REMEDIATION SYSTEM EVALUATION
WYCKOFF/EAGLE HARBOR SUPERFUND SITE
SOIL AND GROUND WATER OPERABLE UNITS
BAINBRIDGE ISLAND, WASHINGTON
Report of the Remediation System Evaluation,
Site Visit Conducted at the Wyckoff/Eagle Harbor Superfund Site
October 8, 2003
Final Report
March 1, 2005
-------
Office of Solid Waste and EPA 542-R-05-013
Emergency Response March 2005
(5102G) www.epa.gov/tio
www.clu-in.org/optimization
Remediation System Evaluation (RSE)
Wyckoff/Eagle Harbor Superfund Site
Soil and Ground Water Operable Units
Bainbridge Island, Washington
-------
NOTICE
Work described herein was performed by GeoTrans, Inc. (GeoTrans) for the U.S. Environmental
Protection Agency (U.S. EPA). Work conducted by GeoTrans, including preparation of this report, was
performed under Dynamac Corporation Prime Contract No. 68-C-02-092, Work Service Requests No.
ST-1-20 and ST-1-15. Mention of trade names or commercial products does not constitute endorsement
or recommendation for use.
-------
EXECUTIVE SUMMARY
A Remediation System Evaluation (RSE) involves a team of expert hydrogeologists and engineers,
independent of the site, conducting a third-party evaluation of site operations. It is a broad evaluation
that considers the goals of the remedy, site conceptual model, above-ground and subsurface performance,
and site exit strategy. The evaluation includes reviewing site documents, visiting the site for up to 1.5
days, and compiling a report that includes recommendations to improve the system. Recommendations
with cost and cost savings estimates are provided in the following four categories:
• improvements in remedy effectiveness
• reductions in operation and maintenance costs
• technical improvements
• gaining site closeout
The recommendations are intended to help the site team identify opportunities for improvements. In
many cases, further analysis of a recommendation, beyond that provided in this report, may be needed
prior to implementation of the recommendation. Note that the recommendations are based on an
independent evaluation by the RSE team, and represent the opinions of the RSE team.
The Wyckoff/Eagle Harbor Superfund site is located on the eastern side of Bainbridge Island, in central
Puget Sound, Washington along Eagle Harbor. The site includes the inactive 57-acre Wyckoff wood-
treating facility, contaminated sediments in adjacent Eagle Harbor, and other upland sources of
contamination to the harbor, including a former shipyard. The wood-treating facility operated from the
early 1900s through 1988, when the plant shut down. The site consists of four operable units, but this
RSE pertains only to the soil and ground water operable units (OU2 and OU4). A P&T system has
operated at the site as an interim remedy for over a decade, and a pilot study of steam enhanced
contaminant recovery was attempted at the site between late October 2002 and April 2003. Technical
problems were encountered during the steam injection pilot test, and the site team is now faced with a
decision as to how to proceed with the remedy given both technical and fiscal considerations. The RSE
team was specifically asked by EPA Region 10 and OSRTI to consider options for moving forward,
ranging from containment only through full-scale steam injection.
Rather than providing recommendations for the operating interim P&T system in the four above-
mentioned categories, the RSE team is providing recommendations or ideas to consider as the final
remedy is chosen, designed, and implemented. In Section 6.0 of this report, the RSE team outlines a road
map for a final remedy and highlights what the RSE team believes are high priority items.
The RSE team believes that the best approach for this site is to initially focus efforts on hydraulically
isolating the contamination underlying the Former Process Area, including the installation of an
upgradient barrier wall and low permeability cap to minimize the amount of water requiring treatment,
and the implementation of a new groundwater treatment system based, in part, on pilot testing of one or
more new approaches. Enhanced monitoring of groundwater in the lower aquifer is also recommended in
conjunction with these efforts, to improve the potential to detect current or future impacts to that aquifer.
It is also recommended that the site team continue to monitor the seeps along the eastern beach, and take
remedial actions if isolation efforts do not stop the seeps and a feasible remedial alternative is identified.
Once these high priority items are addressed and implemented, the site team could then reconsider
aggressive mass removal and the technologies that might be availableat that time (potentially including
-------
but not limited to additional efforts related to steam injection). However, the RSE team believes the most
cost-effective approach is to design and implement the new groundwater treatment system associated
with hydraulic isolation (discussed above) independent of such efforts. This will reduce the potential of
over-designing the groundwater treatment system required for hydraulic isolation. Cost/benefit
evaluations for subsequent testing or implementation of more aggressive source removal would need to
incorporate costs that might be required to further upgrade the groundwater treatment system above and
beyond the treatment system associated with hydraulic isolation. The RSE team also believes that, if
more aggressive source removal technologies are considered in the future, the costs and benefits of
installing additional recovery wells and tying them into the P&T system should be included as a potential
alternative.
During the RSE site visit there was discussion about armoring the existing sheet pile wall to prevent
scour and extend the life of the wall, and there was also discussion about adding a second sheet pile wall
inside the existing wall to create an "attenuation zone" that could be monitored. The RSE team felt that,
if armoring was pursued, then the interior sheet pile wall would not be necessary because the armoring
could likely be constructed in a manner to allow for an attenuation zone that could be monitored.
Subsequent to the RSE site visit, the RSE team was informed that armoring may not be feasible due to
potential impacts to the intertidal zone. It is likely that a variety of alternatives will need to be
considered in the future regarding this issue, and this issue is not addressed in detail in this RSE report.
The RSE team's suggestions for simplifying the new groundwater treatment system could save EPA as
much as $4 million relative to current estimates, while maintaining a protective remedy. This would
represent a savings of approximately 20% relative to the preliminary three-year costs that have been
estimated to date. If pilot test results of the recommended changes do not support the carbon usage
assumptions of the RSE team (that are based on published isotherms plus a safety factor), savings might
be lower. Additional savings would also result beyond the three year period by operating a simplified
and automated treatment system. Table 7-1 summarizes the cost and protectiveness implications of the
recommendations discussed in Section 6.0 of this report.
11
-------
UPDATE
Substantial progress has occurred at the site since the submission of the revised draft RSE report in
February 13, 2004, and this final version. The site team has prioritized remedial efforts and has been
focusing on the high priority items, such as establishing hydraulic isolation of the contamination in the
Former Process Area. The Region plans to finalize a ROD Amendment or BSD in summer 2005,
following public comment. A public proposal for remedial action is scheduled for 2005 with final
decision in FY06.
The following efforts have either taken place or are in progress as of the finalization of this report. RSE
recommendations that correspond to these efforts are indicated in parentheses.
• The site team is proceeding with the high priority items first (as described below) with focus on
hydraulic isolation. The site has not yet omitted thermal remediation as an option. A pilot
summary report for the steam pilot study is being finalized and should be done in January 2005.
In addition, an engineering evaluation for a full-scale thermal system is being prepared. Thus
far, that evaluation suggests that hydraulic isolation will still be needed after thermal remediation
to meet standards. Experts believe that heating would require both steam and electrical resistive
heating. A 7.5 MW power plant for both electrical resistive heating and steam would likely be
required. Other requirements would likely include a cooling tower and a treatment plant that
could treat up to 350 gpm. (See discussion in 6.1)
• The site team conducted a pilot test to bypass the aeration basin (biological treatment) and use
GAC only, yielding favorable results for eliminating the aeration basin from the treatment train.
In addition, tests are underway to address recent problems with multimedia filters. The tests
include using hypochlorite to reduce fouling and trying other filter types, such as walnut shells,
spent carbon, bag filters, etc. Through contract renegotiation, the site team has been able to
reduce the subcontractor O&M costs by approximately $30,000 per month. (See discussion in
6.2.1)
• A 50% design for a water treatment plant was submitted in November 2004. The design includes
the DAF unit, mulitmedia filters, and GAC. The design includes contingencies for options to
multimedia filters, and although the design does not include a biological treatment system, there
is room in the treatment facility footprint if biological treatment is eventually required. A total
cost of $5 million is estimated for the treatment plant design and construction. This cost does not
include overhaul of the extraction system. Construction is scheduled for October 2005, and the
site team is planning on using the existing boiler building to house the future treatment plant.
(See discussion in 6.2.1 and 6.2.6)
• EPA and the State have agreed that installation of an upgradient sheet pile wall and a low
permeability cap are high priorities that would limit the amount of water entering the
contaminated Former Process Area and therefore limit the amount of water that would need to be
treated. A preliminary water budget analysis suggests that with these features, the ground water
extraction rate would be approximately 10 to 11 gpm. An alternatives evaluation has been
drafted for the upgradient cutoff wall. Work has not yet begun on the cap design, which is
complicated due to unknowns associated with the final remedy. (See discussion in 6.2.2, 6.2.3,
and 6.2.4)
in
-------
The site team is conducting lower aquifer monitoring and has included transducers in various
monitoring well pairs to get a better idea of hydraulic gradients for a water budget analysis.
Further augmentation of monitoring in the lower aquifer will likely occur in the future. Recent
data suggest that the hydraulic isolation is adequate and the extraction system does not need to be
upgraded at this time. (See discussion in 6.2.7)
IV
-------
PREFACE
This report was prepared as part of a project conducted by the United States Environmental Protection
Agency (U.S. EPA) Office of Superfund Remediation and Technology Innovation (OSRTI). The
objective of this project is to conduct Remediation System Evaluations (RSEs) at selected pump and treat
(P&T) systems that are jointly funded by EPA and the associated State agency. The project contacts are
as follows:
Organization
Key Contact
Contact Information
USEPA Office of Superfund
Remediation and Technology
Innovation
(OSRTI)
Jennifer Griesert
1235 S. Clark Street, 12th floor
Arlington, VA 22202
Mail Code 5201G
phone: 703-603-8888
griesert.jennifer@epa.gov
Dynamac Corporation
(Contractor to U.S. EPA)
Daniel F. Pope
Dynamac Corporation
3601 Oakridge Boulevard
Ada, OK 74820
phone: 580-436-5740
fax: 580-436-6496
dpope@dynamac.com
GeoTrans, Inc.
(Contractor to Dynamac)
Doug Sutton
GeoTrans, Inc.
