD Amendment requirements, and specific institutional controls being considered for implementation during and following remedial action.

8.1 Institutional Control Definitions

As defined in USEPA (2005a) and the ROD Amendment, institutional controls are non-engineered instruments, such as administrative and legal controls, that may be included as part of the remedial action to minimize the potential for human health or ecological exposure to sediment contamination and ensure the long-term integrity of the remedy. The term “institutional control” generally refers to measures intended to affect human activities in such a way as to reduce exposure to hazardous substances, often by limiting certain uses. USEPA guidance on institutional controls is provided in OSWER Directive 9355.0-74FS-P, Institutional Controls: A Site Manager’s Guide to Identifying, Evaluating, and Selecting Institutional Controls at Superfund and RCRA Corrective Action Cleanups (USEPA 2000a).

Institutional controls are typically grouped into the following categories (EPA 2005a):

• Land use restrictions and maintenance agreements (such as easements and restrictive covenants), which may involve legal instruments placed in the chain of title of certain properties;

• Governmental controls including permit conditions for future actions; and

• Informational devices including signage and fish consumption advisories that may be required until remedial action objectives (RAOs) are met.

General institutional control requirements for OUs 2 to 5, as described in the ROD Amendment, are summarized in Section 8.2 below. Specific institutional controls being considered for implementation during the OUs 2 to 5 remedial action are outlined in Section 8.3. The ICIAP to be provided as part of the 60 Percent RD submittal will provide additional details on institutional controls, and will more specifically identify the parties responsible for implementation of the individual institutional control (i.e., federal, state or local authorities or private entities) for implementation, enforcement, and monitoring.


81

8.2 ROD Amendment Requirements

The ROD Amendment requires institutional controls in OUs 2 to 5 to maintain the integrity of the remedy, including protection of engineered caps and reducing potential exposure in monitored natural recovery areas where residual contamination will remain after completion of remedial actions. Institutional controls will be detailed in the forthcoming ICIAP (provided as part of the 60 Percent Design submittal), which shall include maps of the restricted areas, depicting areas that do not allow unlimited use/unrestricted exposure and areas where institutional controls have been implemented along with a schedule for updating them. In addition, the ICIAP will identify reporting requirements associated with each institutional control including periodic certifications regarding the status and effectiveness of the institutional controls.

The ROD Amendment discussed the use of governmental and/or property use institutional controls such as a Regulated Navigation Area (RNA), and designating areas (including appropriate buffers) where use restrictions will be required. The ROD Amendment outlined the following types of institutional controls that may be considered during RD:

• Water use restrictions (e.g., limitations on anchoring, spudding, dragging or conducting salvage operations, establishment of "no wake" areas and other operating restrictions for commercial and non-commercial vessels which could potentially disturb the riverbed or the engineered remedy, or restricting public access to remedial construction areas);

• Construction limitations (e.g., restrictions on dredging and other excavation activities such as laying cable in aquatic areas);

• Monitoring and maintenance requirements for certain areas (e.g., dams); and

• Providing additional information to the public to assure protectiveness of the remedy (e.g., fish consumption advisories).

The ROD Amendment anticipated some localized impacts to engineered caps such as anchoring activities, since such disturbances are not expected to compromise the overall effectiveness of the remedy (see Section 8.3.1 below). As discussed in Section 6 of this 30 Percent Design Report, engineered caps located within federal navigation channels have been designed such that the top of the cap is at least 2 feet below the authorized depth of the navigation channel, obviating the need for ongoing enforcement of institutional controls in these areas.

Specific institutional controls being considered for implementation during the OUs 2 to 5 remedial action are outlined in the following section.

8.3 Specific Institutional Controls under Consideration for OUs 2 to 5

As outlined above, the following types of institutional controls are being considered for OUs 2 to 5:


82

• Water Use Restrictions

o Establishment of "no wake" areas or other vessel operating restrictions

o Limitations on anchoring, spudding, dragging and salvage operations

o Limitations on public water access during remedial construction

• Construction Limitations

o Limitations on dredging (e.g., for marina expansion or maintenance)

o Limitations on other in-water excavation activities (e.g., laying cable)

o Limitations on pier construction (e.g., pile placement and extrusion)

• Monitoring and Maintenance

o Engineered cap monitoring and maintenance

o Natural recovery monitoring

o Dam maintenance and removal contingencies

• Public Information

o Fish consumption advisories

Each of these institutional control categories is discussed in the sections below.

8.3.1 Water Use Restrictions

Capping designs for OUs 2 to 5 were developed to ensure protection from worst-case hydrodynamic flows including floods, seiches, wind waves, boat wakes, propeller wash, and ice scour forces that could occur within the remedial action area (e.g., capable of withstanding a 100-year or greater event; see Appendix D of the BODR, and Section 6 and Appendix B of this 30 Percent Design Report). The armor layer designs summarized in this report were developed to resist worst-case shear stresses with an appropriate factor of safety, concurrently providing both an erosion protection and bioturbation layer. While gravel-sized (e.g., 1- to 2-inch) armor provides protection from potential erosion throughout most of OUs 2 to 5, slightly larger armor stone is specified in limited areas with water depths between 3 and 4 feet to ensure long-term protection. No caps will be constructed in less than 3 feet of water or in potential frazil ice areas to ensure their stability. For capping areas within the OU 4B federal navigation channel, where large cargo ships occasionally call on docks up to the Fort Howard turning basin, engineering evaluations concluded that an armor layer constructed of larger stones (6 to 9 inches; i.e., quarry spalls) are needed to ensure that cap erosion does not occur under anticipated worst-case conditions. Section 6.1.4 of this 30 Percent Design Report provides additional details on the development of protective cap designs for specific areas of OUs 2 to 5.

Since the cap designs developed for OUs 2 to 5 will resist worst-case shear stresses and other potential erosional events with an appropriate factor of safety, as outlined above, there is no need to establish no-wake or restricted vessel speed zones to ensure protection of engineered caps in the Lower Fox River.


83

Anchoring, spudding, dragging and salvage operations are often restricted in sediment capping areas to provide an additional precaution against activities that could potentially compromise the integrity of the armor layer (EPA 2005a). However, the cap designs developed for OUs 2 to 5 include placement of a target thickness of at least 7 inches of an armor layer comprised of gravel or quarry spalls (as appropriate for the particular location). Considering the range of anchor types for recreational vessels that operate within the Lower Fox River, and also considering the bearing strength of the designed armor layer, recreational vessel anchors as may be deployed in capping areas are not expected to penetrate through the relatively coarse-grained armor layer. Moreover, should an anchor penetrate through the armor layer, the “self-healing” behavior of the cap upon withdrawal of the anchor is anticipated to maintain the integrity of the cap (e.g., see Palermo et al. 1998). In addition, the OUs 2 to 5 caps will be constructed in net depositional environments within the river, such that new sediment will begin accumulating on the cap surface immediately following construction. The clean sediment layer accumulating on the cap will reduce the anchor-related impacts and increase the overall effectiveness of the cap over the long term.

Thus, while further prohibitions on the anchoring of vessels within capping areas of OUs 3 and 4 would provide additional assurance of the long-term protectiveness of engineered caps, such restrictions are unlikely to be necessary to ensure the effectiveness of the remedy. The U.S. Coast Guard already enforces restrictions on anchoring, dredging and other related activities within and/or beneath the navigation channel, and these restrictions are expected to continue in perpetuity as part of ongoing channel operations.

As discussed above, the ROD Amendment anticipated some localized impacts to engineered caps such as anchoring activities, since such disturbances are not expected to compromise the overall effectiveness of the remedy. Nevertheless, for caps located in areas with a relatively high density of recreational craft use or otherwise positioned in possible anchoring, spudding, dragging or salvage operation areas, localized institutional controls will be used to minimize potential cap disturbances. To provide appropriate protection from anchoring and related activities in such localized high use areas, the 60 Percent Design submittal will include localized modifications of cap designs and/or localized institutional controls, as appropriate, such as regulated navigation areas (RNAs). Finally, long-term post-construction cap monitoring will target higher use areas of the river, to ensure that the caps continue to be protective (see Sections 8.3.3 and 10.2 below).

During the course of remedial construction, public access to dredging and capping operations will need to be restricted for health and safety reasons. The remedial contractor will be responsible for setting up and patrolling the construction boundaries to secure the equipment operation areas. The remedial contractor will also be responsible for coordinating such temporary waterway closures with U.S. Coast Guard and Port officials to minimize disruption to commerce and recreation. Again, these details will be provided as part of the 60 Percent Design submittal.


84

8.3.2 Construction Limitations

It will be necessary to set permanent restrictions on certain shoreline development activities to ensure the long-term integrity of caps. This may include restrictions on setting utility or cable corridors, construction or removal of fixed-post docks, or any other construction activity that would otherwise disturb the integrity of the cap (Palermo et al. 2002). Within engineered capping areas, use restrictions on nearshore areas may be necessary to maintain the integrity of the cap. For example, construction of boat ramps, retaining walls, or marina development could potentially expose subsurface contamination and compromise the long-term effectiveness of the OUs 2 to 5 remedy. In addition, restrictions on permits for installing utilities would further ensure the overall protectiveness of the OUs 2 to 5 remedy. These restrictions can be implemented through agencies such as WDNR and USACE that have permitting authority over construction activities in the aquatic environment, including programs that require permits to be obtained for dredging and filling.

Various chapters of the Wisconsin Administrative Code (WAC) and the Code of Federal Regulations (CFR) contain technical or administrative requirements to ensure the appropriate management of in-water construction activities, including activities performed in OUs 2 to 5 capping areas. Wisconsin Statutes Chapter 30 requires the issuance of permits for the construction of any structure on the bed of navigable water. Wisconsin's public water regulations are founded on the Public Trust Doctrine, the body of state law that guides how WDNR regulates construction projects in navigable waters (http://dnr.wi.gov/org/water/fhp/waterway). The program has been in place for decades, and requires permits for activities conducted waterward of the ordinary high water mark (OHWM). In May 2005, new rules went into effect to implement statute changes, including expediting permit decisions and other modifications.

Wisconsin Statutes Chapter 30 requires a general permit or an individual permit for future development or construction projects that affect the bed of a navigable waterway. General permits are currently available for limited situations, such as maintenance dredging of a previously dredged area under NR 345 WAC, provided the project meets relatively restrictive eligibility standards. However, substantive future development or construction projects within capped areas will require an individual Chapter 30 permit. Because these projects are not pre-approved designs, a detailed application and review process is required by WDNR. During the permit process, WDNR reviews sediment quality data available within the project area, among other habitat and water quality information, and during this process can apply permit requirements such as construction mitigation measures to continue to ensure the long-term integrity of caps. Individual permits require a comment period, and the general public and stakeholders are notified by a newspaper notice and mailing to interested parties. For example, under the current permitting program, permit applications located within OUs 2 to 5 capping areas would normally be forwarded to USEPA for review and comment. In this context, USEPA would provide additional assurances that proposed future construction activities would not disturb the integrity of the cap, and would assist WDNR in developing appropriate construction mitigation requirements.


85

There are also a range of federal laws that place restrictions on and require permits to be obtained for dredging, filling, or other construction activities in the aquatic environment. These include Section 404 of the Clean Water Act, Title 33 United States Code (U.S.C.) Section 1344, and Sections 9 and 10 of the Rivers and Harbors Act of 1899, 33 U.S.C. 401 and 403, which require federal permitting for any construction that would impact the course, capacity, or condition of navigable waters of the U.S. The 401/404 regulations are typically implemented by the USACE, but may also be implemented by USEPA. Under the Section 404(b)(1) guidelines, 40 CFR 230.10(b), no discharge (i.e., excavation of caps) shall be allowed if it:

• Causes or contributes to violations of water quality standards, pursuant to Section 401 of the Clean Water Act (CWA), after consideration of local dilution and dispersion; or

• Violates any applicable toxic effluent standard or discharge prohibition under Section 307 of the CWA.