2 Paragon Way
Freehold, NJ 07728
phone: 732-409-0344
fax: 732-409-3020
dsutton@geotransinc.com
-------
TABLE OF CONTENTS
EXECUTIVE SUMMARY i
UPDATE iii
PREFACE v
TABLE OF CONTENTS vi
1.0 INTRODUCTION 1
. 1 PURPOSE 1
.2 TEAM COMPOSITION 1
.3 DOCUMENTS REVIEWED 2
.4 PERSONS CONTACTED 2
.5 SITE LOCATION, HISTORY, AND CHARACTERISTICS 3
1.5.1 LOCATION AND HISTORY 3
1.5.2 POTENTIAL SOURCES 5
1.5.3 HYDROGEOLOGIC SETTING 6
1.5.4 RECEPTORS 6
1.5.5 DESCRIPTION OF GROUND WATERPLUME 7
2.0 CURRENT STATUS OF REMEDY 8
2.1 REMEDY OVERVIEW 8
2.2 CONTAINMENT SHEET PILE WALL 8
2.3 P&T EXTRACTION SYSTEM 8
2.4 P&T TREATMENT SYSTEM 9
2.5 PILOT INJECTION AND EXTRACTION SYSTEM 10
2.6 STEAM SYSTEM 10
2.7 MONITORING PROGRAM 10
3.0 REMEDY OBJECTIVES, PERFORMANCE, AND CLOSURE CRITERIA 12
3.1 REMEDY OBJECTIVES AND CLOSURE CRITERIA 12
3.2 TREATMENT PLANT DISCHARGE CRITERIA 13
4.0 FINDINGS AND OBSERVATIONS FROM THE RSE SITE VISIT 15
4.1 FINDINGS 15
4.2 SUBSURFACE PERFORMANCE AND RESPONSE 15
4.2.1 NAPL AND PLUME CAPTURE 15
4.2.2 CONTAMINANT LEVELS 16
4.2.3 STEAM INJECTION PILOT RESULTS 16
4.3 COMPONENT PERFORMANCE 17
4.3.1 CONTAINMENT SHEET PILE WALL 17
4.3.2 P&T EXTRACTION SYSTEM 17
4.3.3 TREATMENT SYSTEM 17
4.3.4 PILOT STEAM INJECTION AND EXTRACTION SYSTEM 18
4.3.5 STEAM GENERATION SYSTEM 18
4.4 COMPONENTS OR PROCESSES THAT ACCOUNT FOR MAJORITY OF OU2 AND OU4 COSTS 19
4.5 REGULATORY COMPLIANCE 21
4.6 TREATMENT PROCESS EXCURSIONS AND UPSETS, ACCIDENTAL CONTAMINANT/REAGENT RELEASES
21
4.7 SAFETY RECORD 21
vi
-------
5.0 EFFECTIVENESS OF THE SYSTEM TO PROTECT HUMAN HEALTH AND THE ENVIRONMENT . 22
5.1 GROUND WATER 22
5.2 SURFACE WATER 22
5.3 AIR 22
5.4 SOILS 23
5.5 WETLANDS AND SEDIMENTS 23
6.0 RECOMMENDATIONS 24
6.1 ROAD MAP FOR A FINAL REMEDY 24
6.2 HIGH PRIORITY, SHORT-TERM ITEMS 26
6.2.1 SIMPLIFY THE EXISTING TREATMENT SYSTEM (PILOT TEST ALTERNATIVES) 26
6.2.2 INSTALLUPGRADIENTSHEETPILE 27
6.2.3 REMOVE STEAM INJECTION/EXTRACTION SYSTEM AND APPLY CAP 28
6.2.4 MONITOR DRAWDOWN IN FORMER PROCESS AREA AND CONDUCT A WATER BUDGET
ANALYSIS 28
6.2.5 UPGRADE EXTRACTION SYSTEM 29
6.2.6 REPLACE THE EXISTING TREATMENT PLANT 29
6.2.7 POTENTIALLY ENHANCE MONITORING FOR THE LOWER AQUIFER 30
6.3 OTHER RELATED ITEMS 31
7.0 SUMMARY 33
List of Tables
Table 7-1. Cost summary table
List of Figures
Figure 1-1. Wyckoff/Eagle Harbor Operable Units
Figure 1-2. Schematic of OU2 (Ground Water) and OU4 (Saturated Soils)
Figure 1-3. Conceptual Model of NAPL Migration Prior to Remedy Implementation
Figure 1-4. Conceptual Depiction of Hydrogeology at the Site
Figure 2-1. OU2 Layout with Sheet Pile and Well Locations
Vll
-------
1.0 INTRODUCTION
1.1 PURPOSE
During fiscal years 2000, 2001, and 2002 Remediation System Evaluations (RSEs) were conducted at 24
Fund-lead pump and treat (P&T) sites (i.e., those sites with pump and treat systems funded and managed
by Superfund and the States). Due to the opportunities for system optimization that arose from those
RSEs, EPA OSRTI has incorporated RSEs into a larger post-construction complete strategy for Fund-
lead remedies. During fiscal years 2003 and 2004, RSEs at up to eight Fund-lead sites are planned in an
effort to improve or optimize the sites. An independent EPA contractor is conducting these evaluations,
and representatives from EPA OSRTI are attending the RSEs as observers.
The Remediation System Evaluation (RSE) process was developed by the US Army Corps of Engineers
(USAGE) and is documented on the following website:
http://www.environmental.usace.armv.mil/library/guide/rsechk/rsechk.html
An RSE involves a team of expert hydrogeologists and engineers, independent of the site, conducting a
third-party evaluation of site operations. It is a broad evaluation that considers the goals of the remedy,
site conceptual model, above-ground and subsurface performance, and site exit strategy. The evaluation
includes reviewing site documents, visiting the site for one to one and a half days, and compiling a report
that includes recommendations to improve the system. Recommendations with cost and cost savings
estimates are provided in the following four categories:
• improvements in remedy effectiveness
reductions in operation and maintenance costs
• technical improvements
gaining site closeout
The recommendations are intended to help the site team identify opportunities for improvements. In
many cases, further analysis of a recommendation, beyond that provided in this report, might be needed
prior to implementation of the recommendation. Note that the recommendations are based on an
independent evaluation by the RSE team, and represent the opinions of the RSE team.
The Wyckoff/Eagle Harbor Superfund Site was selected by EPA OSRTI based on a recommendation
from EPA Region 10. In particular, the RSE team has been asked to provide input on the remedial
approach at the site, particularly the use of thermal technologies to remediate the site or the use of
engineered barriers and P&T to provide containment. This report provides a brief background for the
site, a summary of observations made during a site visit, and recommendations regarding the remedial
approach. Approximate costs and cost savings associated with the recommendations are also discussed.
-------
1.2
TEAM COMPOSITION
The team conducting the RSE consisted of the following individuals:
Rob Greenwald, Hydrogeologist, GeoTrans, Inc.
Peter Rich, Civil and Environmental Engineer, GeoTrans, Inc.
Doug Sutton, Water Resources Engineer, GeoTrans, Inc.
The RSE team was also accompanied by the following observers:
Jennifer Griesert from EPA OSRTI
1.3
DOCUMENTS REVIEWED
Author
CH2M Hill
CH2M Hill
US EPA
USAGE
USAGE
US EPA
US EPA
US EPA
SCS Engineers
SCS Engineers
CH2M Hill
USAGE
CH2M Hill
USAGE
Date
6/1996
8/1996
12/1997
4/1998
3-4/1999
2/2000
7/2002
9/2002
2002 - 2003
2002 - 2003
5/2003
8/2003
9/2003
10-12/2003
Title
Groundwater Extraction System Assessment Report No. 2
Groundwater Extraction System Assessment Report No. 1
EPA Public Meeting: Proposed Plan for Cleanup of
Contaminated Soil and Groundwater at the Former Wyckoff
Wood Treatment Facility
Offshore Field Investigation Report for the Barrier Wall
Design Project
In-situ Thermal Technology Advisory Panel Meeting Minutes
Wyckoff/Eagle Harbor Superfund Site Soil and Groundwater
Operable Units Record of Decision
Wyckoff/Eagle Harbor Superfund Site Steam Injection
Treatability Study
Five-Year Review, Wyckoff/Eagle Harbor Superfund Site
Wyckoff/Eagle Harbor Superfund Site Thermal Remediation
Pilot O&M Project Monthly Reports (July 2002, October
2002, November 2002, December 2002, March 2003, and
August 2003)
Wyckoff/Eagle Harbor Superfund Site Thermal Remediation
Pilot O&M Project Monthly Chemical Data Reports
(November 2002, December 2002, March 2003, and August
2003)
Wyckoff Steam Pilot Deficiency List
2002-2003 Year 8 Environmental Monitoring Report
Process and Instrumentation Diagrams
http://www.wyckoffsuperfund.com/
-------
1.4 PERSONS CONTACTED
The following individuals associated with the site were present for the visit:
Mary Jane Nearman, Remedial Project Manager, EPA Region 10
Dan Gravning, EPA HQ
Don Heyer, Project Manager, CH2M Hill
Ken Scheffler, Process Engineer, CH2M Hill
Cliff Leeper, Lead Plant Operator, OMI
David Roberson, Plant Operations, SCS Engineers
1.5 SITE LOCATION, HISTORY, AND CHARACTERISTICS
1.5.1 LOCATION AND HISTORY
The Wyckoff/Eagle Harbor Superfund site is located on the eastern side of Bainbridge Island, in central
Puget Sound, Washington along Eagle Harbor. The site includes the inactive 57-acre Wyckoff wood-
treating facility, contaminated sediments in adjacent Eagle Harbor, and other upland sources of
contamination to the harbor, including a former shipyard. The site is currently divided into the following
four operable units, which are depicted in Figure 1-1.
• Operable Unit 1 (OU1) is called East Harbor and has subtidal and intertidal sediments that are
contaminated by poly-aromatic hydrocarbons (PAHs).
• Operable Unit 2 (OU2) consists of 18 acres of unsaturated soil on the Wyckoff property that is
contaminated with PAHs, pentachlorophenol (PCP), and dioxins/furans.
• Operable Unit 3 (OU3) is called West Harbor and consists of metals contaminated sediments and
upland sources.
• Operable Unit 4 (OU4) consists of the contaminated ground water and saturated soil underlying
the soil operable unit (OU2).
The wood-treating facility operated from the early 1900s through 1988, when the plant shut down and the
Wyckoff Company was renamed Pacific Sound Resources. Investigations by EPA began in 1971, when
there were reports of oil on the nearby beaches. The site was listed on the National Priorities List in July
1987. The East and West Harbor operable units have largely been addressed.
This RSE pertains only to the soil and ground water operable units (OU2 and OU4). Figure 1-2 is a
schematic that depicts the area included in OU2 and OU4. The following chronology consists of
excerpts from the September 2002 Five-Year Review and provides a brief summary of activities related
to these two specific operable units.
-------
Approximate Date
January 1990
June 1992 - April 1994
November 1993
September 1994
November 1994
January - June 1995
June - December 1995
November 1997
July 1998
1998-1999
April 1999
September 1999
February 2000
May 2000
February 2001
February 2002
September 2002
October 2002
March 2003
Activity
A groundwater P&T system began operation.
EPA removed approximately 29,000 tons of creosote sludges, disposed of 100,000
gallons of contaminated oils, disposed of 430 cubic yards of asbestos, installed 300
feet of sheet piling, repaired and constructed 150 feet of bulkhead, and recycled
steel from on-site structures.
EPA took control of the P&T system and then made improvements in 1994.
EPA issued an Interim ROD for the Groundwater Operable Unit that included
replacing the treatment system, upgrading the extraction system, installing a
physical barrier, and sealing on-site production wells.
EPA and the Washington Department of Ecology signed the State Superfund
Contract (SSC) for the interim groundwater remedy.
EPA sealed and abandoned 12 on-site production wells.
The seven original extraction wells were replaced by eight new extraction wells.
Other plant upgrades were also made.
EPA issued a "final" Proposed Plan that preferred containment as the cleanup
strategy for soil and ground water.
EPA completed the design for a replacement treatment plant. The plant was not
constructed pending a final decision regarding the ground water remedy.
Long-term O&M associated with the containment strategy concerned the State.
EPA evaluated thermal technologies for possible application at Wyckoff.
EPA completed the Focused Feasibility Study Comparative Analysis of
Containment and Thermal Technologies.
EPA completed conceptual design for thermal remediation and issued a second
Proposed Plan to replace the previous one issued in 1997.
EPA issued a Record of Decision for the soil and ground water remedies
conditionally selecting steam injection as the cleanup remedy.
EPA and Washington Department of Ecology signed the State Superfund Contract
for the soil and ground water remedies.
Over 1,800 lineal feet of sheet pile was installed around the Former Process Area
(two acres of beach were created to mitigate habitat loss) and over 530 lineal feet
of sheet pile was installed within a one-acre area of the site for the steam injection
pilot.
In the pilot area, a vapor cap, 16 injection wells, and seven extraction wells were
installed. Approximately 600 thermal monitoring devices, a boiler building, and
production well were also installed. Soil cleanup of the Former Log Storage/Peeler
Area was completed.
Modifications of the treatment system were made and the boiler system was
installed, including water softeners, heat exchangers, a thermal oxidizer,
compressors, pumps, etc.
Pilot steam injection began. Operation reached approximately 25% capacity with
approximately 50% up-time. Ground water extraction in the Former Process Area
continued during the steam pilot.
Due to technical problems, steam injection was discontinued for further evaluation.
Ground water extraction from the Former Process area continued.
-------
As of the time of the RSE site visit, site team is faced with decision as to how to proceed with the
remedy. The RSE team was specifically asked by EPA Region 10 and OSRTI to consider options for
moving forward, ranging from containment only through full-scale steam injection, given both technical
and fiscal considerations. This RSE report provides a summary of the findings from the site visit and the
RSE team's recommendations for moving forward with the remedy.
1.5.2 POTENTIAL SOURCES
Facility operations included the storage and use of creosote, pentachlorophenol (PCP), and other
hazardous materials, such as solvents and gasoline. Facility operations also included management and
treatment of process wastes and storage of both treated and untreated wood. The release of contaminants
into the environment resulted from daily operations through leaks and spills. OU2 (soil operable unit) is
divided into three areas: the Log Storage/Peeler Area, the Former Process Area, and the Well CW01
Area, which are all depicted in Figure 1-2. The greatest magnitude of contamination is found in the
Former Process Area. Remediation of the Log Storage/Peeler Area has been completed. The
contaminated soils were excavated from this area and permanently placed in the Former Process Area for
remediation during steam injection.
Non-aqeous phase liquid (NAPL) is present at the site in large quantities (estimates suggest that over one
million gallons of NAPL are present in subsurface). The following text from the 2000 ROD describes
the conceptual model for historical NAPL migration at the site and its role as a continuing source of
ground water contamination. This conceptual model is also depicted in Figure 1-3.
LNAPL accumulates at the water-table surface and continues to migrate laterally, eventually
emerging as intertidal seeps in Eagle Harbor and Puget Sound.
DNAPL continues migrating downward. Lateral movement may occur through high-permeability
gravel and cobble zones, or during temporary accumulation on fine-grained layers in the aquifer.
In shoreline areas, downward migration of some DNAPL may be slowed or halted as it encounters
brackish ground water with approximately the same density.
Along the northwest shoreline, DNAPL appears to be perched on clay and silt beds within the
upper aquifer, and has been observed to move laterally through the bulkheads, discharging into the
Log Rafting Area. This discharge appears to have been occurring for several decades,
contemporaneously with sedimentation; the result is several feet of NAPL saturated harbor-bottom
sediments in the Log Rafting Area.
DNAPL entered the lagoon which extended from the Log Rafting Area into the Tram Loading
Area [not shown in RSE report figures], either from the upper aquifer, from surface discharges, or
from treated logs placed in the lagoon. This discharge was apparently contemporaneous with
sedimentation and filling, resulting in as much as 10 feet of NAPL-saturated soil at the bottom of
the old lagoon, now covered with clay fill.
Most of the DNAPL migrates downward through the upper aquifer until it encounters the relatively
low-permeability aquitard layer. The aquitard layer dips toward the north and east. The DNAPL
builds up above the aquitard, forming large accumulations in depressions in the aquitard, and
generally migrating down-dip toward Eagle Harbor.
Small amounts of the DNAPL continue to migrate downward into fractures or sandy zones in the
aquitard. Data from the current explorations indicate that continued downward migration of
-------
DNAPL occurs primarily in the central portion of the site, where the aquitard contains numerous
sand beds and lenses.
Based on the data collected to date, it appears that NAPL has not reached the lower aquifer.
NAPL undergoes dissolution as it encounters ground water in the upper aquifer, resulting in dissolved
contamination. The aqueous-phase contaminants are then transported with the ground water flow, laterally
toward Eagle Harbor. Downward advective transport of dissolved contamination through the aquitard is
unlikely under natural conditions, since the hydraulic head is higher in the lower aquifer than it is in the
upper aquifer.
Although NAPL is now primarily confined to the Former Process Area behind 1,800 linear feet of sheet
pile, source material exists as seeps up to 100 feet beyond the sheet pile wall along the beach east of the
Former Process Area. It is hoped that the current remedy (sheet pile and pumping) will contain the
NAPL and reduce the magnitude of the seeps or eliminate them altogether because natural habitats in the
area would likely be destroyed by active remediation in that area.
1.5.3 HYDROGEOLOGIC SETTING
Figure 1-4 conceptually depicts the hydrogeology at the site. Ground water at the site is approximately 5
to 15 feet below ground surface (bgs). This water table forms the upper limits of the upper aquifer. The
upper 5 to 10 feet of this upper aquifer consists of both fill and native materials that overlay marine sand
containing small amounts of interbedded gravel, silt, and clay. These marine sands range in thickness of
5 to 70 feet. In general, the upper aquifer thickness is approximately 20 feet at the southern (upland)
edge of the site and 75 feet at the northern edge of the site along the harbor. Separating the upper aquifer
from the semi-confined lower aquifer is a relatively impermeable layer that slopes from north to south.
This layer is composed of silt and glacial till and ranges in thickness from 4 to 40 feet. Although it is
considered an aquitard, site documents indicate that this layer has interbedded lenses of sand or other
material that might provide the potential for downward migration of DNAPL. The lower aquifer, which
consists primarily of sand, with small amounts of silt, clay, and gravel, extends to a depth of
approximately 200 feet bgs. Clay layers underlay the lower aquifer and separate it from deeper, potable
water aquifers that range from 200 to 1,500 feet bgs. A 1,800-foot long sheet pile wall along the
shoreline in the Former Process Area extends from a few feet above the tidal high into the glacial till
aquitard, nearly 95 feet bgs.
In general, ground water flows from the upland area at the southern boundary of the site toward the
harbor and is affected by the tidal cycle. Flow may be downward from upper to the lower aquifer at the
upland boundary, but measurements of hydraulic head beneath the Former Process Area generally
suggests upward flow in the area of ground water contamination. The flux of water into and out of the
upper and lower aquifers is not known, but significantly higher flux is expected during the rainy season,
which includes fall, winter, and spring. With the sheet pile in place, preliminary estimates by the site
team suggest that the flux of water into the upper aquifer underlying the Former Process Area is
approximately 35 gpm during the summer (dry season) and over 80 gpm during the remainder of the year
(rainy season). This increase comes from both upland flow that is recharge by precipitation and
infiltration of precipitation through the approximate 8-acre extent of the Former Process Area.
Prior to the installation of the sheet pile wall, the upper aquifer underlying the process area was brackish
with total dissolved solids exceeding 10,000 mg/L, resulting in a Class III designation. The upper aquifer
upgradient of the Former Process Area and the entire lower aquifer in the vicinity of the site has lower
total dissolved solids and has a Class II designation. The sheet pile between the harbor and the Former
Process Area combined with pumping from within the Former Process Area has resulted in a decrease of
-------
the salinity and total dissolved solids in the upper aquifer. The site team expects the upper aquifer
underlying Former Process Area to be predominately non-saline water within a few years.
1.5.4 RECEPTORS
The site is currently secured with a fence. Therefore, current receptors of soil contamination include
underlying ground water, site workers, and visitors. The site workers and visitors adhere to a site-
specific health and safety plan that limits their exposure to soil contamination. Given that future land use
will likely include a park that is open to the public, future receptors to soil contamination would likely
include both ground water and the public. This potential exposure would presumably be eliminated by
the selected remedy.
The primary receptor of ground water contamination is the surface water and sediments of Eagle Harbor
and Puget Sound. Ground water use is present in the area, but is limited to upgradient or side-gradient
locations within the lower or deep aquifers. Historical quarterly ground water sampling at the nearby
supply wells have consistently shown no detectable concentration. The sampling has been discontinued
given that site-related contamination is apparently beyond the influence of these wells.
1.5.5 DESCRIPTION OF GROUND WATER PLUME
The ground water plume is primarily confined to the upper aquifer beneath the Former Process Area.
However, limited ground water contamination has been reported in at least one well in the lower aquifer.
Although no measurable NAPL was detected in the wells, NAPL has been observed on a probe used to
gauge one of the lower aquifer wells. As ground water in the upper aquifer flows toward the harbor, it
discharges to surface water. Therefore, the interface between the subsurface and the harbor marks the
end of the ground water plume. The sheet pile wall has been installed along the shoreline. Some
contamination has historically migrated beyond the location of the wall and some remains in the Former
Process Area. To the east, some of this contamination outside of the wall continues to discharge to the
surface in the form of seeps.
-------
2.0 CURRENT STATUS OF REMEDY
2.1 REMEDY OVERVIEW
The long-term remedy for the ground water and soil operable units targets the Former Process Area. The
current components of the remedy include the following:
• a sheet pile wall installed along the outer boundary of the Former Process Area to help contain
NAPL and contaminated ground water
• a ground water extraction and NAPL recovery system consisting of nine extraction wells
throughout the Former Process Area (only six are active)
• a treatment system to separate the extracted NAPL from ground water and treat the ground water
• pilot injection and extraction systems targeted in a one-acre parcel of the Former Process Area
for steam enhanced recovery
• a boiler system used to produce steam for injection and to destroy non-condensable vapors
recovered during steam enhanced extraction pilot study
Figure 2-1 depicts the OU2 area and indicates portions of the remedy including the treatment system, the
extraction wells, and steam pilot test area.
2.2 CONTAINMENT SHEET PILE WALL
The containment sheet pile wall extends continuously along the western, northern, and eastern
boundaries of the Former Process Area. The southern boundary is open to flow from upgradient. The
sheet pile extends vertically from a few feet above sea level to approximately 95 feet bgs where it is
keyed into the aquitard. Every other seam is welded to reduce the flow of ground water between the
joints, and joint observation points have been installed to monitor flow through eight of the unwelded
remaining seams. No sealant was used during construction of the wall because the use of steam would
likely melt the sealant.
2.3 P&T EXTRACTION SYSTEM
The current extraction system consists of six active extraction wells that extract a total of approximately
35 gpm from the upper aquifer. The wells are eight inches in diameter, are constructed of stainless steel,
and include a 4-foot DNAPL sump. Each of the wells is outfitted with separate above-ground
progressive cavity ground water and NAPL pumps. The ground water pumps operate continuously and
pump water through above-ground HPDE pipe to the head of the treatment system. The NAPL pumps
are operated manually on a periodic basis. Information regarding the active wells from October 2002 is
summarized in the following table.
-------
Extraction
Well
1
2
3
4
5
6
7
8
9
Totals
Average Ground Water
Extraction Rate
(gpm)
4.2
Recovered DNAPL
11/2002
(gallons)
12
Recovered LNAPL
11/2002
(gallons)
42
Inactive
Inactive
5.9
7.5
6.4
0
0
17
0
0
0
Inactive
5.3
4.0
33.4
17
36
82
0
0
42
Total Recoverd
NAPL
(gallons)
7,668
5,663
771
145
9,678
6,440
102
3,588
3,206
37,261
2.4
P&T TREATMENT SYSTEM
The treatment system was originally installed in 1990 by the responsible party but EPA took over
responsibility for O&M in 1993 and made upgrades in 1995. A replacement treatment system was
designed in 1998, but that system was never installed due to uncertainty at the time in the future use of
steam to enhance recovery. The treatment system includes a dissolved air flotation (DAF) unit (which
replaced a nonfunctioning depurator in 2002), a pumping tank, activated sludge system, clarifier,
multimedia filters, and three 8,000-pound GAC vessels. Water is discharged to the harbor in accordance
with an NPDES permit.
The system treats an average flow rate of approximately 35 gpm. In the absence of steam injection,
influent PAH and PCP concentrations are approximately 15 mg/L and 500 ug/L, respectively. This flow
rate and influent concentrations yield a mass removal rate from the dissolved phase of approximately 6.5
pounds per day as calculated below.
35 gal. 3.785 L 15.5 mg. 1440 min.
x x x x
2.2 Ibs. 6.5 Ibs.
min.
gal.
L
day
lxlO°mg day
The DAF unit removes approximately 90% of the PAHs (primarily naphthalene). The aeration basin and
GAC remove the remainder of the contamination (less than one pound per day). The DAF also removes
product. The October 2002 Monthly Operations Report, written prior to the steam pilot study, reported
that approximately 22,000 gallons of NAPL had been recovered from treatment plant operations (i.e., use
of the DAF and its predecessors). In sum, nearly 60,000 gallons of NAPL have been recovered from the
NAPL recovery wells and treatment plant operations.