Thus, existing federal permits will further ensure that proposed future construction activities will not disturb the integrity of the caps to be constructed in OUs 2 to 5. Between the existing state and federal permit programs, appropriate limitations or construction mitigation requirements are expected to be placed on dredging (e.g., for marina expansion or maintenance), other in-water excavation (e.g., laying cable), and pier construction (e.g., pile placement and extrusion) to ensure the long-term protectiveness of caps constructed as part of the OUs 2 to 5 remedy.

8.3.3 Monitoring and Maintenance

As discussed in the ROD Amendment, long-term monitoring and maintenance will be performed to ensure the physical integrity of the cap and the permanent containment of the underlying sediment contaminants. Monitoring events will occur at a pre-determined schedule, with a combination of frequent monitoring of target areas and periodic monitoring of other capped areas integrated into the overall program. Monitoring may be keyed to relatively large storm events (e.g., 50-year storms), and may also be triggered by other potential occurrences such as water levels below the long-term low-water elevation. Monitoring and maintenance plans will be developed in more detail as part of the 60 Percent Design submittal, and specifically as part of the Operations, Monitoring and Maintenance Plan (OMMP; see Section 10.2).

In order for an institutional control to be required, the control must be necessary to ensure the long-term protection of human health and the environment in light of reasonably anticipated circumstances. For example, given the ownership and regular inspection and maintenance of the De Pere dam by the USACE, no basis currently exists to anticipate that this dam will fail or be removed. There are already a number of compelling reasons for the USACE and others to continue to maintain the De Pere dam, such as providing a lamprey barrier, hydropower capability, water supplies, and recreational use.


86

Even if the De Pere dam were removed in the future and soft sediment deposits in OU 3 upstream of the dam were subject to erosion and transport, the degree to which OU 3 sediment would be transported downstream would depend on the manner in which the dam might be removed. The most likely scenario is that the dam would be removed in a controlled manner; if so, little or no sediment would be transported downstream. However, even if dam removal caused a wholesale movement of sediment, the average PCB concentration of that sediment would likely be low. Under the remedy described in the ROD Amendment, approximately 2.0 million cy of recent (post-dam) soft sediments would be retained in OU 3 after remedy construction, along with a relatively small amount (0.2 million cy) of sediment cap material. These sediments would contain an average PCB concentration of about 0.4 ppm, well below the RAL of 1.0 ppm. As a result, institutional controls to prevent the removal of De Pere dam are unlikely to be necessary to ensure the long-term protection of human health or the environment.

In addition, an institutional control may be deemed to be in place already if another agency has responsibility for conducting an activity or enforcing a prohibition and existing laws or regulations require an environmental review before that program is changed (USEPA 2005). For example, USACE and WDNR currently have responsibilities related to operation and maintenance of dams that are part of the Fox River Navigational System. White Paper 4, issued along with the 2003 ROD, describes Wisconsin regulatory and environmental review requirements associated with proposals for dam removal. White Paper 4 also notes that the USACE has continued to operate, inspect, and maintain the De Pere dam. The institutional control evaluation to be presented in the ICIAP will also consider the implications of the De Pere dam’s listing on the National Register of Historic Places and the considerable amount of infrastructure and recent residential development along OU 3 that depends on continued operation of the dam.

8.3.4 Public Information and Advisories

Fish consumption advisories are informational devices, and an existing WDNR fish consumption advisory is already in place in the Lower Fox River and has been incorporated into the ROD Amendment. That is, due to the elevated concentrations of PCBs detected in fish tissue from the Lower Fox River and Green Bay, the WDNR issued consumption advisories in 1977 and 1987 for fish and waterfowl, respectively. General fish consumption advisories are currently in effect for 7 species of fish located in the Lower Fox River from Little Lake Butte des Morts to the De Pere dam, for 13 species of fish located from the De Pere dam to the mouth of Green Bay, and 11 species of fish located in Green Bay.

In 1984, WDNR initiated its wildlife contaminant monitoring program. Results of the monitoring program indicated that elevated PCB concentrations were present in waterfowl species harvested by sportsmen from Green Bay. WDNR then developed procedures for issuing consumption advisories for waterfowl, and issued a waterfowl consumption advisory for mallard ducks in 1987. The advisory for mallards was issued for mallards taken in the “Lower Fox River from Lake Winnebago at Neenah and Menasha downstream, including Little Lake Butte des Morts, to the northeast city to the northeast city limits of Kaukauna”, and the “Lower Fox River from the De Pere Dam to the River’s mouth at Green Bay,


87

and lower Green Bay south of a line from Point au Sable west to the west shore of Green Bay”. The advisory has remained in place since its issuance. The advisories are issued each year in the annual hunting guide distributed by the WDNR. The federal Food and Drug Administration threshold level for poultry of 3 ppm wet-weight PCBs on a fat basis has been adopted by WDNR for the Lower Fox River.

WDNR’s fish and waterfowl advisory programs are expected to continue in the future, so there is no reason to require an independent advisory program as part of the OU 2-5 remedy. In addition, an independent advisory program could create a risk of contradictory advice to the public. A long-term monitoring plan will be developed for the 60 Percent Design submittal that will describe the water and biological tissue monitoring program in the years following the remedial action, to verify that the remedial action was effective at reducing risk to humans, mammals, birds, fish, and invertebrates (see Section 10.3). One possibility is to require that the data collected in the long-term monitoring plan (LTMP) be forwarded to the WDNR staff responsible for the fish and waterfowl advisory programs.

Section 9 – Construction Schedule and Sequencing

88

9 CONSTRUCTION SCHEDULE AND SEQUENCING This section describes the general operations sequencing and the likely durations of OUs 2 to 5 remedial actions. In general, the contractor will be responsible for determining the construction sequence and schedule with consideration of required sequencing of several key activities. The contractor will also be responsible for coordinating with other site users, including the USACE regarding plans for navigational dredging within the channel, to ensure limited interruptions of normal site use.

The preliminary construction schedule was developed assuming an average in-water construction window of approximately 180 days between May 1 and October 31, consistent with the discussion in Section 4.1.2. Variations in the seasonal opening and closing date of this construction window are expected from year to year. Assuming the contractor works 6 days per week (5 days per week for dredging and 1 day per week for maintenance) during this construction window, an average of approximately 130 working days per year are available to implement the remedial action. During other times of the year, cold weather will prevent water-based work. The schedule presented herein pertains only to construction activities and do not include activities occurring prior to Response Agency approval of initiation of any construction-related tasks.

Finally, in a large, multi-year construction project such as this, it is expected that some activities will take longer than projected and others will take less time than projected. As a result, the actual schedule, especially for years after the first year, is likely to differ from the schedule presented here.

9.1 Operations Sequencing

The sequence of operations must take into consideration the thickness of dredge cuts, extent of capping, and duration of work, type of equipment, and methods for verifying that the remedial actions achieved the required objectives. The sequence of operations will likely include the following general considerations:

• Dredging will be performed in a general upstream to downstream sequence throughout the remedial action

• Dredge areas will be divided into dredge certification units (DCUs) for verifying compliance with required dredge elevations, sized based on the contractor’s operational considerations and other logistical considerations. Each DCU will be comprised of several smaller dredge management units (DMUs) for controlling the work.

• During the dredging operations, the contractor will perform daily bathymetric progress surveys to verify accuracy of their work. These daily progress surveys will be used to help adjust the contractor’s operations to comply with the project specifications.

• Once dredging within a DCU (or DMU) is completed, as reported by the contractor, a third-party independent bathymetric survey will be conducted to confirm that the area has been adequately completed. If the surveying indicates that dredging has not been completed as required, additional dredging and subsequent surveying will be performed until compliance can be verified.


89

Details of the construction verification plan and compliance criteria will be included in the CQAPP.

• Once the area has been accepted as complete, post-construction sampling will be conducted in accordance with the CQAPP.

• Capping and sand cover (including residual sand cover) will be performed in a general upstream to downstream sequence following the completion of dredge activities. A suitable offset distance will be maintained between dredging and capping operations to reduce the potential for recontamination of placed caps and sand covers.

• Cap and sand cover areas will be divided into cap certification units (CCUs) and sand cover certification units (SCCUs) for assessing compliance. These certification units may be further subdivided into cap management units (CMUs) and sand cover management units (SCMUs) for controlling the work, as described in the CQAPP.

• During the cap and sand cover placement operations, the contractor will perform daily bathymetric progress surveys to verify accuracy of their work. These daily progress surveys will be used to help adjust the contractor’s operations to comply with the project specifications.

• Once cap or sand cover placement is complete within a given area, as reported by the contractor, verification sampling/monitoring will be performed to assess compliance with the specifications. If the verification monitoring indicates that cap or sand cover placement has not been completed as required, additional material placement and subsequent surveying will be performed until compliance can be verified. Details of the construction verification plan and compliance criteria will be presented in the CQAPP.

9.2 Construction Schedule

The estimated total duration for completion of required remedial construction in OU’s 2 5 is 10 calendar years, excluding any time, if necessary, for re-dredging due to post-dredging residuals and to meet project objectives (see Figure 9-1). Preliminary estimates indicate re dredging activities could extend the project duration by an additional 2 calendar years. In addition, the 10 calendar year time-period excludes an initial year of preparatory activities. Consistent with the 2007 Unilateral Administrative Order for OUs 2 to 5 issued by the Response Agencies, “Year 1” is 2008 when final design and procurement of certain equipment will be performed. “Year 2” is when such equipment will be installed and also the first year of in-water actions, including installation of sheetpile at the former Shell Property, mechanical dredging and capping in OU 2 and upper OU 3, and mechanical dredging to provide access channels required to perform sheetpile installation. Full-scale hydraulic dredging will begin in 2010 (“Year 3”).

As discussed in this 30 Percent Design Report, the OUs 2 to 5 remedial action is expect to utilize a 12-inch hydraulic dredge as the primary means of sediment removal in OU 3 through 5. Additional equipment, including small hydraulic dredge(s) may be utilized for accessing isolated or small deposits to maximize operational efficiency, although only one hydraulic dredge will likely operate at any given time. In addition, a mechanical removal operation will likely be used to remove OU 2 sediments and a portion of upper OU 3 sediments, as well as near shore deposits throughout OUs 3 through 5 that cannot


90

be efficiently accessed with the hydraulic dredge. Dredged sediment will be dewatered at the Shell Property and/or other equivalent staging facility. Dewatered sediment will then be hauled by truck to the landfill. Nearshore surficial sediments within the Shell Property sheetpile footprint, including sediments that are potentially subject to TSCA disposal requirements, will be removed during Year 3, after the overlying non-TSCA sediments are dredged. In addition, some sediment potentially subject to TSCA disposal requirements that are present near the offloading facility may be dredged during Year 2 or early in Year 3 as part of the deepening of the access berth at the offloading facility. Capping will be completed concurrently with dredging sequenced to minimize the potential for recontamination of cap surfaces from adjacent dredging.