-------
2.5 PILOT INJECTION AND EXTRACTION SYSTEM
The pilot injection and extraction system includes 16 steam injection wells and seven extraction wells in
a one-acre parcel of the Former Process Area. The extraction wells are completed to the aquitard
(approximately 25 to 30 feet bgs in this part of the Former Process Area) and have 20-foot screened
intervals. The injection wells are also completed to the aquitard have 5-foot screened intervals. All
piping between the wells, steam system, and treatment plant is above ground.
To semi-isolate this system from the rest of the Former Process Area, approximately 530 feet of sheet
pile was installed on the western, northern, and eastern boundaries of the one-acre parcel. As with the
containment sheet pile wall, the southern boundary is open to flow from upgradient. The wall extends
from a few feet above ground surface to approximately 35 feet deep where it is keyed into the aquitard,
which is shallower in this location than it is along the outer boundaries of the Former Process Area. An
HDPE vapor cap is installed at the surface to help contain contaminant vapors that are released from the
subsurface during steam injection. When operating at 25% of capacity at near steady-state conditions in
March 2003, approximately 2,000 pounds of steam per hour were injected through the injection network
and approximately 20 gpm was extracted through the extraction network. No extraction was occurring at
wells EW-1 and EW-7. The pilot steam project extracted approximately 2,200 gallons of NAPL in under
six months, but the pilot project never achieved levels of steam delivery and temperature distribution
originally planned for the test due to a variety of site-specific complications that were encountered.
2.6 STEAM SYSTEM
The steam system includes the following items:
• an extraction well upgradient of the former process area and approximately 300 feet deep that
provides clean water for generating steam
• a water softening system to condition the water for the boiler
• a heat exchanger that condenses recovered vapors and preheats water prior to the boiler
• a boiler to generate steam and combust the recovered vapors that were not condensed
• vapor recovery pumps to recover contaminant vapors from the subsurface
• a thermal oxidizer to combust non-condensable recovered vapors (never used)
• a total organic carbon (TOC) analyzer to quantify recovered contamination
Uncondensed vapors are destroyed by the boiler flame, and the condensed vapors are sent to the
treatment plant.
2.7 MONITORING PROGRAM
The monitoring program for OU2/OU4 includes sampling of water quality from approximately 10 to 12
wells within the Former Process Area and weekly process monitoring. The weekly monitoring consists
10
-------
of approximately eight samples throughout the treatment plant that are analyzed for PAHs and PCP plus
a few other samples that are analyzed for oil and grease or solids. The weekly samples include the
effluent samples that are needed to meet the NPDES permit requirements. Quarterly biomonitoring is
also conducted on the effluent as directed by the NPDES permit. Laboratory analysis is provided by the
EPA Regional Laboratory at no cost to the site.
11
-------
3.0 REMEDY OBJECTIVES, PERFORMANCE, AND CLOSURE
CRITERIA
3.1 REMEDY OBJECTIVES AND CLOSURE CRITERIA
The remedial action objectives (RAOs) for the soil operable unit, as identified in the February 2000 ROD
are as follows:
prevent human exposure through direct contact (ingestion, inhalation, or dermal contact) with
contaminated soil
prevent storm water runoff containing contaminated soil from reaching Eagle Harbor
The soil Log Storage/Peeler Area has already been addressed. The Former Process Area is the only
remaining area of the site where contaminated soils still need to be addressed.
The RAOs for the ground water operable unit, as identified in the February 2000 ROD are as follows:
reduce the NAPL source and the quantity of NAPL leaving the upper aquifer beneath the Former
Process Area sufficiently to protect marine water quality, surface water, and sediments (e.g.,
ensure the quantity of NAPL leaving the site will not adversely affect aquatic life and sediments)
ensure contaminant concentrations in the upper aquifer ground water leaving the Former Process
Area will not adversely affect marine water quality, and aquatic life in surface water and
sediment
protect humans from exposure to ground water containing contaminant concentrations above
MCLs
protect the ground water outside the Former Process Area and in the lower aquifers, which are
potential drinking water sources
The ROD states that site-specific ground water contaminant concentrations are to be met at the mudline.
However, the addition of the sheet pile wall introduces another potential point of compliance. For the
upper portion of the sheet pile wall, where the surface water is present on the exterior of the wall, the
interior side of the sheet pile wall is considered the point of compliance. This is because the sheet pile is
relatively thin (less than one inch) and there is no expected attenuation of the contaminants over this
short distance. For the lower portion of the sheet pile wall, where sediments are present on the exterior
of the wall, the mudline (i.e., the interface between the sediments and the surface water) is the point of
compliance.
The marine water quality, surface water, and marine sediments in Eagle Harbor are the media of primary
concern. The following table summarizes the ground water cleanup standards for the Wyckoff site.
These standards are the most stringent of the State and Federal marine water quality standards, risk-based
surface water standards for human consumption of organisms, and calculated pore-water maximums
based on Sediment Management Standards for marine sediments.
12
-------
Contaminant of Concern
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo(a)anthrancene
Chrysene
Benzo(b)fluoranthene
Benzo(k)fluoranthene
Benzo(a)pyrene
Dibenzo(a,h)anthracene
Benzo(g,h,i)perylene
Indeno( 1,2,3 -cd)py rene
HP AH
Pentachlorophenol
Ground Water Cleanup Level* (ug/L)
83
3
3
9
3
15
0.0296
0.0296
0.0296
0.0296
0.0296
0.007
0.0296
0.254
4.9
* Where there is no cleanup level specified for a certain chemical, benzo(a)pyrene will be used as
an indicator chemical during remediation. Ground water cleanup levels will be measured at the
point of compliance.
3.2
TREATMENT PLANT DISCHARGE CRITERIA
An NPDES permit provides the criteria for the discharge from the treatment plant. The criteria for PAHs
and PCP are provided below. The table below does not provide the criteria for metals, inorganics, or
biomonitoring. Sampling for all parameters (except biomonitoring) is required on a weekly basis.
Contaminant of Concern
Total of 16 PAHs
Individual PAHs
Naphthalene
Acenaphthylene
Acenaphthene
Discharge Criteria
(ug/L)
20
4
4
4
13
-------
Contaminant of Concern
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo(a)anthrancene
Chrysene
Benzo(b)fluoranthene
Benzo(k)fluoranthene
Benzo(a)pyrene
Dibenzo(a,h)anthracene
Benzo(g,h,i)perylene
Indeno( 1 ,2,3 -cd)pyrene
Pentachlorophenol
Discharge Criteria (ug/L)
2
2
2
2
2
2
2
2
2
2
2
2
2
6
For some of the PAHs, the cleanup standards are orders of magnitude lower than the discharge standards.
The RSE team notes this disconnect between these cleanup and discharge standards, particularly since
the cleanup standards are based on marine water quality and the discharge standards apply to marine
water.
14
-------
4.0 FINDINGS AND OBSERVATIONS FROM THE RSE SITE VISIT
4.1 FINDINGS
The RSE team observed a knowledgeable and competent site team led by an effective, motivated, and
organized EPA RPM. The observations provided below are not intended to imply a deficiency in the
work of the system designers, system operators, or site managers but are offered as constructive
suggestions in the best interest of the EPA and the public. These observations obviously have the benefit
of being formulated based upon operational data unavailable to the original designers. Furthermore, it is
likely that site conditions and general knowledge of ground water remediation have changed over time.
4.2 SUBSURFACE PERFORMANCE AND RESPONSE
4.2.1 NAPL AND PLUME CAPTURE
The OU2 and OU4 contamination is located primarily within the Former Process Area and is surrounded
on three sides by 1,800 feet of sheet pile. The sheet pile separates the Former Process Area from Eagle
Harbor and Puget Sound but leaves it open to ground water flow from the upgradient highlands. Because
the sheet pile is not completely impermeable (leaks at the seals), containment of dissolved contamination
can only be provided if inward hydraulic gradients are established on all sides of and beneath the Former
Process Area. For this to occur, more water should be extracted from the Former Process Area than
enters it. Because flow from Eagle Harbor and Puget Sound are limited by the barrier wall, extraction is
primarily needed to offset ground water flow from upgradient, infiltration from the surface, and
infiltration from the underlying formation. The P&T system currently pumps approximately 35 gpm, and
a hydrogeologic investigation is underway to determine how much water should be extracted to offset
incoming flow. Preliminary estimates suggest that 35 gpm might be sufficient during the dry season (i.e.,
summer) but that more than 80 gpm might be required during the wet season (i.e., fall, winter, and
spring). Therefore, it is likely that the current system does not provide containment of dissolved
contamination during three of the four seasons of the year. Containment may also not be sufficient
during low tide, because at low tide, the water elevation in Eagle Harbor and Puget Sound is likely much
less than the water table in the Former Process Area.
Even if hydraulic gradients can establish containment of dissolved contamination, mobile DNAPL can
potentially migrate through sand beds or other permeable material dispersed throughout the glacial till
aquitard that separates the upper and lower aquifers beneath the Former Process Area. To date, a limited
amount of contamination has been found in the lower aquifer; however, a more extensive sampling
program could identify additional contamination. If contamination in the lower aquifer is in fact limited
after nearly a century of wood treating operations at the former facility (and remediation that has only
taken place over the last decade), it suggests that migration through the aquitard is possible but likely
limited. Of the 10 monitoring wells sampled during November 2002, five of them (CW-05, CW-09, CW-
15, 99CD-MW02, and 99CD-MW04) are screened in the upper portion of the lower aquifer (i.e., below
the aquitard). The other five wells are screened in the upper aquifer. The lower-aquifer wells show
detections of dissolved contamination. Few of the detections are above the cleanup criteria, and these are
primarily associated with CW-015. These detections indicate contamination has migrated to the lower
aquifer, but it is unknown to what degree the migration occurred in the past and to what degree it
15
-------
continues to occur. Although no measurable NAPL has been found in the lower aquifer, one of the
probes used to gauge a lower-aquifer well had NAPL on it when it was retrieved from the well.
Despite efforts to contain contamination within the Former Process Area, there is NAPL contamination
in the form of seeps that is present outside of the sheet pile along the Puget Sound shoreline. This area is
marked on Figure 2-1. The site team reports that when the tide is low, a sheen is often visible. This
contamination was discovered during the offshore field investigation for the design of the barrier wall
and is documented in the associated report dated April 1998. No specific remedial actions are underway
to address these seeps, and it is not clear that a feasible remedy exists for this contamination given the
natural habitat that would be destroyed by remedial activities. It is hoped that containment of the upper
aquifer beneath the Former Process Area will cut off the NAPL migration pathway and reduce the
magnitude of the seeps or eliminate them completely.
4.2.2 CONTAMINANT LEVELS
Contaminant levels in the Former Process Area are significantly above marine water quality standards as
would be expected given the presence of NAPL.
4.2.3 STEAM INJECTION PILOT RESULTS
A full report of the steam pilot test is forthcoming from site contractors. The summary provided herein is
intended to provide information summarized to the RSE team that is pertinent to the recommendations
contained in this RSE report.
The steam pilot test operated primarily between late October 2002 and April 2003, a total of
approximately 160 days. On approximately 50 of those 160 days, no steam was injected due to technical
issues or problems resulting from use of this experimental technology. The steam system only operated
at full capacity for about 30 days. The system appeared to operate at a relatively steady rate between
mid-January and late March of 2003 (approximately 2.5 months), but this was a level of steaming
substantially below the intended levels for the pilot test. This steady rate consisted of approximately
2,000 pounds per hour of steam injected and approximately 22 gpm of total fluids extracted. Over this
portion of the pilot, approximately 1,500 gallons of NAPL were removed according to total organic
carbon analysis of the effluent presented in the August 2003 Monthly Report. An additional 700 gallons
were removed during periods before and after this approximate 2.5-month period of steady operation.
Approximately 60,000 gallons are estimated to be present in the pilot test area. Additional mass removal
would have been expected from the NAPL and vapor phase recovery if the aquifer was sufficiently
heated. However, the current treatment processes were overwhelmed even with the limited steaming, and
if full heating of the pilot area did occur, treatment capacity would not have been available to address the
extracted contaminant mass.
Temperatures in some areas of the pilot test reached above 100 degrees Celsius, but temperatures
immediately below the cap (where LNAPL would be expected) or along the aquitard (where DNAPL
would be expected) were generally between 40 and 80 degrees Celsius. This limited the amount of
NAPL recovery, and was due to a series of site-specific complications associated with the stratigraphy of
the site, the capacity of the existing liquid and vapor treatment systems, and fouling of equipment and
pipes from crystallization of naphthalene.
16
-------
4.3 COMPONENT PERFORMANCE
4.3.1 CONTAINMENT SHEET PILE WALL
The containment sheet pile wall has performed as expected. It has reduced the exchange between the
upper aquifer and Eagle Harbor. The site team has already observed a decrease in the salinity of the
upper aquifer because ground water has been extracted, flow of salt water from the Harbor has been
reduced, and fresh water is entering from both upgradient and from the upward flux from the lower
aquifer. The primary concern regarding this sheet pile wall is scouring and corrosion of the portion in
the intertidal zone. Although the wall is supposed to last for 50 years, the site team believes that this
corrosion and scouring will likely reduce the lifetime of this portion of the wall if it is not protected. The
site team is currently considering strategies to protect the wall, such as placement of sand and rip rap
armor along its exterior side. Subsequent to the RSE visit, the RSE team was informed that placing
armor along the exterior side may not be feasible due to impacts that would result to the intertidal zone.
4.3.2 P&T EXTRACTION SYSTEM
The P&T extraction system has functioned as expected. Sacrifical zinc anodes are placed on each pump
to avoid corrosion in the brackish water. The piping from the extraction system to the treatment plant are
cleaned approximately once every 18 months due to biofouling. The piping and pumps are above
ground, which could provide a complication for future land use if pumping is to continue indefinitely.
4.3.3 TREATMENT SYSTEM
Although the activated sludge tank has been very effective for PAH and PCP removal, it is in poor
condition and the existing system as a whole is overly complex and difficult to operate for treating a
relatively low flow rate and manageable PAH and PCP concentrations. Additionally, the system has had
excessive corrective maintenance requirements due to degradation or corrosion because of chemical
incompatibility. For example, during October 2002, two portions of the above-ground piping cracked
(likely due to corrosion) causing contaminated liquid to discharge to the treatment pad surface. The site
contractor also indicates a number of improvements that are required for current treatment system. Some
of those are listed below.
The existing wet well tank on the discharge of the aeration basin may fail causing the treatment
system to be shut down. A new parallel tank should be added to provide a contingency.
New diffusers should be provided for the aeration basin.
The existing aeration basin blowers are worn out and need to be replaced.
A new polymer feed system is required, a temperature controlled polymer storage tank is
required (current storage location is unacceptable).
Pumps require rebuilding and seals require replacement.
Heat tracing is marginal and offers protection to the plant for temperatures down to
approximately 25 degrees Fahrenheit, and improvements are necessary.
Electrical system upgrades are needed
17
-------
A number of other issues would need to be addressed if steam injection is to be used again in the future.
These issues include (but are not limited to) replacing the undersized aeration basin and clarifier,
providing knockout chambers for removal of naphthalene crystals, and improving solids handling
capabilities. The existing groundwater treatment plant was a rate limiting step with regard to ground
water and NAPL extraction during the steam pilot test.
In general, the existing groundwater treatment system needs to be replaced whether or not the remedy
calls for P&T only or enhanced recovery with steam injection. However, the treatment system
components and expense could differ greatly on whether or not the system is designed to accommodate
influent associated with steam injection. A design for a modified system was proposed in 1998, but a
replacement treatment plant was not constructed due to the pending change in the preferred ground water
remedy. Approximately $250,000 of upgrades were made to help the system provide treatment during
the steam injection pilot, but the system was generally inadequate in terms of capacity during the steam
pilot test and suffered from a number of technical problems.
4.3.4 PILOT STEAM INJECTION AND EXTRACTION SYSTEM
The pilot steam injection and extraction system never reached the expected capacity and performance as
a result of several site-specific technical problems. First, the high concentrations of naphthalene in the
vapor phase led to crystalization and fouling of the extraction lines and a naphthalene wax buildup on the
extraction pump screens. Second, the bottom of the aquifer, where DNAPL is located, did not reach the
high temperatures that were expected and therefore did not increase DNAPL recovery along the aquitard
as high as expected. Finally, the rate of extraction was limited by the capacity of the existing treatment
system. In particular, the aeration basin and clarifier were undersized for treating the extracted
contaminant mass.
4.3.5 STEAM GENERATION SYSTEM
The steam generation system also encountered site-specific technical problems that prevented the pilot
program from reaching the expected capacity. Some of those problems were as follows:
• Naphthalene crystalization caused fouling at a number of locations.
• The liquid ring vacuum pumps had seals that were incompatible with naphthalene, and the
moisture knock out tank was too small causing the seal to become displaced.
• The heat exchanger seals for the vapor condenser melted.
• Using the boiler for destroying off-gas may result in dioxin production and emission.
• The thermal oxidizer was never used because the blower would extinguish the flame.
According to the site contractor, a number of other site-specific issues directly associated with the steam
system would need to be addressed prior to using it further.
18
-------
4.4
COMPONENTS OR PROCESSES THAT ACCOUNT FOR MAJORITY OF
OU2 AND OU4 COSTS
Site activities are currently funded through a trust established by Pacific Sound Resources.
Approximately $ 100 million has been spent to date on all four operable units. Of the remaining funds in
the trust, approximately $4 million are allocated to this site, and after that $4 million is spent, the site
activities will be funded by EPA and the State. It is difficult to project accurate cost estimates for future
activities because the steam pilot test has ended, the existing groundwater treatment system requires
replacement, the further use of steam is uncertain, and a number of other site improvements are required.
If no further steam injection is conducted, the site contractors preliminarily estimate that approximately
$20 million might be spent over the next three years for OU2 and OU4. A breakdown of those estimated
costs, as provided by the site contractor, is provided in the following table. Although these expenditures
are expected to be somewhat representative of costs for fiscal years 2004 through 2006, they should not
be used to extrapolate costs beyond fiscal year 2006.
With regard to steam injection, approximately $10 million was spent on attempting the steam pilot test.
If full-scale steam injection is used in the future, the cost of the remedy might be $60 to $80 million in
addition to the above-mentioned $20 million. Such cost estimates would obviously be a function of the
time estimate assumed to reach site closeout as a result of the steam technology. If full-scale steam did
not achieve site closeout, additional costs would then be required for continued P&T operation that
would likely be needed to provide containment of the remaining contamination.
Estimated Costs for Fiscal Years 2004 through 2006 (without Steam Injection)
Item Description
Site Improvements
Upgrades to treatment plant
Mothball steam equipment
Upgradient sheet pile wall
Bentonite seal for upper portion of outer sheet pile wall
Low permeability site cap
Shoreline improvements
New water treatment plant
New/rebuilt extraction wells and discharge piping
Removal of steam equipment and pilot wells
Site Improvements Subtotal
O&M and Engineering
Carbon usage
Disposal of materials
Operator and sampling labor, utilities, materials, etc.
Engineering and construction management
O&M and Engineering Subtotal
Total Estimated Cost
Estimated Cost
$630,000
$60,000
$1,180,000
$850,000
$3,580,000
$2,090,000
$3,500,000
$900,000
$140,000
$12,930,000
$30,000
$400,000
$3,960,000
$2,350,000
$6,740,000
$19,670,000
Note: These costs do not include costs for USAGE oversight and contract management.
19
-------
The items in the above table are briefly discussed below:
Site Improvements
Upgrades to the current treatment plant - These expenditures are deemed necessary by the site contractor
to keep the current treatment plant operating effectively while a new treatment plant is designed and
constructed.
Mothball the steam pilot equipment - This includes a $40,000 estimate to properly store the equipment
and $20,000 to fix the thermal oxidizer.
Upgradient sheet pile wall - Installing this wall would prevent ground water flow from the highlands
from entering the Former Process Area. With installation of this wall, the Former Process Area would be
completely enclosed by sheet pile, thereby reducing the amount of water that would require extraction
and treatment. The cost is based on the costs of installing the original 1,800 feet of sheet pile adjusted
for the shallower sheet pile depth required at the upgradient location.
Bentonite seal for upper portion of outer wall - A 30-inch wide slurry wall is proposed along the upper
portion of the sheet pile along the shoreline to provide an attenuation zone between the Former Process
Area and surface water. Because no attenuation zone currently exists where sediments are not present on
the outside of the sheet pile, the inside of the sheet pile is considered the compliance point. With the
addition of this slurry, an attenuation zone would exist and the compliance point would theoretically be
the outside of the slurry wall. Because sediments offer an attenuation zone, this slurry would only be
placed along portions of the wall where surface water (and not sediments) is present on the outside of the
sheet pile.
Low permeability site cap - This item includes both grading and contouring along with the installation of
a low permeability cap to prevent precipitation from infiltrating into the contaminated area enclosed by
the sheet pile and to prevent exposure to the public during future land use. The cost of the site grading
and contouring is estimated at approximately $500,000, and the cost of the low permeability cap is
estimated at over $3 million.
Shoreline improvements - Approximately $1.7 million is estimated for using sand and rip rap to protect
the outer sheet pile from corrosion. An additional $300,000 is estimated for beach mitigation and
meeting requirements associated with the Shoreline Management Act.
New water treatment plant - A cost of $3.5 million is expected for the cost of a new water treatment
plant. This cost estimate assumes a treatment capacity of approximately 25 gpm, which is lower than the
treatment capacity of the existing system.
New/rebuilt extraction wells and outfall - Approximately $900,000 is estimated for upgrading the
extraction system, the extraction system piping, and the discharge outfall to be inaccessible to the public
in the future.
Removal of steam equipment and pilot wells - Approximately $140,000 is estimated to remove the steam
equipment and abandon the wells in the pilot area. This estimated cost includes estimated refunds
associated with selling the boiler and unused fuel.
20
-------
O&M and Engineering
Carbon usage - The cost assumes replacement of one 8,000-pound unit per year at approximately $1.10
per pound, which is consistent with current costs and usage.
Disposal of materials - Approximately $400,000 is estimated for disposing of material already on site as
well as drummed materials, recovered product, and biosolids through fiscal year 2006 (a three-year
period).
Treatment plant O&M - Based on an estimated cost of $110,000 per month, the contractor estimates that
the cost for P&T operation for three years (i.e., through fiscal year 2006) is just under $4 million. The
treatment plant is operated by at least two people, eight hours per day, seven days per week. Other
contributors to this $110,000 per month costs might include monthly operations reports and sample
collection.
Engineering and construction management - The contractor estimates approximately $2.35 million for
engineering and construction management over the next three years. The contractor assumes that $1
million will be required in fiscal year 2004, $800,000 will be required in fiscal year 2005, and $550,000
will be required in fiscal year 2006.
4.5 REGULATORY COMPLIANCE
The site regularly meets its treated water discharge criteria. If steaming were to continue, further studies
would be necessary to evaluate the compliance of air discharges from the combustion of the off-gas with
the boiler and/or thermal oxidizer.
4.6 TREATMENT PROCESS EXCURSIONS AND UPSETS, ACCIDENTAL
CONTAMINANT/REAGENT RELEASES
Although leaks have occurred from pipes and seals, no significant contaminant mass is known to have
discharged to the environment as a result of the OU2 and OU4 activities.
4.7 SAFETY RECORD
The site team has an excellent safety record. The contractors stated that no reportable incidents occurred
during the steam pilot test. A site-specific health and safety program is in place to minimize exposure of
workers to site-related contamination.