Figure 9-1 presents a Gantt chart illustrating the schedule for the OUs 2 to 5 remedial action. As discussed in Section 4.1.2, a 6-day work week has been assumed for all in-water construction work, which will include 5 days per week for dredging and 1 day per week for maintenance. As more specifically described in other sections of this 30 Percent Design Report and in Section 5 of the BODR, the OUs 2 to 5 remedial action will include the following general task schedule:

• Staging Facility Preparation. The former Shell Property will be developed as a staging area. Preparation for the work will include the design and procurement of the dewatering and water treatment system discussed in Section 5. On-site preparation will include grading the site and installing utilities and foundations, as well as constructing waterfront structures (e.g., bulkhead line berthing area). As part of the waterfront preparation, remedial action will be performed inside the bulkhead line. Following site grading and foundation installation, the mechanical dewatering and water treatment facility will be constructed on the property to handle the hydraulically dredged material. A separate passive dewatering facility (e.g., stockpile and amendment area with sumps) may be constructed to handle any mechanically dredged material. Separate stockpile areas will also be established for the dewatered dredge material, sand material removed, and capping materials. To facilitate water access to the site, the remedy includes installation of a sheetpile bulkhead wall adjacent to the former Shell Property to accommodate floating equipment (e.g., dredges, barges, import material delivery vessels, etc.). Mechanical dredging will be required to provide access for the installation of sheetpile wall (see bleow). Staging area construction is discussed in Section 3. Roughly 20 months will be required to design, procure, and install the dewatering and water treatment equipment and prepare the Shell Property for the project. A separate staging area in OU 2 or upper OU 3 may be necessary to support mechanical dredging operations in those areas. Site preparation of this area would occur in Year 1 or Year 2, prior to the initiation of mechanical dredging in OU 2.

• Mobilization of Dredging Equipment. The hydraulic and mechanical dredge equipment as well as support equipment will be mobilized to the site. Cap placement equipment will also be mobilized. Mobilization of dredging equipment is anticipated to require 12 weeks, including fabrication of the hydraulic dredge slurry pipeline. At the end of each construction season a period of approximately 3 to 4 weeks will be required to demobilize and winterize equipment. A similar amount of time will be required at the beginning of each construction season to re-mobilize equipment and set up operations. The schedule assumes that the time for both operations will occur outside of the 180-day construction period.

• Sediment Removal at Staging Area. Some sediment (likely including approximately 13,500 cy of non-TSCA and 1,500 cy of TSCA sediment) that is present near the staging/offloading facility


91

will be dredged during Year 2 as necessary to facilitate sheet pile wall installation. Dredging of TSCA sediments (and overlying non-TSCA sediments) within the sheetpile enclosure will occur in Year 3, following installation of the sheetpile wall, which is expected to consume the majority of the Year 2 construction season. Dredging in front of the sheetpile wall also may be performed ahead of full scale dredging in OU 4 as part of the deepening of the access berth at the offloading facility. It should be noted that planned dredging depths inside and in front of the sheetpile wall may be refined during subsequent design pending further consideration of future property use and berthing requirements. This material will likely be dredged mechanically since the mechanical dewatering system will not be fully fabricated.

• Removal of Sediments Subject to TSCA Disposal Requirements. A hydraulic dredge will be used to remove the majority of remaining sediments subject to TSCA disposal requirements. The sediment will be pumped to the Shell Property, dewatered, and hauled as necessary to the TSCA landfill. Subsequent dredging of subsurface sediments subject to TSCA disposal requirements will occur as the overlying non-TSCA sediment is removed. Hydraulic dredging of the sediments subject to TSCA disposal requirements is anticipated to occur at a rate of approximately 3,200 cy per day.

• Non-TSCA Mechanical Sediment Removal—OU 2/OU 3. Sediments within OU 2 (approximately 24,5,500 cy) and a small portion of the OU 3 (approximately 20,000 cy) will be dredged using mechanical equipment. Based on the dredge plans developed for the 30 Percent Design and the weather dependent in-water construction season, it is expected that the contractor would likely employ two 12-hour shifts per day, 6 days per week in order to ensure the OU 2 and upper OU 3 work is completed in a single construction season. Under the proposed operating schedule, and considering the difficult shoreline dredging operations and other logistical constraints, an average dredge production rate of approximately 675 to 700 cy per 24-hour day is expected, equating to a total dredging duration of roughly 14 weeks (approximately 3 months). This work, along with the planned capping in OU 2 (estimated 7 to 8 week duration), represents most of the available Year 2 (2009) construction window (see below). Sand cover placement in the upper portion of OU 3 may also begin in Year 2, to the extent time is available to do so.

• Non-TSCA Hydraulic Sediment Removal—OU 3. One hydraulic dredge will be used to remove the non-TSCA sediment in OU 3 that was not removed with mechanical equipment (anticipated hydraulic removal total of approximately 201,000 cy). Hydraulic dredging of the sediment is anticipated to occur at a rate of approximately 3,200 cy per day. The dredge will remove the OU 3 non-TSCA sediment in approximately 13 weeks.

• Non-TSCA Hydraulic Sediment Removal—OU 4/OU 5. One hydraulic dredge will be used to remove non-TSCA sediment in OUs 4 and 5 (approximately 3.4 Mcy) generally working from upstream to downstream. Hydraulic dredging of the sediment is anticipated to occur at a rate of approximately 3,200 cy per day. The dredge will remove the OUs 4 and 5 non-TSCA sediment in approximately 211 weeks (8 years). Sediment dewatering (including desanding and beneficial use), water treatment, and off-site disposal activities will be performed concurrent with dredging operations.

• Post-Dredge Residual Management. Dredge residual management will likely be required to meet the overall SWAC goals specified in the RODs, and may include MNR, placement of a 6-inch layer of sand on the dredge surface, or re-dredging to attempt to remove the settled materials. Based on an initial evaluation of post-dredge residuals, the following residuals management actions have been assumed for this 30 Percent Design, although actual post-dredge residuals management decisions will be made based on post-dredge monitoring conducted at the time of construction (see Section 4.6.3 and Section 10.1):


92

o Re-dredging of residual contamination: Re-dredging of post-dredge residuals in the form of a nominal 1-foot cut was assumed to occur over approximately 25 percent (or approximately 155 acres) of the OUs 2 to5 dredge area. Only one re-dredging effort (i.e., not multiple passes, which typically results in diminishing returns with each successive pass) has been assumed for this estimate. Therefore a total of approximately 250,000 cy of re-dredging was assumed at a rate of approximately 1,060 cy per day for a total duration of approximately 47 weeks of re-dredging. This re-dredging will likely be performed at various times during the project duration and therefore is shown to extend throughout, and beyond, the period anticipated for required dredging in OUs 3 and 4. Note that the estimated re-dredge production rate is approximately 30 percent that of the full-scale rate, with commensurately higher unit costs.

o Residuals sand cover: Approximately 50 percent (approximately 210 acres) of the OUs 2 to 5 dredge area is anticipated to contain post-dredge residuals exceeding the RAL, and will be suitable for management with a 6-inch sand cover or engineered cap, consistent with the ROD Amendment. Cover and cap material will likely be applied using mechanical equipment. Placement of residual sand covers is anticipated to occur at a rate of 900 cy per 12-hour day in OUs 2 and 3 and at a rate of 2,160 cy per 24-hour day in OUs 4 and 5. The total duration of residual cover placement will be approximately 8 weeks for OU 3 and 24 weeks for OUs 4 and 5.

• In situ Capping of Shoreline Areas. In situ capping of sediments exceeding the 1 ppm RAL may be performed along shoreline areas where dredging could not be completed without adversely impacting the stability of the existing slopes. Capping would occur at a rate of approximately 675 cy per 12-hour day per crew in OU 2/3 and 1,620 cy per 24 hour day per crew in OU 4/OU5. Duration of approximately 26 weeks will be required for the entire shoreline cap placement. These activities will occur near the completion of respective dredging activities.

• Cap Placement. Caps will consist of a lower sand section with either gravel or quarry spall armoring (depending on the location as described in Section 6). Placement of all cap materials is expected to occur at a rate of approximately 900 cy per 12-hour day in OUs 2 and 3 and at a rate of 2,160 cy per 24-hour day in OUs 4 and 5. Cap construction in OU 2, 3 and 4/5 is expected to require 55 weeks, and 89 weeks, respectively.

• Demobilization and Site Restoration. Demobilization and site restoration will involve removing equipment from the staging and work areas and restoring the site to its original condition before construction of the staging area commenced. However, it is expected that uncontaminated paved areas and the bulkhead constructed at the Shell Property staging facility will be left in place.

9.3 Contracting Strategy

The two primary contracting approaches currently under consideration to implement the OUs 2 to 5 remedial action are:1) design-bid-build; and 2) design-build. Each of these contracting strategies is summarized below.

• Design-Bid-Build. Under this contracting approach, the remedial design would be developed to a level that would allow for competitive bidding from multiple contractors on a relatively well-defined scope of work. However, specifications for many components of the work would be developed as “performance-based” to allow bidders flexibility in the means and methods for completing the


93

required work, in order to facilitate the development of cost-effective bids. For instance, contract documents would provide the required limits and extent of dredging, but would not specify the equipment for removal. This contracting method tends to be most favorable when significant modifications to the design are not anticipated after contract award, and when constructability issues are expected to involve only typical field challenges. Modifications to the design following award of the contract would require amendments or change orders.

• Design-Build. Under this contracting approach, contractor procurement would be initiated during the design process and the selected contractor would contribute to development of the design, potentially beginning with the 60 Percent Design submittal. The preliminary (e.g., 30 Percent level) design documents used for evaluation and selection of the design-build team would necessarily be more performance-based than the design-bid-build option outlined above, allowing bidders to identify the most-cost effective means and methods to meet OUs 2 to 5 remedy requirements. This contracting approach tends to be more favorable when the schedule does not allow time to accommodate the design-bid-build approach and/or when significant uncertainty exists in the design and construction. It may also provide for significant flexibility and creativity during the contracting phase to consider more innovative approaches.

It may be appropriate to divide the OUs 2 to 5 remedial action work into more than one contract to facilitate efficient and cost effective implementation. For instance, initial site preparation activities (e.g., sheet pile installation) at the former Shell Property in OU 4 could potentially be separated as a distinct contract from the remainder of the remedial action work. This process may result in using multiple contract methods to complete the overall project.

Regardless of the contracting strategy, Response Agency review and approval would be required at the Preliminary (30 Percent), Intermediate (60 Percent), Pre-final (90 Percent) and Final (100 Percent) Design phases, consistent with the RD Work Plan.

Section 10 – Monitoring, Maintenance, and Adaptive Management

94

10 MONITORING, MAINTENANCE, AND ADAPTIVE MANAGEMENT Detailed construction (short-term) and post-construction (long-term) monitoring, maintenance and adaptive management plans will be prepared as part of the Intermediate (60 Percent) Design documents, which are currently scheduled to be submitted to the Response Agencies in May 2008, depending in part on the timing of the approval of this 30 Percent Design report. Construction monitoring activities, including water quality monitoring and sediment confirmation sampling, will be specified in the CQAPP, to be provided as part of the 60 Percent Design. Long-term cap performance monitoring will be specified in the OMMP, while long-term plans for monitoring of water and biota in the Lower Fox River and Green Bay will be specified in the LTMP, both of which will be included in the 60 Percent Design. The Adaptive Management Plan will set forth a process and procedures to modify the cleanup and monitoring plan as appropriate in response to new information and experience during initial remediation activities in OUs 2 to 5. The basic components and framework of these plans are described in this section.

10.1 Construction Monitoring

Water column monitoring during construction activities, including dredging, capping, and dredged material disposal activities will be described in the CQAPP. Air monitoring activities will also be described. These monitoring activities will be specified to ensure construction BMPs are being properly implemented and to prevent construction activities from unduly impacting the Lower Fox River or Green Bay.