21
-------
5.0 EFFECTIVENESS OF THE SYSTEM TO PROTECT HUMAN
HEALTH AND THE ENVIRONMENT
5.1 GROUND WATER
Ground water use is present in the area, but is limited to upgradient or side-gradient locations within the
lower or deep aquifers. Historical quarterly ground water sampling at the nearby supply wells have
consistently shown no detectable contaminant concentrations. Therefore, site conditions appear
protective of human health with respect to ground water use.
The primary receptor of ground water contamination is the surface water and sediments of Eagle Harbor
and Puget Sound. There are two likely mechanisms for contamination to enter these surface water bodies
under current conditions. The first is the contamination that already exists outside of the sheet pile in the
form of seeps along the eastern side of the Former Process Area in Puget Sound. These seeps may
become inactive in the future due to the presence of the sheet pile, which will mitigate a potential route
for continued NAPL migration. The second is the potential for contamination (dissolved or NAPL) to
seep from the Former Process Area through the sheet pile seams into either Eagle Harbor or Puget Sound.
Because of the pumping, inward gradients are likely established between the surface water and the inside
of the sheet pile (during the majority of the tidal cycle) and the amount of contamination seeping through
the sheet pile is likely minimal relative to the contamination already outside of the sheet pile.
The lower aquifer is also a potential receptor of contamination. To date, a limited amount of
contamination has been found in the lower aquifer. Because upward gradients are established from the
lower to the upper aquifer, any contamination in the lower aquifer is likely due to DNAPL that may have
migrated downward under the influence of gravity. Based on current monitoring, impacts to the lower
aquifer appear to be minimal. However, as discussed in Section 4.2.1, the current monitoring network
may need to be expanded to improve confidence in that conclusion.
5.2 SURFACE WATER
As discussed above, surface water is the primary receptor of site-related contamination. The RSE team
has not evaluated surface water concentrations. The remedy is not currently protective of surface water,
but the site team knows this and is working toward a remedy that will hopefully be protective of surface
water.
5.3 AIR
With the exception of naphthalene, site contaminants are not particularly volatile, and air is not one of
the primary routes of exposure to ground water or saturated soil contamination. If steam is used in the
future, the system should be designed to incorporate adequate vapor off-gas treatment.
22
-------
5.4 SOILS
Soil contamination is present in saturated zone and potentially in the unsaturated zone. A health and
safety plan is currently in place to minimize conduct with contaminated soils, and the final remedy
should provide adequate protectiveness.
5.5 WETLANDS AND SEDIMENTS
Wetlands are not associated with this site, and sediments are potentially threatened as contaminated
ground water migrates toward surface water (see above).
23
-------
6.0 RECOMMENDATIONS
Cost estimates provided herein have levels of certainty comparable to those done for CERCLA
Feasibility Studies (-307+50%), and these cost estimates have been prepared in a manner consistent with
EPA 540-R-00-002, A Guide to Developing and Documenting Cost Estimates During the Feasibility
Study, July 2000.
RSEs generally classify recommendations for operating remedies into four categories:
• improvements in remedy effectiveness
reductions in operation and maintenance costs
• technical improvements
gaining site closeout
This site, however, has an interim remedy that will shortly be replaced by a final remedy. Therefore,
rather than provide recommendations associated with the operating interim remedy, the RSE team is
providing ideas to consider as the final remedy is chosen, designed, and implemented. EPA specifically
asked the RSE team for its position on the merits of remediation via full-scale steam injection versus
long-term containment with P&T and physical barriers. The following sections describe the ideas and
opinions regarding the final remedy for OU2 and OU4 that the RSE team developed based on the site
visit and a review of background documents. Consideration is given to both protectiveness and cost-
effectiveness.
A table summarizing the cost and protectiveness implications of the recommendations is provided in
Table 7-1, which is located at the end of Section 7.0.
6.1 ROAD MAP FOR A FINAL REMEDY
Based on the reviewed information, the RSE site visit, and subsequent discussion with the site team, the
RSE team believes that the best approach for this site is to initially focus efforts on hydraulic isolation of
the contamination underlying the Former Process Area, including the installation of an upgradient barrier
wall and low permeability cap to minimize the amount of water requiring treatment, and the
implementation of a new groundwater treatment system based, in part, on pilot testing of one or more
new approaches. Enhanced monitoring of groundwater in the lower aquifer is also recommended in
conjunction with these efforts, to improve the potential to detect current or future impacts to that aquifer.
It is also recommended that the site team monitor the active seeps along the eastern beach, and take
remedial actions if the seeps persist and a feasible remedial alternative is identified.
Once these high priority items are addressed and implemented, the site team could then reconsider
aggressive mass removal and the technologies that might be available at that time (potentially including
but not limited to additional efforts related to steam injection). However, the RSE team believes the most
cost-effective approach is to design and implement the new groundwater treatment system associated
with hydraulic isolation (discussed above) independent of such efforts. This will reduce the potential of
over-designing the groundwater treatment system required for hydraulic isolation. Cost/benefit
evaluations for subsequent testing or implementation of more aggressive source removal would need to
incorporate costs that might be required to further upgrade the groundwater treatment system above and
24
-------
beyond the treatment system associated with hydraulic isolation. The RSE team also believes that, if
more aggressive source removal technologies are considered in the future, that the costs and benefits of
installing additional recovery wells and tying them into the P&T system should be included as a potential
alternative.
During the RSE site visit there was discussion about armoring the existing sheet pile wall to prevent
scour and extend the life of the wall, and there was also discussion about adding a second sheet pile wall
inside the existing wall to create an "attenuation zone" that could be monitored and provide a backup
barrier in case of premature failure of the outer sheet pile wall. The RSE team felt that, if armoring was
pursued, then the interior sheet pile wall would not be necessary because the armoring could likely be
constructed in a manner to allow for an attenuation zone that could be monitored. Subsequent to the RSE
site visit, the RSE team was informed that armoring may not be feasible due to potential impacts to the
intertidal zone. It is likely that a variety of alternatives will need to be considered in the future regarding
this issue, and this issue is not addressed in detail in this RSE report.
As is evident from the pilot test, steam injection requires a substantial level of effort and money that can
potentially detract from hydraulic isolation and other important issues. The existing treatment system is
currently in a state of disrepair and failure of some components could disrupt the extraction and treatment
required for continued hydraulic isolation. Furthermore, the cost of non-routine maintenance of the
existing system is rising. The site contractor indicates that over $600,000 is required in the short-term to
keep the existing system operational. The long-term cost estimate for full-scale steam application is $60
to $80 million. The RSE team believes that, if the focus were to return to steam injection as the initial
priority at this point, then the funding and level of effort that is required to address the other priority
items suggested above may be diverted to address steam-specific considerations. Therefore, the other
high priority items should be addressed first.
The RSE team, however, does not necessarily suggest that steam enhanced recovery or other aggressive
remediation should never be used at the site. While the site moves forward with providing a protective
remedy based on the high-priority items suggested above, technological progress will likely be made in
the application of steam enhanced recovery and other aggressive source remediation approaches. The
site team can reconsider more aggressive source removal options in the future. The following is a partial
list of items that should be considered when evaluating whether or not to purse more aggressive source
removal in the future:
• Will more aggressive source removal action shorten the estimate of cleanup duration, and if so,
what is the uncertainty in that estimate?
• What is the estimated life-cycle cost impact of more aggressive source removal, and how does
that calculation depend on the cleanup time estimate and related uncertainty of that estimate?
• Will more aggressive source removal require modifications to the new groundwater treatment
system, and have those modifications been accounted for in the cost estimates?
• If further steam injection is considered, will the problems encountered in the pilot test (such as
heating the lower part of the aquifer and the naphthalene crystallization) be overcome with
alternative engineering, and have additional site-specific complications been anticipated?
• Is there an increased potential to mobilize DNAPL flow downward to the lower aquifer as a
result of the more aggressive source removal?
25
-------
• Are increased health and safety protocols included to account for any increased potential hazards
associated with more aggressive source removal?
• Are there potential negative impacts from more aggressive source removal (such as noise, odor,
etc.)? Can such impacts (if any) be mitigated through reasonable engineering efforts, and if not,
do the potential benefits of the action outweigh these impacts?
The site team can decide whether or not to pursue more aggressive source removal in the future by
evaluating these (and potentially other) considerations as part of a detailed cost/benefit analysis and an
evaluation of potential costs versus available resources.
6.2 HIGH PRIORITY, SHORT-TERM ITEMS
To improve hydraulic isolation of site contamination expeditiously and cost-effectively, the RSE team
recommends addressing the following high priority, short-term items below.
6.2.1 SIMPLIFY THE EXISTING TREATMENT SYSTEM (PILOT TEST ALTERNATIVES)
The contractor indicates that the treatment system requires approximately $630,000 in upgrades/repairs
for it to operate reliably over the next year or two while decisions are made regarding other remedial
components. This assumes continued operation of the biosystem with a three-year cost of approximately
$4 million. A greatly simplified system, however, might provide adequate treatment at reduced capital
and annual costs. This simplified system would bypass the biological treatment components and use the
GAC units to treat the effluent from the DAF unit. The reduction in costs would be dependent on GAC
usage rates and other factors.
The RSE team was provided with a number of chemical analysis reports that provide the analytical
results from process water sampling. Two of those reports (October 2002 and August 2003) were from
periods where the steam injection was not occurring and therefore are seen as representative of the
concentrations that would be expected at the site with the current extraction system. The data from these
two reports suggest that the effluent from the dissolved air flotation (DAF) unit (i.e., the influent to the
rest of the treatment plant) has less than 1,500 ug/L of PAHs and PCP combined and approximately 10
mg/L of total suspended solids. As shown in the calculation below, this influent concentration and the
current flow rate of under 40 gpm translates to a mass loading of less than one pound of PAHs and PCP
per day after the DAF.
40 gal. 3.785 L 1,500 ug. 1440 mrn. 2.2 Ibs. 0.72 Ibs.
X X X X g =
min. gal. L day Ix 10 mg day
For treating the DAF effluent, the theoretical GAC usage rate calculated with published isotherms is
approximately 10 pounds of GAC per pound of contaminants in the DAF effluent. A more conservative
estimate of the GAC usage rate might be 20 pounds of GAC to one pound of contaminants in the DAF
effluent. Using this usage rate of 20 to 1, approximately 8,000 pounds of GAC might be required per
year. The Wyckoff GAC units are approximately 8,000-pound units; therefore, according to these
calculations, one replacement would be required per year. Actual GAC usage, however, does not always
follow the published isotherms due to solids loading to the GAC or other factors. Site-specific data
would provide a more accurate representation of GAC usage. Site-specific GAC usage data in the
absence of the biosystem is not available. Therefore, the RSE team recommends proceeding with a pilot
26
-------
test in which the GAC usage rates are evaluated based on treating the effluent from the DAF unit without
the biosystem operating.
If the pilot suggests that the GAC usage rate is sufficiently low, the cost of operation and maintenance
could be significantly reduced by permanently bypassing the biosystem. For example, one full time
operator should be sufficient for operating a treatment plant with a DAF, filters, and GAC, but given the
age of this system, the potential need for repairs, manual NAPL recovery, and other responsibilities at the
site, a full time technician might also be appropriate. The labor requirements for system operation
should therefore decrease from five full time employees to two full time employees. Other costs savings
should be realized as well. The utility costs should decrease because the aeration blowers for the
biological system would not be needed, and the disposal costs should decrease because the biosolids
would not be generated. Until the treatment system is replaced, operator labor costs might be $250,000
per year and electrical might be $20,000 per year. Other costs for ground water sampling, project
management, O&M reporting, and analytical costs (for OU2/OU4) should be well under $200,000 per
year for this simpler system. The GAC cost would depend on the usage rate and would cost
approximately $9,000 per changeout. The above calculations suggested that a single changeout per year
might be possible, but a more conservative estimate would be appropriate. If GAC changeouts were
required quarterly ($36,000 per year), monthly ($108,000 per year), or biweekly ($234,000 per year), the
total annual O&M cost would range from $506,000 (quarterly changeouts) to $704,000 (biweekly
changeouts).
For a three year period, this translates to approximately $1.5 million to $2.1 million and represents a
substantial decrease from the approximately $4 million estimated by the contractors using the current
system. Furthermore, eliminating the biological treatment component and simplifying the system should
eliminate most of the $630,000 in capital expenditures that would have been necessary to keep the
current treatment system operating over the next few years. Process monitoring might also be reduced
for this simpler system. This reduction in process sampling would reduce the workload on the plant
operators and would prevent the Regional EPA laboratory from analyzing unnecessary samples.
Therefore, a pilot test of the alternate treatment approach discussed above is strongly recommended by
the RSE team to reduce O&M costs over the next three years and to serve as a basis for designing and
implementing a cost-effective long-term groundwater treatment system that will replace the current
groundwater treatment system.
6.2.2 INSTALL UPGRADIENT SHEET PILE
Installing the upgradient sheet pile wall will help isolate the contamination in all horizontal directions.
This should significantly reduce the amount of ground water entering the system and should therefore
decrease the amount of water that requires extraction and treatment. Without the upgradient wall, the
site team's preliminary estimate is that approximately 80 gpm or more would require extraction. With
the upgradient sheet pile wall, the influx of ground water should be reduced to limited flow through the
sheet pile seams, upflow from the underlying aquifer, and recharge from infiltration. With this limited
influx, pumping at current levels may be sufficient to maintain an inward gradient through all or most of
the tidal cycle. The cost of installing the wall would be offset by savings from designing a lower
capacity treatment system and reduced O&M costs over an indefinite number of years.
6.2.3 REMOVE STEAM INJECTION/EXTRACTION SYSTEM AND APPLY CAP
Steam activities should not take place for a number of years for the reasons stated in Section 6.1.
Therefore, it is reasonable to remove the steam injection and extraction system and associated steam
27
-------
equipment from the site. This would allow a suitable cap to be placed on the Former Process Area. The
RSE team defers to the site team's estimate of approximately $140,000 for these activities.
Once the steam injection and extraction system have been removed, the surface will be ready for surface
grading, contouring, and applying a cap. By applying the cap, the site team will be able to work on clean
surface material and will be able to make extraction system upgrades (new wells and piping) through
clean surface material. If the cap is delayed until after the extraction system upgrades, then the site team
will be working in potentially contaminated soil and future repairs or replacement of the piping or wells
would require work in potentially contaminated soil. Special surface requirements for the cap (i.e.,
asphalt or vegetation) that are dependent on future land use and development plans could be delayed until
the extraction system upgrades are made.
Another role of the cap is to reduce pumping requirements by reducing infiltration of precipitation. The
approximate precipitation rate for the area is approximately 36 inches per year, and the area encircled by
the sheet pile would be approximately eight acres (350,000 square feet). This translates to approximately
15 gpm of precipitation per year into the encircled area. Only a portion of this would infiltrate, and
constructing a low permeability cap with adequate management for surface runoff should further reduce
the infiltration rate.
The cost for the cap will vary depending on the type of cap, and the type of cap will depend on the
anticipated future land use. Of the three caps proposed by the site contractor, the asphalt cap should be
the least expensive and will be preferable with operating and maintaining an active extraction system
within the cap limits. The RSE team estimates the cost of an asphalt cap at approximately $1,500,000
including surface grading and assuming a 3-inch wearing layer, a 6-inch base course, and eight acres of
coverage. The site team also provided two options that would allow for a vegetated cap. One includes a
clay layer and the other includes a geomembrane. Relative to an asphalt cap, a geomembrane cap and
(depending level of quality assurance) a clay cap would require more effort to repair/maintain during
extraction system repairs. Both cap options would have similar maintenance requirements and the
geomembrane cap would offer slightly better infiltration reductions than the clay cap. After $500,000 in
surface grading, each of these two caps might cost $2,000,000 for eight acres of coverage. However, the
RSE team does not know the availability of clay in the area, and the lack of availability would increase
cost of the clay cap and make the geomembrane cap preferable to clay cap. The difference between these
preliminary cost estimates and those provided by the site contractor (see Section 4.4 of this report) is
likely due to additional contingency in site contractor's estimate and additional oversight and engineering
costs beyond the $2.35 million of engineering that the contractor estimates for 2004 through 2006 (see
Section 4.4 of this report). Because of the difference between the RSE cost estimates ($1.5 million to
$2.5 million, excluding oversight) and the site contractor's preliminary cost estimates ($3.6 million), the
RSE team recommends that EPA carefully review the revised cost estimate that will be provided by the
site contractor in the near future.
6.2.4 MONITOR DRAWDOWN IN FORMER PROCESS AREA AND CONDUCT A WATER BUDGET
ANALYSIS
With the upgradient wall and cap installed, the site team should be able to conduct a fairly accurate water
budget analysis. Water levels in the area encircled by the wall can be monitored and correlated with the
extraction rate, rain events, and changes in water elevations of the lower aquifer. If water levels within
the encircled area increase, it is an indication that inflow exceeds extraction, and if water levels decrease
within the encircled area, it is an indication that extraction exceeds inflow. This water budget analysis
should give the site team a relatively accurate idea as to the extraction rate necessary to maintain an
inward hydraulic gradient across the sheet pile and therefore an approximate treatment capacity for the
28
-------
new treatment system. The water budget analysis should be conducted during both the dry and rainy
seasons so that the water budget accounts for the upper and lower limits of pumping that would be
required. This water budget analysis should not require additional wells, but would require monitoring
water levels, pumping rates, and precipitation over perhaps six months as well as data analysis.
6.2.5 UPGRADE EXTRACTION SYSTEM
With the upgradient wall and cap in place and the water budget analysis conducted, the site team should
then have the appropriate information to upgrade the extraction system. Existing wells may be
abandoned or replaced, and/or additional wells may be added to provide the extraction rate necessary for
cleanup. The wells could also be placed and constructed to maximize NAPL recovery. Because of
future development opportunities, the site team should consider placing extraction piping under ground.
Manually controlling NAPL collection may also be impractical depending on the future land use. The
site team may want to consider the use of pneumatic submersible pumps for total fluids recovery.
Pump tests should be conducted from these new wells for the following reasons:
• confirm that the wells are capable of providing the intended extraction rate
• confirm that the intended and/or actual extraction rate yields the intended hydraulic results (i.e.,
inward gradients across the sheet pile
• provide information regarding the characteristics of the influent so that the treatment plant can be
designed accordingly
The RSE team defers to the site contractor's preliminary estimate of approximately $900,000 for upgrade
to the extraction system and the discharge outfall.
6.2.6 REPLACE THE EXISTING TREATMENT PLANT
The site team should begin consideration of alternate approaches technologies for the treatment plant,
and the pilot test suggested in 6.2.1 should provide valuable information. In addition, once information
from the water budget analysis is available, the site team should have a better understanding of the future
extraction rates and required treatment capacity. The pump tests described in Section 6.2.3 would also
provide valuable information. Ideally, the treatment plant should have sufficient capacity to maintain an
inward gradient through the sheet pile and an upward gradient from the lower to upper aquifer. With the
upgradient wall in place, a treatment plant with 50 gpm capacity might be sufficient and is assumed in
the preparation of this recommendation. Even with this increased flow rate the effluent from the DAF
would likely have less than one pound per day of contamination. The system should be designed with
automation and low annual costs in mind. If pilot testing supports it, the ideal system would have the
DAF unit (or a standard oil/water separator), sediment filters, organoclay (only if needed), and GAC
followed by discharge. If pilot testing does not support the use of that type of system, a fixed-film
bioreactor (such as the BioTrol model 12K4 used at the MacGillis and Gibbs Fund-lead site in
Minnesota) could be considered in place of the current biosystem. The MacGillis and Gibbs system
treats similar constituents and is designed for 50 gpm and influent PCP concentrations of 10 mg/L.
Once the system reaches steady-state operation, the total annual O&M costs (including labor, utilities,
materials, sampling, laboratory analysis, reporting, and project management) should be well under
$500,000 per year. A system similar to the "ideal system" discussed above operates at the Bayou
Bonfouca Superfund Site (a Fund-lead site that has recently been turned over to the State) in Slidell,
29
-------
Louisiana. The Bayou Bonfouca system has an oil/water separator instead of a DAF unit, but is designed
to treat up to 50 gpm and handles influent concentrations over 6,000 ug/L of PAHs. Under EPA-lead and
USAGE oversight, annual O&M costs for the system (including sampling, project management, USAGE
oversight, reporting, etc.) were approximately $400,000 per year. Over the past two years, the State of
Louisiana has operated the site for approximately $260,000 per year, with the majority of savings
reportedly due to eliminating USAGE oversight.
The new long-term treatment system would likely be smaller than the current system and, as the site team
suggested, could be moved back toward the tree line to allow for more flexibility in determining future
land use. The design, construction, and start-up for either of the above-mentioned types of treatment
plants should be well under the site contractor's estimate of $3.5 million, which is for a more complex
system. For the simpler systems described in this section, the RSE team suggests that a cost of under $2
million might be appropriate for design, construction, and 3-month startup.
6.2.7 POTENTIALLY ENHANCE MONITORING FOR THE LOWER AQUIFER
To monitor the effectiveness of hydraulic isolation, it is appropriate to continue monitoring the lower
aquifer to determine if contamination in either separate or dissolved phase is migrating from the upper to
the lower aquifer. The current monitoring program includes sampling at five wells (CW-05, CW-09,
CW-15, 99CD-MW02, and 99CD-MW04) in the lower aquifer. In addition to these five wells, it appears
that there are four other wells (CW-02, CW-12, EW-C1, and 99CD-MW01) that are also screened in the
lower aquifer. These wells appear to provide relatively good coverage of the lower aquifer, and it may be
appropriate to sample some or all nine of these wells, perhaps once or twice per year, especially for those
wells where there is known to be overlying contamination in the upper aquifer. It may also be
appropriate to install up to three additional wells in the lower aquifer; however, the benefits of adding
these wells should be weighed against the potential for breaching the aquitard. Appropriate locations for
these new wells might be between CW-09 and CW-02 (i.e. along the eastern boundary of the Former
Process Area), below CW-13 (i.e., along the western boundary of the Former Process Area), and below
CW-08 (i.e., along the northwestern boundary of the Former Process Area).
If contamination in these wells remains relatively stable or decreases and an upward gradient is
maintained between the lower and upper aquifers, then the site team can have reasonable confidence that
vertical hydraulic isolation is effective. If substantial increases in contamination become evident, the
eventual fate of that contamination should be determined so that the necessary remedial actions can be
implemented.
If the site team opts to install the additional wells, the installation might require up to $75,000, including
preparation of a work plan and providing oversight. Sampling of these additional wells should not
significantly increase annual costs because a full time operator and technician would be on-site to
conduct the sampling and the analysis is provided by the Regional laboratory at no cost to the site.
6.3 OTHER RELATED ITEMS
Once the above items are addressed, the site team can continue with other site-related activities. Some of
these activities are described below.
30
-------
Alter LTM approach to emphasize monitoring for hydraulic isolation
This item is necessary to evaluate the effectiveness of hydraulic isolation of the contamination and will
be an ongoing activity. The gradients can be monitored by measuring water levels in wells inside the
area encircled by sheet pile and comparing them to water levels measured from either sampling points
within armoring if it is installed (see Section 6.2.5) or the elevation of the surface water.
Water quality sampling in the upper aquifer beneath the Former Process Area should be kept to a
minimum. The site is not expected to clean up for decades, and water quality sampling in this area of
known contamination will not provide useful information. Rather, monitoring efforts in this area should
likely consist of measuring water levels and NAPL thickness. However, water quality sampling should
continue in the lower aquifer (see Section 6.2.7), and if an attenuation zone is included along the sheet
pile, then water quality sampling should be conducted at those locations as well.