One of the primary CQAPP elements is the design of a sediment confirmation sampling program to confirm the attainment of the RAL. If the RAL is not met within a remedial action certification area, a range of response actions may be appropriate. The section below presents a summary of plans developed to verify the performance of dredging, capping, and sand cover placement within OUs 2 to 5 relative to the RAL. The verification plans and performance criteria presented below were developed consistent with the ROD Amendment and build on similar plans and criteria that have been utilized for the Phase 1 and OU 1 projects, and address the following issues relevant to verification sampling and contingency response actions:

Post-Dredge Verification

• Verification that dredging has been completed to the limits and extents required by the design;

• Verification that dredging has achieved the RAL for PCBs in dredge only areas;

• Post-dredge PCB concentrations for which placement of a minimum 6-inch sand cover would be a suitable management response action, consistent with the provisions of the ROD Amendment, to permanently address post-dredge residuals exceeding the RAL, without the need for further engineering evaluations or post-construction monitoring; and


95

• Description of specific elements of the more detailed post-dredge engineering evaluations that would be performed in ‘real time’ to inform expedited contingency response decisions (e.g., redredging, engineered capping, or sand cover placement) within a given certification area.

Sand Cover and Engineered Cap Placement Verification

• Verification that sand covers (placed either as a primary remedial action or as a secondary post-dredge residuals management response) have been placed to the limits and grades required by the design; and

• Verification that engineered caps have been placed to the limits and grades required by the design in accordance with the specifications.

Each of these sampling compliance criteria and contingency response action issues is addressed in the sections below.

10.1.1 Post-Dredge Verification

The ROD Amendment adopted sediment removal (i.e., dredge-only) as the primary remedial approach for addressing sediment exceeding the RAL in OUs 2 to 5. This section discusses how post-dredge data for dredge-only areas will be evaluated to determine the nature and extent of and management alternatives for dealing with dredging residuals, both generated and undisturbed, using guidance provided in the ROD Amendment.

Dredge Certification Unit Areas. Consistent with the approach currently being utilized on the Phase 1 Project, dredge areas within OUs 2 to 5 will be subdivided based on location and operational considerations into DMUs. Several DMUs will be combined into DCUs for verifying compliance with performance criteria described in the ROD Amendment.

The appropriate size (acreage) range of a given DCU will be based on consideration of the following:

• The spatial scale of concern for exposure to sediment PCBs, as described in the 2003 ROD and ROD Amendment;

• The practical frequency of agency-approved certification decisions (i.e., allowing for appropriate data review and discussion periods); and

• The spatial scale of variability of PCB concentrations within the Lower Fox River, as determined from detailed geostatistical analyses of sediment PCB data (see discussion below).

As discussed in the 2003 ROD (WDNR and USEPA 2003), the spatial scale of concern for exposure to sediment PCBs is tied to the SWAC measured throughout the OU. The total areas of OU 3 and OU 4 are approximately 950 and 1,175 acres, respectively. In developing the RAL, USEPA and WDNR used core


96

sampling data collected on a spatial density of approximately 1 core per 5 acres (roughly 400 cores collected over the 2,000-acre OU 3 & 4 area) to determine the relationship between the RAL and the 0.25 ppm SWAC criterion. Thus, PCB exposure within a 5-acre RAL decision unit would be consistent with the spatial scale of concern as originally defined in the ROD.

Consistent with decision time-frames utilized for the Phase 1 Project and as discussed in the BODR, the optimal frequency of agency-approved certification decisions during remedial action (i.e., allowing for appropriate data review and discussion periods) is approximately every two weeks. Assuming a total OUs 2 to 5 dredge area of approximately 700 acres, a seasonal dredging period of 26 weeks/year, and a dredging duration of approximately 9 years (Shaw and Anchor 2006), on average a 6-acre dredge unit would need to be certified every two weeks during remedial action. It is important to note that the size of a dredge unit for certification will vary depending on numerous physical setting and operational considerations for a particular area including the water depth, dredge equipment, depth of contamination, and the nature and extent of debris. Grouping of DMUs and DCUs will also consider geographical similarities (e.g., channels, near shore, etc).

The spatial scale of variability of PCB concentrations within the Lower Fox River to minimize errors associated with sample interpolation between points was evaluated in the RD Work Plan based on detailed geostatistical analyses of the available sediment PCB data (Shaw and Anchor 2004). Spatial correlation analysis of surface and subsurface PCB concentrations in OU 3 and OU 4 revealed an asymmetrical correlation structure oriented along the axis of the river, with longitudinal and transverse correlation distances of 1,640 and 660 feet, respectively. Based on these analyses, the average spatial scale of variability of PCB concentrations in OUs 2 to 5 is approximately 25 acres. Sample grids of approximately one-half the distance of the correlation scale, or approximately one sample every 6 acres, were used during RD sampling to ensure precise delineation of PCB deposits in OUs 2 to 5 (smaller grids were established along the 1 ppm boundary and in areas of high concentration).

Based on these considerations, DCUs within the Lower Fox River will typically have a size of approximately 6 acres. However, for particular DCUs, site-specific factors may dictate otherwise (i.e., based on the contractor’s operational plans, a DCU may only encompass an area of 1 to 2 acres). Separate from Agency-approved certification decisions, the size range of a given DMU will be developed by the contractor based on operational factors associated with dredging in order to facilitate timely and cost-effective completion of activities within individual work units. The specific size and grouping of DMUs will be determined considering the contractor’s operational plans, also considering geographic similarities within and between DMUs.

A preliminary example layout of DCUs, DMUs, and verification sampling locations for a representative 30-acre dredge area within OU 4 is provided in Figure 10-1. As detailed in the accompanying 30 Percent Design plan set, the average anticipated cut thickness including overdepth allowance in this area is approximately 3.5 feet. At an average production rate of approximately 3,200 cy per day using a single


97

12-inch hydraulic cutterhead dredge (Shaw and Anchor 2006), a two-week production period in this area equates to an average DCU size of 6.8 acres, which was rounded down to 6 acres to achieve five equally-sized DCUs. The preliminary sampling plan presented in Figure 10-1 assumed a total of 15 DMUs, or approximately one DMU per 2 acres. Further refinement of the spatial scale of post-dredge verification sampling to be used during initial OUs 2 to 5 remedial actions, along appropriate adjustments to the plan as remedial action proceeds (e.g., relaxing sampling density over time as uncertainties diminish) is planned for the 60 Percent Design submittal, incorporating RD Team and A/OT input.

Post-Dredge Monitoring and Sampling. After bathymetric monitoring confirms that the target dredge elevation, within an appropriate tolerance, has been achieved in at least 95 percent of a given DMU, sediment grab and/or core samples will be collected to characterize residual sediments and to inform the selection of appropriate management actions if residual PCB concentrations above the RAL are identified.

Consistent with the SAP Addendum for the Phase 1 Project (Shaw et al. 2007), post-dredge sampling will initially be conducted at approximately five locations within each DMU. In general, at least one of the post-dredge sampling cores in each DMU will be co-located with the RD sampling locations (i.e., “primary” stations), if such an RD sampling location exists within that DMU. The remaining four post-dredge samples in each DMU (i.e., “secondary” stations) will be distributed along a general grid pattern (e.g., 5 points on a die) to the extent possible, in order to provide an appropriately stratified and unbiased sampling design. Note that final DMUs and DCUs will not be developed until the time of construction based on the contractor’s planned operations.

Post-Dredge Evaluation. Sediment cores (potentially supplemented with surface grab samples) will be collected from each post-dredge sampling location within a given DMU. Sediment cores/grab samples collected at each sampling location will be sectioned into 6-inch intervals, and corresponding depth intervals (e.g., 0-6 inch and 6-12 inch segments) will be composited within each DMU (i.e., one primary plus four surrounding secondary stations). In addition, aliquots of each individual surface and subsurface sample (primary and secondary stations) used to create the composite will be archived for potential future testing. Initially, the composited surface (0 to 6 inch) samples representative of a given DMU will be analyzed for total solids, dry density, and total PCBs. If the average surface (0 to 6-inch) composite sediment concentration within a given DCU exceeds 1.0 ppm PCBs, deeper composite sample intervals (e.g., 6-12 inches) comprised of the primary and secondary cores will be submitted for PCB analysis until the vertical extent of composited sediments exceeding 1.0 ppm is determined.

Consistent with the ROD Amendment, the post-dredge sampling data will be evaluated initially against the following criteria:

• If the average surface (0 to 6 inch) composite sediment PCB concentration in a given DCU is less than 1.0 ppm (predicted using preliminary geostatistical analyses to occur in approximately 25 percent of the OUs 2 to 5 post-dredge DCUs; see Section 4), no post-dredge management (including sand cover) will be needed.


98

• If the average surface (0 to 6 inch) composite sediment PCB concentration in a given DCU is greater than 1.0 ppm, an initial screening using the following criteria will be performed to determine if a sand cover may be suitable for residuals management, consistent with the ROD Amendment:

• Arithmetic mean of all 0 to 6-inch DMU composite samples within a DCU ≤ 10 ppm; and

• Arithmetic mean of all DMU composite samples within a DCU for layers below the upper 0- to 6-inch interval ≤ 1.0 ppm.

Based on preliminary geostatistical analyses, both of these post-dredge sand cover suitability criteria are anticipated to be met in approximately 50 percent of the OUs 2 to 5 post-dredge DCUs; see Section 4).

• If the average surface (0 to 6 inch) composite sediment PCB concentration in a given DCU is greater than 10 ppm, and/or if the average subsurface (layers below 6 inches) composite sediment PCB concentration in a given DCU is greater than 1.0 ppm, the entities performing the remedy will work collaboratively with the A/OT to determine the appropriate extent of areas requiring additional response. Based on preliminary geostatistical evaluations, this condition is anticipated in the remaining approximately 25 percent of the OUs 2 to 5 post-dredge DCUs (see above). An engineering evaluation will be conducted to determine the most appropriate residual management action(s). The engineering evaluation will consider:

• Calculation of the percent PCB mass removed to date within each DCU, along with updated mass per unit area calculations that provide increased statistical confidence relative to pre-dredge estimates of PCB mass (see additional detail in Attachment A-3 of Appendix A, of this report);

• Practicability, technical feasibility, cost-effectiveness, and implementability factors (e.g., layer thickness, PCB concentration, and density); and

• Consideration of the residual management (if any) in adjacent DCUs.

For areas requiring an engineering evaluation, updated mass per unit area calculations (as well as percent PCB mass removal estimates) will be developed for each DCU on an area-weighted average basis.

In order to expedite response actions, it may be appropriate to focus supplemental response actions in a DCU that requires them on individual DMUs (or portions thereof). To inform the engineering evaluation, individual 5-point composite samples representing each DMU within the DCU may be analyzed to determine the spatial and vertical extent of sediment exceeding the 1.0 ppm RAL. In the event that PCB concentrations exceeding 1.0 ppm are identified in subsurface (below 6 inches) composite samples on a recurring basis, the entities conducting the remedial action and the Response Agencies will discuss additional evaluation of individual archived aliquot samples from the primary and secondary sample locations, as appropriate.

Potential response and adaptive management actions (e.g., balancing the need for real-time decisions, maintaining productivity, and statistical rigor) will be refined during development of the 60 Percent Design submittal, incorporating A/OT input. The following section further describes the process by which post-dredge engineering evaluations would be conducted and the need for an expedited decision-making process.


99

Post-Dredge Expedited Engineering Evaluations. For those DCUs that exceed the initial screening criteria summarized above, a more detailed post-dredge engineering evaluation will be performed on a case-by-case basis to identify post-dredge contingency response decisions appropriate within all or a portion of a given DCU. Possible management actions include placement of six or more inches of sand cover; placement of an engineered cap; additional dredging (followed by resampling); and additional dredging immediately followed by placement of a sand cover or engineered cap. Because such contingency response decisions will need to be made in ‘real time’ to facilitate certification of DCUs on an approximate 2-week frequency (and likely more rapidly in certain remedial action scenarios), such evaluations will need to be performed on an expedited basis. The section below describes key elements of the expedited post-dredge engineering evaluation.

The initial step of the expedited engineering evaluation will be analysis (likely on a quick turn-around basis) of individual aliquot samples from DCU and/or DMU composites, in order to provide a detailed characterization of the spatial extent and variability of PCB concentrations within a given DCU. These data will then be compiled into tabular and graphical summaries of layer thicknesses, PCB concentrations, densities, and PCB mass distributions within the DCU. Available information concerning remedial actions in adjacent DCUs, shoreline stability, presence of hardpan or bedrock immediately underlying the residual sediment layers, and the presence of debris and infrastructure within the DCU will also be compiled.

The compiled data will be used to develop cost-effectiveness and implementability factors specific to that DCU, including evaluations of sediment dredgeability and consistency with adjacent remedial actions. If only a thin layer of sediment is present that contains relatively low PCB concentrations, there may be little technical rationale supporting a re-dredging response, as discussed in the ROD Amendment. On the other hand, higher PCB concentrations and/or thicker layers may indicate a need to redredge. The volume of sediment with PCB concentrations below the RAL that would likely be removed if dredging were to be selected as the response action is also an important element of the evaluation. The design of prospective engineered caps will be consistent with that used for areas remediated by capping or by partial dredging followed by capping, as described in the ROD Amendment, BODR and in this 30 Percent Design report. Operational efficiencies will be developed consistent with standard accepted dredging and cover/capping practices. The overall cost effectiveness of each response action under consideration will include capital and operation, maintenance, and monitoring (OM&M) costs. Items common to all response actions or indirect costs would not normally be included in the cost effectiveness evaluation.

One of the more important elements of the expedited engineering evaluation is the assessment of overall PCB mass per unit area and mass removal achieved within the DCU. As discussed in the BODR (Shaw et al. 2006) and in Patmont and Palermo (2007), the limitations of modern dredging equipment in removing contaminated sediments have recently been documented. Post-dredge sediment residuals, are now understood as inevitable due to the inability of existing dredging equipment to remove all


100

contaminated sediment within a dredge prism. In dredging-only remedies, post-dredge sediment residuals can make cleanup goals difficult to achieve.

Typically, modern dredging equipment can consistently achieve 91 to 98 percent removal of the targeted “neatline” PCB mass within the last production cut (Patmont and Palermo 2007); greater removal has proven to be difficult and/or technically infeasible. Moreover, performance requirements for multiple passes of the dredge to achieve a low residual concentration have often been inefficient and costly, with little or no discernable benefit in the form of reduced generated residual concentrations, thicknesses or risk. The relatively low density characteristic of generated residuals, combined with the observation that dredging is less efficient for such low density materials, suggests that the cleanup pass option has limitations when applied to generated residual management. In consideration of the documented limitations of modern dredging equipment, metrics based on remaining mass per unit area will be used as one consideration in the engineering evaluation, along with information on residual thickness, concentration, density and ‘nearest neighbor’ response actions. Where the residual PCB mass per unit area is relatively low within a DCU (i.e., similar to the mass per unit area identified for cover actions in nearby areas), and/or when at least 90 percent of the original PCB mass has been removed from a given DCU, the engineering evaluation will give additional consideration to management options other than redredging. An evaluation process similar to that used for the Phase 1 Project is being considered for application to the larger OUs 2 to 5 remedial action.

The determination of post-dredge contingency response decisions appropriate within all or a portion of a given DCU within OUs 2 to 5 will be performed as a collaborative undertaking between the entities conducting the remedial action and the Response Agencies. As discussed above, such contingency response decisions will need to be made on an expedited basis. Additional engineering evaluations will be performed during subsequent stages of RD, including standardized actions for post-dredge management of undisturbed and generated residuals, considering feasibility and cost-effectiveness. Further design details will be developed as part of the Draft CQAPP and Draft Adaptive Management Plan to be provided as part of the 60 Percent Design submittal.

10.1.2 Engineered Cap and Sand Cover Placement Verification

Engineered caps and sand covers will be placed in areas where the criteria set forth in the ROD Amendment for employing such alternatives are met. This section discusses the methodology that will be used to confirm that the constructed engineered caps and sand covers meet the minimum design criteria provided in Chapter XI.(A)2 of the ROD Amendment.

The BODR (Shaw and Anchor 2006) included a preliminary delineation of cap areas on a Thiessen polygon basis using design criteria (PCB concentrations, cap thickness, armor stone size, etc.) available at that time. Subsequent engineering evaluations performed as part of this 30 Percent Design included refining that delineation based on the extent of contaminated sediment suitable for engineered capping as predicted by the geostatistical model with consideration of refined engineering analyses relative to the


101

design criteria. The total OUs 2 to 5 cap area as refined in this 30 Percent Design does not exceed the total area identified in the BODR.

Consistent with the approach currently being utilized for the Phase 1 and OU 1 projects, placement of the minimum specified thicknesses of sand cover and each layer of engineered caps will be verified through a combination of accurate material placement records (to compute the delivered thickness), physical measurements, and bathymetric surveying. Note that the specific method and measurement technique utilized to verify cap and cover placement will be dependent on the thickness and material type of the layer being verified (e.g., ranging from a 6-inch sand layer to an 18-inch quarry spall layer) and on the local site conditions (e.g., water depth, size of area, etc).

A combination of the various measurement techniques will provide the accuracy and spatial coverage necessary to confirm adequate placement to the required limits and extents. In the case of verification using the computed delivered thickness in combination with a set of discrete physical measurements, a statistical analysis of the combined data set from both measurement techniques will be performed to ensure with statistical confidence that the minimum thickness has been achieved within a given area.

While measurement of the amount of material placed will provide the primary verification metric for engineered caps and sand covers, focused surface sediment sampling and PCB analyses will also be performed during installation of Cap B (minimum 16-inch thickness) and Cap C (minimum 33-inch thickness) caps to verify that clean isolation layers have been placed. While PCB analysis is not a necessary component of verification for the Cap A (minimum 13-inch thickness), since this design is based on a presumed mixing of the sand with the underlying sediment, a limited program of PCB analysis of core segments in the Cap A areas will be conducted to confirm RD assumptions on cap placement behavior. This program is anticipated to be limited in extent and duration, and will focus on sampling of representative areas. Details of PCB sampling and analysis during cap construction will be developed as part of the 60 Percent Design submittal.

Engineered Cap and Sand Cover Pilot Areas. Engineered cap and sand cover placement activities, respectively, will be initiated within discrete “pilot” test areas covering approximately 1 acre. The intent of these pilot test areas will be for the Engineer and/or A/OT to observe the contractor’s method(s) of material placement for compliance with the performance criteria, to assess the contractor’s proposed quality control method, and confirm required thickness and extent of materials. The contractor may need to adjust its production methods based on the results of the pilot tests.

Engineered Cap and Sand Cover Certification Units. Following completion of placement within the initial pilot test areas for engineered caps and sand covers, full scale production will begin. Similar to the subdivision of dredge areas into dredge certification units, sand cover and engineered cap areas will be divided into SCCUs and CCUs for assessing compliance. SCCUs and CCUs will be further subdivided


102

into SCMUs and cap management units CMUs, where several management units will be combined into a certification unit for assessing compliance.

The delineation of management and certification units will be based in part on the relative size and position of contiguous cover and cap areas (see accompanying 30 Percent Design plans), as well as on the typical production rate for sand cover and engineered cap placement. Furthermore, delineation of management and certification units will also consider the various cap design thicknesses such that CCUs will likely consist of areas with consistent cap thicknesses (i.e., separate CCUs for each type of engineered cap: Cap A [min 13-inch], Cap B [min16-inch], Cap C [min 33-inch], and shoreline caps).

Separate from Agency-approved certification decisions, the size range of a given SCMU or CMU will be developed by the contractor based on operational factors associated with material placement in order to facilitate timely and cost-effective completion of activities within individual work units. The specific size of SCMUs and CMUs will be determined considering the contractor’s operational plans.

Verification of Placed Thickness. As discussed above, several techniques may be used to verify the thickness of cap and cover materials placed within OUs 2 to 5 depending on the material type and thickness as well as the local site conditions. Verification techniques for the various caps and sand covers are discussed below.

Sand Cover. The thickness of sand covers placed as a stand-alone remedial action or for post-dredge residuals management will be verified through a combination of accurate material placement records and physical sampling as summarized below:

• The volume of material placed within each SCMU will be accurately tracked such that the thickness of sand placed within that SCMU can be computed.

• Following completion of sand cover placement within a given SCMU, the thickness of placed sand will be physically measured in samples collected at randomly-selected locations using a hydraulic Power Grab, piston core, “catch pan”, or other appropriate equipment capable of collecting a relatively undisturbed sample of the sand cover. Alternatively, the thickness of sand cover could be measured in-place by divers using survey rods placed prior to sand cover placement. Selection of the most appropriate measurement technique will be determined prior to or during the initial stages of construction and may be varied as construction proceeds depending on local site conditions. Some random thickness measurement locations will also be placed on the outer limits of (and slightly beyond) the SCMUs to verify adequate thickness across the entire design footprint.

During the initial stages of full-scale sand cover placement, a regression-based statistical approach will be used to develop a relationship between the thickness determined through material placement records and the mean measured thickness of the sand cover placed, consistent with the approach currently being utilized for the Phase 1 and OU 1 projects. As each additional SCMU is sampled, the correlation between the placed and measured thicknesses will be refined based on the expanding set of available data; the


103

refined relationship will be used to estimate SCMU-specific mean thicknesses. For each SCCU, the lower tolerance limit of all sand cover thickness measurements will be determined.

Compliance with the minimum sand cover thickness for each SCCU will be based on verification of the minimum specified cover thicknesses (6 inches including any mixing with underlying sediment) over at least 80 percent of the area (of a given SCCU) with 90 percent confidence or greater. To meet these statistical criteria, it is anticipated that the minimum thickness criterion will typically need to be exceeded in approximately 95 percent or more of the physical samples of the sand cover (a similar result is anticipated in capping areas; see below).

The mixed layer will be defined as being comprised of both existing sediment and placed sand where the sand comprises at least half of the matrix, as determined through visual observations and/or physical characterization in the field. It is expected that the A/OT and Respondents’ representatives in the field will work closely to develop a field screening procedure to measure the mixed layer thickness.

Initially, it is anticipated that physical measurements of the placed sand cover thickness will be performed within each SCMU. However, assuming the correlation between the two measurement methods is verified, and provided the method of placement has not been significantly modified, the frequency of physical sampling may be reduced as construction proceeds to expedite the process of verifying compliance with the minimum required thicknesses. More intensive sampling would be resumed following extended periods of inactivity or significant modifications to the material placement method. An adaptive management approach will be used in the field to select which SCMUs will be sampled and the number of physical samples collected for each based on the goal of demonstrating compliance with the specified statistical confidence.