The current monitoring costs for OU2/OU4 were not broken out of the total annual O&M costs; however,
it is very likely that a monitoring program as described above would be similar in cost to the current
monitoring program, which includes water quality sampling and analysis.
Monitor seeps along the eastern beach
The seeps along the eastern beach appear to provide the greatest threat to protectiveness because they are
the only source of unaddressed contamination that appears to remain outside of the sheet pile area. It is
presumed that the installation of the sheet pile and isolation efforts will reduce the magnitude of the
seeps or eliminate them altogether over time. Monitoring of the seeps should continue to determine if
isolation activities are effective at addressing the seeps.
If contaminant isolation efforts are effective at reducing or eliminating the seeps within the next few
years, then the site team may wish to address the contaminated surface sediments by dredging and/or
capping, if such actions are determined to be feasible. If isolation efforts are not effective at reducing or
eliminating the seeps within the next few years, then the site team may wish to investigate alternative
solutions. It may be that NAPL is migrating out of the Former Process Area from areas that have not
been sampled. It may also be that substantial sources of NAPL are already present outside of the Former
Process Area that are yet to be discovered. Properly addressing the seeps (if they are not addressed by
isolation efforts) would require further investigation and is beyond the scope of this RSE. Once again, if
the seeps continue, they are likely the greatest threat to protectiveness at the site and efforts should likely
be focused on them.
Consider adding extraction points for enhanced recovery
The site team, State, and community have expressed interest in enhanced NAPL recovery with the hope
that the site can be remediated in a shorter time frame. It is not certain that NAPL recovery, with or
without steam, will allow the site to be remediated in a reasonable time frame (i.e., less than 30 years).
However, if enhanced recovery is desired once hydraulic isolation of the contamination is achieved (i.e.,
the system is protective), the RSE team would recommend that additional recovery wells be considered
to augment product recovery to the extent the additional pumping does not compromise the treatment
system or significantly add to the life-cycle cost. In other words, the benefits of some additional product
recovery in this manner may not be considered to be great enough to merit substantial additional cost.
31
-------
7.0 SUMMARY
The RSE team observed a knowledgeable and competent site team led by an effective, motivated, and
organized EPA RPM. The observations provided in this report are not intended to imply a deficiency in
the work of the system designers, system operators, or site managers but are offered as constructive
suggestions in the best interest of the EPA and the public. These observations obviously have the benefit
of being formulated based upon operational data unavailable to the original designers. Furthermore, it is
likely that site conditions and general knowledge of ground water remediation have changed over time.
The RSE team believes that the best approach for this site is to initially focus efforts on hydraulic
isolation of the contamination underlying the Former Process Area, including the installation of an
upgradient barrier wall and low permeability cap to minimize the amount of water requiring treatment,
and the implementation of a new groundwater treatment system based, in part, on pilot testing of one or
more new approaches. Enhanced monitoring of groundwater in the lower aquifer is also recommended in
conjunction with these efforts, to improve the potential to detect current or future impacts to that aquifer.
It is also recommended that the site team monitor the active seeps along the eastern beach, and take
remedial actions if the seeps persist and a feasible remedial alternative is identified.
Once these high priority items are addressed and implemented, the site team could then reconsider
aggressive mass removal and the technologies that might be available at that time (potentially including
but not limited to additional efforts related to steam injection). However, the RSE team believes the most
cost-effective approach is to design and implement the new groundwater treatment system associated
with hydraulic isolation (discussed above) independent of such efforts. This will reduce the potential of
over-designing the groundwater treatment system that is associated with the hydraulic isolation efforts.
Cost/benefit evaluations for subsequent testing or implementation of more aggressive source removal
would need to incorporate costs that might be required to further upgrade the groundwater treatment
system above and beyond the treatment system associated with hydraulic isolation. The RSE team also
believes that if more aggressive source removal technologies are considered in the future, the costs and
benefits of installing additional recovery wells and tying them into the P&T system should be included as
a potential alternative.
During the RSE site visit there was discussion about armoring the existing sheet pile wall to prevent
scour and extend the life of the wall, and there was also discussion about adding a second sheet pile wall
inside the existing wall to create an "attenuation zone" that could be monitored. The RSE team felt that,
if armoring was pursued, then the interior sheet pile wall would not be necessary because the armoring
could likely be constructed in a manner to allow for an attenuation zone that could be monitored.
Subsequent to the RSE site visit, the RSE team was informed that armoring may not be feasible due to
potential impacts to the intertidal zone. It is likely that a variety of alternatives will need to be
considered in the future regarding this issue, and this issue is not addressed in detail in this RSE report.
The RSE team's suggestions for simplifying the new groundwater treatment system could save EPA as
much as $4 million relative to current estimates, while maintaining a protective remedy. This would
represent a savings of approximately 20% relative to the preliminary three-year costs that have been
estimated to date. If pilot test results of the recommended changes do not support the carbon usage
assumptions of the RSE team (that are based on published isotherms plus a safety factor), savings might
be lower. Additional savings would also result beyond the three year period by operating a simplified
and automated treatment system. Table 7-1 summarizes the cost and protectiveness implications of the
recommendations discussed in Section 6.0 of this report.
32
-------
Table 7-1. Cost summary Table
RSE Recommendation
6. 1 . 1 A Road Map to a Final Remedy
6.2.1 Simplify existing treatment plant
6.2.2 Install upgradient sheet pile
6.2.3 Remove steam
injection/extraction system and apply
cap
6.2.4 Upgrade extraction system
6.2.5 Conduct water budget analysis
6.2.6 Replace the existing treatment
plant
6.2.7 Augment monitoring in lower
aquifer
6.3 Other related items
Estimated Potential
Change in Cost*
($45 million)
to
($65 million)
($2.5 million)
no change
potential savings
no change
no change
~($ 1.5 million)
$75,000
Not quantified
Notes
The cost comparison ($15 million vs. $60 million to $80 million) includes an optimized P&T
system providing hydraulic isolation versus the projected steam injection costs over a 5-year
period until site closure. The P&T cost of $15 million assumes 30 years of operation at
$750,000 per year, a discount rate of 5%, and some contingency. Although the P&T system
would likely operate longer than 30 years, the costs beyond 30 years are neglected because
they are small contributions to the life-cycle cost when discounted. The steam injection cost
of $60 million to $80 million was provided by the site team. This cost projection and the
ability to reach closure in five years is optimistic.
The indicated savings results from operating a simplified system over a three year period.
Additional potential savings is likely possible from simplifying the system rather than making
some anticipated repairs.
This recommendation is consistent with that provided by the site team.
This recommendation is consistent with that provided by the site team; however, a comparison
of the preliminary cost estimates provided by the RSE team and the site contractor suggest that
potential savings may be possible.
This recommendation is consistent with that provided by the site team.
This recommendation can likely be conducted as part of the $2.35 million in engineering costs
that are expected over the next three years.
The indicated potential cost savings reflects the design, construction, and startup of a simpler
system than the one anticipated by the site team. Substantial savings in annual O&M costs
would also likely be realized if the new treatment system follows the RSE recommendation
rather than the more complex system that has been anticipated by the site team.
Additional wells in the lower aquifer might be appropriate to monitor hydraulic isolation. The
site team would need to weigh the benefits of adding new lower-aquifer wells against the
potential for breaching the aquitard. The RSE team provides three potential well locations.
The RSE team suggests the following be considered after those in Section 6.2 are addressed:
• Alter LTM to focus on hydraulic isolation
• Monitor active seeps along the eastern beach
• Consider adding extraction points for enhanced recovery
* Cost changes in parentheses indicate a decrease (i.e., cost savings)
33
-------
FIGURES
34
-------
FIGURE 1-1. WYCKOFF/EAGLE HARBOR OPERABLE UNITS.
CITY OF
BAINBRIDGE ISLAND
EAST HARBOR OU
(OU1)
WEST HARBOR OU
(OU3)
FORMER WYCKOFF FAC L TY
WYCKOFF SATURATED SOIL
AND GROUND WATER
(OU4)
WYCKOFF
UNSATURATEDSOIL
(OU2)
NOT TO SCALE
(Note: This figure is taken from the September 2002 Five-Year Review.)
-------
FIGURE 1-2. SCHEMATIC OF OU2 (GROUND WATER) AND OU4 (SATURATED SOILS).
WELL CWOI
AREA
UPPER AQUIFER
GROUNDWATER
BENEATH FORMER
LOG STORAGE/
PEELER AREA
UPPER AQUIFER BENEATH
FORMER LOG STORAGE/
PEELER AREA
FORMER PROCESS AREA
• UPPER AQUIFER GROUNDWATER
'BENEATH FORMER PROCESS AREA,
NOT TO SCALE
CONFINING LAYER
(AQUITARD)
(Note: This figure is taken from the September 2002 Five-Year Review.)
-------
FIGURE 1-3. CONCEPTUAL MODEL OF NAPL MIGRATION PRIOR TO REMEDY IMPLEMENTATION.
LNAPL DISCHARGE AREA
20 -1
FORMER BULKHEAD
(H-PILE WITH
TIMBER LAGGING)
MEAN SEA LEVEL (NGVD
NON-MARINE CLAY
COARSE GRAVELLY LENS
NAPL SEEPAGE FROM BULKHEAD -K.
SILTY OR CLAYEY
LENS (AQUJTARD)
DNAPL POOLS ON
HARBOR BOTTOM
GLACIAL SAND, SILT AND GRAVEL (AQUITARD)
SURFCAL MAR NE SED MENT
MARINE SAND o
AND GRAVEL
MARINE SILT
FLUVIAL DEPOSITS (LOWER AQUIFER)
/- LNAPL SEPARATES
/ AT WATER TABLE
NOTES:
1. MODEL IS NOT TO SCALE. APPROXIMATE VERTICAL EXAGGERATION = 5X.
2. THE CONCEPTUAL MODEL SHOWN ON THIS DRAWING IS REPRESENTATIVE
OF THE NORTHWEST SHORELINE AND LOG RAFTING AREA ONLY. GEOLOGY,
HYDROLOGY, AND NAPL MIGRATION MECHANISMS MAY BE DIFFERENT FOR
OTHER SHORELINE AND OFFSHORE AREAS.
3. NAPL LEAKAGE IS SHOWN DIRECTLY THROUGH THE BULKHEAD, BASED ON
DIVERS' OBSERVATIONS. THERE IS A POSSIBILITY, HOWEVER, OF ADDITIONAL
LEAKAGE UNDER THE BULKHEAD ONTO THE SURFICIAL MARINE SEDIMENTS.
(Note: This figure is taken from the Offshore Field Investigation Report for the Barrier Wall Design Project. USAGE, April, 1998.)
-------
FIGURE 1-4. CONCEPTUALIZATION OF HYDROGEOLOGY BENEATH THE FORMER PROCESS AREA
Groundwater flow
from hillside
Lower Aquifer
(Class II)
Groundwater
extraction
wells
\
txisting
Sheetpile
_ ;J GW Level
Precipitation (varies with sd.
Upper Aquifer
(Class 111)
ieafra
" ut>*
"''*
Less (-low
Largest (-low
Current Conditions
High Tide
Low Tide
Minor leakage
through sheetpile
wall "
Beach Sediments
Wyckoff Bainbrtdge Island
(Note: This figure was taken from a presentation prepared by the site contractor, CH2M Hill, presented at the RSE site visit.)
-------
FIGURE 2-1. OU2 LAYOUT WITH SHEET PILE AND WELL LOCATIONS.
-N-
FORMER
PROCESS AREA
STEAM
SYSTEM
BUILDING
GROUNDWATER
TREATMENT
PLANT
TREES
0 200
SCALE IN FEET
400
LEGEND
A EXTRACTION WELL
'••« SHEET PILE
------- |