If compliance is not verified for a SCCU, additional sand cover material will be placed in SCMUs having the lowest placed thicknesses (as based on the volume of material delivered, adjusted using the regression equation). Selection of SCMUs for additional sand placement will be based on an evaluation of data relative to the compliance metric. Additional sand volume will be added as necessary to achieve compliance, according to the regression and the compliance criteria. Additional sampling will then be conducted to reassess compliance. Contingency plans for addressing poor correlations between thickness measurements will be developed during subsequent remedial design and/or adaptive management phases in order to more fully utilize information gained from early remedial actions and other relevant projects.

Engineered Caps A and B. Similar to the discussion above for sand covers, the thickness of engineered Caps A and B (minimum of 13- and 16-inch, respectively) will be verified through a combination of accurate material placement records and physical sampling as summarized below.

• The volume of material placed to construct each layer of the engineered caps (e.g., lower sand chemical isolation layer and overlying gravel armor layer) will be will be accurately tracked such that the placed thickness of each layer within a given CMU can be computed.


104

• Following completion of placement of each layer (i.e., sand or gravel) within a given CMU, the thickness of placed material will be physically measured at randomly-selected locations. Samples of the sand layer may be collected using a hydraulic Power Grab, piston core, “catch pan”, or other appropriate equipment capable of collecting a relatively undisturbed sample of the sand. Due to the difficulty in collecting a representative core sample of gravel armor layer, physical measurements of the gravel thickness will be limited to catch pans, survey stakes, or comparison of pre- and post-cap placement bathymetric surveys. Selection of the most appropriate measurement technique will be determined prior to or during the initial stages of construction and may be varied as construction proceeds depending on local site conditions.

Consistent with the approach described above for sand covers, a regression-based statistical approach will be developed during initial cap placement activities in the first several CCUs. Using adaptive management, the frequency of physical sampling may be reduced as construction proceeds to expedite the process of verifying compliance with the minimum required thicknesses.

Similar to the sand covers, the metric for verifying compliance of the engineered cap layer thicknesses referenced in the ROD Amendment will be delivery of material to meet the minimum required layer thickness (including mixing with underlying sediment) over at least 80 percent of the area (of a given CCU) with 90 percent confidence or greater. Compliance will be evaluated through accurate material placement records and using a regression-based correlation with in-place measurements. Core samples will also allow direct comparisons with individual layer thickness performance criteria developed using USEPA guidance, as detailed in Shaw and Anchor (2006).

If compliance is not verified for a CCU, additional sand and/or gravel material will be placed in CMUs having the lowest placed thicknesses (as based on the volume of material delivered, adjusted using the regression equation). Sufficient volume will be added to achieve compliance, according to the regression and the compliance criteria. Additional sampling will then be conducted to reassess compliance. Again, assuming the correlation between the measurement methods is verified, and provided the method of placement has not been significantly modified, the frequency of physical sampling may be reduced as construction proceeds to expedite the process of verifying compliance with the minimum required thicknesses. More intensive sampling would be resumed following extended periods of inactivity or significant modifications to the material placement method. An adaptive management approach will be used in the field to select which CMUs will be sampled and the number of physical samples collected for each based on the goal of demonstrating compliance with the specified statistical confidence.

Engineered Cap C and Armor Shoreline Caps. Consistent with the discussion above for engineered caps A and B, the thickness of engineered cap C (minimum 33-inch) and armored shoreline caps will be verified through a combination of accurate material placement records, physical sampling (where feasible), and bathymetric surveying as summarized below.


105

• The volume of material placed to construct each layer of the engineered caps (e.g., sand chemical isolation layer and armor layer) will be will be accurately tracked such that the delivered thickness of each layer for a given CMU can be computed.

• Following completion of placement of the sand layer within a given CMU, the thickness of placed material will be physically measured at randomly-selected locations. Samples of the sand layer may be collected using a hydraulic Power Grab, piston core, “catch pan”, or other appropriate equipment capable of collecting a relatively undisturbed sample of the sand.

• Following verification of compliance of the sand layer thickness and prior to placement of the armor layer, a bathymetric or survey stake survey will be conducted. A second survey will then be conducted following placement of the armor layer and compared to the pre-placement survey to verify compliance with the required armor layer thickness. In shallow water shoreline areas, it may be necessary to perform the bathymetric surveys using manual survey techniques as opposed to acoustical methods appropriate for deeper water areas.

Consistent with the discussion above for engineered caps A and B, compliance for the sand/filter layer of the cap C areas will be evaluated through accurate material placement records and using a regression-based correlation with in-place measurements. Compliance with geotechnical “filter criteria” (e.g. to avoid mixing of armor and underlying materials) can be achieved for typical gradations that comply with the armor stone sizes recommended to resist erosion. Although core sampling is generally not feasible for the armor layer due to the size of the material, a similar regression-based statistical evaluation using bathymetric or stake survey points within a CMU, relative to the computed delivered thickness, will be used to assess compliance of the armor layer thickness with the 80 percent coverage with 90 percent confidence metric.

10.2 Post-Construction Operation, Maintenance, and Monitoring

Certain elements of the remedial action will require long-term monitoring and/or maintenance, and such activities will be covered under the OMMP. This will include, for example, post-construction monitoring and maintenance of capped areas to ensure the cap remains physically stable (i.e., does not erode) and chemically protective. The OMMP will also include long-term monitoring of natural recovery areas. The objectives of the OMMP monitoring programs (to be detailed as an element of the 60 Percent Design submittal) are to confirm the remedial actions performed in OUs 2 to 5 achieve the performance standards specified in the ROD Amendment, including verification of the effectiveness of engineered caps and sediment natural recovery. The OMMP will also identify points of compliance for the remedial action and potential contingency response actions that could be implemented in the event that the remedial actions do not meet their respective performance standards.

10.2.1 Cap Performance Monitoring and Maintenance

The ROD Amendment requires long-term monitoring of any caps that are installed in OUs 2 to 5 to ensure their long-term integrity and protectiveness. The long-term monitoring will include:

• Hydrographic surveys and core sampling.


106

• Monitoring for physical integrity.

• Monitoring for chemical containment.

• Cap enhancement and/or removal.

The objectives of the cap monitoring program are to detect and evaluate any changes in the physical or chemical properties of the cap that would compromise its integrity. The physical integrity of the cap will be monitored to ensure the cap thickness does not diminish by erosion. In addition, the chemical integrity of the cap will be monitored to ensure that chemicals of concern in the underlying sediments do not migrate through the cap and into the river. Given that completion of capping is anticipated to take several years, monitoring will occur independently within the four general work areas: OU 2, OU 3, OU 4A, and OU 4B. The “Year 0” trigger for post-construction cap monitoring in a given area will occur when cap construction is completed within that area.

Physical Integrity of Sediment Caps. An initial post-construction hydrographic survey of the capped areas will be performed immediately following completion of the remedial action. As discussed above, a limited number of sediment cores (or other techniques suitable for measuring the thickness of placed caps with consideration of armor stone size) will also be collected through the completed cap to correlate to the hydrographic survey. This initial post-construction survey and correlated thickness measurements will verify that cap placement specifications have been met, and establish the baseline (Year 0) cap condition for the subsequent assessment of long-term changes in cap thickness.

Post-construction hydrographic surveys of the capped area will be completed and a limited number of cores (or other suitable measurements) will be collected during Years 2, 4, and 9 following completion of the remedial action, such that the results could be incorporated into the 5 and 10-year CERCLA reviews. In addition, hydrographic surveys will be performed as soon as possible following any flood event with a recurrence interval of 50 years or more. Hydrographic surveys may also be performed following major river construction events (e.g. new bridge construction) or significant changes in waterway use (e.g. channel reauthorization, etc.). Hydrographic survey results and core samples collected during cap monitoring events will be analyzed to determine cap thickness and integrity and compliance with minimum water depth criteria for capping. Monitoring details will be developed during the 60 Percent Design.

To the extent possible, survey data will be collected along the same transects from year to year to ensure comparable data are collected. Changes in bathymetry over time will be evaluated to identify areas of potentially significant erosion, deposition, or consolidation. Initially, erosion will be estimated based on elevation loss in hydrographic surveys or from regularly-scheduled cores collected in cap areas. In capped areas with significant elevation loss (to be defined in greater detail as part of the 60 Percent Design submittal), follow-on sediment cores will be collected to determine whether the elevation loss has occurred as a result of erosion or settlement based on visual assessment of cap thickness in the core


107

sample, and in consideration of core compaction. If the required thickness of cap material is present in the core sample, it will be determined that settlement has occurred rather than erosion.

In addition to hydrographic surveys, bank surveys will be performed during low-water conditions to monitor caps placed on river banks and side-slope areas. The bank surveys would include:

• Field reconnaissance for evidence of erosional features (i.e., presence of gullies, escarpments, slumps, etc.).

• Monitoring elevation changes using stakes embedded in the cap.

• Follow-up land surveying as necessary.

If the low-water field surveys show significant cap erosion has occurred on the banks, follow-up hydrographic and/or diver surveys may be conducted in the adjacent areas of the river to determine whether the erosion extends into deeper water.

A decision tree will be developed during the 60 Percent Design to evaluate physical integrity monitoring results. This decision tree may include the following general evaluation criteria:

• If a specific percentage (to be determined during RD) of the capping area is at or above design grade ("target thickness"), no maintenance action will be necessary.

• If less than the specified percentage of the area is at or above grade, cores (or other suitable measurements) will be collected and the post-cap bathymetry reviewed on a CCU basis to determine if the area has more than the required minimum cap thickness and would be expected to self-level through hydrodynamic forces such that the specified percentage of the area would be at or above grade. The need for cap maintenance would be determined on a case-by-case basis.

If cap erosion is confirmed by collected cores (or other suitable measurements), possible response actions can include:

• No action if the chemical analytical results indicate the remaining cap thickness is sufficient to prevent “breakthrough” of contaminants.

• Repair area of erosion (re-establish cap thickness).

• Armor area of erosion.

• Enact managerial or institutional controls, such as changes to vessel operations, to help control any further cap erosion.

Consistent with CERCLA requirements, the Response Agencies and the entities conducting the remedial action will evaluate cap performance and the need for and scope of continued cap monitoring as part of the five-year review process.


108

Chemical Integrity of Sediment Caps. Core samples collected during cap monitoring events will be analyzed for PCB contamination within 6 inch intervals to determine whether contamination is being effectively contained and isolated from biota. Cap monitoring is outlined below.

Surface samples will be collected in all capped areas where sampling of the substrate is feasible (e.g., it may not be feasible to collect samples of large armor stones). Consistent with the procedures used to sample post-dredge areas (see Section 10.1 above), composite grab samples will be collected and analyzed. Composite confirmation samples will typically be comprised of five individual grab samples, although fewer individual grabs may suffice in smaller capped areas. An aliquot of each individual grab sample will be separately archived for possible future analysis in case it is determined that finer spatial resolution is needed. Surface sediment sampling in the capped area will be completed during Years 1, 4, and 9 following completion of the remedial action.

An estimated 340 acres in OUs 2 to 5 will be remediated by either capping or a combination of dredging and capping (excluding shoreline areas to be evaluated during subsequent design phases). These acres would be included in long-term cap integrity monitoring. The specific number and location of cap monitoring samples, including surface samples and subsurface cores, will be developed as part of the 60 Percent Design submittal.

The purpose of cap surface sampling is to monitor for chemical breakthrough of the cap. If any composite sample results are above 1.0 ppm PCBs, the individual grab samples which comprise those composite samples will be analyzed. If the individual sample results confirm the composite sample result, a sediment core will be collected in that area, as described below.

Sediment cores extending through the cap and into underlying sediments (approx. four-foot cores) may be collected in representative capped areas containing surface sediments exceeding 1.0 ppm PCBs (where sampling through armor stone is feasible), with sampling targeted towards areas with significantly above-average river currents and chemical gradients.

The cores will be sub-sectioned and analyzed for PCBs and TOC. The core profiles will be analyzed for evidence of chemical migration through the cap, and evidence of “breakthrough” in which the chemical has migrated through the full thickness of the cap and into the overlying river. If breakthrough has occurred, or is predicted to occur before the design life of the cap is reached, additional response actions will be evaluated.

Cap Maintenance and Response Actions. Records of previously completed capping projects are available and may be used to estimate the expected level of maintenance required for the OUs 2 to 5 engineered caps. Maintenance records from the caps that have been in place for more than 15 years (e.g., a number of estuarine and river caps constructed in the Pacific Northwest; e.g., Sumeri 1996) indicate that maintenance of approximately 5 percent of the total cap area may be needed after 2 years of monitoring. After appropriate modifications to the armor stone size, further cap maintenance has not been required at


109

any of these capping sites (up to 20 years after construction) following initial maintenance. Therefore maintenance of engineered caps to be placed in OUs 2 to 5 may be expected over roughly 5 percent of the total cap area. For the purpose of the BODR cost estimates (Shaw and Anchor 2006), localized maintenance was assumed necessary at 2, 4, 9, and 30 years following construction and at 20-year intervals thereafter, based on the possible occurrence of large but infrequent storms.

If monitoring data indicate that caps placed in an area no longer meets its original as-built design criteria and that degradation of the cap in the area may result in an actual or threatened release of PCBs at or from the area, the Response Agencies and the entities conducting the remedial action will collectively perform supplemental evaluations and identify additional response activities that may be appropriate for consideration in the area. For example, sediment cores collected for evaluation of chemical breakthrough will be evaluated on a case-by-case basis. If chemical profiles in sediment cores indicate cap breakthrough may have occurred, possible response options can include:

• Increasing the thickness of the cap to ensure cap integrity.

• Increasing the frequency and intensity of cap monitoring.

• If monitoring or other information shows a pattern of cap degradation in multiple areas, then additional response activities may be considered, including cap enhancement (e.g., application of a thicker sand layer or stone layer or use of larger armor stone) or cap and underlying contaminated sediment removal.

Again, details will be provided as part of the 60 Percent Design submittal.

10.2.2 Natural Recovery Monitoring

Natural recovery areas in OU 2 and OU 5 will be monitored to verify the effectiveness of natural recovery in terms of reducing surface sediment concentrations of PCBs, to confirm ROD predictions of natural recovery over the next decade or more. Co-location of sampling locations with prior RI/FS stations will be performed to support assessments of natural recovery over time. As practical, natural recovery monitoring, cap monitoring and water/biota sampling will be coordinated to take place during the same year, conducted approximately one year prior to the scheduled CERCLA five-year reviews, so that the most up-to-date information will be available to inform the process and to better scope future monitoring efforts and strategies.

Sampling and analysis procedures will be developed as part of the 60 Percent Design submittal. Surface sediment (0 to 10 cm) natural recovery monitoring will initially occur during Year 4 following completion of remedial actions. Those stations that are below the RAL during the Year 4 sampling event will not be sampled further as a part of the OMMP. Those stations exceeding the RAL during the Year 4 sampling will be resampled during Year 9, and potentially also during subsequent 5-year events.


110

10.3 Long Term Monitoring

Long-term plans for monitoring of water and biota in the Lower Fox River and Green Bay will be specified in the LTMP, which will describe the monitoring program in years following completion of remedial actions, to verify that the remedial action is effective at reducing risk to human health and the environment. Data collected under the LTMP will be used to help determine the magnitude and extent of reductions in PCB concentrations over time in response to the remedial action. The basic components and framework of this plan, which will be included as an element of the 60 Percent Design submittal, are outlined below.

Water and fish tissue are the media of interest for long-term monitoring of risk reduction to humans and wildlife as a result of the sediment remedial action. Both media will be sampled and analyzed at a number of stations from Lake Winnebago to Green Bay. The water monitoring plan will generally consist of systematic monthly sampling over the course of an entire year, excluding the winter months of December through March, when PCB concentrations and loads are small (these four months combined contribute approximately 10 percent of the total annual PCB load). The fish tissue monitoring plan will include sampling of three different species at nine different stations, in replicates of 5 composite samples (for ecological and “early indicator” forage fish species) or 15 individuals (for human health species) at each station. Both water and fish tissue monitoring programs are scheduled to include a Year 0 (construction completion) event, and will be implemented on a five-year cycle thereafter during the post-construction period. The OUs 2 to 5 area may be divided into smaller management units to allow monitoring activities to begin in certain reaches of the river as remedial actions are completed, even though work may be ongoing in other downstream reaches.

The details of the LTMP are currently being developed through a collaborative Work Group process. The draft plan submitted to the Intergovernmental Parties was used as a starting point for development of the LTMP, modified as appropriate based on the Agency-approved Baseline Monitoring Plan and ongoing Work Group discussions. Baseline water and fish tissue quality conditions in the Lower Fox River and Green Bay were characterized during 2006-2007 to provide a point of comparison for long-term (post-remediation) monitoring data. Interpretation of the baseline monitoring data is ongoing.

10.3.1 LTMP Objectives

As discussed above, the overall objective of the LTMP is to evaluate the rate and magnitude of risk reduction that occurs in response to the sediment remedial action. The combined baseline and long-term monitoring data is expected to provide the Response Agencies with sufficient information to determine whether the implemented remedy meets risk reduction success criteria. Specific objectives of the LTMP include the following:

• Monitor Progress toward Achieving RAOs.

• Characterize Bioaccumulation Pathways.


111

Each of these objectives is briefly described below.

Monitor Progress toward Achieving RAOs. Long-term monitoring data will be collected to evaluate progress toward achieving the RAOs of reduced risk to humans and wildlife, as presented in the RODs (WDNR/EPA 2002, 2003). More specifically, these RAOs include the following:

• Verify that sediment remedial actions in OUs 2 to 5 result in substantive reductions in water column and fish tissue PCB concentrations. The RODs identified water and fish tissue as key exposure media through which bioaccumulation may occur.

• Verify that sediment remedial actions in OUs 2 to 5 result in substantive reductions of PCB loadings to Green Bay. Decreased loadings from the Lower Fox River will help facilitate natural recovery processes in Green Bay.

Characterize Bioaccumulation Pathways. As determined in the Baseline Risk Assessment (Retec 2002), the primary exposure pathway for humans and wildlife to become exposed to PCBs in OUs 2 to 5 is through consumption of PCB-contaminated fish. Therefore, the LTMP will focus on monitoring risk reductions to humans and wildlife (including fishermen as well as fish-eating mammals and birds) as a result of the remedial action, by monitoring PCB concentrations in an appropriate selection of fish species, fish size/age, and preparation methods (i.e. whole fish or fillets) which are relevant to these receptors.

10.3.2 Water Quality Monitoring Plan

In general, water monitoring stations will be sited near the boundaries of the OUs such that the net PCB contribution from each OU, and the effectiveness of the remedy in each OU, can be evaluated. In addition, multiple water quality monitoring stations will be sited in OU 2 and OU 5. The monitoring locations will be identical to the stations occupied during the Baseline Monitoring Program, and consistent to the extent possible with stations occupied during past and ongoing monitoring programs.

Water Quality Monitoring Stations. Water column samples will be collected and analyzed at 1 upstream reference location in Lake Winnebago, 6 stations along OUs 1 to 4, and 3 stations in Green Bay (OU 5), for a total of 10 stations. These stations are described below:

• Lake Winnebago (upstream reference station). Just above Neenah and Menasha Channels.

• OU-1. Downstream of LLBDM and above the first Appleton Dam.

• OU-2A. Reach between Lock 4 and Cedars Lock.

• OU-2B. Reach between Lock 5 and Rapide Croche Lock.

• OU-2C. Above Little Rapids Dam.


112

• OU-3. Above DePere Dam.

• OU-4. Near the USGS river gage (Oil Depot gage); approximately 1,300 meters upstream from the mouth, and beyond the influence of upstream transport of bay water under seiche conditions. This station will be co-located with a USGS water quality monitoring station.

• OU-5A. Zone II/Zone III Boundary.

• OU-5B. Zone III South.

• OU-5C. Zone III North.

Water Quality Monitoring Schedule. During each scheduled monitoring year, sampling will be performed on a monthly basis for the entire year, excluding the winter months of December through March (8 monthly sampling events total). The sampling schedule will be “systematic” in design (i.e., predetermined sampling at regular intervals), to provide representative and unbiased coverage. Specific runoff events will not be targeted but a representative range of flows will likely be captured during the course of the monitoring program.

During five-year monitoring events, water quality monitoring may be reduced or eliminated in favor of fish tissue monitoring since fish tissue is the primary medium of exposure to PCBs for humans and wildlife. For example, if fish tissue concentrations recover at a rate and magnitude consistent with expectations outlined in the ROD and ROD Amendment, the scope of the water quality monitoring program (i.e., number of sampling stations and/or seasonal sampling frequency) may be reduced or eliminated. Alternatively, if fish tissue recovery trends fall significantly behind expectations, water column monitoring may be reinstated. A more detailed description of the conditions under which water column monitoring would be reduced, eliminated, or reinstated will be provided in the LTMP (to be provided as part of the 60 Percent Design submittal).

Water Quality Analytical Parameters. During each specified monitoring year, eight rounds of water column samples (once a month, excluding December through March) will be collected at ten stations, for a total of 80 samples plus quality control samples. All water column samples will be analyzed for the following:

• PCB Congeners (209 total) by EPA Method 1668A (high-res GC/MS).

• Total Suspended Solids (TSS) by EPA Method 160.2.

• Total Organic Carbon (TOC) by EPA Method 415.1.

• Temperature and turbidity (in situ field measurements)


113

10.3.3 Fish Tissue Monitoring Plan

Fish monitoring stations include an upstream reference site (Lake Winnebago), six stations in the Lower Fox River, and two stations in Green Bay (9 stations total). One sampling station is assigned to each OU, except OU 2 which has three sampling stations because of its length and dam controls, and Green Bay which has two stations to better describe the gradient between the river mouth and the deeper portions of the bay. Exact locations will be determined in the field based on species availability, habitat, and seasonal migration patterns. Because of these variables, it can be expected that different species will be collected from different parts of the OUs.

Fish Tissue Monitoring Stations. Fish will be collected from the following locations:

• Lake Winnebago (upstream reference station).

• OU-1. Little Lake Butte de Morts.

• OU-2A. Reach between Lock 4 and Cedars Lock.

• OU-2B. Reach between Lock 5 and Rapide Croche Lock.

• OU-2C. Above Little Rapids Dam.

• OU-3. Reach above DePere Dam.

• OU-4. Reach from DePere Dam to about 1,000 meters upstream from the mouth, above the influence of bay water.

• OU-5A (Green Bay). Shallow inner bay, inside of Little Tail Point.

• OU-5B (Green Bay). Deeper middle bay, between Little Tail Point and Sturgeon Bay.

Fish Tissue Monitoring Schedule. Long-term fish sampling will be conducted during the same seasonal windows used for the baseline events to minimize seasonal differences in fish tissue concentrations between monitoring years. The primary fishing season runs from August 15 through September 15 but can be extended into the fall if necessary. Monitoring will be discontinued when specific remediation goals or other performance criteria have been achieved (e.g., attainment of specific target tissue levels which do not pose a risk to humans or wildlife or achieving equilibrium concentrations with current source inputs). A more detailed description of the conditions under which monitoring would be terminated will be provided in the LTMP.

Fish Species and Target Size Ranges. Target fish species were selected based on a number of criteria:

• Presence of fish consumption advisories;

• Popular recreational fishery;

• Key species evaluated in Human Health or Ecological Risk Assessments (Retec 2002c);

• Common food source for upper-level animals (i.e., fish-eating mammals, birds); and


114

• Elevated PCB concentrations in recent monitoring data.

Five target fish species were selected to address three different monitoring objectives:

• Protection of human health (walleye, smallmouth bass);

• Protection of wildlife (carp, drum); and

• Early indication of river recovery (young gizzard shad).

All five fish species were collected and analyzed during the baseline (i.e. pre-construction) monitoring event. In the post-construction period, fish tissue monitoring will be reduced to three species—a human health index (walleye), an ecological index (carp or drum, depending on the OU), and a young forage fish species (gizzard shad). Smallmouth bass and carp or drum (depending on the OU) will be reserved as alternate species if collection of any of the three primary species becomes problematic.

Fish Tissue Analysis. During long-term monitoring events, the monitoring program will consist of one human health indicator species (walleye – 15 individuals at 9 stations), one ecological indicator species (carp or drum – 5 composite samples at 9 stations) and one “early indicator” forage fish species (gizzard shad – 5 composite samples at 9 stations), for a total of 225 fish tissue analyses, not including quality control samples. To ensure consistency with past and ongoing monitoring programs, analytical methods will follow procedures used by the Wisconsin State Lab of Hygiene (SLOH) to the extent possible. Specifically, fish tissue samples will be analyzed according to the following methods:

• Tissue Extraction by SLOH Method.

• PCB Aroclors by EPA Method 8082.

• Lipid Content by gravimetric method (EPA 2000).

In addition, walleye samples from Station OU 2C will also be analyzed for:

• Mercury (EPA Method 7471).

10.4 Adaptive Management

The Adaptive Management Plan will set forth a process and procedures to modify the cleanup and monitoring plan as appropriate in response to new information and experience during initial remediation activities in OUs 2 to 5. The basic components and framework of this plan, which will be included as an element of the 60 Percent Design submittal, include the following:


115

• General Considerations and Guidance

• Relevant case studies (e.g., Everglades; Lavaca Bay; Grasse River)

• Risk management framework (e.g., NRC)

• Developing USEPA guidance (e.g., Triad program)

• New Information

• Phase 1 Project

• Pilot OUs 2 to 5 Dredging, Capping, and Cover Projects

• Integration with Other Related Projects

• USACE navigation channel maintenance dredging

• Development & maintenance projects (e.g., marinas)

• Modifications to Remedial Action Sequencing

• Evolving construction schedules

• Dredging, capping, and cover sequencing

• Future Changes in Site Conditions

• Physical changes (sedimentation, new structures)

• Declines in sediment PCB concentrations prior to initiating OUs 2 to 5 remedial actions (e.g., changes to no action as a result of natural recovery from upstream actions)

• Remedial Action Verification Sampling

• Reduction in sampling frequency over time

• Post-dredge verification

• Post-cap & post-cover verification

• Dredge Residual Management

• Presumptive engineering evaluations

• Refinement of sand cover criteria

• Role of SWAC calculations

• Post-Construction Elements

• Cap monitoring and maintenance

• Natural recovery monitoring

• LTMP monitoring

• Changes to Remedial Design

• Administrative Mechanisms


116

10.5 Monitoring, Maintenance, and Adaptive Management Schedule

The scheduling of various construction tasks is discussed in Section 9. The remedial action is expected to take approximately nine years to complete. At the completion of the remedial action, a hydrographic condition survey will be performed to characterize the starting elevation of capped areas and to provide a basis to evaluate cap erosion in subsequent surveys. At that time, a round of water quality and aquatic biota sampling will also be performed to support the objectives of the LTMP.

During the post-construction period, cap performance monitoring will be conducted to ensure the physical and chemical integrity of the cap is maintained. Cap performance monitoring will initially be conducted on a relatively frequent basis, with completion of Year 0, Year 2, and Year 4 events following remedial action to provide early feedback after the completion of the remedial action. Cap performance monitoring will then be reduced to ten-year or twenty-year intervals pending favorable results during the initial monitoring events (with contingency monitoring following flood events with 50-year recurrence interval or greater).

Sampling of water quality and fish biota will initially be performed on a five-year cycle to monitor progress toward achieving RAOs which are focused on reducing risks to humans and wildlife. When appropriate, cap monitoring and water/fish biota sampling can be coordinated to take place during the same year. To the extent practical, both cap monitoring and water/biota sampling will be conducted one year prior to the scheduled CERCLA five-year reviews, so that the most up-to-date information will be available to inform the process and to better scope future monitoring efforts and strategies.

Section 11 ‐ References

117

11 REFERENCES

Anchor and LimnoTech, 2006a, Evaluation of Geostatistical Methods for Delineating Remediation Boundaries in Operable Unit 3, Lower Fox River Remedial Design, OUs 2‐5. February 28, 2006.

Anchor and LimnoTech, 2006b, Evaluation of Geostatistical Methods for Delineating

Remediation Boundaries in Operable Unit 4, Lower Fox River Remedial Design, OUs 2‐5. February 28, 2006.

Anchor and LimnoTech, 2006c, Supplemental Information on Geostatistical Performance

Metrics, Operable Units 3 and 4, Lower Fox River Remedial Design, OUs 2‐5. February 28, 2006.

Anchor and LimnoTech, 2006d, Updated Kriging Results for Operable Unit 4, Lower Fox River

Remedial Design, OUs 2‐5. March 24, 2006. Blaauw, H.G., and E.J. van de Kaa., 1978. Erosion of Bottom and Sloping Banks Caused by the

Screw Race of Maneuvering Ships. Paper presented at the 7th International Harbour Congress, Antwerp, Belgium. May 22‐26, 1978.

Bhowmik, N.G., Soong, T.W., Reichelt, W.F., and Seddik, N.M.L., 1991. Waves generated by recreational traffic on the Upper Mississippi River system, Research Report 117, Department of Energy and Natural Resources, Illinois State Water Survey, Champaign, Illinois.

Bridges, T., Ells, S., Hayes, D., Mount, D., Nadeau, S., Palermo, M., Patmont, C., and Schroeder,

P. In preparation. Relating the 4 Rs of Environmental Dredging: Resuspension, Release, Residual, and Risk.

Foth & Van Dyke, 2001. Fox River Deposit N Appendix to Summary Report. Prepared for

Wisconsin Department of Administration and Wisconsin Department of Natural Resources. April 2001.

Foth & Van Dyke, 2005. Lower Fox River Operable Unit 1 Remedial Action‐2004 Remedial

Summary Report, Appendix D Draft. February, 2005. Kreibel, D. L. and Seelig, W. N., 2005. An Empirical Model for Ship‐Generated Waves. 5th

Internatioinal Symposium Waves 2005. Madrid Spain.


118

Palermo, M. R., J. E. Clausner, M. P. Rollings, G. L. Williams, T. E. Myers, T. J. Fredette, and R. E. Randall, 1998a. Guidance for Subaqueous Dredged Material Capping. Technical Report DOER‐1. United States Army Corps of Engineers, Waterways Experiment Station, Vicksburg, Mississippi. Website: http://www.wes.army.mil/el/dots/doer/pdf/doer‐1.pdf.

Palermo, M. R., J. Miller, S. Maynord, and D. Reible, 1998b. Assessment and Remediation of

Contaminated Sediments (ARCS) Program Guidance for In Situ Subaqueous Capping of Contaminated Sediments. EPA 905/B‐96/004. Prepared for the Great Lakes National Program Office, United States Environmental Protection Agency, Chicago, Illinois. Website: http://www.epa.gov/glnpo/sediment/iscmain.

Palermo, M.R., T.A. Thompson, and F. Swed, 2002. White Paper No. 6B – In‐Situ Capping as a

Remedy Component for the Lower Fox River. Prepared for Wisconsin Department of Natural Resources. December 2002.

Patmont, C. and Palermo, M., 2007. Case Studies of Environmental Dredging Residuals and

Management Implications. Proceedings, 4th International Conference on Remediation of Contaminated Sediments, January 22‐25, 2007, Savannah, GA, USA.

Permanent International Association of Navigation Congresses, 1987. ʺGuide‐lines for the

design and construction of flexible revetments incorporating geotextiles for inland waterways.ʺ Working Group 4 of the Permanent Technical Committee, Brussels.

Reible, D.D., Kiehl‐Simpson, C., Marquette, A., 2004. “Modeling Chemical Fate and Transport in

Sediment Caps.” Technical Presentation 380‐D. New York, NY: American Institute of Chemical Engineers.

Shaw and Anchor, 2006. Final Basis of Design Report; Lower Fox River and Green Bay Site.

Prepared for Fort James Operating Company, Inc. and NCR Corporation by Shaw Environmental and Infrastructure, Inc. and Anchor Environmental, LLC. June 16, 2006.

Sorensen, R., 1997. Prediction of Vessel‐Generated Waves with Reference to Vessels Common to

the Upper Mississippi River System. Lehigh University and Coastal Hydraulics Laboratory of the US Army Engineer Waterways Experiment Station. ENV Report 4. December.

Sorensen, R. M., and Weggel, J. R., 1984. A Development of ship wave design information.ʺ

Proceedings of the 19th Conference on Coastal Engineering, Houston TX 3‐7 September 1984. Billy L. Edge, ed., American Society of civil Engineers, NY III 3227‐43.

USEPA and WDNR, 1998. Technical Memorandum 2a: Simulation of historical and projected

total suspended solids loads and flows to the Lower Fox River, N.E. Wisconsin, with the Soil and Water Assessment Tool (SWAT), August 19.


119

USEPA and WDNR, 2003. Record of Decision, Operable Units 3, 4, and 5. Lower Fox River and

Green Bay Wisconsin. June 2003. USEPA and WDNR, 2007. Record of Decision Amendment: Operable Unit 2 (Deposit DD),

Operable Unit 3, Operable Unit 4, and Operable Unit 5 (River Mouth). Lower Fox River and Green Bay Superfund Site. June.

USEPA, 2004. Administrative Order on Consent, In the Matter of Lower Fox River and Green

Bay Site. Respondents for Operable Units 2, 3, 4, and 5: Fort James Operating Company, Inc. and NCR Corporation. March.

USEPA 2005. Contaminated Sediment Remediation Guidance for Hazardous Waste Sites.

OSWER Publication 9355.0‐85. United States Environmental Protection Agency, Office of Solid Waste and Emergency Response. Website: http://www.epa.gov/superfund/resources/sediment/guidance.htm.

Verhey, H.J., 1983. The Stability of Bottom and Banks Subjected to the Velocities in the Propeller

Jet behind Ships. Presented at the 8th International Harbour Congress Antwerp, Belgium, June 13‐17, 1983.

WDNR, 2000. Fox River Remediation Air Monitoring Report, Ambient PCBs During SMU 56/57

Demonstration Project, August – November, 1999. Publication Number: PUBL‐AM‐310‐00. Wisconsin Department of Natural Resources, Bureau of Air Management. Madison, Wisconsin. October, 2000.

LOWER FOX RIVER 30 PERCENT DESIGN

Documents